Minimally Invasive

Foregut Surgery for

Malignancy

Principles and Practice

Steven N. Hochwald
Editor
Moshim Kukar
Associate Editor

123
Minimally Invasive Foregut Surgery
for Malignancy
Steven N. Hochwald
Editor
Moshim Kukar
Associate Editor

Minimally Invasive
Foregut Surgery
for Malignancy
Principles and Practice
Editor Associate Editor
Steven N. Hochwald, MD, FACS Moshim Kukar, MD
Department of Surgical Oncology Department of Surgical Oncology
Roswell Park Cancer Institute Roswell Park Cancer Institute
Buffalo, NY Buffalo, NY
USA USA

Videos to this book can be accessed at

http://www.springerimages.com/videos/978-3-319-09341-3.

ISBN 978-3-319-09341-3 ISBN 978-3-319-09342-0 (eBook)

DOI 10.1007/978-3-319-09342-0
Springer Cham Heidelberg New York Dordrecht London

Library of Congress Control Number: 2014957156

© Springer International Publishing Switzerland 2015

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed. Exempted from this
legal reservation are brief excerpts in connection with reviews or scholarly analysis or material
supplied specifically for the purpose of being entered and executed on a computer system, for
exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is
permitted only under the provisions of the Copyright Law of the Publisher's location, in its
current version, and permission for use must always be obtained from Springer. Permissions for
use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable
to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility
for any errors or omissions that may be made. The publisher makes no warranty, express or
implied, with respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Foreword

It has taken several decades for physicians and surgeons to accept that the man-
agement of the cancer patient is no longer discipline based but disease based.
The majority of advances that have come about have been the result of
increased knowledge and understanding of pathogenesis, molecular diagno-
sis, natural history, and prognostic factors for progression and survival. This
knowledge, accompanied by controlled trials, has allowed the integration of
meaningful treatment. No longer can the oncologist expect patient manage-
ment to be solely driven by his or her own discipline.
With such knowledge-based care, traditional views of surgical approaches
can now be maximized, balancing morbidity against outcome. The rapid evo-
lution of minimal access techniques has quickly demonstrated that morbidity
can be minimized over and above more radical classical approaches. The
perioperative advantages of such approaches were easy to define. Initially,
these were confined to the benefits of solely ablative procedures such as hys-
terectomy or cholecystectomy done through minimal incisions with low risk
and low morbidity. In situations where cancer care involved more extensive
procedures, particularly those requiring reconstruction, progress was slower
but is clearly being made. Once it was established that a minimal access
approach could provide adequate oncological resection with similar lymph
node yield where appropriate, then those advantages were confirmed. Internal
reconstruction techniques were defined, and the benefits of minimal access
surgery seen in the perioperative period could then be examined in the con-
text of long-term outcome. This has now been established such that the peri-
operative benefits are associated, when done well, with equivalent long-term
outcome. The ability of surgeons to utilize the technical improvements in
vision and robotic instruments is expanding exponentially.
The present text by Drs. Kukar and Hochwald brings together this combina-
tion of knowledge-based treatment with minimal access techniques as they apply
to foregut surgery for malignancy. The authors have assembled an international
cast, many of whom have been leaders in bringing such techniques to the fore.
For any physicians involved in the surgical management of foregut malig-
nancy, this text will be required reading.

NY, USA Murray F. Brennan, MD

v
Preface

Malignancies of the stomach and esophagus remain devastating for the

patient and challenging for the treating physician. Worldwide, these cancers
remain a major health concern, and due to varied presentation on different
continents, surgical practice and expertise varies considerably. The aggres-
sive biology of these tumors coupled with the advanced stage at presentation
in many patients mandates multidisciplinary care. Such care is frequently
associated with the careful integration of radiotherapy and chemotherapy
either before or after surgery. No matter what the approach, esophagectomy
or gastrectomy is associated with measurable deficits for the patient and the
need for physical as well as functional rehabilitation.
Minimally invasive surgical treatment of these malignancies allows for
more rapid return of preillness physical strength due to a reduced physiologi-
cal insult resulting from smaller incisions, more rapid mobilization of the
patient, decreased narcotic use, and shorter hospital time. Up until recently,
minimally invasive surgery for esophageal or gastric malignancy was not con-
sidered mainstream due to fear of inadequacy of oncological resection in the
face of advanced disease at presentation. Technological advancements and the
published results of surgical pioneers in these areas have allowed for rapid
progress in minimally invasive esophagogastric surgical approaches that
closely emulate and potentially improve upon the traditional open approaches.
Furthermore, the development of robotic surgery platforms offers great prom-
ise for refinements of techniques and outcomes in the near future.
This global, comprehensive work captures the brilliant progress made in
the minimally invasive surgical care of gastric and esophageal cancer patients.
There is much to learn from our physician colleagues as patterns of disease
presentation have led to the development of distinct and regional experts in
minimally invasive treatment approaches. No existing book on this topic has
assembled essential background chapters discussing tumor biology and treat-
ment approaches as well as comprehensive technique chapters complemented
extensively by high-definition videos illustrating salient surgical points. To
accomplish this, we have assembled an international group of experts that
discuss and demonstrate every major minimally invasive surgical and endo-
scopic treatment modality including the use of endoscopic submucosal dis-
section and robotic surgery for stomach and esophageal cancer. In order to
give the reader an opportunity to visualize several different approaches for
one operation, minimally invasive esophagectomy with cervical anastomosis

vii
viii Preface

is described utilizing both laparoscopic and robotic approaches. In addition,

minimally invasive esophagectomy with intrathoracic anastomosis is
described utilizing both an EEA circular stapler in an end-to-side fashion as
well as the use of a linear stapler to create a side-to-side anastomosis. Both
robotic and laparoscopic approaches to gastric cancer surgery are extensively
reviewed and described.
The work is divided into background chapters useful for current treatment
recommendations, while technique chapters enriched with multiple figures
demonstrate the various minimally invasive surgical approaches. Finally, as
visual demonstrations of techniques are essential for more widespread adap-
tation, each technique chapter is accompanied by at least one video demon-
strating the critical portion of the procedure.
We have learned much during our assembly of the outstanding contribu-
tions and from watching superb videos available through an online link. We
are confident that you will look to this book as an integrated state-of-the-art
invaluable resource.

Buffalo, NY, USA Moshim Kukar, MD

Buffalo, NY, USA Steven N. Hochwald, MD

Internet Access to Video Clips: The owner of this text will be able to
access these video clips through Springer with the following Internet link:
http://www.springerimages.com/videos/978-3-319-09341-3.
Contents

1 Pathogenesis of Esophageal Cancer . . . . . . . . . . . . . . . . . . . . . . . 1

Charles LeVea
2 Endoscopic Treatment of Premalignant and Early
Esophageal Malignancy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Toshitaka Hoppo and Blair A. Jobe
3 The Volume-Outcome Relationship, Standardized
Clinical Pathways, and Minimally Invasive Surgery
for Esophagectomy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Sheraz R. Markar and Donald E. Low
4 Goals of Surgical Therapy for Esophageal Cancer. . . . . . . . . . . 35
A. Koen Talsma, J. Shapiro, Bas P.L. Wijnhoven,
and J. Jan B. Van Lanschot
5 Optimization of Patients for Esophageal Cancer Surgery . . . . . 51
Wesley A. Papenfuss and Todd L. Demmy
6 Pathogenesis of Gastric Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Fátima Carneiro and Heike I. Grabsch
7 Standards for Surgical Therapy of Gastric Cancer . . . . . . . . . . 73
Roderich E. Schwarz
8 Endoscopic Submucosal Dissection for Gastric Cancer:
Its Indication, Technique, and Our Experience . . . . . . . . . . . . . 93
Hiroki Sato and Haruhiro Inoue
9 Multimodality Therapy in Gastric Cancer . . . . . . . . . . . . . . . . . 105
Usha Malhotra and Mei Ka Fong
10 Laparoscopic Transhiatal Esophagectomy
for Esophageal Cancer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Dido Franceschi, Elizabeth Paulus, and Danny Yakoub
11 Laparoscopic and Thoracoscopic Esophagectomy
with EEA Anastomosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
R. Taylor Ripley, David D. Odell, and James D. Luketich
12 Laparoscopic and Thoracoscopic Esophagectomy
with Side-Side Thoracic Anastomosis . . . . . . . . . . . . . . . . . . . . . 137
Kfir Ben-David and Isaac P. Motamarry

ix
x Contents

13 Laparoscopic and Thoracoscopic Transhiatal

Esophagectomy with Cervical Anastomosis . . . . . . . . . . . . . . . . 147
Moshim Kukar and Steven N. Hochwald
14 Laparoscopic and Thoracoscopic Esophagectomy
with Colonic Interposition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Christopher Armstrong, Monica T. Young,
and Ninh T. Nguyen
15 Thoracolaparoscopic Esophagectomy in the Prone
Position for Carcinoma of the Esophagus . . . . . . . . . . . . . . . . . . 165
C. Palanivelu, Palanivelu Praveen Raj,
Palanisami Senthilnathan, and R. Parthasarathi
16 Thoracoscopic Enucleation of Esophageal Benign Tumors . . . . 177
Yusuke Kimura, Akira Sasaki, Toru Obuchi, Takeshi Iwaya,
Yuji Akiyama, Masafumi Konosu, Fumitaka Endo,
Koki Otsuka, Hiroyuki Nitta, Keisuke Koeda,
and Go Wakabayashi
17 Minimally Invasive Feeding Tube and Esophageal
Stent Placement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Erin Schumer and Robert C.G. Martin II
18 Robotic Utilization in Esophageal Cancer Surgery . . . . . . . . . . 189
Richard van Hillegersberg, Roy J.J. Verhage,
Pieter C. van der Sluis, Jelle P.H. Ruurda,
and A. Christiaan Kroese
19 Minimally Invasive Intragastric Surgery. . . . . . . . . . . . . . . . . . . 199
Didier Mutter and Marius Nedelcu
20 Laparoscopic Partial Gastrectomy. . . . . . . . . . . . . . . . . . . . . . . . 205
Georgios Rossidis
21 Principles and Practice of Laparoscopic Gastrectomy
with Gastroduodenostomy (Billroth I). . . . . . . . . . . . . . . . . . . . . 213
Sang-Hoon Ahn and Hyung-Ho Kim
22 Laparoscopic Subtotal Gastrectomy with
Gastrojejunostomy and D2 Lymphadenectomy . . . . . . . . . . . . . 223
Joshua Ellenhorn
23 Laparoscopic Proximal Gastrectomy
with Double Tract Anastomosis . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Yukinori Kurokawa, Noriko Wada, Shuji Takiguchi,
and Yuichiro Doki
24 Laparoscopy-Assisted Total Gastrectomy . . . . . . . . . . . . . . . . . . 243
Nobuhiko Tanigawa, Sang-Woong Lee, and George Bouras
25 Robotic Utilization in Gastric Cancer Surgery . . . . . . . . . . . . . . 261
Kaitlyn J. Kelly and Vivian E. Strong

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
Contributors

Sang-Hoon Ahn, MD Department of Surgery, Seoul National University

Bundang Hospital, Seongnam, Gyeonggi, Republic of Korea
Yuji Akiyama, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Christopher Armstrong, MD, FRCSC Department of General Surgery,
Rockyview General Hospital and South Health Campus,
University of Calgary, Calgary, AB, Canada
Kfir Ben-David, MD Department of Surgery, University of Florida
Health, Gainesville, FL, USA
George Bouras, BMBS, BMedSci, FRCS Department of Surgery
and Cancer, Imperial College London, London, UK
Fátima Carneiro, MD, PhD Department of Pathology, IPATIMUP and
Medical Faculty, Centro Hospitalar de São João, Porto, Portugal
Todd L. Demmy, MD Department of Thoracic Surgery, Roswell Park
Cancer Institute, Buffalo, NY, USA
Yuichiro Doki, MD Department of Gastroenterological Surgery,
Osaka University Graduate School of Medicine, Suita, Osaka, Japan
Joshua Ellenhorn, MD Department of Surgery, Cedars-Sinai
Medical Center, Los Angeles, CA, USA
Fumitaka Endo, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Mei Ka Fong, PharmD Department of Pharmacy, Roswell Park
Cancer Institute, Buffalo, NY, USA
Dido Franceschi, MD Department of Surgery, University of Miami
Hospital and Jackson Memorial Hospital, Miami, FL, USA
Heike I. Grabsch, MD, PhD, FRCPath Department of Pathology,
Maastricht University Medical Center, AZ Maastricht, The Netherlands
Steven N. Hochwald, MD, FACS Department of Surgical Oncology,
Roswell Park Cancer Institute, Buffalo, NY, USA

xi
xii Contributors

Toshitaka Hoppo, MD, PhD Department of Surgery, Institute for the

Treatment of Esophageal and Thoracic Disease, The Western Pennsylvania
Hospital, Pittsburgh, PA, USA
Harohiro Inoue, MD, PhD Digestive Disease Center,
Showa University, Northern Yokohama Hospital, Yokohama, Japan
Takeshi Iwaya, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Blair A. Jobe, MD, FACS Department of Surgery, Institute for the
Treatment of Esophageal and Thoracic Disease, The Western Pennsylvania
Hospital, Pittsburgh, PA, USA
Kaitlyn J. Kelly, MD Department of Surgery, University of California,
San Diego, La Jolla, CA, USA
Hyung-Ho Kim, MD, PhD Department of Surgery, Seoul National
University Bundang Hospital, Seongnam, Gyeonggi, Republic of Korea
Yusuke Kimura, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Keisuke Koeda, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Masafumi Konosu, MD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
A. Christiaan Kroese, MD Division of Anesthesiology,
Intensive Care and Emergency Medicine, University Medical Center
Utrecht, Utrecht, The Netherlands
Moshim Kukar, MD Department of Surgical Oncology,
Roswell Park Cancer Institute, Buffalo, NY, USA
Yukinori Kurokawa, MD Department of Gastroenterological Surgery,
Osaka University Graduate School of Medicine, Suita, Osaka, Japan
Sang-Woong Lee, MD, PhD Department of General
and Gastroenterological Surgery, Osaka Medical College,
Takatsuki, Osaka, Japan
Charles LeVea, MD, PhD Department of Pathology and Laboratory
Medicine, Roswell Park Cancer Institute, Buffalo, NY, USA
Donald E. Low, FACS, FRCS(c) Department of General, Thoracic,
and Vascular Surgery, Virginia Mason Medical Center, Seattle, WA, USA
James D. Luketich, MD Department of Cardiothoracic Surgery,
University of Pittsburgh Medical Center, UPMC Presbyterian,
Pittsburgh, PA, USA
Usha Malhotra, MD Department of Medicine, Roswell Park
Cancer Institute, Buffalo, NY, USA
Contributors xiii

Sheraz R. Markar, MRCS(Eng), MA, MSc Department of Thoracic

Surgery, Virginia Mason Medical Center, Seattle, WA, USA
Robert C.G. Martin II , MD, PhD, FACS Division of Surgical
Oncology, University of Louisville, Louisville, KY, USA
Isaac P. Motamarry, MD Department of General Surgery,
University of Florida Shands, Gainesville, FL, USA
Didier Mutter, MD, PhD, FACS Department of Digestive
and Endocrine Surgery, IRCAD, IHU, University Hospital of Strasbourg,
Nouvel Hôpital Civil – Pôle Hépato-Digestif, Hôpitaux Universitaires
de Strasbourg, Strasbourg, Alsace, France
Marius Nedelcu, MD Department of Digestive and Endocrine Surgery,
University Hospital of Strasbourg, Nouvel Hôpital Civil – Pôle
Hépato-Digestif, Hôpitaux Universitaires de Strasbourg,
Strasbourg, Alsace, France
Nihn T. Nguyen, MD Department of Surgery, University of California
Irvine Medical Center, Orange, CA, USA
Hiroyuki Nitta, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Toru Obuchi, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
David D. Odell, MD, MMSc Department of Cardiothoracic Surgery,
University of Pittsburgh Medical Center, UPMC Presbyterian, Pittsburgh,
PA, USA
Koki Otsuka, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
C. Palanivelu, MS, MCh, FRCS, FACS Division of Oesophagastric
Surgery, GEM Hospital and Research Centre, Coimbatore, TN, India
Wesley A. Papenfuss, MD Department of Surgical Oncology,
Roswell Park Cancer Institute, Buffalo, NY, USA
R. Parthasarathi, MS, FMAS Division of Advanced Minimally
Invasive and GI Surgery, GEM Hospital and Research Centre,
Coimbatore, TN, India
Elizabeth Paulus, MD Division of Surgical Oncology,
University of Miami – Miller School of Medicine/Jackson
Memorial Hospital, Miami, FL, USA
Palanivelu Praveen Raj, MS, DNB(GI), DNB(SGI), FALS, FMAS
Department of Surgical Gastroenterology, GEM Hospital
and Research Centre, Coimbatore, TN, India
R. Taylor Ripley, MD Division of Thoracic Surgery, Memorial
Sloan-Kettering Cancer Center, New York, NY, USA
xiv Contributors

Georgios Rossidis, MD Department of Surgery, University of Florida,

Gainesville, FL, USA
Jelle P.H. Ruurda, MD, PhD Department of Surgery, University
Medical Center Utrecht, Utrecht, The Netherlands
Akira Sasaki, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Hiroki Sato, MD Digestive Disease Center, Showa University,
Northern Yokohama Hospital, Yokohoma, Japan
Erin Schumer, MS, MD Department of General Surgery,
University of Louisville, Louisville, KY, USA
Roderich E. Schwarz, MD, PhD, FACS Department of Surgical
Oncology, Indiana University Health Goshen Center for Cancer Care,
Goshen, IN, USA
Palanisami Senthilnathan, MS, DNBGen, MRCSEd, DNB GISurg,
FACS Department of Surgical Gastroenterology, GEM Hospital
and Research Centre, Coimbatore, TN, India
J. Shapiro, MD Department of Surgery, Erasmus Medical Center,
Rotterdam, CA, The Netherlands
Vivian E. Strong, MD Department of Surgery, Memorial
Sloan-Kettering Cancer Center, New York, NY, USA
Shuji Takiguchi, MD Department of Gastroenterological Surgery,
Osaka University Graduate School of Medicine, Suita, Osaka, Japan
A. Koen Talsma, MD, MSc Department of Surgery, Erasmus
Medical Center, Rotterdam, CA, The Netherlands
Nobuhiko Tanigawa, MD, FACS Department of Surgery,
Tanigawa Memorial Hospital, Ibaraki, Osaka, Japan
Richard van Hillegersberg, MD, PhD Department of Surgery,
University Medical Center Utrecht, Utrecht, The Netherlands
J. Jan B. Van Lanschot, MD, PhD Department of Surgery,
Erasmus Medical Center, Rotterdam, CA, The Netherlands
Pieter C. van der Sluis, MD, MSc Department of Surgery,
University Medical Center Utrecht, Utrecht, The Netherlands
Roy J.J. Verhage, MD, PhD Department of Surgery, University
Medical Center Utrecht, Utrecht, The Netherlands
Noriko Wada, MD Department of Gastroenterological Surgery,
Osaka University Graduate School of Medicine, Suita, Osaka, Japan
Go Wakabayashi, MD, PhD Department of Surgery, Iwate Medical
University School of Medicine, Morioka, Iwate Prefecture, Japan
Contributors xv

Bas P.L. Wijnhoven, MD, PhD Department of Surgery,

Erasmus Medical Center, Rotterdam, CA, The Netherlands
Danny Yakoub, MD, PhD Division of Surgical Oncology,
University of Miami – Miller School of Medicine/Jackson
Memorial Hospital, Miami, FL, USA
Monica T. Young, MD Department of General Surgery,
University of California Irvine Medical Center, Orange, CA, USA
 Pathogenesis
of Esophageal Cancer 1
Charles LeVea

Introduction Table 1.1 Risk factors for esophageal cancer [4–11]

Risk factor Risk
Worldwide, esophageal cancer is most common Tobacco 4–8×
in Asia, primarily Northern Iran, Central Asia, Alcohol 1.3–8×
and Northern China. The majority of esophageal Alcohol dehydrogenase 4×
cancers in Asia, histologically, are squamous cell Fruits and vegetables 0.53–0.56×
carcinoma. In Western countries only a third of Poverty 8×
carcinomas represent squamous cell carcinoma, Gastroesophageal reflux 7.7×
where the majority of cancers, histologically, are Obesity 3.1×
adenocarcinoma. The differences in pathogenesis Helicobacter pylori infection 0.56–1.1×
of esophageal cancer in Asia versus Western
countries may be due to differences in environ-
mental and lifestyle habits. Genetic differences subtypes, such as undifferentiated carcinoma,
may also come into play as China becomes more adenosquamous carcinoma, adenoid cystic carci-
westernized but, even with the westernization, noma, neuroendocrine carcinomas, and small cell
Barrett’s esophagus and adenocarcinoma remain carcinoma, also occur. This chapter will focus on
uncommon [1, 2]. the pathogenesis of squamous cell carcinoma and
Approximately 18,000 Americans will develop adenocarcinoma, the two most common cancers
esophageal carcinoma in 2013 [3]. Esophageal of the esophagus (Table 1.1).
carcinoma encompasses a variety of histological
subtypes. Worldwide, the predominant subtype is
squamous cell carcinoma. However, in the United Squamous Cell Carcinoma
States, it is estimated that only one-third of
patients will develop squamous cell carcinoma. Worldwide squamous cell carcinoma is the most
The majority of the remainder of the patients common cancer of the esophagus. Tobacco use,
will develop adenocarcinoma. However, rarer alcohol consumption, genetic abnormalities in
the enzymes that metabolize alcohol, caustic
injury to the esophagus, infrequent consumption
of fruits and vegetables, and poverty have been
C. LeVea, MD, PhD implicated in the pathogenesis of squamous cell
Department of Pathology and Laboratory Medicine,
carcinoma (Table 1.2). Each of these risk factors
Roswell Park Cancer Institute,
Elm & Carlton Streets, Buffalo, NY 14263, USA for the development of squamous cell carcinoma
e-mail: [email protected] of the esophagus will be discussed below.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 1
DOI 10.1007/978-3-319-09342-0_1, © Springer International Publishing Switzerland 2015
2 C. LeVea

Table 1.2 Factors involved in the pathogenesis of include p53, p14ARF, p16INK4a, cyclin D1, epidermal
squamous cell carcinoma of the esophagus
growth factor receptors, COX-2, retinoic acid,
Tobacco smoke –> polycyclic aromatic hydrocarbons retinoic acid receptor beta2, and the fragile histi-
and N-nitrosamines –> DNA adducts, methylation, and dine triad.
chromosomal translocations
P53 is a cellular stress sensor, a tumor sup-
Alcohol –> metabolized in liver and oral bacteria –>
acetaldehyde –> covalent DNA bonds pressor, which normally functions to maintain
Alcohol –> squamous mucosa cytochrome P450 the integrity of cellular DNA. Loss of function of
induction –> reactive oxygen species –> lipid p53 in esophageal squamous cell carcinoma
peroxidation and oxidative cell injury –> DNA adducts occurs in approximately 50–60 % of Japanese
Alcohol dehydrogenase mutation –> inefficient patients [19–21] making the tumor cells unable
metabolism of alcohol –> increased acetaldehyde in
blood stream –> covalent DNA bonds to enter into apoptosis or senescence. The tumor
Genes affected –> p53, p14ARF, p16INK4a, cyclin D1, cells cannot repair the tobacco-mediated DNA
EGFR, COX-2, retinoic acid, retinoic acid receptor damage, and the result is dysregulated cellular
beta 2, and the fragile histidine triad division [22, 23]. P14 ARF blocks MDM2-
Other factors –> caustic injury due to lye, infrequent mediated degradation of p53, leading to increased
consumption of raw fruits and vegetables, poverty
expression of p53. Tobacco smoke causes the
p14ARF promoter to be methylated, silencing
Tobacco Use and Squamous Cell expression of p14ARF, which results in decreased
Carcinoma of the Esophagus p53 expression in about 60 % of patients with
esophageal squamous cell carcinoma [24].
Cigarette smoke contains polycyclic aromatic Loss of protein expression of the cyclin-
hydrocarbons and N-nitrosamines, which dependent kinase inhibitor, p16INK4a, has been
have been shown to be carcinogenic [12, 13]. observed early in the development of squamous
Cigarette smoke has a number of other carcino- cell carcinoma of the esophagus. This occurs pre-
gens, but polycyclic aromatic hydrocarbons and dominantly through loss of heterozygosity of the
N-nitrosamines are the most important in regard p16INK4a gene or through silencing of the p16INK4a
to esophageal squamous cell carcinoma devel- promoter by methylation [25]. P16INK4a proteins
opment. The mechanisms of carcinogenesis by normally function to inhibit cyclin-dependent
tobacco smoke may include formation of DNA kinase 4 and cyclin-dependent kinase 6, prevent-
adducts, silencing of genes by methylation, and ing cellular division. Loss of p16INK4a allows cel-
chromosomal translocations [14]. lular division of squamous cell tumors by
Tobacco smoke causes cancer through the for- allowing the cells to progress unchecked through
mation of covalent bonds between the carcinogen G1 to S phase of the cell cycle [17].
and cellular DNA, producing DNA adducts. The While p16INK4a normally inhibits cyclin-
more DNA adducts formed, the more likely per- dependent kinase 4 and cyclin-dependent kinase
manent mutations, in genes in cellular division 6, cyclin D1 activates cyclin-dependent kinase
regulating pathways, occur. When DNA adducts 4/ cyclin-dependent kinase 6 leading to progres-
are bypassed incorrectly by DNA polymerases, sion through the cell cycle. Tobacco has been
permanent mutation in genes that deregulate cel- shown to increase levels of cyclin D1 in vitro
lular division is formed [15, 16]. [26], thus, facilitating cell cycle progression.
Hypermethylation of promoters and intra- Other signaling molecules that have been linked
genic hypermethylation can silence the transcrip- to the development of squamous cell carcinoma
tion of genes, and DNA translocations can lead to include overexpression of epidermal growth factor
mutational activation or to silencing of growth- receptors and associated overexpression of COX-2
regulating genes [17, 18]. and Her2/neu overexpression [27–29]. Retinoic
A number of genes regulating cellular division acid and retinoic acid receptor beta2 induction can
have been implicated in the pathogenesis of squa- downregulate epidermal growth factor receptor
mous cell carcinoma of the esophagus. These expression. Tobacco smoke can suppress retinoic
1 Pathogenesis of Esophageal Cancer 3

acid receptor beta2 by methylating the retinoic acid primarily hydrogen peroxide and superoxide
receptor beta2 gene promoters [30]. This may be a anions. The reactive oxygen species cause lipid
tobacco-mediated mechanism contributing to over- peroxidation and other forms of oxidative injury
expression of epidermal growth factor receptor to the cell, which leads to DNA adducts [37, 40].
and, possibly, COX-2 and Her2/neu in squamous The resulting DNA adducts can cause permanent
cell carcinoma. mutations.
The fragile histidine triad gene, encoding a Chronic alcohol consumption also results in
tumor suppressor, has been shown to be inacti- aberrant gene regulation through ineffective pro-
vated in squamous cell carcinoma of the esopha- moter methylation (hypomethylation). Alcohol
gus [31, 32]. The mechanism of inactivation inhibits the synthesis of S-adenosyl-L-methionine,
occurs through silencing of the gene, by promoter the donor group used for the methylation of pro-
methylation, or silencing through genome insta- moter regions [41, 42]. Hypomethylated genes
bility/chromosome translocations [33, 34]. can be aberrantly transcribed, dysregulating cel-
lular division [37].
Patients with squamous cell carcinoma of the
Alcohol Consumption and Squamous esophagus, who consumed alcohol more than
Cell Carcinoma of the Esophagus four times a week, demonstrated decreased levels
of retinoic acid receptor gamma in their non-
In the liver, ethanol is metabolized by alcohol neoplastic squamous mucosa when compared to
dehydrogenase. The acetaldehyde generated by control patients, who consumed one drink a week
alcohol dehydrogenase has been shown to be car- or less [43]. Retinoic acid through its receptor’s
cinogenic in squamous cell carcinoma of the activation leads to decreased expression of epi-
esophagus [35]. Additionally, oral bacteria dermal growth factor receptors. By decreasing
metabolize ethanol to acetaldehyde resulting in a retinoic acid receptor expression, alcohol may
10–100 times higher concentration of acetalde- dysregulate growth by increasing expression and
hyde in the oral cavity [36, 37]. The acetaldehyde activation of epidermal growth factor receptor
in the saliva comes into direct contact with the signaling pathways.
squamous mucosa of the esophagus upon swal-
lowing, directly adding to the amount of acetal-
dehyde that the squamous mucosa is already Alcohol Dehydrogenase Mutation
being exposed to via the blood during alcohol and Squamous Cell Carcinoma
consumption. of the Esophagus
Acetaldehyde forms covalent bonds with
DNA, and the resulting DNA adducts can escape Ethanol is metabolized into acetaldehyde by
cellular DNA repair mechanisms causing detri- alcohol dehydrogenase, and acetaldehyde is fur-
mental mutations in growth-regulating genes ther metabolized to acetate by aldehyde dehydro-
[38]. In addition to directly causing mutations in genase. Prevalent in East Asians are the
DNA, acetaldehyde indirectly causes DNA muta- ADH2*1/2*1 alleles of alcohol dehydrogenase
tions by binding to enzymes involved in DNA and the ALDH2*2 alleles of aldehyde dehydro-
repair and DNA methylation. Alterations in these genase [35]. The concentration of acetaldehyde
enzymes lead to mutations and aberrant regula- in the bloodstream is increased by both of these
tion of genes [37]. enzymes. The ADH2*1/2*1 allele encodes a
Esophageal squamous mucosa from patients superactive form of alcohol dehydrogenase pro-
with chronic alcohol consumption demonstrated ducing acetaldehyde quicker. The ALDH2*2
induction of cytochrome P450 2E1 (CYP2E1) allele of aldehyde dehydrogenase produces an
when compared to the squamous mucosa from a inactive enzyme slowing the removal of acetalde-
teetotaler control group [39]. The cytochrome hyde from the blood. The formation of acetalde-
P450 system generates reactive oxygen species, hyde DNA adducts is mutagenic.
4 C. LeVea

Caustic Injury and Squamous Cell fruit and vegetable intake, low socioeconomic
Carcinoma of the Esophagus status has an independent effect [47]. Whether
this independent effect can be explained by poor
The first association of a lye burn and squamous dental care or other nutritional or environmental
cell carcinoma of the esophagus was reported in factors needs to be further investigated.
1904 by Telesky [44]. The average interval
between a caustic burn to the esophagus and the
development of squamous cell carcinoma is Adenocarcinoma
approximately 40 years [44]. Chemical injury
from a caustic chemical, such as lye, leads to The incidence of adenocarcinoma of the esopha-
fibrosis with stricture of the esophagus in the area gus has been increasing in Western countries
of injury. The narrowed lumen causes an obstruc- over the last few decades [48]. Environmental
tion during swallowing, and the constant irrita- factors are most likely to have caused the increase
tion leads to repeated injury, inflammation, and in adenocarcinoma incidence, as it is unlikely
repair, which, over time, leads to carcinogenesis. that genetic risk/predisposition to adenocarci-
For similar reasons, achalasia is a risk factor for noma has changed so abruptly. There is a gender
developing squamous cell carcinoma. Why lye influence on the development of adenocarcinoma
injury leads to squamous cell carcinoma and why of the esophagus, as, in the United States, men
the caustic injury from acid reflux (to be dis- are six times more likely to develop esophageal
cussed more below) leads to adenocarcinoma of adenocarcinoma than women [48]. Up to 13 % of
the esophagus is unclear. adenocarcinomas of the esophagus may be due to
patients inheriting a genetic predisposition.
Genetic predisposition as well as the environ-
Infrequent Consumption of Raw mental influences of gastroesophageal reflux dis-
Fruits and Vegetables and Squamous ease, obesity, and Helicobacter pylori infection
Cell Carcinoma of the Esophagus on the development of adenocarcinoma of the
esophagus will be discussed.
A number of studies [45, 46] have shown an
inverse relationship between raw fruit and vege-
table consumption and the risk of squamous cell Genetic Factors and Adenocarcinoma
carcinoma of the esophagus. Lower consumption of the Esophagus
of vegetables and fruits is associated with a
higher risk of squamous cell carcinoma. Odds Three candidate genes containing germline muta-
ratios were adjusted for alcohol consumption, tions were identified in patients with esophageal
tobacco use, and gender. The mechanism of the adenocarcinoma: MSR1, ASCC1, and CTHRC1
protective effect of fruit and vegetables is unclear, [49]. MSR1 encodes the class A macrophage
but it may be related to the vitamins and minerals scavenger receptor, whose protein function
contained in the foods. becomes disrupted by the germline mutation. The
MSR1 mutation suggests a link between esopha-
geal adenocarcinoma and inflammation. ASCC1
Poverty and Squamous Cell encodes activating signal cointegrator 1, which
Carcinoma of the Esophagus activates NF kappa B, serum response factor, and
activating protein 1 [50]. Therefore, ASCC1 is
The development of squamous cell carcinoma of another signaling molecule putatively linking
the esophagus is strongly associated with low inflammation to growth signal transduction path-
income. While the majority of the risk of devel- ways. Another germline mutation was found in
oping squamous cell carcinoma of the esophagus CTHRC1, a protein expressed during tissue
can be explained by alcohol, tobacco, and low repair processes, called collagen triple helix
1 Pathogenesis of Esophageal Cancer 5

repeat containing 1 protein [51]. CTHRC1 Gender Influence

signaling regulates TGF beta pathways, thus, is and Adenocarcinoma
an additional protein linking inflammation/repair of the Esophagus
processes to control of cellular proliferation [52].
Patients with a single gene polymorphism in Worldwide, there is a male predominance for
the matrix metalloproteinase gene family, specif- developing adenocarcinoma. In the United States,
ically MMP1 1G/2G, have a higher risk of devel- the association is even stronger with a 3:1 (Native
oping esophageal adenocarcinoma [53]. Matrix American) to 9:1 (Caucasian) ratio between men
metalloproteinase proteins are involved in extra- and women [58]. Thus, female sex hormones
cellular matrix and basement membrane degrada- may have a protective effect. This is supported by
tion. A synergistic effect of gastroesophageal the delayed development of adenocarcinoma on
reflux disease combined with the MMP1 1G/2G average by 17 years in women when compared to
polymorphism increases the risk of developing men [59]. Another interesting observation is the
esophageal adenocarcinoma [53]. protective effect of breastfeeding on esophageal
A decreased local secretion of epidermal growth adenocarcinoma. Increased duration of breast-
factor (EGF) has been associated with the develop- feeding is correlated with a reduced risk of devel-
ment of esophageal adenocarcinoma [54]. The oping esophageal adenocarcinoma [60]. More
EGF 5′ UTR G/G genotype confers an increased research is required to determine the hormonal
risk of developing adenocarcinoma and is associ- mechanisms involved.
ated with low EGF levels. Interestingly, epidermal
growth factor receptor levels in these patients are
overexpressed, possibly caused by lack of an inhib- Gastroesophageal Reflux Disease
itory effect on EGF receptor expression due to the and Adenocarcinoma
low circulating EGF hormone levels. of the Esophagus
Vascular endothelial growth factor is involved
in the regulation of angiogenesis. The variant T Gastroesophageal reflux is an important risk fac-
allele of the VEGF gene in the +936CT/TT poly- tor for the development of esophageal adenocarci-
morphism is associated with increased risk of noma. When compared to the risk of people
esophageal adenocarcinoma, especially in smok- without reflux symptoms developing adenocarci-
ers [55]. Carriers of the T allele of VEGF have noma, an individual experiencing reflux symp-
higher levels of VEGF. VEGF-induced angio- toms on a weekly basis has a lower risk of
genesis has been shown to be an early event in developing adenocarcinoma (5-fold risk) than
esophageal adenocarcinoma development [55]. someone experiencing daily reflux symptoms
Interleukin-18 is a cytokine, whose inflam- (7-fold risk) [61]. Reflux of the acid contents of
matory responses have been linked to antitumor the stomach into the esophagus causes caustic
immunity [56]. The single-nucleotide polymor- damage to the esophageal squamous mucosa.
phism, IL-18-607 C/A in its promoter, is associ- This leads to injury of the squamous mucosa and
ated with the development of Barrett’s esophagus acute and chronic inflammation. Repair does not
and esophageal adenocarcinoma. Alternatively, involve scarring as seen with lye but, rather,
the IL-18RAP rs917997C allele is associated involves glandular metaplasia (Barrett’s metapla-
with a protective effect on Barrett’s esopha- sia) of the esophagus. Barrett’s esophagus is when
gus from developing into adenocarcinoma. The the squamous mucosa is replaced by intestinal-
DNA repair protein O(6)-methylguanine-DNA type glandular epithelium containing goblet cells.
methyltransferase, which repairs DNA adducts, Further reflux damage results in further injury
has a variant single-nucleotide polymorphism – with subsequent chronic inflammation and repair.
rs12268840, when homozygous, which is asso- The resulting increased cellular turnover makes
ciated with an increased risk for esophageal the mucosa susceptible to mutations in growth-
adenocarcinoma [57]. regulating genes, which leads to glandular
6 C. LeVea

dysplasia. Low-grade glandular dysplasia may caused by obesity has been correlated with length
lead to high-grade glandular dysplasia and esoph- of Barrett’s esophagus [72–74].
ageal adenocarcinoma [62, 63]. Alternatively, instead of being caused by this
The majority of people with Barrett’s esophagus obesity-related metabolic syndrome, Barrett’s
do not progress to esophageal adenocarcinoma. esophagus may be a response to increased acid
Neoplastic transformation of Barrett’s esophagus reflux caused by the increased intra-abdominal
can be difficult to identify, as dysplasia can be focal. pressure due to intra-abdominal obesity. There is
Thus, a number of biopsies are required to prevent a direct correlation between increased body mass
sampling errors and false-negative results [63, 64]. index and increased esophageal reflux [75]. The
Low-grade glandular dysplasia has a low rate of increased esophageal reflux or a combination of
progression to esophageal adenocarcinoma [65]. risk factors associated with reflux and the meta-
Even high-grade glandular dysplasia progresses to bolic syndrome of obesity may lead to the devel-
esophageal adenocarcinoma only 10–60 % of the opment of esophageal adenocarcinoma.
time [66, 67].
The future of predicting which patients
with Barrett’s esophagus are at higher risk of Helicobacter Pylori Infection
progressing to esophageal adenocarcinoma and and Adenocarcinoma
which have a low risk of progression may be of the Esophagus
with molecular and chromosomal markers.
Chromosome instability, demonstrated by a Helicobacter pylori infection occurs in 50 % of the
combined panel of abnormalities encompassing worldwide population and commonly colonizes
9p loss of heterozygosity (LOH), 17p LOH, and the stomach of children [76]. Up to a 50 % decrease
DNA aneuploidy or DNA tetraploidy in Barrett’s in esophageal adenocarcinoma risk has been
esophagus, predicted subsequent development of attributed to Helicobacter pylori infection [77].
esophageal adenocarcinoma, relative risk = 38.7, One possible mechanism includes Helicobacter
and a 5-year cumulative risk of developing ade- pylori infection leading to gastric atrophy. The
nocarcinoma of 79.1 %. Those patients without reduction in the acidity and volume of gastric con-
any demonstrable chromosome instability in tents leads to an associated decrease in esophageal
their Barrett’s esophagus had 0 % cumulative reflux disease.
incidence of adenocarcinoma at 8 years [68].
Molecular markers of chromosome instability in
Barrett’s esophagus would be useful to determine Acute and Chronic Inflammation
patients that would benefit from close clinical and Esophageal Carcinoma
surveillance.
Acute and chronic inflammation may provide the
mechanisms common to the development of
Obesity and Adenocarcinoma esophageal carcinoma. In both squamous cell
of the Esophagus carcinoma and adenocarcinoma, reactive oxygen
species generated by acute and chronic inflam-
Obesity is a strong risk factor for developing mation can be mutagenic. An esophageal inflam-
esophageal adenocarcinoma [69]. The risk is matory reaction is seen in response to smoking,
even greater for people with central and intra- alcohol consumption, lye injury, chronic reflux,
abdominal obesity [70, 71]. Various mechanisms and obesity.
for obesity-related cancer have been proposed, Acute and chronic esophageal inflammation
including increased levels of endogenous sex causes intracellular oxidative stress [78, 79].
hormones, leptin, plasminogen activator inhibi- Increased serum levels of inflammatory cyto-
tor-1, and IGF-1, and decreased adiponectin, and kines, such as Il-6, TNF-alpha, C-reactive pro-
chronic inflammation. This metabolic syndrome tein, and leptin, have been observed in patients
1 Pathogenesis of Esophageal Cancer 7

with chronic reflux and obesity [80, 81]. When 15. Hoeijmakers JH. Genome maintenance mechanisms
for preventing cancer. Nature. 2001;411:366–74.
the inflammatory reaction is localized to the
16. Mangerich A, Dedon PC, Fox JG, et al. Chemistry
esophageal mucosa, the oxidative stress can lead meets biology in colitis-associated carcinogen-
to DNA damage and mutagenesis. Further esis. Free Radic Res. 2013;47(11):958–86. doi:
genomic alterations can lead to DNA and chro- 10.3109/10715762.2013.832239. Epub 2013 Oct 4.
17. Belinsky SA. Silencing of genes by promoter hyper-
mosomal instability and the development of
methylation: key event in rodent and human lung can-
esophageal adenocarcinoma or squamous cell cer. Carcinogenesis. 2005;26:1481–7.
carcinoma [82, 83]. 18. Ehrlich M, Lacey M. DNA methylation and differen-
tiation: silencing, upregulation and modulation of
gene expression. Epigenomics. 2013;5(5):553–68.
19. Egashira A, Morita M, Kakeji Y, et al. p53 gene muta-
References tions in esophageal squamous cell carcinoma and
their relevance to etiology and pathogenesis: results in
1. Chai J, Jamal MM. Esophageal malignancy: a growing Japan and comparisons with other countries. Cancer
concern. World J Gastroenterol. 2012;18(45):6521–6. Sci. 2007;98:1152–6.
2. Huang Q, Fang DC, Zhang J, et al. Barrett’s 20. Oki E, Zhao Y, Yoshida R, et al. The difference in p53
esophagus-related diseases remain uncommon in mutations between cancers of the upper and lower gas-
China. J Dig Dis. 2011;12(6):420–7. trointestinal tract. Digestion. 2009;79 Suppl 1:33–9.
3. National Cancer Institute at the National Institutes of 21. Egashira A, Morita M, Yoshida R, et al. Loss of p53 in
Health USA. 2013. http://www.cancer.gov/cancertop- esophageal squamous cell carcinoma and the correla-
ics/types/esophageal. Accessed 7 Oct 2013. tion with survival: analyses of gene mutations, protein
4. Bosman FT, Carneiro F, Hruban RH, Theise ND. WHO expression, and loss of heterozygosity in Japanese
Classification of Tumours, IARC. WHO Classification patients. J Surg Oncol. 2011;104(2):169–75.
of Tumours. 2010;3:18. 22. Efeyan A, Serrano M. P53: guardian of the genome
5. Bagnardi V, Rota M, Botteri E, et al. Light alcohol and policeman of the oncogenes. Cell Cycle. 2007;
drinking and cancer: a meta-analysis. Ann Oncol. 2013; 6(9):1006–10.
24:301–8. 23. Jackson SP, Bartek J. The DNA-damage response in
6. IARC Monographs on the Evaluation of Carcinogenic human biology and disease. Nature. 2009;461(22):
Risks to Humans IARC Monographs, WHO Press. 1071–8.
2012;100:381. 24. Ito S, Ohga T, Saeki H, et al. Promoter hypermethyl-
7. Zhang GH, Mai RQ, Huang B. Meta-analysis of ation and quantitative expression analysis of
ADH1B and ALDH2 polymorphisms and esophageal CDKN2A (p14ARF and p16INK4a) gene in esopha-
cancer risk in China. World J Gastroenterol. 2010; geal squamous cell carcinoma. Anticancer Res. 2007;
16(47):6020–5. 27:3345–53.
8. Morris Brown L, Hoover R, Silverman D, et al. 25. Tokugawa T, Sugihara H, Tani T, et al. Modes of
Excess incidence of squamous cell esophageal cancer silencing of p16 in development of esophageal
among Black Men: role of social class and other risk squamous cell carcinoma. Cancer Res. 2002;62:
factors. Am J Epidemiol. 2001;53(2):114–22. 4938–44.
9. Lagergren J, Bergstrom R, Lindgren A, et al. 26. Hu H, Zhang S, Zhu S. Influence of aspirin and
Symptomatic gastroesophageal reflux as a risk factor cigarette smoke extract on the expression of cyclin D1
for esophageal adenocarcinoma. N Engl J Med. 1999; and effects of cell cycle in esophageal squamous cell
340(11):825–31. carcinoma cell line. Dis Esophagus. 2009;22:310–6.
10. Morris Brown L, Swanson CA, Gridley G, et al. 27. Kuwano H, Kato H, Miyazaki T, et al. Genetic altera-
Adenocarcinoma of the esophagus: role of obesity tions in esophageal cancer. Surg Today. 2005;35:7–18.
and diet. J Natl Cancer Inst. 1995;87(2):104–9. 28. Moraitis D, Du B, De Lorenzo MS, et al. Levels of
11. Islami F, Kamangar F. Helicobacter pylori and esoph- cyclooxygenase-2 are increased in the oral mucosa of
ageal cancer risk – a meta-analysis. Cancer Prev Res smokers: evidence for the role of epidermal growth
(Phila). 2008;1(5):329–38. factor receptor and its ligands. Cancer Res. 2005;65:
12. Boffetta P, Jourenkova N, Gustavsson P. Cancer risk 664–70.
from occupational and environmental exposure to 29. da Costa NM, Soares Lima SC, de Almeida Simao T,
polycyclic aromatic hydrocarbons. Cancer Causes et al. The potential of molecular markers to improve
Control. 1997;8:444–72. interventions through the natural history of oesopha-
13. Kamangar F, Strickland PT, Pourshams A, et al. High geal squamous cell carcinoma. Biosci Rep. 2013;
exposure to polycyclic aromatic hydrocarbons may 33(4):art e00057.
contribute to high risk of esophageal cancer in 30. Wang XD, Liu C, Bronson RT, et al. Retinoid signal-
Northeastern Iran. Anticancer Res. 2005;25:425–8. ing and activator protein-1 expression in ferrets given
14. Xu XC. Risk factors and gene expression in esopha- beta-carotene supplements and exposed to tobacco
geal cancer. Methods Mol Biol. 2009;471:335–60. smoke. J Natl Cancer Inst. 1999;91:60–6.
8 C. LeVea

31. Mori M, Mimori K, Shiraishi T, et al. Altered expres- 49. Orloff M, Peterson C, He X, et al. Germline mutations
sion of Fhit in carcinoma and precarcinomatous lesions in MSR1, ASCC1, and CTHRC1 in patients with
of the esophagus. Cancer Res. 2000;60:1177–82. Barrett esophagus and esophageal adenocarcinoma.
32. Chava S, Mohan V, Shetty PJ, et al. Immunohistochemical JAMA. 2011;306:410–9.
evaluation of p53, FHIT, and IGF2 gene expression in 50. Jung DJ, Sung HS, Goo YW, et al. Novel transcription
esophageal cancer. Dis Esophagus. 2012;25(1):81–7. coactivator complex containing activating signal coin-
33. Zochbauer-Muller S, Fong KM, Maitra A, et al. 5′ tegrator 1. Mol Cell Biol. 2002;22(14):5203–11.
CpG island methylation of the FHIT gene is corre- 51. Durmus T, LeClair RJ, Park KS, et al. Expression analy-
lated with loss of gene expression in lung and breast sis of the novel gene collagen triple helix repeat contain-
cancer. Cancer Res. 2001;61:3581–5. ing-1 (Cthrc1). Gene Expr Patterns. 2006;6(8):935–40.
34. Jeong YJ, Jeong HY, Lee SM, et al. Promoter meth- 52. LeClair R, Lindner V. The role of collagen triple helix
ylation status of the FHIT gene and Fhit expression: repeat containing 1 in injured arteries, collagen
association with HER2/neu status in breast cancer expression, and transforming growth factor beta sig-
patients. Oncol Rep. 2013;30(5):2270–8. naling. Trends Cardiovasc Med. 2007;17(6):202–5.
35. Yokoyama A, Kato H, Yokoyama T, et al. Genetic poly- 53. Cheung WY, Zhai R, Bradbury P, et al. Single nucleo-
morphisms of alcohol and aldehyde dehydrogenases and tide polymorphisms in the matrix metalloproteinase
glutathione S-transferase M1 and drinking, smoking, gene family and the frequency and duration of gastro-
and diet in Japanese men with esophageal squamous cell esophageal reflux disease influence the risk of esopha-
carcinoma. Carcinogenesis. 2002;23:1851–9. geal adenocarcinoma. Int J Cancer. 2012;131:2478–86.
36. Homann N, Tillonen J, Meurman JH, et al. Increased 54. Menke V, Pot RG, Moons LM, et al. Functional
salivary acetaldehyde levels in heavy drinkers and single-nucleotide polymorphism of epidermal growth
smokers: a microbiological approach to oral cavity factor is associated with the development of Barrett’s
cancer. Carcinogenesis. 2000;21:663–8. esophagus and esophageal adenocarcinoma. J Hum
37. Seitz HK, Stickel F. Molecular mechanisms of alco- Genet. 2012;57:26–32.
hol mediated carcinogenesis. Nat Rev Cancer. 2007;7: 55. Zhai R, Liu G, Asomaning K, et al. Genetic polymor-
599–612. phisms of VEGF, interactions with cigarette smoking
38. Bartsch H. DNA adducts in human carcinogenesis: exposure and esophageal adenocarcinoma risk.
etiological relevance and structure – activity relation- Carcinogenesis. 2008;29:2330–4.
ship. Mutat Res. 1996;340:67–79. 56. Babar M, Ryan AW, Anderson LA, et al. Genes of the
39. Millonig G, Wang Y, Homann N, et al. Ethanol-induced interleukin-18 pathway are associated with susceptibility
carcinogenesis in the human esophagus implicates to Barrett’s esophagus and esophageal adenocarcinoma.
CYP2E1 induction and generation of carcinogenic Am J Gastroenterol. 2012;107:1331–41.
DNA-lesions. Int J Cancer. 2011;128:533–40. 57. Doecke J, Zhao ZZ, Pandeya N, et al. Australian
40. Homann N, Seitz HK, Wang XD, et al. Mechanisms Cancer Study. Polymorphisms in MGMT and DNA
in alcohol-associated carcinogenesis. Alcohol Clin repair genes and the risk of esophageal adenocarci-
Exp Res. 2005;29:1317–20. noma. Int J Cancer. 2008;123:174–80.
41. Baylin SB. DNA methylation and gene silencing in 58. Cook MB, Chow WH, Devesa SS. Oesophageal cancer
cancer. Nat Clin Pract Oncol. 2005;2 Suppl 1:S4–11. incidence in the United States by race, sex, and histo-
42. Obeid R. The metabolic burden of methyl donor defi- logic type, 1977–2005. Br J Cancer. 2009;101:855–9.
ciency with focus on the betaine homocysteine meth- 59. Derakhshan MH, Liptrot S, Paul J, et al. Oesophageal
yltransferase pathway. Nutrients. 2013;5:3481–95. and gastric intestinal-type adenocarcinomas show the
43. Bergheim I, Wolfgarten E, Bollschweiler E, et al. same male predominance due to a 17 year delayed
Role of retinoic acid receptors in squamous-cell carci- development in females. Gut. 2009;58:16–23.
noma in human esophagus. J Carcinog. 2005;8:4–20. 60. Cronin-Fenton DP, Murray LJ, Whiteman DC, et al.
44. Hopkins RA, Postlethwait RW. Caustic burns and carci- Barrett’s Esophagus, Adenocarcinoma Consortium
noma of the esophagus. Ann Surg. 1981;194(2):146–8. (BEACON) Investigators. Reproductive and sex hor-
45. Link LB, Potter JD. Raw versus cooked vegetables monal factors and oesophageal and gastric junction
and cancer risk. Cancer Epidemiol Biomarkers Prev. adenocarcinoma: a pooled analysis. Eur J Cancer.
2004;13(9):1422–35. 2010;46:2067–76.
46. Chen YK, Lee CH, Wu IC, et al. Food intake and the 61. Rubenstein JH, Taylor JB. Meta-analysis: the associa-
occurrence of squamous cell carcinoma in different tion of oesophageal adenocarcinoma with symptoms
sections of the esophagus in Taiwanese men. Nutrition. of gastro-oesophageal reflux. Aliment Pharmacol
2009;25(7–8):753–61. Ther. 2010;32:1222–7.
47. Brown LM, Hoover R, Silverman D, et al. Excess 62. Sharma P. Barrett’s esophagus. N Engl J Med.
incidence of squamous cell esophageal cancer among 2009;361:2548–56. DOI: 10.1056.
US Black men: role of social class and other risk fac- 63. Spechler SJ. Barrett esophagus and risk of esophageal
tors. Am J Epidemiol. 2001;153(2):114–22. cancer: a clinical review. JAMA. 2013;310(6):627–36.
48. Edgren G, Adami HO, Weiderp E, et al. A global 64. Wang KK, Sampliner RE. Updated guidelines 2008
assessment of the oesophageal adenocarcinoma epi- for the diagnosis, surveillance and therapy of Barrett’s
demic. Gut. 2013;62:1406–14. esophagus. Am J Gastroenterol. 2008;103:788–97.
1 Pathogenesis of Esophageal Cancer 9

65. Gatenby P, Ramus J, Caygill C, et al. Routinely diag- 75. Jacobson BC, Somers SC, Fuchs CS, et al. Body-mass
nosed low-grade dysplasia in Barrett’s oesophagus: a index and symptoms of gastroesophageal reflux in
population based study of natural history. Histopathology. women. N Engl J Med. 2006;354:2340–8.
2009;54:814–9. 76. Conteduca V, Sansonno D, Lauletta G, et al. H. pylori
66. Reid BJ, Li X, Galipeau PC, et al. Barrett’s oesopha- infection and gastric cancer: state of the art (review).
gus and oesophageal adenocarcinoma: time for a new Int J Oncol. 2013;42:5–18.
synthesis. Nat Rev Cancer. 2010;10:87–101. 77. Whiteman DC, Parmar P, Fahey P, et al. Australian
67. Schnell TG, Sontag SJ, Chejfec G, et al. Long-term non- Cancer Study. Association of Helicobacter pylori
surgical management of Barrett’s esophagus with high- infection with reduced risk for esophageal cancer is
grade dysplasia. Gastroenterology. 2001;120:1607–19. independent of environmental and genetic modifiers.
68. Galipeau PC, Li X, Blount P, et al. NSAIDs modulate Gastroenterology. 2010;139:73–83.
CDKN2A, TP53, and DNA content risk for future 78. Grisham MB, Jourd’heuil D, Wink DA. Review arti-
esophageal adenocarcinoma. PLoS Med. 2007;4:e67. cle: chronic inflammation and reactive oxygen and
69. Hoyo C, Cook MB, Kamangar F, et al. Body mass nitrogen metabolism–implications in DNA damage
index in relation to oesophageal and oesophagogastric and mutagenesis. Aliment Pharmacol Ther. 2000;14
junction adenocarcinomas: a pooled analysis from the Suppl 1:3–9.
International BEACON Consortium. Int J Epidemiol. 79. Sihvo EI, Salminen JT, Rantanen TK, et al. Oxidative
2012;41:1706–18. stress has a role in malignant transformation in
70. Edelstein ZR, Farrow DC, Bronner MP, et al. Central adi- Barrett’s oesophagus. Int J Cancer. 2002;102:551–5.
posity and risk of Barrett’s esophagus. Gastroenterology. 80. Calle EE, Kaaks R. Overweight, obesity and cancer:
2007;133:403–11. epidemiological evidence and proposed mechanisms.
71. Corley DA, Kubo A, Zhao W. Abdominal obesity and Nat Rev Cancer. 2004;4:579–91.
the risk of esophageal and gastric cardia carcinomas. 81. Trayhurn P, Bing C, Wood IS. Adipose tissue and adi-
Cancer Epidemiol Biomarkers Prev. 2008;17:352–8. pokines–energy regulation from the human perspec-
72. Ryan AM, Healy LA, Power DG, et al. Barrett esophagus: tive. J Nutr. 2006;136(7 Suppl):1935S–9.
prevalence of central adiposity, metabolic syndrome, and 82. Turker MS, Gage BM, Rose JA, et al. A novel signa-
a proinflammatory state. Ann Surg. 2008;247:909–15. ture mutation for oxidative damage resembles a muta-
73. van Kruijsdijk RC, van der Wall E, Visseren tional pattern found commonly in human cancers.
FL. Obesity and cancer: the role of dysfunctional adi- Cancer Res. 1999;59:1837–9.
pose tissue. Cancer Epidemiol Biomarkers Prev. 83. Picardo SL, Maher SG, O’Sullivan JN. Barrett’s to
2009;18(10):2569–78. oesophageal cancer sequence: a model of inflammatory-
74. Doyle SL, Donohoe CL, Finn SP, et al. IGF-1 and its driven upper gastrointestinal cancer. Dig Surg. 2012;29:
receptor in esophageal cancer: association with ade- 251–60.
nocarcinoma and visceral obesity. Am J Gastroenterol.
2012;107:196–204.
 Endoscopic Treatment
of Premalignant and Early 2
Esophageal Malignancy

Toshitaka Hoppo and Blair A. Jobe

Introduction and T1a cancer is unlikely (<2 %), esophagectomy

may be unreasonably invasive [7–9]. However,
Barrett’s esophagus (BE) is a well-known risk patients with HGD and T1a cancer have a chance
factor for esophageal adenocarcinoma, and for cure of disease, although overall prognosis of
progression of metaplasia through dysplasia to esophageal cancer is poor with 5-year survival of
adenocarcinoma is a widely accepted theory of approximately 15 % despite multidisciplinary
esophageal carcinogenesis [1, 2]. High-grade approaches including chemoradiation and surgi-
dysplasia (HGD) has a high risk of progression to cal therapy [10, 11]. Therefore, it is extremely
cancer, and esophageal resection (esophagectomy) important to determine what the best approach is
has been recommended as a standard surgical for this population to accomplish cure without
therapy to treat HGD based on the previous residual or recurrent disease, while minimizing
studies demonstrating that the incidence of con- the postoperative morbidity and mortality.
comitant invasive cancer in the surgically resected With the introduction of endoscopic
specimens of patients with biopsy-proven HGD surveillance programs, patients with HGD and
has been reported to be approximately 40 % [3, T1a cancer have been increasingly discovered.
4]. Esophagectomy is one of the most invasive Accumulating data have demonstrated that highly
surgeries in the upper gastrointestinal tract and is selected patients with HGD and T1a cancer with
associated with high mortality and morbidity low risk or no risk of lymph node involvement can
even with the recent refinement of surgical tech- be treated with esophageal-preserving approaches
niques and perioperative care [5, 6]. Given that including endoscopic ablation (radiofrequency
lymph node involvement in patients with HGD ablation and cryotherapy) and resection (endo-
scopic mucosal resection and submucosal dis-
section) with equivalent oncological outcomes
as surgical resection [12, 13]. Esophageal-
T. Hoppo, MD, PhD (*)
preserving approaches include any endolumi-
Department of Surgery, Institute for the Treatment
of Esophageal and Thoracic Disease, The Western nal procedure that is performed in an attempt
Pennsylvania Hospital, 4800 Friendship Avenue, to completely eradicate disease, while preserv-
Suite 4600, Pittsburgh, PA 15224, USA ing the anatomical structure of esophagus. The
e-mail: [email protected]
recent advances in endoscopic technology and
B.A. Jobe, MD, FACS therapeutic techniques have made esophageal-
Department of Surgery, Institute for the Treatment
preserving approach real. The guideline put forth
of Esophageal and Thoracic Disease,
The Western Pennsylvania Hospital, by the American College of Gastroenterology
Pittsburgh, PA, USA (2008) states that esophagectomy is no longer

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 11
DOI 10.1007/978-3-319-09342-0_2, © Springer International Publishing Switzerland 2015
12 T. Hoppo and B.A. Jobe

the necessary treatment response to HGD [14]. Table 2.1 Low- and high-risk factors to consider for
endoscopic resection of high-grade dysplasia (HGD) and
By contrast, esophageal-preserving options have
intramucosal adenocarcinoma
caused more confusion in the decision-mak-
ing among health care providers. The optimal Indications (low risk) High risk
Unifocal (limited or focal), Multifocal HGD, HGD
management of HGD and T1a cancer remains
flat HGD with nodules
controversial. In this chapter, we focus on esoph- Type I, IIa < 2 cm, IIb, Type I, II > 3 cm, type III
ageal-preserving therapy to treat HGD and T1a IIc < 1 cm
cancer. Well- or moderately Poorly differentiated
differentiated adenocarcinoma
adenocarcinoma
Patient Selection Based on Risk Lesions limited to the Invasion into the
mucosa (m) submucosa (sm)
Stratification No lymphovascular Presence of
invasion lymphovascular invasion
Appropriate patient selection is crucial for Type I, polypoid type; IIa, flat, elevated; IIb, level with the
esophageal-preserving therapy, and patients mucosa. IIc, slightly depressed; III: ulcerated type
at high risk of lymph node involvement and/
or potential progression to cancer or presence
of concomitant cancer need to be accurately metastatic disease. Several macro- and microscopic
identified and excluded from candidates for findings including submucosal invasion (T1b),
esophageal-preserving therapy. Careful endo- squamous-type histology, lymphovascular invasion
scopic examination of esophageal epithelium (L+ or V+), poor differentiation, and a nodule
with extensive biopsies for tissue diagnosis is the >3 cm in diameter have been recognized as high-risk
first step to esophageal-preserving therapy. High- factors for lymph node involvement [12, 13, 15, 16].
quality endoscopic images are required to detect Furthermore, multifocal HGD has a significant risk
questionable, subtle mucosal abnormalities. of concomitant cancer ranging from 60 to 78 %
Several new endoscopic technologies (e.g., opti- [17–20]. By contrast, low-risk factors include uni-
cal coherent tomography, autofluorescent imag- focal (limited or focal) or flat HGD [17–20], type I,
ing, confocal laser endomicroscopy) combined IIa <2 cm, IIb, IIc <1 cm, well or moderately
with enhancement techniques (e.g., narrowband differentiated adenocarcinoma, mucosal cancer
imaging, chromoendoscopy) have been investi- (m), and no lymphovascular invasion (L- and V-)
gated; however, none of them has been routinely [12, 13]. Risk factors for HGD and T1a adenocarci-
used in general practice. For the evaluation of noma are summarized in Table 2.1.
accurate risk stratification, the mucosal and sub- Esophageal squamous cell cancer appears to
mucosal layers have been subdivided into thirds be biologically more aggressive than adenocarci-
with each third going deeper into the esophageal noma, and the risk of lymph node involvement is
wall. Currently, T1 cancers have six different higher in patients with squamous cell cancer.
layers of invasion: T1m1–m3 (m1 = limited to Patients with intraepithelial cancers (m1) and can-
the epithelial layer, m2 = invades lamina propria, cers invading the lamina propria (m2) have almost
m3 = invades into but not through muscularis no risk of lymph node involvement [21– 23],
mucosae) and T1sm1–sm3 (different thirds of whereas the risk of lymph node involvement in
the submucosa). cancers invading the muscularis mucosa (m3) and
the submucosa (sm) ranges from 0 to 10 % [23]
and from 50 to 55 % [22], respectively. For
HGD and/or T1a Esophageal Cancer patients with esophageal squamous cell cancer,
esophageal-preserving therapy can be indicated
Overall, esophageal-preserving therapy can be indi- only for superficial (m1 and m2) cancers with
cated for HGD and/or T1a adenocarcinoma with well-to-moderate differentiation and no lympho-
low risk or no risk of lymph node involvement or vascular invasion. Patients with m3 cancers could
2 Endoscopic Treatment of Premalignant and Early Esophageal Malignancy 13

Table 2.2 Indications for endoscopic resection of esoph- tasis after esophageal-preserving therapy for
ageal squamous cell carcinoma (SCC)
T1sm1 appears to be lower than the mortality
Indications (low risk) High risk rate of esophagectomy, suggesting that patients
No consensus on the maximal size with low-risk T1sm1 adenocarcinoma could be
Well- or moderately Poorly differentiated SCC a candidate for esophageal-preserving therapy,
differentiated SCC
particularly when poor functional status and
Limited to the lamina Invasion into the deeper
propria (m1–2) layer than the muscularis
comorbid conditions make esophagectomy too
mucosa (m3, sm) risky. T1sm2 and sm3 adenocarcinoma and all
No lymphovascular Presence of lymphovascular T1b squamous cell carcinomas are associated
invasion invasion with a high rate of lymph node involvement, and
esophagectomy should be considered [28, 29].
It is noted that these data may not be transfer-
be candidates for esophageal-preserving therapy able to patients at all centers delivering therapy
if there are no further risk factors for lymph node because these data were achieved within high-
involvement. Risk factors for esophageal squa- volume, experienced centers.
mous cell cancer are summarized in Table 2.2.

Clinical Staging
T1b Esophageal Cancer
Accurate clinical staging is essential for
Once tumors invade the submucosal layer, the esophageal-preserving therapy and it is extremely
probability of lymph node involvement is expo- important to exclude patients with potential lymph
nentially increased due to the abundant submu- node involvement and/or metastatic disease.
cosal lymphatic networks [24, 25]. Therefore, Therefore, all patients require positron-emission
esophagectomy has been recommended as a tomography/computed tomography (PET/CT),
standard of care for patients with T1b esopha- endoscopic ultrasound (EUS), and diagnostic
geal cancer. A recent review using the pooled endoscopic mucosal resection (EMR) for staging
data of 7,645 patients with T1b submucosal purposes, when esophageal-preserving therapy
esophageal cancer has demonstrated that the is considered. Since approximately 25 % of all
overall rate of lymph node involvement in T1b patients with esophageal cancer have metastatic
cancers was 37 %; however, there was a sub- disease identified by PET/CT and this yield is
stantial difference between T1sm1 and T1sm2/3 far superior to the combination of EUS and CT
adenocarcinoma (6 % vs. 23 %/58 %, respec- scan [30, 31], PET/CT has been utilized to assess
tively), suggesting that highly selected patients metastatic disease.
with T1sm1 adenocarcinoma could be candi- EUS has been utilized to assess tumor depth
dates for esophageal-preserving therapy [26]. (T-stage) and lymph node involvement (N-stage).
This is further supported by the most recent A recent meta-analysis has demonstrated that the
study involving 66 patients with low-risk T1sm1 pooled sensitivity and specificity of EUS to diag-
cancer (macroscopically polypoid or flat lesion, nose T1 stage cancer was 81.6 and 99.4 %,
well-to-moderate differentiation and no lympho- respectively [32], suggesting that EUS cannot
vascular invasion) demonstrating that complete accurately differentiate T1a from T1b esophageal
remission was achieved in 97 % of patients with cancers. To improve the diagnostic yield of
small nodules ≤2 cm, and long-term remission T1 stage tumors, a high-frequency EUS mini-
without any metachronous disease was achieved probe (20 or 30 MHz) has been introduced and
in 90 %. There were no tumor-related deaths and investigated. A recent retrospective study demon-
the estimated 5-year survival was 84 %, although strated that the overall accuracy, sensitivity, and
one patient developed lymph node metastasis specificity to differentiate T1b from T1a cancers
[27]. The risk of developing lymph node metas- with high-frequency miniprobes were 73.5, 62,
14 T. Hoppo and B.A. Jobe

and 76.5 %, respectively, suggesting that even the compared to 19 % in the control group (no RFA)
high-frequency miniprobe still has a limited (p <0.001) and patients who underwent RFA
accuracy for the diagnosis of T1a cancer [33]. had significantly less disease progression and
Other meta-analysis has demonstrated that fewer cancers developed during the follow-up of
pooled sensitivity and specificity of EUS for 12 months [35]. Either an ablation balloon cath-
regional lymph node involvement were 76.4 and eter (Barrx™ 360 RFA Balloon Catheter) for cir-
72.4 %, respectively, suggesting EUS is also not cumferential ablation or an endoscopic mounted
satisfactory for the assessment of N-staging [34]. device (Barrx™ 90, 60, Ultra Long RFA Focal
It is important to understand the limitation of Catheter) for focal ablation can be selected based
EUS in the staging process. on the length, extension, and location of disease
Since EUS is not reliable for T- and N-staging, (Fig. 2.1). This system delivers a high-power,
a diagnostic EMR for the staging purpose is ultrashort burst of ablative energy to the abnor-
essential to exclude any possibility of submuco- mal esophageal epithelium, and the delivered
sal (T1b) or deeper invasion (>T2) and predict energy provides uniform treatment to a depth of
potential lymph node involvement based on com- approximately 500 μm. Therefore, the depth of
plete histological assessment. EMR provides treatment is limited to the mucosal layer, thereby
complete specimens including both mucosa and significantly reducing the risk of stricture forma-
submucosa for histological assessment of lateral tion. The rate of stricture formation was reported
and deep margins, thereby determining the accu- to be 6 % [35]. A further follow-up study demon-
rate T-stage (i.e., differentiating T1a from T1b). strated that patients’ quality of life significantly
A positive lateral margin can be addressed with improved after the RFA treatment, although most
additional endoscopic intervention, while a posi- patients were worried about esophageal cancer
tive deep margin should be considered for and esophagectomy before the RFA treatment
esophagectomy. [36]. Due to the limited depth of treatment, RFA
is not indicated for invasive cancer.

Endoscopic Ablation
Cryotherapy
The purpose of endoscopic ablation therapy is to
eradicate disease by ablating (burning or freez- Cryotherapy involves the topical application by
ing) the affected epithelium of the esophagus. spraying aerosolized liquid nitrogen or carbon
Currently, radiofrequency ablation (RFA) and dioxide onto the abnormal esophageal epithe-
cryotherapy have been primarily performed as lium, providing intracellular disruption and isch-
endoscopic ablation therapy. The common draw- emia while preserving the extracellular matrix
back of ablation therapy is that there is no speci- and thereby minimizing fibrosis. A prospective
men available for histological assessment. study involving 98 patients with HGD has dem-
onstrated that 97 % had complete eradication
of HGD with no esophageal perforation [28].
Radiofrequency Ablation (RFA) Current cryotherapy devices require a venting
system such as a nasogastric tube to help exces-
RFA using the Barrx™ Ablation System sive nitrogen gas escape out of the esophagus
(Covidien, Sunnyvale, CA) has been most com- and stomach, thus preventing perforation of the
monly used as ablation therapy, since the mul- gastrointestinal tract. Furthermore, cryotherapy
ticenter, randomized, sham-controlled trial is associated with several issues including its
involving 127 patients with Barrett’s esophagus nonuniform application using a handheld cathe-
demonstrated that 81 % of patients with HGD ter, the fogging of scope lens, and the prolonged
had complete eradication of dysplasia with RFA duration of the therapy. Currently, a novel
2 Endoscopic Treatment of Premalignant and Early Esophageal Malignancy 15

Fig. 2.1 Radiofrequency

ablation therapy. The upper
panels show the Barrx™
(left) and Barrx™ systems
(right). The lower panel
shows endoscopic findings
of pre- and post-treatment
for intestinal metaplasia BarrxTM 360
(top), low-grade dysplasia BarrxTM 90
(middle), and high-grade
dysplasia (bottom)
(Permission for use Pre-treatment Follow-up
granted by Cook Medical
Incorporated, Bloomington,
Indiana)
Intestinal metaplasia

Low-grade metaplasia

High-grade metaplasia

through-the-scope cryoballoon device, which resection can provide specimens for the complete
does not require a venting system and potentially histological assessment including depth of cancer
delivers a uniform and reproducible ablation, invasion, degree of cellular differentiation, and
has been under investigation, and further study involvement of lymphatics or vessels. Currently,
to evaluate the safety and efficacy of this device endoscopic mucosal resection (EMR) is used for
is awaited. the lesions less than 2 cm, and endoscopic submu-
cosal dissection (ESD) is recommended for en bloc
removal of lesions larger than 2 cm. The major
Endoscopic Resection complication associated with endoscopic resection
is stricture formation, especially when more than
The goal of endoscopic resection is to completely 75 % of the esophageal circumference is involved
remove the entire segment of abnormal esophageal in a single setting [29]. Small clinical series have
epithelium, thereby curing HGD and T1a cancers. reported that the stricture rate after circumferential
Unlike endoscopic ablation therapy, endoscopic EMR is up to 80 % [37, 38].
16 T. Hoppo and B.A. Jobe

a b

c d

Fig. 2.2 Four types of endoscopic mucosal resection (EMR) techniques. (a) Snare polypectomy. (b) Strip biopsy
technique. (c) The cap resection technique. (d) The ligate-and-cut technique (From Soetikno et al. [39] with permission)

Endoscopic Mucosal Resection (EMR) clear plastic cap (either straight or oblique
shaped) is attached to the tip of endoscope. The
EMR has been commonly used as both a oblique-shaped caps are usually used for esopha-
diagnostic and therapeutic tool. There are primar- geal lesions, while the straight caps are most
ily two techniques: cap resection technique and commonly used for the lesions in the stomach
ligate-and-cut technique (Fig. 2.2) [39]. A ran- and colon. The mucosal-submucosal complex is
domized trial comparing these two techniques then sucked into a cap mounted on an endoscope,
has demonstrated no difference in safety and effi- creating a pseudopolyp. The pseudopolyp is then
cacy between the techniques [40]. Both tech- resected by being captured at its base with a cau-
niques are initiated by injecting normal saline tery snare which is positioned inside the cap [41].
into the submucosal space to lift the lesions away For the ligate-and-cut technique, the only differ-
from the muscularis propria. The injection needle ence to the cap technique is to deploy a band to
should be inserted into a submucosal space at the create a pseudopolyp [26, 27]. Currently, there is
sharp angle to avoid transmural penetration of the a novel multiband mucosectomy device avail-
needle. Injected saline acts as a “safety cushion” able, which uses a specially designed multiple-
between the mucosa and muscularis propria to band ligator and allows endoscopists to perform
reduce the risk of unexpected complications such ligation and subsequent immediate resection
as perforation during the procedure. Additional without removal of the endoscope by passing a
injection of saline is sometimes required because cautery snare through the ligator handle. The
the injected saline disappears within a few retrieved specimen is pinned to a piece of cork
minutes. Difficulty lifting up the lesion by sub- and placed into preservative solution prior to pro-
mucosal injection suggests invasion into the mus- cessing for histological assessment. EMR is indi-
cularis propria. For the cap resection technique, a cated for lesions less than 2 cm in diameter.
2 Endoscopic Treatment of Premalignant and Early Esophageal Malignancy 17

Table 2.3 Risk factors potentially associated with recurrent cancers with patients prior to the initia-
recurrence after endoscopic resection of early esophageal
tion of endoscopic interventions.
cancer
Risk factors for recurrence after endoscopic resection
of early esophageal cancer
Endoscopic Submucosal
1. Piecemeal resection
2. Long-segment BE
Dissection (ESD)
3. No ablation therapy of BE after CR
4. Time until CR achieved > 10 months ESD has been established as an advanced endo-
5. Multifocal neoplasia scopic resection technique to accomplish en bloc
BE Barrett’s esophagus, CR complete remission
resection of lesions larger than 2 cm in diameter.
ESD is expected to provide more accurate histo-
logical assessment for the lateral and deep
Although en-bloc resection is ideal in any margins of lesions and thus prevent or minimize
situation, piecemeal resection of large lesions the development of metachronous lesions. ESD
(>2 cm) is acceptable; however, several studies employs the same concept of EMR but requires
have shown that piecemeal EMR is associated some modifications for en bloc resection of a
with incomplete resection and compromised large lesion. Each step is summarized in Fig. 2.3.
histological assessment, likely causing the devel- ESD is initiated by a mucosal marking around the
opment of metachronous lesions [42, 43]. lesion by using electrocautery, thus easily identi-
An early retrospective study from a single fying the location of the entire lesion after the
institution demonstrated that 98 % of patients submucosal injection (Fig. 2.4a). Subsequently, a
with HGD and T1a adenocarcinoma (n = 115) solution is injected into the submucosal space to
achieved complete response to EMR; however, lift the lesion away from the muscularis propria.
30 % of patients developed metachronous can- The injection solutions for ESD include normal
cers during a mean follow-up of 34 months [42]. saline, glycerol, and sodium hyaluronate. Sodium
In a further study from the same institution, sev- hyaluronate stays longer in the submucosal space
eral factors including piecemeal resection, long- than normal saline or glycerol and may be ideal
segment BE, no ablation therapy of BE after a for ESD. Diluted sodium hyaluronate (approxi-
complete response, multifocal neoplasia, and mately 0.5 % solution) is usually mixed with epi-
time until complete response >10 months were nephrine (0.01 mg/ml) and indigo carmine
found to be associated with frequent tumor recur- (0.04 mg/ml). The mucosal cutting is then per-
rence after endoscopic resection (Table 2.3) [43]. formed to create the entry to the submucosal
Based on these results, combination therapy space by using a specialized endoscopic electo-
involving focal EMR to resect nodules followed cautery called “needle knife” (Fig. 2.4b, c).
by RFA to treat any residual flat Barrett’s epi- Several types of needle knives having different
thelium has been investigated to minimize the shaped tips are available, depending on the pref-
development of recurrent disease. A recent multi- erence of endoscopists (Fig. 2.5). Once the sub-
center, prospective study to evaluate the efficacy mucosal space is entered, tension and
of this combination therapy has demonstrated counter-tension are maintained by a cap mounted
that 95 % of patients with HGD or T1a adenocar- on the tip of endoscope, which is placed in the
cinoma (n = 24) had a complete response to the plane between the mucosal-submucosal complex
combination therapy and no recurrence occurred and the muscularis propria. The needle knife is
during a median follow-up of 22 months [44]. then introduced through the endoscopic working
These studies emphasize the importance of channel, and the attachments and bridging ves-
intensive surveillance, the risk of metachronous sels between the two layers are dissected. At the
lesions after endoscopic resection, the need for completion of this procedure, the lesion can be
post-intervention intensive surveillance, and resected en bloc regardless of its size and the
the necessity of discussing the possibility of remaining thin layer of sm3 can be seen over the
18 T. Hoppo and B.A. Jobe

a d
Lesion
Mucosa

Submucosa Muscularis propria

b e

c f

Fig. 2.3 Schematic representation of endoscopic around the mucosal markings. (e) Submucosal dissection
submucosal dissection. (a) Mucosal markings for the inci- with a needle knife through the cap attached on the tip of
sion line. (b) Submucosal injections of a solution. endoscope. (f) En bloc resection of the tumor. M mucosa,
(c) Complete elevation of the lesion by injecting a solu- SM submucosa, MP muscularis propria
tion into the submucosal space. (d) Mucosal incision
2 Endoscopic Treatment of Premalignant and Early Esophageal Malignancy 19

a b

c d

Fig. 2.4 Endoscopic submucosal dissection of early sodium hyaluronate, submucosal dissection can be
esophageal squamous cell carcinoma. (a) Chromoendoscopy performed. (c) Submucosal dissection is performed using
shows the presence of an irregular unstained area in the the needle knife. (d) The tumor is resected en bloc. A thin
middle esophagus. The markings are performed using an layer of sm3 was observed over the muscle layer. (e) The
electrocautery. (b) After the submucosal injection of resected specimen was spread out and pinned on a flat cork

resected area (Fig. 2.4d). It is important to be resected throughout the procedure. For this
maintain this thin layer of sm3, especially when purpose, it may be better to start with a partial
repair of a perforation is required. ESD is a mucosal incision rather than a circumferential
“one-person” procedure and does not allow for mucosal incision, maintaining the continuity of
assistant hands. It is therefore important to main- the lesion to the esophageal epithelium as “coun-
tain adequate counter-traction on the mucosa to ter-traction,” and mucosal incision and
20 T. Hoppo and B.A. Jobe

Fig. 2.5 Different types of needle knife for endoscopic hook knife. Lower left: triangle-tip knife (TT knife).
submucosal dissection. Upper left: insulation-tipped dia- Lower right: dual knife
thermic electrosurgical knife (IT knife). Upper right:

submucosal dissection should be repeated step by ESD, a Japanese group reported that en bloc
step. The advantage of gravity should be consid- resection and complete resection were achieved in
ered; submucosal dissection should be started 100 and 88 % of patients, respectively, and the
from the upper portion of the lesion so that the 5-year cause-specific survival of patients with
dissected mucosa is pulled down by gravity, T1a cancers was 100 % [45]. Major complica-
spontaneously exposing the submucosal layer. It tions including perforation occurred in 4 % of
is also worth considering repositioning patients patients, and 18 % developed benign esophageal
to obtain the advantage of gravity. stricture requiring dilation [45]. This suggests that
ESD is expected to be superior to EMR ESD could be a reasonable option for cure of
because of the availability of en bloc resection HGD and T1a cancers. The perforation rate dur-
specimens, although no randomized controlled ing ESD is reported to be higher than that during
study comparing ESD with EMR is available. EMR (4–10 % vs. 0.3–0.5 %, respectively)
Since ESD has not been routinely performed to [46–50]. Perforation is easily identified during the
treat patients with HGD and T1a cancers, the procedure, and a small perforation can be
available data to show the efficacy of ESD in this addressed by deploying endoclips [41, 42]. A large
setting are limited. In a study to evaluate the long- perforation requires an emergent surgery to avoid
term outcomes of 84 patients with superficial peritonitis and/or mediastinitis. Furthermore,
esophageal squamous cell cancer who underwent stricture formation is the other major complication
2 Endoscopic Treatment of Premalignant and Early Esophageal Malignancy 21

of ESD. Strictures are more likely to occur after reasonable to continue intensive surveillance
ESD for esophageal lesions (up to 26 %) [29, 43, every 3 months up to 1 year following esophageal-
44]. It should be noted that ESD for BE and preserving therapy, and then consider antireflux
esophageal adenocarcinoma may be technically surgery if there is no evidence of recurrence.
more difficult than ESD for gastric cancer or Long-term endoscopic surveillance per the ASGE
esophageal squamous cell cancer because of its guidelines is still required even after antireflux
location in the distal esophagus close to the gas- surgery is performed. BE refractory to endo-
troesophageal junction and the submucosal scar- scopic intervention may be caused by persistent,
ring due to reflux-induced inflammation. Because significant acid exposure to the treated area, and
of the high rates of perforation and stricture and early antireflux surgery may be considered.
steep learning-curve, ESD has not been widely Upper endoscopy has been performed under
accepted especially for esophageal lesions in the conscious sedation, causing significant direct
United States. (e.g., personnel, facility) and indirect (e.g., off
work, third-party transportation) costs. In addi-
tion, most complications are associated with con-
Management After scious sedation. With the recent advances in optic
Esophageal-Preserving Therapy technology, small-caliber endoscopes have been
developed and introduced to perform transnasal
It should be emphasized that intensive endo- endoscopy. Transnasal endoscopy can be per-
scopic surveillance and strict acid suppression formed in the office setting without intravenous
with high-dose proton pump inhibitors (PPI) and sedation, and our previous study has demon-
nocturnal H2 blockers after intervention are criti- strated the equivalent efficacy and accuracy and
cal for successful esophageal-preserving therapy. better patients’ tolerance compared to conven-
Strict acid suppression establishes an acid-free tional endoscopy [52]. There is a small-caliber
environment in the treated area, thus facilitating endoscope available, which has a disposable
the healing process to the normal “neosquamous” sheath with an incorporated coaxial biopsy chan-
lining. There has been no consensus on surveil- nel placed over it. Therefore, there is no need for
lance protocols following esophageal-preserving post-procedure endoscope processing as required
therapy; however, the guidelines issued by the for sedated endoscopy, and the cost can be sig-
American Society for Gastrointestinal Endoscopy nificantly reduced. This technology may allow us
(ASGE) states that surveillance endoscopy for perform a low-cost, safe, and intensive surveil-
patients with HGD should undergo every lance of patients who undergo esophageal-
3 months for at least 1 year with multiple large preserving therapy.
capacity biopsy specimens obtained at 1 cm
intervals. After 1 year, if there is no detection of
recurrence, the interval of surveillance may be Long-Term Outcome
lengthened if there are no dysplastic changes on of Esophageal-Preserving Therapy
two subsequent endoscopies performed at
3-month intervals [51]. Since BE is caused by The long-term outcome data of esophageal-
long-term acid exposure to the distal esophagus, preserving therapy to treat HGD and T1a cancers
a surgical repair of underlying gastroesophageal are still limited. In a single large prospective
reflux disease (GERD) should be considered in study involving 349 patients with HGD and T1a
order to eliminate all acid exposure to the esoph- adenocarcinoma who underwent esophageal-
ageal epithelium, thus liberating patients from preserving therapies such as ablation and endo-
acid suppression therapy. Although the ASGE scopic resection, 96.6 % achieved complete
guideline states that antireflux surgery should not response and only 3.7 % required surgery during
be advised with the rationale that the procedure a mean follow-up of 5 years without tumor-
will prevent esophageal cancer [51], it may be related deaths. In addition, patients who
22 T. Hoppo and B.A. Jobe

underwent RFA for persistent or recurrent BE reducing the temperature below 32 °C, the
had a lower incidence of metachronous neoplasia cultured cells spontaneously detach from the
compared to those who did not undergo RFA culture plate without any proteolytic enzyme as a
(16.5 % vs. 28.3 %) [43]. This suggests that addi- “cell sheet”. Transplantable, autologous epithelial
tional ablation therapy for remaining BE may cell sheets have been applied onto the treated area
minimize the development of metachronous neo- after aggressive ESD and shown to successfully
plasia. The rates of bleeding and stenosis were 12 minimize stricture formation in the clinical setting
and 4.3 %, respectively. [59]. Regenerative medicine approaches may
Several studies have demonstrated a low com- make more aggressive endoscopic resection pos-
plication rate and a good disease-specific 5-year sible, although further studies are required.
survival rate for the endoscopic resection of
esophageal squamous cell carcinoma [53–55]. Conclusion
The most recent retrospective cohort study Highly selected patients with HGD and T1a
involving 51 patients with either squamous dys- cancers can be treated endoscopically. Based
plasia or T1a squamous cell cancer who under- on the risk stratification and accurate clinical
went repeated EMR until complete local staging, patients with potential lymph node
remission was achieved, 91 % of patients had involvement and/or metastatic disease need to
complete remission, and the disease-specific be excluded. At present, patients with HGD
5-year survival rate was 95 %. Minor bleeding and T1a adenocarcinoma or early squamous
occurred in 17 % of patients and 3 patients (6 %) cell carcinoma (m1 and m2) with low risk or no
developed mild stenosis requiring dilation, risk of lymph node involvement can be a candi-
although there was no perforation. During the date for esophageal-preserving therapy. Highly
follow-up period, local disease recurrence was selected patients with T1sm1 adenocarcinoma
observed in 26 % of patients [53]. could be a candidate for esophageal-preserving
therapy, particularly when poor functional
status and comorbid conditions make esopha-
Future Prospective gectomy too risky. Intensive surveillance and
strict acid suppression therapy are required
With the advances in endoscopic technologies after esophageal-preserving therapy. Since BE
and techniques, aggressive endoscopic resection is caused by GERD, antireflux surgery should
is technically feasible. However, the main con- be considered once the treated area is deter-
cern is a post-procedure stricture formation. To mined to be stable.
prevent or minimize stricture formation after
aggressive endoscopic resection, novel approaches
such as biologic scaffold materials composed of
xenogeneic extracellular matrix (ECM) and cell- References
sheet technology have been investigated in the
1. Jankowski JA, Wright NA, Meltzer SJ, et al. Molecular
context of regenerative medicine. ECM has been
evolution of the metaplasia-dysplasia-adenocarcinoma
shown to remodel the default tissue via neo-epi- sequence in the esophagus. Am J Pathol. 1999;154(4):
thelialization rather than scar formation [54, 55] 965–73.
and to successfully minimize a stricture formation 2. Werner M, Mueller J, Walch A, Hofler H. The molecu-
lar pathology of Barrett’s esophagus. Histol Histopathol.
after aggressive, circumferential endoscopic
1999;14(2):553–9.
resection both in preclinical [56] and clinical set- 3. Collard JM. High-grade dysplasia in Barrett’s esopha-
tings [57]. Cell-sheet technology is a tissue engi- gus. The case for esophagectomy. Chest Surg Clin N
neering approach to attempt reconstructing tissue Am. 2002;12(1):77–92.
4. Falk GW, Rice TW, Goldblum JR, Richter JE. Jumbo
without scaffold [58]. Epithelial cells obtained
biopsy forceps protocol still misses unsuspected can-
from patients’ oral mucosal tissue are cultured on cer in Barrett’s esophagus with high-grade dysplasia.
temperature-responsive culture plates. By Gastrointest Endosc. 1999;49(2):170–6.
2 Endoscopic Treatment of Premalignant and Early Esophageal Malignancy 23

5. Birkmeyer JD, Siewers AE, Finlayson EV, et al. 21. Araki K, Ohno S, Egashira A, Saeki H, Kawaguchi H,
Hospital volume and surgical mortality in the United Sugimachi K. Pathologic features of superficial esoph-
States. N Engl J Med. 2002;346(15):1128–37. ageal squamous cell carcinoma with lymph node and
6. Halm EA, Lee C, Chassin MR. Is volume related to distal metastasis. Cancer. 2002;94(2):570–5.
outcome in health care? A systematic review and meth- 22. Eguchi T, Nakanishi Y, Shimoda T, et al.
odologic critique of the literature. Ann Intern Med. Histopathological criteria for additional treatment
2002;137(6):511–20. after endoscopic mucosal resection for esophageal
7. Oh DS, Hagen JA, Chandrasoma PT, et al. Clinical cancer: analysis of 464 surgically resected cases. Mod
biology and surgical therapy of intramucosal adeno- Pathol. 2006;19(3):475–80.
carcinoma of the esophagus. J Am Coll Surg. 2006; 23. Tajima Y, Nakanishi Y, Ochiai A, et al. Histopathologic
203(2):152–61. findings predicting lymph node metastasis and prog-
8. Rice TW, Blackstone EH, Adelstein DJ, et al. Role of nosis of patients with superficial esophageal carci-
clinically determined depth of tumor invasion in the noma: analysis of 240 surgically resected tumors.
treatment of esophageal carcinoma. J Thorac Cardiovasc Cancer. 2000;88(6):1285–93.
Surg. 2003;125(5):1091–102. 24. Rice TW. Pro: esophagectomy is the treatment of
9. Rice TW, Zuccaro Jr G, Adelstein DJ, Rybicki LA, choice for high-grade dysplasia in Barrett’s esopha-
Blackstone EH, Goldblum JR. Esophageal carcinoma: gus. Am J Gastroenterol. 2006;101(10):2177–9.
depth of tumor invasion is predictive of regional lymph 25. Sepesi B, Watson TJ, Zhou D, et al. Are endoscopic
node status. Ann Thorac Surg. 1998;65(3):787–92. therapies appropriate for superficial submucosal
10. Pohl H, Sirovich B, Welch HG. Esophageal adenocar- esophageal adenocarcinoma? An analysis of esopha-
cinoma incidence: are we reaching the peak? Cancer gectomy specimens. J Am Coll Surg. 2010;210(4):
Epidemiol Biomarkers Prev. 2010;19(6):1468–70. 418–27.
11. Siegel R, Naishadham D, Jemal A. Cancer statistics, 26. Gockel I, Sgourakis G, Lyros O, et al. Risk of lymph
2012. CA Cancer J Clin. 2012;62(1):10–29. node metastasis in submucosal esophageal cancer: a
12. Ell C, May A, Pech O, et al. Curative endoscopic resec- review of surgically resected patients. Expert Rev
tion of early esophageal adenocarcinomas (Barrett’s Gastroenterol Hepatol. 2011;5(3):371–84.
cancer). Gastrointest Endosc. 2007;65(1):3–10. 27. Manner H, Pech O, Heldmann Y, et al. Efficacy,
13. Pech O, May A, Gossner L, et al. Curative endoscopic safety, and long-term results of endoscopic treatment
therapy in patients with early esophageal squamous- for early-stage adenocarcinoma of the esophagus with
cell carcinoma or high-grade intraepithelial neoplasia. low-risk sm1 invasion. Clin Gastroenterol Hepatol.
Endoscopy. 2007;39(1):30–5. 2013;11(6):630–5.
14. Wang KK, Sampliner RE. Updated guidelines 2008 28. Shaheen NJ, Greenwald BD, Peery AF, et al. Safety
for the diagnosis, surveillance and therapy of Barrett’s and efficacy of endoscopic spray cryotherapy for
esophagus. Am J Gastroenterol. 2008;103(3):788–97. Barrett’s esophagus with high-grade dysplasia.
15. Bolton WD, Hofstetter WL, Francis AM, et al. Impact Gastrointest Endosc. 2010;71(4):680–5.
of tumor length on long-term survival of pT1 esopha- 29. Katada C, Muto M, Manabe T, Boku N, Ohtsu A,
geal adenocarcinoma. J Thorac Cardiovasc Surg. 2009; Yoshida S. Esophageal stenosis after endoscopic
138(4):831–6. mucosal resection of superficial esophageal lesions.
16. Stein HJ, Feith M, Bruecher BL, Naehrig J, Sarbia M, Gastrointest Endosc. 2003;57(2):165–9.
Siewert JR. Early esophageal cancer: pattern of lym- 30. Gananadha S, Hazebroek EJ, Leibman S, et al. The util-
phatic spread and prognostic factors for long-term ity of FDG-PET in the preoperative staging of esopha-
survival after surgical resection. Ann Surg. geal cancer. Dis Esophagus. 2008;21(5):389–94.
2005;242(4):566–73; discussion 573–5. 31. Meyers BF, Downey RJ, Decker PA, et al. The utility
17. Buttar NS, Wang KK, Sebo TJ, et al. Extent of high- of positron emission tomography in staging of poten-
grade dysplasia in Barrett’s esophagus correlates with tially operable carcinoma of the thoracic esophagus:
risk of adenocarcinoma. Gastroenterology. 2001;120(7): results of the American College of Surgeons Oncology
1630–9. Group Z0060 trial. J Thorac Cardiovasc Surg. 2007;
18. Levine DS, Haggitt RC, Blount PL, Rabinovitch PS, 133(3):738–45.
Rusch VW, Reid BJ. An endoscopic biopsy protocol 32. Puli SR, Reddy JB, Bechtold ML, Antillon D, Ibdah JA,
can differentiate high-grade dysplasia from early ade- Antillon MR. Staging accuracy of esophageal cancer
nocarcinoma in Barrett’s esophagus. Gastroenterology. by endoscopic ultrasound: a meta-analysis and sys-
1993;105(1):40–50. tematic review. World J Gastroenterol. 2008;14(10):
19. Schnell TG, Sontag SJ, Chejfec G, et al. Long-term 1479–90.
nonsurgical management of Barrett’s esophagus with 33. Chemaly M, Scalone O, Durivage G, et al. Miniprobe
high-grade dysplasia. Gastroenterology. 2001;120(7): EUS in the pretherapeutic assessment of early esopha-
1607–19. geal neoplasia. Endoscopy. 2008;40(1):2–6.
20. Weston AP, Sharma P, Topalovski M, Richards R, 34. Sgourakis G, Gockel I, Lyros O, Hansen T,
Cherian R, Dixon A. Long-term follow-up of Barrett’s Mildenberger P, Lang H. Detection of lymph node
high-grade dysplasia. Am J Gastroenterol. 2000;95(8): metastases in esophageal cancer. Expert Rev Anticancer
1888–93. Ther. 2011;11(4):601–12.
24 T. Hoppo and B.A. Jobe

35. Shaheen NJ, Sharma P, Overholt BF, et al. 48. Kato M. Endoscopic submucosal dissection (ESD) is
Radiofrequency ablation in Barrett’s esophagus with being accepted as a new procedure of endoscopic
dysplasia. N Engl J Med. 2009;360(22):2277–88. treatment of early gastric cancer. Intern Med. 2005;
36. Shaheen NJ, Peery AF, Hawes RH, et al. Quality of 44(2):85–6.
life following radiofrequency ablation of dysplastic 49. Ono H. Early gastric cancer: diagnosis, pathology,
Barrett’s esophagus. Endoscopy. 2010;42(10):790–9. treatment techniques and treatment outcomes. Eur J
37. Seewald S, Ang TL, Omar S, et al. Endoscopic muco- Gastroenterol Hepatol. 2006;18(8):863–6.
sal resection of early esophageal squamous cell can- 50. Yokoi C, Gotoda T, Hamanaka H, Oda I. Endoscopic
cer using the Duette mucosectomy kit. Endoscopy. submucosal dissection allows curative resection of
2006;38(10):1029–31. locally recurrent early gastric cancer after prior endo-
38. Soehendra N, Seewald S, Groth S, et al. Use of modi- scopic mucosal resection. Gastrointest Endosc. 2006;
fied multiband ligator facilitates circumferential EMR 64(2):212–8.
in Barrett’s esophagus (with video). Gastrointest 51. Hirota WK, Zuckerman MJ, Adler DG, et al. ASGE
Endosc. 2006;63(6):847–52. guideline: the role of endoscopy in the surveillance of
39. Soetikno RM, Gotoda T, Nakanishi Y, Soehendra N. premalignant conditions of the upper GI tract.
Endoscopic mucosal resection. Gastrointest Endosc. Gastrointest Endosc. 2006;63(4):570–80.
2003;57(4):567–79. 52. Jobe BA, Hunter JG, Chang EY, et al. Office-based
40. May A, Gossner L, Behrens A, et al. A prospective unsedated small-caliber endoscopy is equivalent to
randomized trial of two different endoscopic resection conventional sedated endoscopy in screening and sur-
techniques for early stage cancer of the esophagus. veillance for Barrett’s esophagus: a randomized and
Gastrointest Endosc. 2003;58(2):167–75. blinded comparison. Am J Gastroenterol. 2006;101(12):
41. Inoue H, Endo M, Takeshita K, Yoshino K, Muraoka Y, 2693–703.
Yoneshima H. A new simplified technique of endoscopic 53. Ciocirlan M, Lapalus MG, Hervieu V, et al.
esophageal mucosal resection using a cap-fitted panen- Endoscopic mucosal resection for squamous prema-
doscope (EMRC). Surg Endosc. 1992;6(5):264–5. lignant and early malignant lesions of the esophagus.
42. May A, Gossner L, Pech O, et al. Local endoscopic Endoscopy. 2007;39(1):24–9.
therapy for intraepithelial high-grade neoplasia and early 54. Inoue H, Fukami N, Yoshida T, Kudo SE. Endoscopic
adenocarcinoma in Barrett’s oesophagus: acute-phase mucosal resection for esophageal and gastric cancers.
and intermediate results of a new treatment approach. J Gastroenterol Hepatol. 2002;17(4):382–8.
Eur J Gastroenterol Hepatol. 2002;14(10):1085–91. 55. Takeshita K, Tani M, Inoue H, et al. Endoscopic treat-
43. Pech O, Behrens A, May A, et al. Long-term results ment of early oesophageal or gastric cancer. Gut.
and risk factor analysis for recurrence after curative 1997;40(1):123–7.
endoscopic therapy in 349 patients with high-grade 56. Nieponice A, McGrath K, Qureshi I, et al. An extra-
intraepithelial neoplasia and mucosal adenocarcinoma cellular matrix scaffold for esophageal stricture pre-
in Barrett’s oesophagus. Gut. 2008;57(9):1200–6. vention after circumferential EMR. Gastrointest
44. Pouw RE, Wirths K, Eisendrath P, et al. Efficacy of Endosc. 2009;69(2):289–96.
radiofrequency ablation combined with endoscopic 57. Badylak SF, Hoppo T, Nieponice A, Gilbert TW,
resection for barrett’s esophagus with early neoplasia. Davison JM, Jobe BA. Esophageal preservation in
Clin Gastroenterol Hepatol. 2010;8(1):23–9. five male patients after endoscopic inner-layer cir-
45. Ono S, Fujishiro M, Niimi K, et al. Long-term out- cumferential resection in the setting of superficial
comes of endoscopic submucosal dissection for cancer: a regenerative medicine approach with a bio-
superficial esophageal squamous cell neoplasms. logic scaffold. Tissue Eng Part A. 2011;17(11–12):
Gastrointest Endosc. 2009;70(5):860–6. 1643–50.
46. Fujishiro M, Yahagi N, Kakushima N, et al. 58. Yamato M, Utsumi M, Kushida A, Konno C, Kikuchi
Endoscopic submucosal dissection of esophageal A, Okano T. Thermo-responsive culture dishes allow
squamous cell neoplasms. Clin Gastroenterol Hepatol. the intact harvest of multilayered keratinocyte sheets
2006;4(6):688–94. without dispase by reducing temperature. Tissue Eng.
47. Gotoda T. A large endoscopic resection by endoscopic 2001;7(4):473–80.
submucosal dissection procedure for early gastric can- 59. Takagi R, Yamato M, Kanai N, et al. Cell sheet tech-
cer. Clin Gastroenterol Hepatol. 2005;3(7 Suppl 1): nology for regeneration of esophageal mucosa. World
S71–3. J Gastroenterol. 2012;18(37):5145–50.
 The Volume-Outcome
Relationship, Standardized Clinical 3
Pathways, and Minimally Invasive
Surgery for Esophagectomy

Sheraz R. Markar and Donald E. Low

Volume-Outcome Relationship that esophagectomy performed at low-volume

Associated with the Surgical centers was associated with an increase in the
Treatment of Esophageal Cancer incidence of inhospital (8.48 % vs. 2.82 %) and
30-day mortality (2.09 % vs. 0.73 %) [8]. There
Esophagectomy remains an important component was insufficient evidence to provide a meaning-
of treatment for locoregional esophageal cancer. ful analysis of the effect of hospital volume on
Despite recent advancements in surgical technique length of hospital stay and postoperative compli-
and perioperative care, esophagectomy remains cations. Furthermore, hospital volume has not
one of the most demanding surgical procedures currently been demonstrated to have any signifi-
associated with a highly variable rate of morbidity cant influence on long-term survival following
and mortality (3–14 %) and a poor overall survival esophagectomy [9]. Therefore, the present pub-
(5 year survival of 20–30 %) [1–3]. Recent studies lished data would suggest that high hospital
examining the volume-outcome relationship for esophagectomy volume is associated with a
esophageal resection suggest that high-volume reduction in mortality; however, there is insuffi-
institutions with a larger caseload and appropriate cient data to comment on other important vari-
infrastructure are better prepared to deliver ables including long-term survival, complications,
consistently high-quality outcomes [4–6]. and quality of life following surgery.
Some studies have suggested that hospital vol- It remains unclear whether hospital or surgeon
ume does not influence perioperative mortality volume is the most important factor determining
following esophagectomy [6, 7]. However, in clinical outcome following esophagectomy.
2012, a pooled analysis of nine relevant publica- Some studies have suggested that surgeon vol-
tions comprising 27,843 patients who underwent ume does not influence clinical outcome follow-
esophagectomy from 2000 to 2011 demonstrated ing esophagectomy [10, 11]; however, others
suggest high surgeon volume is associated with a
reduced incidence of postoperative complica-
S.R. Markar, MRCS(Eng), MA, MSc (*) tions and reduced length of hospital stay [12]. It
Department of Thoracic Surgery, is important to acknowledge the variability in
Virginia Mason Medical Center, esophagectomy surgical approaches (e.g., tran-
1100 Ninth Avenue, Seattle, WA 98111, USA
shiatal, Ivor-Lewis, 3-stage, minimally invasive,
e-mail: [email protected]
etc.), and this heterogeneity may not be accu-
D.E. Low, FACS, FRCS(c)
rately characterized in clinical coding and thus
Department of General, Thoracic, and Vascular Surgery,
Virginia Mason Medical Center, fails to allow for subset esophageal procedural
Seattle, WA, USA analysis to elucidate subtle variations in outcome.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 25
DOI 10.1007/978-3-319-09342-0_3, © Springer International Publishing Switzerland 2015
26 S.R. Markar and D.E. Low

It is therefore possible that the improvements Countries that have used the volume-outcome
seen in postoperative mortality following esopha- relationship as a means to centralize esophageal
gectomy in recent years may either be the result cancer services to high-volume centers have seen
of individual surgeon performance, evolution in this centralization of surgical services translate
surgical technique, or improvements in perioper- into a reduction in perioperative mortality in
ative care associated with high-volume esopha- more recent years. The treatment of esophageal
geal surgical units, e.g., improved intensive care malignancy remains a highly challenging issue
units and monitoring, physiotherapy, and multi- associated with significant pre-, intra-, and post-
disciplinary team input. operative challenges, and high-volume centers
More recently, some groups have highlighted provide the appropriate multidisciplinary infra-
the volume outcome analysis as a means to pro- structure to reduce the potential impact of these
vide a system guiding evidence-based hospital challenges upon clinical outcome.
referral. The Leapfrog Group is a consortium of
public and private healthcare stakeholders in the
USA aiming to leverage purchasing power to Impact of the Application
improve and ensure healthcare quality and inform of Standardized Clinical Pathways
consumers regarding hospital performance [13]. on Outcomes Associated with
However, recent follow-up studies have demon- Esophageal Resection
strated that although a greater proportion of
esophageal resections were performed in hospi- Standardized clinical care pathways allow the
tals meeting a given evidence-based hospital introduction of a targeted goal-directed approach
referral volume metric in the 7 years in which to postoperative recovery following major sur-
Leapfrog has been collecting data, this shift had a gery. They provide a template for all medical
negligible impact on postoperative outcome [14, personnel interacting with these patients and
15]. In countries with a socialized healthcare sys- can outline an individualized goal-directed
tem, e.g., the UK, Netherlands, and Sweden, the treatment approach and recovery for each
demonstrable improvements in clinical outcome patient. A multidisciplinary approach to the for-
associated with high institutional procedural vol- mulation, implementation, and evolution of
ume have driven the push towards centralization standardized care pathways is important to
of esophageal cancer services to high-volume facilitate success.
institutions, which has translated into a reduction Clinical pathways are usually multifaceted
in observed perioperative mortality [16, 17]. and aimed at optimizing every aspect of a
Management of patients with esophageal cancer patient’s treatment including preoperative assess-
at high-volume institutions lends itself towards a ment, procedural selection, intraoperative man-
true multidisciplinary approach to the treatment agement, and postoperative care. These pathways
of these patients. Presentation and assessment of once well established can provide a framework
these patients at a multidisciplinary tumor board, for quality improvement, improving postopera-
with appropriate allocation of neoadjuvant or tive outcomes and reducing costs [18–20].
adjuvant therapies along with attention to addi- Formal clinical care pathways have been success-
tional issues including appropriate staging, nutri- fully introduced in oncological colorectal surgery
tion, and improvement in enrollment in national to provide a targeted goal-directed patient recov-
clinical trials, are important components of the ery, which has translated into a reduction in mor-
multimodality nature of management of these bidity and in length of hospital stay [21–23].
esophageal cancer patients. In the past esophageal surgeons have been
In major oncological procedures such as esoph- hesitant to apply multidisciplinary clinical path-
agectomy, greater procedural volume whether ways to enhance recovery following esophagec-
surgeon or institutional has been shown to be asso- tomy, due to the complex nature of the surgery
ciated with a reduction in perioperative mortality. and the associated rate of morbidity. However, in
3 Standardized Pathways and Volume-Outcome Relationship in Esophagectomy 27

Table 3.1 The effect of enhanced recovery on clinical outcome following esophagectomy
ECP LOS (PC) LOS (ECP) Mortality Mortality Morbidity Morbidity
Author patients (days) (days) (PC) (%) (ECP) (%) (PC) (%) (ECP) (%)
Zehr et al. 96 13.6 ± 6.9 9.5 ± 2.8 3.6 0 – –
(1991–1997) [24]
Cerfolio et al. 90 – 7 (median) – 4.4 – 26.6
(1999–2003)[20]
Low et al. 340 – 11.5 (6–49) – 0.3 – 45
(1991–2006) [25]
Jiang et al. 114 – 7 (5–28) – 2.6 – 16.7
(2006–2007) [26]
Tomaszek et al.a 110 9 (4–107) 7 (5–54) 4.6b 4.6b 42.8b 42.8b
(2004–2008) [27]
Munitiz et al. 74 13 (8–106) 9 (5–98) 5 1 38 31
(1998–2008) [19]
Preston et al. 12 17 (12–30) 7 (6–37) 0 0 75 33.3
(2011–2012) [28]
Li et al. 59 10 (9–17) 8 (7–17) 0 2 62 59
(2009–2011) [29]
Tang et al. 36 15 (IQR: 11 (IQR: 3.7 5.6 25.9 16.7
(2008–2010) [30] 12–24) 8–15.5)
Blom et al. 103 15 (12–26) 14 (11–20) 1 4 68 71
(2008–2010) [31]
PC previous care, ECP esophagectomy clinical pathway
a
Compared a conventional preexisting pathway group to an alternative pathway group
b
Results for both grouped analyzed together

more recent years, there has been expanding physiological testing. The role of the oncology
evidence to suggest that the principles of nurse coordinator has evolved and been assigned
enhanced recovery can be applied to esophagec- greater importance over the past 20 years, as they
tomy, resulting in a reduction in morbidity, length provide a point of contact for the patient during
of hospital stay, and overall costs [19, 20, 24–31] their initial consultation, staging investigations,
(Table 3.1). treatment, and recovery. The importance of the
Esophagectomy clinical pathways optimally initial interview by the nurse coordinator is rou-
are initiated at the time of patient’s initial refer- tinely highlighted in patient satisfaction surveys
ral, where an initial telephone interview with the following treatment.
patient typically done by the cancer nurse spe- All patients presenting with potentially resect-
cialist will help to initiate the process of assess- able esophageal cancer should be discussed at a
ment of the patient’s general physiological fitness multidisciplinary tumor board, and this includes
and nutritional status. Furthermore, this inter- an assessment of patient demographics including
view provides an opportunity to inform patients comorbidities, tumor characteristics, and nutri-
and family regarding the relevant steps in their tional assessment to allow appropriate allocation
clinical staging investigations and allocation to of multimodality treatment.
multimodality therapy and introduce the concept Neoadjuvant chemoradiotherapy followed by
of goal-directed recovery following surgery. The surgery has been shown to improve survival in
oncology nurse coordinator has an important role patients with esophageal cancer when compared
in making initial contact with the patient and in to surgery alone [32–34]. However, patients
coordinating the staging investigations along must be carefully selected and, in some cases,
with appointments with surgery, oncology, and optimized to be able to tolerate the entire course
substitutory dietary services, as well as arranged of treatment involved in trimodality therapy.
28 S.R. Markar and D.E. Low

Table 3.2 Evolution in patient demographics; age and medical comorbidities at Virginia Mason Medical Center,
Seattle, WA, USA (1991–2012)
1991–1996 1997–2002 2003–2007 2008–2012
Variable (Group 1) (Group 2) (Group 3) (Group 4) P value
Case no. 92 159 161 183
Patient age 64 (16–90) 64 (15–89) 66 (32–89) 66 (37–90) 0.17
M:F ratio (%) 74 (80.4) 134 (84.3) 127 (78.9) 141 (77) 0.63
BMI 26 (18–38) 25 (17–41) 26 (18–45) 27 (17–42) 0.03
Charlson (− age) 2 (0–4) 2 (0–6) 2 (0–5) 2 (0–7) 0.005
Charlson (+age) 4 (0–7) 4 (0–9) 5 (1–8) 5 (0–10) 0.02
ASA 3 (1–4) 3 (2–4) 3 (2–4) 3 (1–5) 0.07
Arrhythmia (%) 9 (9.8) 11 (6.9) 14 (8.7) 21 (11.5) 0.83
IHD (%) 12 (13.0) 34 (21.4) 19 (11.8) 31 (16.9) 0.51
Diabetes (%) 2 (2.2) 2 (1.3) 3 (1.9) 29 (15.8) 0.0004
Hypertension (%) 11 (12.0) 29 (18.2) 39 (24.2) 90 (49.2) <0.0001
Liver disease (%) 0 (0) 2 (1.3) 3 (1.9) 9 (4.9) 0.03
Renal insufficiency (%) 1 (1.1) 1 (0.6) 6 (3.7) 6 (3.3) 0.43
COPD (%) 7 (7.6) 11 (6.9) 4 (2.5) 19 (10.4) 0.60
DVT/PE (%) 0 (0) 0 (0) 0 (0) 11 (6) 0.02
PVD (%) 1 (1.1) 3 (1.9) 4 (2.5) 8 (4.4) 0.28

Nutritional assessment prior to commencing framework to optimize every aspect of the

multimodality treatment of esophageal cancer is patient’s treatment and recovery.
important to ensure patient compliance and Pretreatment patient education allows
completion of therapy. In patients with major appropriate management of patient expectations
issues with dysphagia, odynophagia ,or loss of and empowers patients and their families to work
appetite resulting in significant loss of weight, with their primary caregivers to achieve treatment
the patient should be considered for placement and recovery landmarks within a goal-directed
of removable endoscopic esophageal stent for pathway. Management of patient expectations is an
obstructive symptoms or pretreatment jejunos- important component of any clinical pathway, as
tomy to facilitate nutritional support during preoperative education must foster patient under-
neoadjuvant treatment. In the current era standing and commitment to the pathway goals.
approximately 20 % of patients at our institution This issue becomes more important as health sys-
receive a pretreatment jejunostomy to address tems move towards centralization of complex can-
potential nutritional issues and improve toler- cer services, which will inevitably result in patients
ance and completion of neoadjuvant treatment. traveling for especially complex surgical proce-
Other groups have reported the successful utili- dures. Within our own institutional esophagectomy
zation of percutaneous radiologically sited gas- series, 48 % of patients travel from more than
trostomy tubes as an alternative to jejunostomy 150 miles and 26 % from more than 400 miles. We
to address these pretreatment nutritional issues also aim to communicate decisions made at multi-
[35]. However, we prefer jejunostomies as we disciplinary tumor board to the patient and primary
believe they are safer, less likely to compromise care practitioner within 24 h of the tumor board
the gastric conduit, and can be placed in con- meeting.
junction with another procedure such as diag- In recent years there have been significant
nostic laparoscopy or port-a-cath placement. changes in the demographics of patients consid-
Routine pretreatment nutritional assessment is ered for surgical resection for esophageal cancer
one illustration of the importance of these clini- at high-volume institutions, with an increase in
cal pathways, especially when they are initiated average age, body mass index, and the incidence
at the time of referral and then form the of medical comorbidities (Table 3.2). Previous
3 Standardized Pathways and Volume-Outcome Relationship in Esophagectomy 29

Fig. 3.1 Esophagectomy clinical pathway (From Markar et al. [38] with permission)

reports of enhanced recovery protocols have Medical Center, Seattle, WA, USA, and has
specifically highlighted the challenges that undergone five revisions to date. These revisions
elderly patients undergoing esophagectomy have specifically involved all members of the
represent. Cerfolio et al. [20] demonstrated that healthcare team including from surgery, anesthe-
75 % of patients over 70 years of age failed their siology, intensive care unit staff, ward nursing,
‘fast track’ protocol. Moskovitz et al. [36] in a dietetics, and cancer nurse coordinators.
series of 31 patients undergoing esophagectomy Specific goals within the pathway that evolved
over the age of 80 years demonstrated significant during the past 20-year period (see Fig. 3.1)
poorer outcomes with a longer length of hospital include:
stay (26 (21.1–30.8) vs. 17.9 (16–19.8)) and a • Improving patient education regarding path-
greater incidence of perioperative mortality way targets
(19.4 % vs. 7.3 %) compared to those under • Adapting surgical approach according to indi-
80 years. However, we have previously pub- vidual presenting patient characteristics
lished from our own institutional series that • Developing approaches to minimizing blood
selected patients over the age of 80 years can loss and perioperative fluid administration
undergo surgical treatment for esophageal can- • Optimizing perioperative pain regimens to
cer within a standardized clinical pathway and maintain targeted postoperative hemodynam-
have a similar clinical outcome to younger ics but facilitating postoperative mobilization
patients, with no incidences of inhospital or goals to ultimately mobilize patients on the
30-day mortality in a series of 32 patients over day of surgery
80 years [37]. • Assessment and monitoring of nutrition prior
A multidisciplinary commitment to the con- to neoadjuvant therapy and esophagectomy
tinued revision of these standardized clinical • Earlier application of enteric feeding and
pathways is important to ensure continued evolu- nasogastric tube removal
tion and improvement in clinical outcomes. A • Modifying targeted discharge goals from
standardized esophagectomy clinical pathway 12–14 days in the early 1990s to 6–7 days in
was first introduced in 1991 at the Virginia Mason the current era.
30 S.R. Markar and D.E. Low

Minimally Invasive vs. Open to methodologies for laparoscopic fashioning of

Esophagectomy the gastric conduit [69, 70]. The variability in
surgical approaches and the variability in the
A minimally invasive surgical approach has been documentation of postoperative outcomes create
shown to reduce physiological stress and improve additional challenges to the comparison of out-
clinical outcome in several major surgical proce- comes with MIE, implementation of MIE within
dures including colorectal, liver, and pancreatic training programs, and the conducting of high-
resections [39–41]. In the UK, there has been a quality randomized controlled trials [71]. To
steady increase in the uptake of minimally inva- date one randomized controlled trial has been
sive esophagectomy (MIE), with 24.7 % of published comparing MIE with open esophagec-
esophageal cancer resections in 2009 being tomy [49] and demonstrated a significantly
performed using a hybrid or completely mini- reduced incidence of pulmonary infection in
mally invasive approach [42]. A robust compara- patients who underwent MIE (9 % vs. 29 %). In
tive review or meta-analysis of the literature the same cohort of patients, the authors of this
regarding the relative merits of MIE compared to trial demonstrated the acute-phase and stress
a standard open approach would be challenging responses were better preserved in the MIE
due to several inherent limitations of the publica- group, which may underline the fewer clinical
tions on this subject. The definition of MIE is manifestations of respiratory infections seen in
highly variable and includes laparoscopic abdom- the MIE group [72]. Furthermore Luketich et al.
inal phase with open thoracotomy, open abdomi- published a large series of over 1,000 minimally
nal phase and thoracoscopic approach to thoracic invasive esophagectomies with a 30-day mortal-
dissection, and totally minimally invasive esoph- ity rate of 1.7 %, median length of hospital stay
agectomy. Together with the continued variation of 8 (6–14) days, and anastomotic leak rate of
in open approaches to esophagectomy, this makes 5 % [73]. These outstanding results from this
direct objective comparison more challenging single-institution study compare favorably to
than in other surgical procedures and the wide- any published open series. This does provide
spread applicability of such comparisons some- evidence to suggest that a minimally invasive
what questionable. approach to esophagectomy can be introduced in
Pooled analyses of the available evidence high-volume centers without deterioration in
have identified potential benefits to MIE over outcomes that was seen with the introduction of
open approaches including reduced overall mor- a minimally invasive approach to other complex
bidity including respiratory morbidity and length surgical procedures including hepatectomy and
of hospital stay; however, a minimally invasive pancreatectomy. The current data suggests that
approach does not appear to influence periopera- in accomplished high-volume centers minimally
tive mortality [43–45]. It is important to note that invasive esophagectomy can be performed safely
these systematic reviews and pooled analyzes are and it may have some advantages with respect to
largely based on poor-quality evidence with sig- pulmonary morbidity, and can produce equal
nificant heterogeneity in reported results oncological outcomes as reflected by lymph
(Table 3.3) and very limited data regarding long- node yields. However, it is important to note that
term survival following MIE. Furthermore, there the average MIE results from the published lit-
is significant heterogeneity in the definition of erature do suggest a high level of variability in
complications following esophagectomy, which clinical outcomes associated with MIE
is an important limitation when attempting to (Table 3.3), and further large national or multi-
draw objective conclusions based upon the lim- institutional randomized controlled trials are
ited existing evidence [68]. required to document the risks, benefits, and
Minimally invasive esophagectomy has also long-term survival following MIE in comparison
been associated with an increase in the incidence to open esophagectomy, before a consensus on
of gastric conduit failure, which may be related this important issue is reached.
 3
Table 3.3 Comparison of minimally invasive vs. open esophagectomy
Inhospital Inhospital
Patient no. (MIE/ Patient no. mortality (MIE/ mortality Anastomotic leak Anastomotic leak Lymph node yield Lymph node yield
Author HMIE) (open) HMIE) (%) (open) (%) (MIE/HMIE) (%) (open) (%) (MIE/HMIE)a (open)a
Bailey et al. [46] 39 31 5.0 6.0 NA NA 16 ± 1.2 17 ± 1.4
Ben-David et al. [47] 100 32 1.0 5.0 4.0 12.5 14 (8–31) NA
Berger et al. [48] 65 53 7.7 7.5 14.0 11.0 20 9
Biereχ et al. [49] 53 56 3.0 2.0 12.0 7.0 20 (3–44) 21 (7–47)
Blazeby et al. [50] 124 68 1.6 3.0 NA NA 20.5–34.7 29
Briez et al. [51] 140 140 1.4 7.1 5.7 4.3 22 (8–53) 22 (6–56)
Gao et al. [52] 96 78 NA 3.8 7.3 7.7 17.8 18.0
Hamouda et al. [53] 51 24 0 0 7.8 8.0 12 (Early Gp) 24
23 (Late Gp)
Javidfar et al. [54] 92 165 3.2 4.2 5.0 4.0 17 (IQR: 11 (IQR: 7–16)
12.5–24.5)
Kunisaki et al. [55] 92 79 NA NA 8.7 NA 34.3 29
Lee et al. [56] 74 64 NA NA 13.5 28.0 14.6 (HMIE) 18.4
14.0 (MIE)
Nafteux et al. [57] 65 101 3.1 2.0 7.7 9.9 12 (HGD/1a) 17 (HGD/1a)
14 (1b) 18 (1b)
Noble et al. [58] 53 53 2.0 2.0 9 4 18 (7–52) 19 (7–50)
Parameswaran et al. [59] 67 19 4.5 5.3 13.4 (GCF) 10.5 (GCF) NA NA
Safranek et al. [60] 75 46 4.0 2.2 14.7 2.2 14 15
Scheepers et al. [61] 50 60 NA NA NA NA 14 (8–24) 10 (8–22)
Schroder et al. [62] 238 181 2.9 6.1 7.6 9.4 27 32.3
Standardized Pathways and Volume-Outcome Relationship in Esophagectomy

Schoppmann et al. [63] 31 31 0 0 3.2 25.8 17.9 ± 7.8 20.5 ± 12.6

Sihag et al. [64] 38 76 NA NA 0 2.6 19 (15–28) 21 (16–27)
Smithers et al. [65] 332 114 2.1 2.6 5.4 8.7 17 (2–59) (HMIE) 16 (1–44)
17 (9–33) (MIE)
Yamasaki et al. [66] 109 107 0 1.9 5.5 3.7 19.3 20.8
Zingg et al. [67] 56 98 3.6 6.1 20.0 12.8 5.7 6.7
Biereχ only randomized controlled trial included
HMIE hybrid minimally invasive esophagectomy, MIE total minimally invasive esophagectomy, NA not available, HGD high-grade dysplasia, GCF gastric conduit failure
a
Lymph node yield given as mean/median ± range (where available)
31
32 S.R. Markar and D.E. Low

References Variation in esophagectomy outcomes in hospitals

meeting Leapfrog volume outcome standards. Ann
Thorac Surg. 2011;91(4):1003–9.
1. Kohn GP, Galanako JA, Meyes MO, Feins RH, Farrell
16. Coupland VH, Lagergren J, Luchtenborg M, Jack RG,
TM. National trends in esophageal surgery – are out-
Allum W, Holmberg L, Hanna GB, Pearce N, Moller H.
comes as good as we believe? J Gastrointest Surg.
Hospital volume, proportion resected and mortality from
2009;13(11):1900–10.
oesophageal and gastric cancer: a population-based
2. National Oesophago-Gastric cancer audit 2013; an
study in England, 2004–2008. Gut. 2013;62(7):961–6.
audit of the care received by people with Oesophago-
17. Dikken JL, van Sandick JW, Allum WH, et al.
Gastric Cancer in England and Wales. https://
Differences in outcomes of oesophageal and gastric can-
catalogue.ic.nhs.uk/publications/clinical/oesoph-
cer surgery across Europe. Br J Surg. 2013;100:83–94.
ago-gastric/nati-clin-audi-supp-prog-oeso-gast-
18. Lovely JK, Maxson PM, Jacob AK, Cima RR,
canc-2013/clin-audi-supp-prog-oeso-gast-2013-rep.
Horlocker TT, Hebl JR, Harmesen WS, Huebner M,
pdf.
Larson DW. Case-matched series of enhanced versus
3. Ra J, Paulson EC, Kucharczuk J, et al. Postoperative
standard recovery pathway in minimally invasive
mortality after esophagectomy for cancer: develop-
colorectal surgery. Br J Surg. 2012;99:120–6.
ment of a preoperative risk prediction model. Ann
19. Munitiz V, Martinez-de-Haro LF, Ortiz A, Ruiz-de-
Surg Oncol. 2008;15:1577–84.
Angulo D, Pastor P, Parrilla P. Effectiveness of a writ-
4. Anderson O, Ni Z, Moller H, Coupland VH, Davies
ten clinical pathway for enhanced recovery after
EA, Allum WH, et al. Hospital volume and survival in
transthoracic (Ivor Lewis) oesophagectomy. Br J Surg.
oesophagectomy and gastrectomy for cancer. Eur J
2010;97:714–8.
Cancer. 2011;47:2408–14.
20. Cerfolio RJ, Bryant AS, Bass CS, Alexander JR,
5. Finks JF, Osbourne NH, Birkmeyer JD. Trends in hos-
Bartolucci AA. Fast tracking after Ivor Lewis esopha-
pital volume and operative mortality for high-risk sur-
gogastrectomy. Chest. 2004;126:1187–94.
gery. N Engl J Med. 2011;364(22):2128–37.
21. Gustafsson UO, Scott MJ, Schwenk W, Demartines N,
6. Kozower BD, Stukenborg GJ. Hospital esophageal
Roulin D, Francis N, McNaught CE, Macfie J,
cancer resection volume does not predict patient mor-
Liberman AS, Soop M, et al. Guidelines for periop-
tality risk. Ann Thorac Surg. 2012;93(5):1690–6.
erative care in elective colonic surgery: Enhanced
7. LaPar DJ, Kron IL, Jones DR, Stukenborg GJ,
Recovery After Surgery (ERAS((R))) Society recom-
Kozower BD. Hospital procedure volume should not
mendations. World J Surg. 2013;37(2):259–84.
be used as a measure of surgical quality. Ann Surg.
22. Aarts MA, Okrainec A, Glicksman A, Pearsall E,
2012;256(4):606–15.
Victor JC, McLeod RS. Adoption of enhanced recovery
8. Markar SR, Karthikesalingam A, Thrumurthy S, Low
after surgery (ERAS) strategies for colorectal surgery at
DE. Volume-outcome relationship in surgery for esoph-
academic teaching hospitals and impact on total length
ageal malignancy: systematic review and meta-analysis
of hospital stay. Surg Endosc. 2012;26(2):442–50.
2000–2011. J Gastrointest Surg. 2012;16:1055–63.
23. Rotter T, Kinsman L, James E, Machotta A, Gothe H,
9. Derogar M, Sadr-Azodi O, Johar A, Lagergren P,
Willis J, Snow P, Kugler J. Clinical pathways: effects
Lagergren J. Hospital and surgeon volume in relation to
on professional practice, patient outcomes, length of
survival after esophageal cancer surgery in a population-
stay and hospital costs. Cochrane Database Syst Rev.
based study. J Clin Oncol. 2013;31(5):551–7.
2010;(3):CD006632.
10. Wouters MW, Gooiker GA, van Sandick JW, Tollenaar
24. Zehr KJ, Dawson PB, Yang SC, Heitmiller RF.
RA. The volume-outcome relation in the surgical
Standardized clinical care pathways for major tho-
treatment of esophageal cancer: a systematic review
racic cases reduce hospital costs. Ann Thorac Surg.
and meta-analysis. Cancer. 2012;118(7):1754–63.
1998;66(3):914–9.
11. Rutegard M, Lagergren J, Rouvelas I, Lagergren P.
25. Low DE, Kunz S, Schembre D, Otero H, Malpass T,
Surgeon volume is a poor proxy for skill in esopha-
His A, Song G, Hinke R, Kozarek RA. Esophagectomy
geal cancer surgery. Ann Surg. 2009;249(2):256–61.
– it’s not just about mortality anymore: standardize
12. Yasunaga H, Matsuyama Y, Ohe K, Japan Surgical
perioperative clinical pathways improve outcomes in
Society. Effects of hospital and surgeon case-volumes
patients with esophageal cancer. J Gastrointest Surg.
on postoperative complications and length of stay after
2007;11:1395–402.
esophagectomy in Japan. Surg Today. 2009;39(7):
26. Jiang K, Cheng L, Wang JJ, Li JS, Nie J. Fast track
566–71.
clinical pathway implications in esophagogastrec-
13. The Leapfrog Group. Fact sheet. Available at www.
tomy. World J Gastroenterol. 2009;15:496–501.
leapfroggroup.org.
27. Tomaszek SC, Cassivi SD, Allen MS, Shen KR,
14. Massarweh NN, Flum DR, Symons RG, Varghese
Nichols 3rd FC, Deschamps C, Wigle DA. An alterna-
TK, Pellegrini CA. A critical evaluation of the impact
tive postoperative pathway reduces length of hospitali-
of Leapfrog’s evidence-based hospital referral. J Am
sation following oesophagectomy. Eur J Cardiothorac
Coll Surg. 2011;212(2):150–9.
Surg. 2010;37:807–13.
15. Varghese Jr TK, Wood DE, Farjah F, Oelschlager BK,
28. Preston SR, Markar SR, Baker CR, Soon Y, Singh S,
Symons RG, MacLeod KE, Flum DR, Pellegrini CA.
Low DE. Impact of a multidisciplinary standardized
3 Standardized Pathways and Volume-Outcome Relationship in Esophagectomy 33

clinical pathway on perioperative outcomes in patients esophagectomy for cancer in England: a population-
with oesophageal cancer. Br J Surg. 2013;100:105–12. based national study. Ann Surg. 2012;255:197–203.
29. Li C, Ferri LE, Mulder DS, Ncuti A, Neville A, Lee L, 43. Butler N, Collins S, Memon B, et al. Minimally inva-
Kaneva P, Watson D, Vassiliou M, Carli F, Feldman LS. sive oesophagectomy: current status and future direc-
An enhanced recovery pathway decreases duration of tion. Surg Endosc. 2011;25:2071–83.
stay after esophagectomy. Surgery. 2012;152:606–14. 44. Sgourakis G, Gockel I, Radtke A, et al. Minimally
30. Tang J, Humes DJ, Gemmil E, Welch NT, Parsons SL, invasive versus open esophagectomy: meta-analysis
Catton JA. Reduction in length of stay for patients of outcomes. Dig Dis Sci. 2010;55:3031–40.
undergoing oesophageal and gastric resections with 45. Nagpal K, Ahmed K, Vats A, et al. Is minimally invasive
implementation of enhanced recovery packages. Ann surgery beneficial in the management of esophageal can-
R Coll Surg Engl. 2013;95:323–8. cer? A meta-analysis. Surg Endosc. 2010;24:1621–9.
31. Blom RL, van Heijl M, Bemelman WA, Hollmann MW, 46. Bailey L, Khan I, Willows E, et al. Open and laparoscop-
Klinkenbijl JH, Busch OR, van Berge Henegouwen MI. ically assisted oesophagectomy: a prospective compara-
Initial experiences of an enhanced recovery protocol in tive study. Eur J Cardiothorac Surg. 2013;43:268–73.
esophageal surgery. World J Surg. 2013;37:2372–8. 47. Ben-David K, Sarosi GA, Cendan JC, et al. Decreasing
32. van Hagen P, Hulshof MC, van Lanschot JJ, et al. morbidity and mortality in 100 consecutive minimally
Preoperative chemoradiotherapy for esophageal or invasive esophagectomies. Surg Endosc. 2012;26:162–7.
junctional cancer. N Engl J Med. 2012;366:2074–84. 48. Berger AC, Bloomenthal A, Weksler B, et al.
33. Sjoquist KM, Burmeister BH, Smithers BM, Zalcberg Oncologic efficacy is not compromised, and may be
JR, Simes RJ, Barbour A, et al. Survival after neoad- improved with minimally invasive esophagectomy.
juvant chemotherapy or chemoradiotherapy for J Am Coll Surg. 2011;212:560–6.
resectable oesophageal carcinoma: an updated meta- 49. Biere SS, van Berge Henegouwen MI, Maas KW,
analysis. Lancet Oncol. 2011;12:681–92. et al. Minimally invasive versus open oesophagec-
34. Courrech Staal EFW, Aleman BMP, Boot H, van tomy for patients with oesophageal cancer: a multi-
Velthuysen MLF, van Tinteren H, van Sandick JW. centre, open-label, randomised controlled trial.
Systematic review of the benefits and risks of neoad- Lancet. 2012;379:1887–92.
juvant chemoradiation for oesophageal cancer. Br J 50. Blazeby JM, Blencowe NS, Titcomb DR, et al.
Surg. 2010;97:1482–96. Demonstration of the IDEAL recommendations for
35. Tessier W, Piessen G, Briez N, et al. Percutaneous evaluating and reporting surgical innovation in mini-
radiological gastrostomy in esophageal cancer patients: mally invasive oesophagectomy. Br J Surg. 2011;98:
a feasible and safe access for nutritional support during 544–51.
multimodal therapy. Surg Endosc. 2013;27:633–41. 51. Briez N, Piessen G, Torres F, et al. Effects of hybrid
36. Moskovitz AH, Rizk NP, Venkatraman E, et al. minimally invasive oesophagectomy on major postop-
Mortality increases for octogenarians undergoing erative pulmonary complications. Br J Surg. 2012;99:
esophagogastrectomy for esophageal cancer. Ann 1547–53.
Thorac Surg. 2006;82:2031–6. 52. Gao Y, Wang Y, Chen L, et al. Comparison of open
37. Markar SR, Low DE. Physiology not chronology, dic- three-field and minimally-invasive esophagectomy
tates outcomes after esophagectomy for esophageal for esophageal cancer. Interact Cardiovasc Thorac
cancer: outcomes in patients 80 years and older. Ann Surg. 2011;12:366–9.
Surg Oncol. 2013;20:1020–6. 53. Hamouda AH, Forshaw MJ, Tsigritis K, et al.
38. Markar SR, Schmidt H, Kunz S, Bodnar A, Hubka M, Perioperative outcomes after transition from conven-
Low DE. Evolution of standardized clinical pathways: tional to minimally invasive Ivor-Lewis esophagec-
refining multidisciplinary care and process to improve tomy in a specialized center. Surg Endosc. 2010;24:
outcomes of the surgical treatment of esophageal can- 865–9.
cer. J Gastrointest Surg. 2014;18(7):1238–46. 54. Javidfar J, Bacchetta M, Yang JA, et al. The use of a
39. Grailey K, Markar SR, Karthikesalingam A, Aboud tailored surgical technique for minimally invasive
R, Ziprin P, Faiz O. Laparoscopic versus open colorec- esophagectomy. J Thorac Cardiovasc Surg. 2012;143:
tal resection in the elderly population. Surg Endosc. 1125–9.
2013;27:19–30. 55. Kunisaki C, Kosaka T, Ono HA, et al. Significance of
40. Venkat R, Edil BH, Schulick RD, Lidor AO, Makary thoracoscopy-assisted surgery with a minithoracot-
MA, Wolfgang CL. Laparoscopic distal pancreatectomy omy and hand-assisted laparoscopic surgery for
is associated with significantly less overall morbidity esophageal cancer: the experience of a single surgeon.
compared to the open technique: a systematic review and J Gastrointest Surg. 2011;15:1939–51.
meta-analysis. Ann Surg. 2012;255:1048–59. 56. Lee JM, Cheng JW, Lin MT, et al. Is there any benefit
41. Fancellu A, Rosman AS, Sanna V, Nigri GR, Zorcolo L, to incorporating a laparoscopic procedure into mini-
Pisano M, Melis M. Meta-analysis of trials comparing mally invasive esophagectomy? The impact on peri-
minimally-invasive and open liver resections for hepato- operative results in patients with esophageal cancer.
cellular carcinomas. J Surg Res. 2011;171:e33–45. World J Surg. 2011;35:790–7.
42. Mamidanna R, Bottle A, Aylin P, et al. Short-term 57. Nafteux P, Moons J, Coosemans W, et al. Minimally
outcomes following open versus minimally invasive invasive oesophagectomy: a valuable alternative to
34 S.R. Markar and D.E. Low

open oesophagectomy for the treatment of early 65. Smithers BM, Gotley DC, Martin I, et al. Comparison
oesophageal and gastro-oesophageal junction carci- of the outcomes between open and minimally invasive
noma. Eur J Cardiothorac Surg. 2011;40:1455–63. esophagectomy. Ann Surg. 2007;245:232–40.
58. Noble F, Kelly JJ, Bailey IS, et al. A prospective 66. Yamasaki M, Miyata H, Fujiwara Y, et al. Minimally
comparison of totally minimally invasive versus open invasive esophagectomy for esophageal cancer: com-
Ivor Lewis esophagectomy. Dis Esophagus. 2013;26: parative analysis of open and hand-assisted laparoscopic
263–71. abdominal lymphadenectomy with gastric conduit
59. Parameswaran R, Titcomb DR, Blencowe NS, et al. reconstruction. J Surg Oncol. 2011;104:623–8.
Assessment and comparison of recovery after open 67. Zingg U, McQuinn A, DiValentino D, et al. Minimally
and minimally invasive esophagectomy for cancer: an invasive versus open esophagectomy for patients with
exploratory study in two centers. Ann Surg Oncol. esophageal cancer. Ann Thorac Surg. 2009;87:911–9.
2013;20:1970–7. 68. Blencowe NS, Strong S, McNair AG, Brookes ST,
60. Safranek PM, Cubitt J, Booth MI, et al. Review of Crosby T, Griffin SM, Blazeby JM. Reporting of short-
open and minimal access approaches to oesophagec- term clinical outcomes after esophagectomy: a sys-
tomy for cancer. Br J Surg. 2010;97:1845–53. tematic review. Ann Surg. 2012;255:658–66.
61. Scheepers JJ, van der Peet DL, Veenhof AA, et al. 69. Wajed SA, Veeramootoo D, Shore AC. Surgical opti-
Influence of circumferential resection margin on mization of the gastric conduit for minimally invasive
prognosis in distal esophageal and gastroesophageal oesophagectomy. Surg Endosc. 2012;26:271–6.
cancer approached through the transhiatal route. Dis 70. Nguyen NT, Nguyen XT, Reavis KM, et al. Minimally
Esophagus. 2009;22:42–8. invasive esophagectomy with and without gastric isch-
62. Schroder W, Holscher AH, Bludau M, et al. Ivor- emic conditioning. Surg Endosc. 2012;26:1637–41.
Lewis esophagectomy with and without laparoscopic 71. Hanna GB, Arya S, Markar S. Variation in the standard
conditioning of the gastric conduit. World J Surg. of minimally invasive esophagectomy for cancer –
2010;34:738–43. systematic review. Semin Thorac Cardiovasc Surg.
63. Schoppmann SF, Prager G, Langer F, et al. Fifty-five 2012;24:176–87.
minimally invasive Esophagectomies: a single centre 72. Maas KW, Biere SS, van Hoogstraten IM, et al.
experience. Anticancer Res. 2009;29:2719–25. Immunological changes after minimally invasive or
64. Sihag S, Wright CD, Wain JC, et al. Comparison of conventional esophageal resection for cancer: a ran-
perioperative outcomes following open versus mini- domized trial. World J Surg. 2014;38(1):131–7.
mally invasive Ivor Lewis oesophagectomy at a single, 73. Luketich JD, Pennathur A, Awais O, et al. Outcomes
high-volume centre. Eur J Cardiothorac Surg. 2012; after minimally invasive esophagectomy: review of
42:430–7. over 1000 patients. Ann Surg. 2012;256:95–103.
 Goals of Surgical Therapy
for Esophageal Cancer 4
A. Koen Talsma, J. Shapiro, Bas P.L. Wijnhoven,
and J. Jan B. Van Lanschot

Introduction Pretreatment Work-Up and Staging

Operative resection of esophageal cancer is Multidisciplinary Approach

probably one of the most challenging procedures
in surgery. Partly this is because it encompasses In patients with esophageal cancer, a great variety
two or even three body compartments: chest and of treatment options are available. For proper
abdomen with or without neck. Moreover, its medical decision making, accurate pretreatment
position immediately adjacent to vital structures staging is of crucial importance. Early (mucosal)
(trachea, bronchi, aorta, and heart) warrants a lesions, for example, can be cured with endo-
careful dissection. With the recent introduction of scopic mucosal resection, thus avoiding conven-
minimally invasive esophagectomy, the operation tional surgery. At the other end of the clinical
has become technically even more demanding. spectrum, accurate pretreatment staging is also
This chapter describes the surgeon’s main goals essential to avoid futile attempts at radical treat-
when performing a potentially curative esopha- ment for patients that are in fact incurable due to
gectomy for esophageal cancer, regardless of the distant metastases and to guide effective pallia-
surgical approach that is chosen. The various tion that can be achieved with endoscopic stent-
indicators that have been identified to promote ing or intraluminal brachytherapy. Discussion of
oncological control in open surgery will be dis- all patients with esophageal malignancies in a
cussed as well as the tools that help to prevent multidisciplinary tumor board is recommended
complications. because it is associated with improved outcomes
In fact, these same goals have to be set for after surgery [1, 2]. In a considerable number of
minimally invasive esophagectomy. patients, the diagnostic work-up or treatment
plan is altered after careful evaluation in a multi-
disciplinary tumor board [3]. Adenocarcinomas
arising at the esophagogastric junction can pose a
specific problem for guiding the choice between
A.K. Talsma, MD, MSc (*) • J. Shapiro, MD neoadjuvant chemo- versus chemoradiotherapy
B.P.L. Wijnhoven, MD, PhD and between subtotal esophagectomy versus
J.J.B. Van Lanschot, MD, PhD extended gastrectomy. At present, Siewert type I
Department of Surgery, Erasmus Medical Center,
15 Gravendykwal 230, Suite H-812,
and II tumors are treated as esophageal cancers
Rotterdam, 2040, 3000 CA, The Netherlands while type III tumors are generally treated as
e-mail: [email protected] gastric cancers.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 35
DOI 10.1007/978-3-319-09342-0_4, © Springer International Publishing Switzerland 2015
36 A.K. Talsma et al.

Patient Selection: Does the General centralization of surgical procedures, early

Condition of the Patient Allow cancer detection, and use of neoadjuvant ther-
for Extensive Surgery? apy, improved patient and tumor selection
based on novel staging modalities accounts for
The pretreatment assessment should not only this improvement [17, 18].
focus on tumor staging but also on optimization Guidelines for pretreatment staging of patients
of the patient’s general condition. The success of with esophageal and junctional cancer recom-
a specific treatment modality does not only mend a number of investigations, including
depend on the tumor stage but also on the fitness endoscopy with biopsy; endoscopic ultrasonogra-
of the patient. Surgery for esophageal and junc- phy (EUS); computed tomography (CT) of neck,
tional cancer has a high risk of postoperative chest, and abdomen; and external ultrasonogra-
(especially pulmonary) complications. Several phy (US) of the neck with fine-needle aspiration
risk scoring systems have been developed as pre- (FNA) of suspected lymph nodes. In addition,
dictors of poor postoperative outcome. These positron emission tomography (PET) can also
scoring systems can be used for the individual be a useful staging modality, albeit not yet man-
patient to guide treatment choice. Moreover, datory in, e.g., Dutch, UK, and US guidelines.
these scoring systems can be used to correct for In case of an advanced tumor above the carina,
case-mix differences when comparing perfor- bronchoscopy is advised to confirm or exclude
mance between hospitals. The prognostic value invasion of the tracheobronchial tree. Clinical
of the available models however is generally lim- and histopathological staging is generally based
ited. Worldwide, the most widely used and most on the tumor/node/metastasis (TNM) classifi-
simple classification is that of the American cation developed by the Union Internationale
Society of Anesthesiologists [4] but has been Contre le Cancer (UICC) and the American Joint
criticized for being subjective. The POSSUM [5] Committee on Cancer (AJCC) [19]. The most
and Charlson score [6] are more comprehensive important change in the latest (7th) edition is
but are also more cumbersome to calculate [7]. that the concept of non-regional lymph nodes has
Several series have shown that POSSUM and been abandoned and that staging of tumors in the
esophageal(O)-POSSUM [8] overestimate post- esophagus, at the esophogastric junction, and in
operative mortality in gastroesophageal cancer the stomach has been harmonized. The number
patients [9–11]. The Portsmouth(P)-POSSUM of positive lymph nodes is now more important
showed less overestimation and may be the most than their location.
useful predictor of likely postoperative mortality
in these types of patients [12]. Older age (e.g., EUS
>80 years) per se is not a contraindication for EUS is superior to any current diagnostic modal-
upper GI surgery, but older patients have ity for imaging of the primary tumor and its
increased postoperative mortality and decreased immediate surroundings (T- and N-stage) due to
long-term survival after esophageal resection for its ability to identify the component layers of the
cancer [13, 14]. Substantial weight loss before esophageal wall [20, 21]. The main problem with
surgery was also a negative prognostic factor in EUS is failure to pass in 1 out of 5 patients [22].
several studies [15, 16]. FNA of suspected nodes is only indicated when
the results will change the treatment plan (e.g.,
radiation field). EUS can identify metastatic
Tumor Selection: Can the Tumor Be lymph nodes at the celiac trunk but is not accu-
Radically Resected and Potentially rate in detecting distant metastases, with the
Cured? exception of hematogenous metastases in the left
liver lobe and left adrenal gland. FNA of the
Over the past decades, long-term survival celiac nodes is technically feasible in 95 % of
results have substantially improved. Besides patients [23].
4 Goals of Surgical Therapy for Esophageal Cancer 37

CT and External US cancer and malignant melanoma. In a British

Spiral CT and external US are used for the study, 96 % of SN biopsies accurately detected
detection of distant hematogenous and lymphatic lymph node metastatic disease [29]. In another
metastases (M-stage). Probably, PET scanning study, however, so-called skip lesions were identi-
can replace US of the neck, although it is gener- fied in 55 % of resected two-field lymphadenec-
ally recommended to confirm suspected lymph tomy specimens [30–32]. Currently, a multicenter
nodes by US-FNA to exclude false positivity of trial in Japan is being performed, in which the
the PET scan (e.g., due to sarcoidosis) [24]. The extent of lymph node dissection during gastric sur-
ability to accurately predict locoregional resect- gery is tailored depending on the SN biopsy [33].
ability is especially important before embarking
upon a thoracoscopic or laparoscopic surgical Restaging
approach to minimize the risk of accidental dam- After completion of neoadjuvant therapy, patients
age. For this purpose, CT continues to play an can be restaged to evaluate response to treatment
important role. Invasion into adjacent organs is and to detect any progression of disease before
unlikely when a periesophageal fat plane can be proceeding to surgery. The assessment of nodal
recognized, but when absent, it cannot be taken disease following chemoradiotherapy by EUS
as absolute evidence of invasion. This accounts and CT is disappointing because viable tumor
for the overestimation of tumor invasion into tra- cannot be readily distinguished from fibrotic tis-
chea, aorta, and pericardium. sue [32, 34]. Studies with PET especially when
measuring SUV before and after chemotherapy
PET have been encouraging [35, 36]. Unfortunately,
PET is a noninvasive imaging technique which is tumor response assessment by PET after neoad-
increasingly used in the staging of various tumor juvant chemoradiotherapy is hampered by
types, including esophageal cancer [25, 26]. The radiation-induced inflammation.
increased glucose metabolism of malignant cells
is the driving force for the uptake of fluorine-18- Future Developments
fluorodeoxyglucose (FDG), which is the most Recently, more research has focused on staging
common radiotracer used for oncological PET techniques that address the biological behavior of
studies. In addition to qualitative staging (esp. tumors which is important in the response to
detection of distant metastases), PET is able to chemoradiotherapy and likelihood of recurrence.
quantify FDG uptake in malignant tissue by cal- This can be achieved by PET scanning with novel
culating the standardized uptake value (SUV) of radiotracers such as (18)F FLT 3-deoxy-3-
the primary tumor. After extensive “conven- fluorothymidine or (11)C-choline [37, 38]. Other
tional” diagnostic work-up, additional PET scan- studies focus on MRI as a potential noninvasive
ning yields a diagnosis of distant dissemination technique for locoregional staging of esophageal
in an additional 10 % of patients, especially in cancer [39]. Encouraging results have been
case of T3 tumors [27]. The simultaneous, com- achieved in the rapidly improving technology of
bined PET and CT scan is able to localize and in vivo intraoperative imaging as well [40].
classify hotspots more accurately than PEt alone.

Intraoperative Staging by Laparoscopy Definitive Chemoradiotherapy:

and Sentinel Node Biopsy An Alternative for Potentially
Although inconsistently applied, a systematic review Curative Resection?
has recommended the use of staging laparoscopy in
junctional cancer patients [28], especially for In recent years two randomized controlled tri-
demonstrating low-volume peritoneal disease. als compared definitive chemoradiotherapy
The value of sentinel node (SN) sampling in (dCRT) to neoadjuvant chemoradiotherapy plus
esophageal cancer is less clear than for, e.g., breast surgery (nCRT + S). Both studies employed a
38 A.K. Talsma et al.

non-inferiority design to test the chance that primary tumor and affected lymph nodes. As
patients in both treatment paradigms have a sig- esophageal cancer easily spreads longitudinally
nificantly different survival. via the submucosal lymphatics, the incidence of
The first study by Stahl et al. [41] included 172 intramucosal and submucosal metastases is
patients between 1994 and 2002 from 11 German reportedly high (Fig. 4.1a, b). The completeness
centers. It compared dCRT (without salvage of resection of the primary tumor and its intramu-
surgery) with nCRT + S for “locally advanced” ral metastases can be described with respect to
(i.e., T3–4, N0-1, M0) esophageal squamous cell the proximal, distal, and circumferential resec-
carcinomas. Two-year survival was 35.4 and tion margin and is a well-known determinant of
39.9 % in the dCRT arm and nCRT + S arm, long-term survival in several studies [43–46].
respectively (P = 0.007). Freedom from local pro- Previous studies have investigated the required
gression was worse in the dCRT arm (40.7 % vs. length of macroscopic proximal and distal resec-
64.3 %, respectively; HR 2.1 P = .003). A signifi- tion margins in order to minimize anastomotic
cant difference was found in treatment-related recurrence. A reasonable margin is 10 cm for
mortality: 3.5 % in the dCRT arm and 12.8 % in larger tumors and 4 cm for more localized tumors
the nCRT + S arm (χ2, P = .03). In summary, there [47]. When only a short proximal resection mar-
was no difference in overall survival; however, gin can be obtained through the thoracic expo-
local failure was more common, and treatment- sure (especially for a squamous cell carcinoma),
related death was less common in the dCRT arm. a cervical extension with subtotal esophagectomy
The second randomized controlled trial is advisable. An adenocarcinoma of the lower
(FFCD 9102) [42] compared dCRT to nCRT + S esophagus requires an extensive sleeve resection
in patients who had an objective clinical response of the lesser curve and fundus to minimize posi-
or an improvement of dysphagia after neoadju- tive distal resection margins.
vant chemoradiotherapy (259/444, 58.3 %). Two- An esophageal resection can be suboptimal
year survival rates for the dCRT arm and due because of an involved circumferential mar-
nCRT + S arm were 39.8 and 33.6 %, respectively gin. The definition of circumferential resection
(P = 0.03, i.e., the chance that the actual differ- margin (CRM) involvement remains controver-
ence is >10 %). Three-month mortality (0.8 % vs. sial. The College of American Pathologists
9.3 %, P = 0.003) favored the dCRT arm, whereas (CAP) and the Royal College of Pathologists
locoregional relapse (43.0 % vs. 33.6 %, HR (RCP) use different definitions for CRM involve-
1.63, P = 0.03) favored the nCRT + S arm. ment. Microscopic tumor involvement (R1 resec-
Both studies suffered from major drawbacks tion) is defined by CAP as tumor found at the cut
(e.g., inadequate power and lack of standardized circumferential resection margin, while it is
chemoradiotherapy protocols), thus precluding defined by RCP as any tumor within 1 mm of the
more general conclusions from these data. This circumferential resection plane. Recently, a sys-
ambiguity towards dCRT is reflected in clinical tematic review was published of 14 studies
practice where in most countries dCRT is involving 2,433 patients. Rates of CRM involve-
reserved only for those patients who are deemed ment were 15.3 and 36.5 % according to the CAP
unfit for surgery. and RCP criteria, respectively. It was shown that
CRM involvement is an important predictor of
poor prognosis and that the CAP criteria had a
Surgical Performance Indicators: On greater (negative) prognostic power than the RCP
Which Parameters Should MIE Be criteria [48]. It can be difficult and time-
Judged? consuming to identify a positive circumferential
resection margin in a large T3 tumor, and it has
Resection Margins been suggested that this should preferably be
done in accordance with the CAP criteria (tumor
The main goal in the surgical treatment for is found at the inked lateral margin of resection)
esophageal cancer is the complete removal of the [49]. There has been a significant decrease in
4 Goals of Surgical Therapy for Esophageal Cancer 39

Fig. 4.1 (a) The a Lymphatics Anatomic layers

lymphatics of the Epithelium
esophagus are distributed Basement membrane
in the form of a Lamina propria
submucosal and a Muscularis mucosa
paraesophageal plexus Submucosa
Submucosal
that can both drain gland
directly into the Lamina propria Muscularis
periesophageal lymph propria
nodes. (b) Longitudinal Submucosal Adventitia
spread of tumor to
involve submucosal Muscularis
propria
lymphatic plexus (From
Elsevier; Raja et al. [116]
with permission)

Regional

Thoracic duct

b Submucosa

Inner

Middle

Deep

CRM involvement especially with the in combination with negative regional lymph
introduction of neoadjuvant chemoradiotherapy nodes, are encountered relatively frequently
[17, 50]. After neoadjuvant chemotherapy CRM [52]. Lymphatic dissemination occurs not only
involvement still has prognostic importance [51]. in a chaotic pattern but also at an early stage.
Some 30 % of the T1b tumors (with infiltration
limited to the submucosa) already have positive
Lymphadenectomy lymph nodes involved [53]. Ideally, a complete
resection of all locoregional nodes draining
As esophageal cancer readily spreads longi- the esophagus should include the two or three
tudinally in the submucosal lymphatics, early fields (see above) in addition to the easily acces-
dissemination to lymph nodes in the chest sible periesophageal and perigastric lymph
and abdomen may be involved in cancer of all nodes (Fig. 4.2). In a survey among surgeons
parts of the esophagus. And even skip metas- around the world, the technically challenging
tases, defined as positive distant lymph nodes three-field lymphadenectomy was performed
40 A.K. Talsma et al.

Fig. 4.2 Extent of

resection and fields of
lymph node dissection
routinely carried out for
cancer of the esophagus
(From Griffin and Raimes Cervical field
[117] with permission) (three field)

Thoracic field
(two field)

Abdominal field
(one field)

routinely by only 12 % of the responders [54]. esophagectomy have similar lymph node
A SEER analysis showed that the median num- retrieval compared to open techniques [56–58].
ber of total lymph nodes resected in over 5,600 For staging purposes it is clear that an extended
esophagectomies was only eight nodes [55]. lymphadenectomy is superior to a limited dissec-
Lymphadenectomy can be performed safely tion. It has, therefore, been suggested by the
during minimally invasive surgery, and it has 7th edition of the TNM staging system that for
been shown that minimally invasive and robotic staging purposes, the total number of resected
4 Goals of Surgical Therapy for Esophageal Cancer 41

and identified lymph nodes should be at least breast cancer, the sentinel node concept has not
15 nodes. The therapeutic impact of an extended become popular in esophageal surgery [29, 31].
lymphadenectomy is still a matter of debate in Several studies have confirmed the higher mor-
esophageal cancer surgery [59]. Some authors bidity after thoracotomy than after transhiatal
state that surgery has reached its limit, while oth- approach: more pulmonary complications, more
ers believe that the course of the disease can be recurrent nerve injuries, and higher early mortal-
influenced positively by aggressive surgery with ity [67–69].
an extended lymphadenectomy. One of the Meta-analysis of the available literature data
hypotheses supporting the benefits of extended did not show differences in survival between tran-
lymphadenectomy is the clearance of microme- shiatal and transthoracic operations. Other studies
tastases that can be present in up to 50 % of compared fields of dissection, for example, the
histology-negative nodes. This hypothesis is sup- single-center studies by Lerut et al. [70] and
ported by the correlation of micrometastases in Altorki et al. [71] that suggested a potential sur-
routine lymph node-negative patients with a poor vival benefit for three-field lymphadenectomy.
outcome [60, 61]. Finally, there are studies that investigated the
More skeptical authors believe that the thera- absolute number of nodes dissected. This has
peutic impact of an increased lymph node harvest led to different recommendations regarding the
per se is limited and it is probably not the type of optimal extent of lymphadenectomy ranging
operation performed that makes a difference but from 16 to 30 nodes. In a population of 4,627
rather the stage of the disease at the time of oper- patients in the Worldwide Esophageal Cancer
ation [56]. According to this view, lymph node Collaboration (WECC), extent of lymphadenec-
metastases are markers of systemic disease and tomy was not associated with increased survival
removal of the primary lesion alone will yield the for patients with extremes of esophageal cancer
same survival [62]. The spurious effect of (TisN0M0 and 7 or more nodes positive) and
extended lymphadenectomy might then be those with well-differentiated pN0 cancer [72].
caused by stage migration which occurs if posi- For all other cancers, 5-year survival improved
tive nodes in the extended field change N-stage. with increasing extent of lymphadenectomy.
This results in the so-called Will Rogers phenom- Based on these WECC data, a stage-dependent
enon or stage purification and leads to unreliable extent of lymphadenectomy was recommended.
stage-by-stage comparisons of survival. For that This is comparable to the findings of the HIVEX
reason some authors prefer to use the lymph node trial that showed a better survival after a transtho-
ratio (i.e., the number of positive nodes over the racic approach in the subgroup of patients with
number of removed nodes) rather than the abso- 1–8 nodes positive [66]. Rizk et al. identified 18
lute number of positive nodes [63, 64]. nodes resected as the minimum necessary for
Several prospective trials have been performed accurate staging and for eliminating an effect of
comparing survival after esophagectomy with or lymphadenectomy on survival [73]. In the study
without extended lymphadenectomy. In the larg- by Altorki et al., effect of lymphadenectomy on
est RCT (HIVEX trial), comparing limited tran- survival was lost after 25 nodes for early stage and
shiatal esophagectomy and extended transthoracic after 16 nodes in stage III and IV cancers [71].
esophagectomy with two-field lymphadenectomy, Peyre et al. investigated an international database
5-year survival was not significantly different [65, of 2,303 esophagectomies in which survival was
66]. The survival benefit of an extended lymphad- maximized with 23 nodes resected [74].
enectomy by a transthoracic approach was limited Nowadays, multimodality treatment of
to a subgroup of patients with low burden of nodal esophageal cancer has been widely accepted. As
disease (1–8 nodes positive on pathological neoadjuvant chemoradiotherapy (CRT) is known
examination of the resection specimen). The iden- to “sterilize” nodes, it is unclear whether the
tification of this group makes the pretreatment recommendations for number of lymph nodes
staging very challenging. Unfortunately, unlike in from the surgery-alone era still stand. Extended
42 A.K. Talsma et al.

lymphadenectomy seems to be beneficial, magnification offered by thoracoscopy might

particularly in patients who are not downstaged decrease complications, but lack of tactile control
regarding pathological tumor depth (ypT) and is probably a contributory factor to the increase
those with persistent nodal metastases (ypN+) of intraoperative injuries. It is unlikely that mini-
[75, 76]. The effect of lymphadenectomy is mally invasive methods will reduce mortality
influenced by tumor response after CRT, and the rates since in experienced centers death after
survival benefit is stronger in patients without a open esophagectomy is already a rare event.
complete pathological response (non-pCR) com- Minimally invasive esophagectomy (MIE) might
pared to those with pCR [77]. be proven superior for other endpoints such as
blood loss, duration of ICU or hospital stay, need
for analgesics, and pulmonary function. The best
Morbidity: Prevention available evidence comes from a recently pub-
of Complications lished RCT (TIME trial) showing that MIE is
accompanied by less pulmonary complications
The typical esophageal cancer patient suffers [87]. This trial has been criticized because of the
from several comorbidities including obesity lack of a clear definition of “pulmonary compli-
(especially in adenocarcinoma) and cardiopul- cations” as the primary endpoint [88]. Moreover,
monary diseases (in both squamous and adeno- an unexplained increase of recurrent nerve inju-
carcinoma) that put the patient at increased risk ries was present in the open group.
for postoperative complications. Serious intraop-
erative and postoperative complications can Respiratory Complications
occur with minimally invasive as well as open Respiratory failure is a major problem after
techniques, also depending on the need of a tho- esophagectomy. Several studies have reported
racic phase of the operation. Overall, complica- that about half of the inhospital deaths after
tion rates are reported in over 50 % of esophagectomy is due to pneumonia, which is the
esophagectomy series, with incidence varying most frequent general complication after surgery
between 17 and 74 % [78, 79]. Postoperative [89]. Preventive measures include preoperative
complications have been directly linked to a vari- respiratory training, cessation of smoking, and
ety of other outcome parameters including mor- continuous postoperative pain control by epi-
tality, readmission rate, early cancer recurrence, dural analgesia in order to avoid restrictive respi-
survival, length of hospital stay, costs and ration and insufficient coughing. Micro-aspiration
resource utilization, and quality of life [80–83]. as a consequence of impaired swallowing coordi-
The most important issues in the management of nation because of a cervical anastomosis also
perioperative complications are prevention and plays a role in the pathophysiology of broncho-
early detection. However, a clear understanding pneumonias. Another reason for postoperative
of the relationships between complications, their respiratory impairment is a large pleural effusion,
recognition, management, and how they influ- which should be drained if provoking extended
ence subsequent mortality is hampered by the atelectasis. Avoiding the need for a combined
lack of standardized definitions [84, 85]. Finally, thoracotomy and laparotomy may potentially
early detection and proper management of post- reduce postoperative pain, ventilator dependence,
operative complications is of crucial importance. and cardiopulmonary complications [90]. In a
It has been shown repeatedly that the so-called study comparing thoracoscopic resection with a
failure to rescue largely explains the difference in historical cohort, the overall incidence of pulmo-
mortality rates between low-volume and high- nary complications was reduced from 33 to 20 %
volume hospitals for complicated surgery includ- [91]. Probably cardiopulmonary complications
ing esophagectomy [86]. do not depend on the incision size only. The ben-
The exact role for minimally invasive tech- efit of smaller port sites that are needed during
niques is still not fully clear. The increased minimally invasive surgery may be offset by the
4 Goals of Surgical Therapy for Esophageal Cancer 43

lengthened time of operation and single-lung Chylothorax

ventilation. The use of a prone position also plays The incidence of accidental thoracic duct leakage
a role but will be discussed elsewhere. can be diminished by intraoperative identifica-
tion and ligation of the duct. Reported incidence
Recurrent Laryngeal Nerve Injury of chylothorax varies between 3 and 10 % and is
More recurrent laryngeal nerve injuries when seen more often in patients who undergo trans-
using thoracoscopy have been reported, which thoracic esophagectomy and in patients who
might be attributed to the use of diathermia. have more positive nodes. Patients with chyle
Others claim that the use of minimally invasive leakage have more pulmonary complications.
techniques has lowered the incidence of hoarse- Conservative therapy (initial parenteral feeding
ness because of the magnified view [87]. and subsequent enteral diet with medium-chain
triglycerides (MCT)) is often successful, but
Anastomotic Leakage operative therapy should be seriously considered
Lack of standardization of definitions is a prob- in patients with a persistently high daily output of
lem when reporting on complications. In a recent more than 2 L after 2 days of optimal conserva-
meta-analysis, anastomotic leakage was reported tive therapy [96].
in most of the publications, but it was defined
in only a minority with 22 differing definitions Cardiac Arrhythmias
[84]. Early disruption of the esophagogastric Cardiac arrhythmias are not uncommon in the
anastomosis is the result of a technical problem postoperative phase. Atrial fibrillation (AF) is
and immediate reexploration is frequently indi- seen in 15–20 % of patients and requires further
cated for correction. Many different suturing investigation because it can be an early manifes-
and (semi-)mechanical techniques have been tation of, e.g., mediastinitis due to intrathoracic
described. The semimechanical side-by-side anastomotic leakage. AF can also be associ-
technique claims a lower leakage rate compared ated with hypervolemia, preexistent pulmonary
to a hand-sewn anastomosis, but has not been or cardiac disease, and dilation of the gastric
tested in a randomized trial [92, 93]. Leakage is conduit.
more frequent in the neck than in the chest, but
the associated mortality might be lower, espe-
cially after a transhiatal approach [94]. If a trans- Mortality and Quality Control
mural necrosis of the gastric conduit is suspected,
this can be diagnosed by endoscopy and when Definitions
present is also an indication for surgery with for- There is an increasing interest in comparing
mation of a cervical esophagostomy, resection institutional performance. For surgical proce-
of the gastric tube, and placement of a feeding dures postoperative mortality rate is generally
jejunostomy. After rehabilitation of the patient, a used, because it is a relatively objective mea-
colonic interposition can be performed at a sec- sure and reflects the summation of the most
ondary stage. Late disruptions become manifest severe postoperative complications. Currently
generally between postoperative day 5 and 10 and it is unclear which definition of postoperative
are most frequently due to ischemia. They can mortality best reflects surgical quality of care.
be managed nonoperatively in most cases with The 30-day operative mortality (30DM) and
aggressive drainage using radiologically guided the inhospital mortality (IHM) after esophageal
drains or endoluminal vacuum therapy [95]. Self- resection are well documented and vary from
expandable stents can be inserted in these situa- 4 % for specialized centers to > 10 % for nation-
tions but can have the disadvantage of migration wide registries [97]. Few studies report on mor-
or further necrosis due to tissue compression tality beyond 30 days. Damhuis et al. however
ultimately leading to, e.g., neoesophago-tracheal showed in the Dutch Cancer Registry that 43 %
fistula formation. of inhospital deaths after surgery for esophageal
44 A.K. Talsma et al.

cancer occurred 30 days or more after the opera- Table 4.1 Several studies over previous decades showing
improved long-term survival after esophageal resection
tion [98]. Therefore, 90-day mortality (90DM)
might be preferred as a performance indicator. Study Randomization Survival
Using a longer time period after the operation for Muller et al., N/A 5-year
1990 [106] survival
defining postoperative mortality may thus pro-
10 %
vide a better definition of quality of surgery [99].
Walsh et al., Multimodality therapy 3-year
Extending the mortality period beyond 30 days 1996 [107] versus surgery survival
and beyond inhospital stay has the advantage 32 %
that patients who die because of surgery-related Hulscher et al. Transthoracic versus 5-year
complications outside the hospital are included 2002, Omloo transhiatal approach survival
et al. 2007 36 %
as well. [65, 66]
Not only short-term outcomes but also long- Van Hagen, 2013 Multimodality therapy 5-year
term survival should be part of the benchmark as [17] versus surgery survival
both aspects are relevant for comparing surgical 47 %
performance. Both surgery-related deaths and
cancer recurrence-related deaths are reflections
of surgical quality of care. Less radical surgical
resections will generally result in lower postop- perioperative clinical pathways [18, 108]. In many
erative morbidity and mortality but will generally countries around the world, it has been decided
give less favorable oncological outcomes. that high-risk surgical procedures such as esopha-
gectomy should be restricted to facilities with a
Case Mix Correction yearly minimum volume [109, 110]. It has been
Even after agreement on a uniform definition of demonstrated that the incidence of postoperative
postoperative mortality, direct comparison of complications is similar across hospitals but that
crude mortality rates between hospitals can be the associated mortality rates are lowest in high-
misleading as they do not take into account the volume centers, which generally show a lower
case-mix difference, i.e., the differences in physi- “failure to rescue” [86, 111]. Centralization is cur-
ological condition and tumor stages of patients. rently implemented widely. Also auditing has
Sophisticated models have been developed for been implemented as a way of improvement of
prediction of 30DM and IHM [8, 14, 67, 100– care. Of course this results in an additional regis-
104] after esophageal surgery, but models for tration burden for the surgeon, but comparing indi-
90DM have been mostly based on large multi- vidual or institutional results with the benchmark
institutional databases with only few parameters has proven valuable in other types of cancer sur-
available [105]. gery, such as for rectal cancer [112, 113]. For
esophageal cancer, variables of interest are, for
Outcome-Volume Relationship example, hospital mortality, radicality (R-status),
and Registration extent of lymph node dissection, length of hospital
Over the past decades, better long-term survival stay, application of neoadjuvant therapy, availabil-
results have been presented, evolving from 18 % ity of PET-CT, and the presence of a well-struc-
5-year survival in the era from 1980 to 1990 to tured MDT. The quality indicators can be divided
48 % in the most recently published RCT in structural, process, and outcome measures,
(Table 4.1) [17, 65, 99, 106, 107]. It is suggested respectively (Table 4.2) [114]. Heterogeneity and
that many factors are responsible for this positive lack of standardized definitions of the outcome of
effect, including large hospital volume, early interest are a problem here as well. In a review of
tumor detection, improved patient selection based esophagectomy outcomes from 164 NSQIP
on novel staging modalities, increased use of neo- (National Surgical Quality Improvement Project)
adjuvant therapy, better surgical and anesthesio- hospitals, it was demonstrated that even following
logical techniques, and improved standardized case mix adjustment, results between centers
4 Goals of Surgical Therapy for Esophageal Cancer 45

Table 4.2 Performance indicators that have been diagnosis of distant dissemination in an
identified in esophageal cancer surgery
additional 10 % of patients, especially in case
Quality-of-care indicators of T3 tumors.
Structural measures • The goals that have been achieved in open
Hospital volume esophageal surgery should also act as targets
Surgeon volume for minimally invasive esophagectomy, being
Centralization a lymph node retrieval of at least 15 nodes, R0
Process measures
resection (>1 mm margin), and operative
Discussion in multidisciplinary board
mortality < 5 %.
Age
• Neoadjuvant chemoradiotherapy decreases
Preoperative quality of life
the incidence of a tumor-positive circumferen-
Staging (FDG-PET vs. FDG-PET)
Lymphadenectomy
tial margin.
Neoadjuvant chemoradiation • Meta-analysis of the available literature data
Surgical approach did not show differences in survival between
Outcome measures transhiatal and transthoracic operations. The
Postoperative complications survival benefit of an extended lymphadenec-
Radicality of resection tomy by a transthoracic approach seems to be
Number of resected lymph nodes limited to a subgroup of patients with low bur-
From Courrech Staal et al. [114] with permission den of nodal disease.
• Overall, complication rates are reported in
over 50 % of esophagectomy series, with inci-
varied by 161 % for 30-day mortality and 84 % for dences varying between 17 and 74 %.
serious morbidity [67]. Postoperative complications have been
Finally, comparing the quality of infrequent directly linked to a variety of other outcome
operations such as esophagectomies is difficult, parameters including mortality, readmission
besides issues of definition and case-mix correc- rate, early cancer recurrence, survival, length
tion, because of another complex element in of hospital stay, resource utilization, and qual-
comparing surgical performance, i.e., the prob- ity of life.
lem of sample size [115]. • It has been suggested that MIE is accompa-
nied by less pulmonary complications.
• The 30-day operative mortality (30DM)
Conclusion/Take Home Messages and the inhospital mortality (IHM) after
esophageal resection vary from 4 % for spe-
• Discussion of all patients with esophageal cialized centers to > 10 % for nationwide
malignancies in a multidisciplinary tumor registries.
board is recommended and is associated with • Many factors are responsible for the better
improved outcomes after surgery. long-term survival rates that have been
• ASA, (O)-POSSUM, and Charlson are the achieved over the previous decades, including
preoperative risk scoring systems that are large hospital volume, early tumor detection,
often used in esophageal surgery. improved patient selection based on novel
• The most important change in the most recent staging modalities, increased use of neoadju-
7th edition of the TNM staging system is that vant therapy, better surgical and anesthesio-
the concept of non-regional lymph nodes has logical techniques, and improved standardized
been abandoned and that staging of esopha- perioperative clinical pathways.
geal cancer has been harmonized with gastric • The lack of standardized definitions of com-
cancer. plications and mortality has hampered out-
• After extensive “conventional” diagnostic come assessment after open and minimally
work-up, additional PET scanning yields a invasive esophagectomy.
46 A.K. Talsma et al.

References 16. Masoomi H, et al. Predictive factors of acute

respiratory failure in esophagectomy for esophageal
malignancy. Am Surg. 2012;78(10):1024–8.
1. Stephens MR, et al. Multidisciplinary team manage-
17. van Hagen P, et al. Preoperative chemoradiotherapy
ment is associated with improved outcomes after
for esophageal or junctional cancer. N Engl J Med.
surgery for esophageal cancer. Dis Esophagus. 2006;
2012;366(22):2074–84.
19(3):164–71.
18. Stein HJ, Siewert JR. Improved prognosis of resected
2. Davies AR, et al. The multidisciplinary team meet-
esophageal cancer. World J Surg. 2004;28(6):520–5.
ing improves staging accuracy and treatment selec-
19. Sobin LH, Compton CC. TNM seventh edition:
tion for gastro-esophageal cancer. Dis Esophagus.
what’s new, what’s changed: communication from the
2006;19(6):496–503.
International Union Against Cancer and the American
3. van Hagen P, et al. Impact of a multidisciplinary
Joint Committee on Cancer. Cancer. 2010;116(22):
tumour board meeting for upper-GI malignancies on
5336–9.
clinical decision making: a prospective cohort study.
20. Kelly S, et al. A systematic review of the staging per-
Int J Clin Oncol. 2013;18(2):214–9.
formance of endoscopic ultrasound in gastro-
4. Keats AS. The ASA classification of physical sta-
oesophageal carcinoma. Gut. 2001;49(4):534–9.
tus–a recapitulation. Anesthesiology. 1978;49(4):
21. van Vliet EP, et al. Staging investigations for oesoph-
233–6.
ageal cancer: a meta-analysis. Br J Cancer. 2008;
5. Prytherch DR, et al. POSSUM and Portsmouth
98(3):547–57.
POSSUM for predicting mortality. Physiological
22. Vickers J, Alderson D. Influence of luminal obstruc-
and Operative Severity Score for the enUmeration of
tion on oesophageal cancer staging using endoscopic
Mortality and morbidity. Br J Surg. 1998;85(9):
ultrasonography. Br J Surg. 1998;85(7):999–1001.
1217–20.
23. Reed CE, et al. Esophageal cancer staging: improved
6. Hall WH, et al. An electronic application for rapidly
accuracy by endoscopic ultrasound of celiac lymph
calculating Charlson comorbidity score. BMC
nodes. Ann Thorac Surg. 1999;67(2):319–21; discus-
Cancer. 2004;4:94.
sion 322.
7. Chandra A, Mangam S, Marzouk D. A review of risk
24. Omloo JM, et al. Additional value of external
scoring systems utilised in patients undergoing gas-
ultrasonography of the neck after CT and PET
trointestinal surgery. J Gastrointest Surg. 2009;13(8):
scanning in the preoperative assessment of patients
1529–38.
with esophageal cancer. Dig Surg. 2009;26(1):
8. Tekkis PP, et al. Risk-adjusted prediction of opera-
43–9.
tive mortality in oesophagogastric surgery with
25. van Westreenen HL, et al. Systematic review of the
O-POSSUM. Br J Surg. 2004;91(3):288–95.
staging performance of 18F-fluorodeoxyglucose
9. Lagarde SM, et al. Evaluation of O-POSSUM in pre-
positron emission tomography in esophageal cancer.
dicting in-hospital mortality after resection for
J Clin Oncol. 2004;22(18):3805–12.
oesophageal cancer. Br J Surg. 2007;94(12):1521–6.
26. Blom RL, et al. PET/CT-based metabolic tumour
10. Lai F, et al. Evaluation of various POSSUM models
volume for response prediction of neoadjuvant
for predicting mortality in patients undergoing elec-
chemoradiotherapy in oesophageal carcinoma. Eur J
tive oesophagectomy for carcinoma. Br J Surg. 2007;
Nucl Med Mol Imaging. 2013;40(10):1500–6.
94(9):1172–8.
27. van Westreenen HL, et al. Limited additional value
11. Bosch DJ, et al. Comparison of different risk-adjustment
of positron emission tomography in staging oesoph-
models in assessing short-term surgical outcome after
ageal cancer. Br J Surg. 2007;94(12):1515–20.
transthoracic esophagectomy in patients with esopha-
28. Gouma DJ, et al. Laparoscopic ultrasonography for
geal cancer. Am J Surg. 2011;202(3):303–9.
staging of gastrointestinal malignancy. Scand J
12. Dutta S, Horgan PG, McMillan DC. POSSUM and
Gastroenterol Suppl. 1996;218:43–9.
its related models as predictors of postoperative
29. Lamb PJ, et al. Sentinel node biopsy to evaluate the
mortality and morbidity in patients undergoing sur-
metastatic dissemination of oesophageal adenocarci-
gery for gastro-oesophageal cancer: a systematic
noma. Br J Surg. 2005;92(1):60–7.
review. World J Surg. 2010;34(9):2076–82.
30. Schroder W, et al. Localization of isolated lymph node
13. Cijs TM, et al. Outcome of esophagectomy for can-
metastases in esophageal cancer–does it influence the
cer in elderly patients. Ann Thorac Surg. 2010;90(3):
sentinel node concept? Hepatogastroenterology. 2007;
900–7.
54(76):1116–20.
14. Koppert LB, et al. Impact of age and co-morbidity
31. Grotenhuis BA, et al. The sentinel node concept in
on surgical resection rate and survival in patients
adenocarcinomas of the distal esophagus and gastro-
with oesophageal and gastric cancer. Br J Surg.
esophageal junction. J Thorac Cardiovasc Surg.
2012;99(12):1693–700.
2009;138(3):608–12.
15. Polee MB, et al. Prognostic factors for survival in
32. Kalha I, et al. The accuracy of endoscopic ultrasound
patients with advanced oesophageal cancer treated
for restaging esophageal carcinoma after chemoradi-
with cisplatin-based combination chemotherapy.
ation therapy. Cancer. 2004;101(5):940–7.
Br J Cancer. 2003;89(11):2045–50.
4 Goals of Surgical Therapy for Esophageal Cancer 47

33. Kitagawa Y, et al. Sentinel node mapping for gastric 47. Skinner DB. En bloc resection for neoplasms of the
cancer: a prospective multicenter trial in Japan. esophagus and cardia. J Thorac Cardiovasc Surg.
J Clin Oncol. 2013;31(29):3704–10. 1983;85(1):59–71.
34. Ribeiro A, et al. Endoscopic ultrasound restaging 48. Chan DS, et al. Systematic review and meta-analysis
after neoadjuvant chemotherapy in esophageal can- of the influence of circumferential resection margin
cer. Am J Gastroenterol. 2006;101(6):1216–21. involvement on survival in patients with operable
35. Westerterp M, et al. Esophageal cancer: CT, endo- oesophageal cancer. Br J Surg. 2013;100(4):456–64.
scopic US, and FDG PET for assessment of response 49. Verhage RJ, et al. How to define a positive circumfer-
to neoadjuvant therapy–systematic review. Radiology. ential resection margin in T3 adenocarcinoma of the
2005;236(3):841–51. esophagus. Am J Surg Pathol. 2011;35(6):919–26.
36. Swisher SG, et al. 2-Fluoro-2-deoxy-D-glucose 50. Sujendran V, et al. Effect of neoadjuvant chemother-
positron emission tomography imaging is predictive apy on circumferential margin positivity and its
of pathologic response and survival after preopera- impact on prognosis in patients with resectable
tive chemoradiation in patients with esophageal car- oesophageal cancer. Br J Surg. 2008;95(2):191–4.
cinoma. Cancer. 2004;101(8):1776–85. 51. Khan OA, Cruttenden-Wood D, Toh SK. Is an involved
37. Smyth EC, Shah MA. Role of (1)(8)F 2-fluoro-2- circumferential resection margin following oespha-
deoxyglucose positron emission tomography in upper gectomy for cancer an important prognostic indicator?
gastrointestinal malignancies. World J Gastroenterol. Interact Cardiovasc Thorac Surg. 2010;11(5):645–8.
2011;17(46):5059–74. 52. Prenzel KL, et al. Prognostic relevance of skip
38. Han D, et al. Comparison of the diagnostic value of metastases in esophageal cancer. Ann Thorac Surg.
3-deoxy-3-18F-fluorothymidine and 18F-fluorode- 2010;90(5):1662–7.
oxyglucose positron emission tomography/com- 53. Westerterp M, et al. Outcome of surgical treatment
puted tomography in the assessment of regional for early adenocarcinoma of the esophagus or gastro-
lymph node in thoracic esophageal squamous cell esophageal junction. Virchows Arch. 2005;446(5):
carcinoma: a pilot study. Dis Esophagus. 2012;25(5): 497–504.
416–26. 54. Boone J, et al. International survey on esophageal
39. Riddell AM, et al. The appearances of oesophageal car- cancer: part I surgical techniques. Dis Esophagus.
cinoma demonstrated on high-resolution, T2-weighted 2009;22(3):195–202.
MRI, with histopathological correlation. Eur Radiol. 55. Schwarz RE, Smith DD. Clinical impact of lymph-
2007;17(2):391–9. adenectomy extent in resectable esophageal cancer.
40. Kijanka M, et al. Rapid optical imaging of human J Gastrointest Surg. 2007;11(11):1384–93; discus-
breast tumour xenografts using anti-HER2 VHHs sion 1393–4.
site-directly conjugated to IRDye 800CW for image- 56. Herbella FA, Patti MG. Minimally invasive esophagec-
guided surgery. Eur J Nucl Med Mol Imaging. tomy. World J Gastroenterol. 2010;16(30):3811–5.
2013;40(11):1718–29. 57. Weksler B, et al. Robot-assisted minimally invasive
41. Stahl M, et al. Chemoradiation with and without sur- esophagectomy is equivalent to thoracoscopic mini-
gery in patients with locally advanced squamous cell mally invasive esophagectomy. Dis Esophagus. 2012;
carcinoma of the esophagus. J Clin Oncol. 2005;23(10): 25(5):403–9.
2310–7. 58. Pennathur A, Luketich JD. Minimally invasive
42. Bedenne L, et al. Chemoradiation followed by sur- esophagectomy: short-term outcomes appear compa-
gery compared with chemoradiation alone in squa- rable to open esophagectomy. Ann Surg. 2012;255(2):
mous cancer of the esophagus: FFCD 9102. J Clin 206–7.
Oncol. 2007;25(10):1160–8. 59. Hulscher JB, et al. Transthoracic versus transhiatal
43. Dexter SP, et al. Circumferential resection margin resection for carcinoma of the esophagus: a meta-
involvement: an independent predictor of survival analysis. Ann Thorac Surg. 2001;72(1):306–13.
following surgery for oesophageal cancer. Gut. 2001; 60. Lagarde SM, et al. Prognostic factors in adenocarci-
48(5):667–70. noma of the esophagus or gastroesophageal junction.
44. Scheepers JJ, et al. Influence of circumferential J Clin Oncol. 2006;24(26):4347–55.
resection margin on prognosis in distal esophageal 61. Izbicki JR, et al. Prognostic value of immunohisto-
and gastroesophageal cancer approached through the chemically identifiable tumor cells in lymph nodes
transhiatal route. Dis Esophagus. 2009;22(1):42–8. of patients with completely resected esophageal can-
45. Rao VS, et al. Comparison of circumferential resec- cer. N Engl J Med. 1997;337(17):1188–94.
tion margin clearance criteria with survival after sur- 62. Orringer MB, Marshall B, Iannettoni MD.
gery for cancer of esophagus. J Surg Oncol. 2012; Transhiatal esophagectomy for treatment of benign
105(8):745–9. and malignant esophageal disease. World J Surg.
46. O’Neill JR, et al. Defining a positive circumferential 2001;25(2):196–203.
resection margin in oesophageal cancer and its 63. Siewert JR, et al. Histologic tumor type is an inde-
implications for adjuvant treatment. Br J Surg. 2013; pendent prognostic parameter in esophageal cancer:
100(8):1055–63. lessons from more than 1,000 consecutive resections
48 A.K. Talsma et al.

at a single center in the Western world. Ann Surg. of patients with and without pathological response.
2001;234(3):360–7; discussion 368–9. Ann Surg Oncol. 2012;19(11):3500–5.
64. van Sandick JW, et al. Indicators of prognosis after 78. Dunst CM, Swanstrom LL. Minimally invasive
transhiatal esophageal resection without thoracotomy esophagectomy. J Gastrointest Surg. 2010;14 Suppl
for cancer. J Am Coll Surg. 2002;194(1):28–36. 1:S108–14.
65. Hulscher JB, et al. Extended transthoracic resection 79. Courrech Staal EF, et al. Systematic review of the ben-
compared with limited transhiatal resection for ade- efits and risks of neoadjuvant chemoradiation for
nocarcinoma of the esophagus. N Engl J Med. oesophageal cancer. Br J Surg. 2010;97(10):1482–96.
2002;347(21):1662–9. 80. Kassin MT, et al. Risk factors for 30-day hospital
66. Omloo JM, et al. Extended transthoracic resection readmission among general surgery patients. J Am
compared with limited transhiatal resection for ade- Coll Surg. 2012;215(3):322–30.
nocarcinoma of the mid/distal esophagus: five-year 81. Hii MW, et al. Impact of postoperative morbidity on
survival of a randomized clinical trial. Ann Surg. long-term survival after oesophagectomy. Br J Surg.
2007;246(6):992–1000; discussion 1000–1. 2013;100(1):95–104.
67. Merkow RP, et al. Short-term outcomes after esoph- 82. Derogar M, et al. Influence of major postoperative
agectomy at 164 American College of Surgeons complications on health-related quality of life
National Surgical Quality Improvement Program among long-term survivors of esophageal cancer
hospitals: effect of operative approach and hospital- surgery. J Clin Oncol. 2012;30(14):1615–9.
level variation. Arch Surg. 2012;147(11):1009–16. 83. Lagarde SM, et al. Postoperative complications after
68. Boshier PR, Anderson O, Hanna GB. Transthoracic esophagectomy for adenocarcinoma of the esophagus
versus transhiatal esophagectomy for the treatment are related to timing of death due to recurrence. Ann
of esophagogastric cancer: a meta-analysis. Ann Surg. 2008;247(1):71–6.
Surg. 2011;254(6):894–906. 84. Blencowe NS, et al. Reporting of short-term clinical
69. Fujita H, et al. Mortality and morbidity rates, post- outcomes after esophagectomy: a systematic review.
operative course, quality of life, and prognosis after Ann Surg. 2012;255(4):658–66.
extended radical lymphadenectomy for esophageal 85. Koch CG, et al. What are the real rates of postoperative
cancer. Comparison of three-field lymphadenectomy complications: elucidating inconsistencies between
with two-field lymphadenectomy. Ann Surg. 1995; administrative and clinical data sources. J Am Coll
222(5):654–62. Surg. 2012;214(5):798–805.
70. Lerut T, et al. Three-field lymphadenectomy for carci- 86. Ghaferi AA, Birkmeyer JD, Dimick JB. Variation in
noma of the esophagus and gastroesophageal junction hospital mortality associated with inpatient surgery.
in 174 R0 resections: impact on staging, disease-free N Engl J Med. 2009;361(14):1368–75.
survival, and outcome: a plea for adaptation of TNM 87. Biere SS, et al. Minimally invasive versus open
classification in upper-half esophageal carcinoma. oesophagectomy for patients with oesophageal can-
Ann Surg. 2004;240(6):962–72; discussion 972–4. cer: a multicentre, open-label, randomised con-
71. Altorki NK, et al. Total number of resected lymph trolled trial. Lancet. 2012;379(9829):1887–92.
nodes predicts survival in esophageal cancer. Ann 88. Law S. Is minimally invasive preferable to open
Surg. 2008;248(2):221–6. oesophagectomy? Lancet. 2012;379(9829):1856–8.
72. Rizk NP, et al. Optimum lymphadenectomy for 89. Tandon S, et al. Peri-operative risk factors for
esophageal cancer. Ann Surg. 2010;251(1):46–50. acute lung injury after elective oesophagectomy.
73. Rizk N, et al. The prognostic importance of the num- Br J Anaesth. 2001;86(5):633–8.
ber of involved lymph nodes in esophageal cancer: 90. Law S, et al. Predictive factors for postoperative pul-
implications for revisions of the American Joint monary complications and mortality after esophagec-
Committee on Cancer staging system. J Thorac tomy for cancer. Ann Surg. 2004;240(5):791–800.
Cardiovasc Surg. 2006;132(6):1374–81. 91. Akaishi T, et al. Thoracoscopic en bloc total esopha-
74. Peyre CG, et al. The number of lymph nodes gectomy with radical mediastinal lymphadenectomy.
removed predicts survival in esophageal cancer: an J Thorac Cardiovasc Surg. 1996;112(6):1533–40;
international study on the impact of extent of surgi- discussion 1540–1.
cal resection. Ann Surg. 2008;248(4):549–56. 92. Orringer MB, Marshall B, Iannettoni MD. Eliminating
75. Stiles BM, et al. Worldwide Oesophageal Cancer the cervical esophagogastric anastomotic leak with a
Collaboration guidelines for lymphadenectomy pre- side-to-side stapled anastomosis. J Thorac Cardiovasc
dict survival following neoadjuvant therapy. Eur J Surg. 2000;119(2):277–88.
Cardiothorac Surg. 2012;42(4):659–64. 93. Collard JM, et al. Terminalized semimechanical
76. Torgersen Z, et al. Prognostic implications of lymph- side-to-side suture technique for cervical esophago-
adenectomy in esophageal cancer after neo-adjuvant gastrostomy. Ann Thorac Surg. 1998;65(3):814–7.
therapy: a single center experience. J Gastrointest 94. van Heijl M, et al. Intrathoracic manifestations of
Surg. 2011;15(10):1769–76. cervical anastomotic leaks after transhiatal and trans-
77. Chao YK, et al. Lymph node dissection after chemo- thoracic oesophagectomy. Br J Surg. 2010;97(5):
radiation in esophageal cancer: a subgroup analysis 726–31.
4 Goals of Surgical Therapy for Esophageal Cancer 49

95. Weidenhagen R, et al. Anastomotic leakage after 106. Muller JM, et al. Surgical therapy of oesophageal
esophageal resection: new treatment options by endo- carcinoma. Br J Surg. 1990;77(8):845–57.
luminal vacuum therapy. Ann Thorac Surg. 2010; 107. Walsh TN, et al. A comparison of multimodal therapy
90(5):1674–81. and surgery for esophageal adenocarcinoma. N Engl J
96. Lagarde SM, et al. Incidence and management of Med. 1996;335(7):462–7.
chyle leakage after esophagectomy. Ann Thorac 108. Low DE, et al. Esophagectomy–it’s not just about
Surg. 2005;80(2):449–54. mortality anymore: standardized perioperative clinical
97. Dikken JL, et al. Effect of hospital volume on postop- pathways improve outcomes in patients with esopha-
erative mortality and survival after oesophageal and geal cancer. J Gastrointest Surg. 2007;11(11):1395–
gastric cancer surgery in the Netherlands between 402; discussion 1402.
1989 and 2009. Eur J Cancer. 2012;48(7):1004–13. 109. Dikken JL, et al. Differences in outcomes of oesoph-
98. Damhuis RA, et al. Comparison of 30-day, 90-day ageal and gastric cancer surgery across Europe.
and in-hospital postoperative mortality for eight dif- Br J Surg. 2013;100(1):83–94.
ferent cancer types. Br J Surg. 2012;99(8):1149–54. 110. Markar SR, et al. Volume-outcome relationship in
99. Jamieson GG, et al. Postoperative mortality follow- surgery for esophageal malignancy: systematic review
ing oesophagectomy and problems in reporting its and meta-analysis 2000–2011. J Gastrointest Surg.
rate. Br J Surg. 2004;91(8):943–7. 2012;16(5):1055–63.
100. Wright CD, et al. Predictors of major morbidity and 111. Patti MG, et al. A hospital’s annual rate of esopha-
mortality after esophagectomy for esophageal cancer: a gectomy influences the operative mortality rate.
Society of Thoracic Surgeons General Thoracic J Gastrointest Surg. 1998;2(2):186–92.
Surgery Database risk adjustment model. J Thorac 112. van Gijn W, et al. Nationwide outcome registrations
Cardiovasc Surg. 2009;137(3):587–95; discussion 596. to improve quality of care in rectal surgery. An
101. Morita M, et al. In-hospital mortality after a surgical initiative of the European Society of Surgical
resection for esophageal cancer: analyses of the Oncology. J Surg Oncol. 2009;99(8):491–6.
associated factors and historical changes. Ann Surg 113. Birgisson H, et al. Improved survival in cancer of the
Oncol. 2011;18(6):1757–65. colon and rectum in Sweden. Eur J Surg Oncol.
102. Lagarde SM, et al. Prognostic nomogram for patients 2005;31(8):845–53.
undergoing oesophagectomy for adenocarcinoma of 114. Courrech Staal EF, et al. Quality-of-care indicators
the oesophagus or gastro-oesophageal junction. Br J for oesophageal cancer surgery: a review. Eur J Surg
Surg. 2007;94(11):1361–8. Oncol. 2010;36(11):1035–43.
103. Lagarde SM, et al. Preoperative prediction of the 115. Dimick JB, Welch HG, Birkmeyer JD. Surgical mor-
occurrence and severity of complications after tality as an indicator of hospital quality: the problem
esophagectomy for cancer with use of a nomogram. with small sample size. JAMA. 2004;292(7):
Ann Thorac Surg. 2008;85(6):1938–45. 847–51.
104. Grotenhuis BA, et al. Validation of a nomogram pre- 116. Raja S, et al. Esophageal submucosa: the watershed
dicting complications after esophagectomy for can- for esophageal cancer. J Thorac Cardiovasc Surg.
cer. Ann Thorac Surg. 2010;90(3):920–5. 2011;142(6):1403–11.
105. Dikken JL, et al. Influence of hospital type on out- 117. Griffin S, Raimes SA. A companion to specialist sur-
comes after oesophageal and gastric cancer surgery. gical practice: oesophagogastric surgery. 4th ed.
Br J Surg. 2012;99(7):954–63. Elsevier; 2009. p 97.
 Optimization of Patients
for Esophageal Cancer Surgery 5
Wesley A. Papenfuss and Todd L. Demmy

Introduction most important aspects. All patients should have

appropriate staging tests including CT of chest,
Surgery of the esophagus continues to be abdomen, and pelvis with integrated PET imaging
challenging with significant morbidity and mor- to exclude metastatic disease. A mid/upper esoph-
tality. Despite advances in anesthetic and surgical agus tumor generally requires bronchoscopy to
techniques, morbidity can be as high as 50 % and rule out tracheal invasion that precludes resection.
mortality as high as 10 %. Directing complex Initial or repeat upper endoscopy may be
gastrointestinal procedures to specialized centers needed to confirm or refine the pathologic diagno-
has improved outcomes significantly [1, 2]. sis, define the proximal extent of Barrett’s esopha-
Esophageal cancer patients are likely to have gus, evaluate neoadjuvant treatment response,
significant comorbid disease including cardiac, document residual esophageal/gastric disease,
respiratory, and hepatic disease, diabetes, and mal- and evaluate the proposed reconstructive conduit.
nutrition. Properly assessing and optimizing these Colonoscopy may be appropriate when consider-
comorbidities is essential to preoperative planning. ing alternative conduits for reconstruction.
This chapter will focus on the medical evalua- Imaging indicating overt metastatic disease or
tion of the esophageal surgery patient including invasion of “unresectable” organs such as the
nutritional optimization in the neoadjuvant setting. pulmonary vein, aorta, and trachea generally
establishes unresectability. Preoperative visceral
angiography may be useful for those at risk for
Surgical Evaluation mesenteric artery stenosis such as patients with
known coronary artery disease [4]. Overall, the
The oncologic evaluation of the esophageal cancer benefits of angiography are controversial.
patient is well described elsewhere [3]. Briefly,
evaluations of anatomy and resectability are the
Medical Evaluation
W.A. Papenfuss, MD
Department of Surgical Oncology, Comorbid conditions are common among
Roswell Park Cancer Institute, patients undergoing esophagectomy for cancer;
Elm & Carlton Streets, Buffalo, NY 14263, USA
for instance, they are at risk for cardiopulmonary
T.L. Demmy, MD (*) disorders, hepatic disease, and variable degrees
Department of Thoracic Surgery,
of malnutrition. The following section discusses
Roswell Park Cancer Institute,
Elm & Carlton Streets, Buffalo, NY, USA the assessment and management of patients with
e-mail: [email protected] coexisting diseases (see Table 5.1)

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 51
DOI 10.1007/978-3-319-09342-0_5, © Springer International Publishing Switzerland 2015
 52

Table 5.1 Preoperative medical evaluation prior to esophageal cancer surgery

Cardiac disease Pulmonary disease Liver disease VTE Nutrition Frailty
Risk factors CAD COPD/asthma ETOH Hx of VTE > 10 % Weight Age > 65
loss
Prior MI Exercise Hepatitis Immobility Albumin < 3.5 Impaired gait
intolerance
Angina Prior lung surgery Stigmata Malignancy BMI < 18 Recent chemotherapy/surgery
Diabetes Varices seen on Chemotherapy Chronic disease
Hypertension imaging
Evaluation EKG Pulmonary Childs score As clinically Dietitian Frailty testing
Echo function testing indicated consult
Exercise stress testing
Intervention Coronary angiography/ Smoking cessation Appropriate Pre-op Physical/occupational therapy
intervention anticoagulation supplement
Risk factor modification Respiratory Inferior vena cava TPN Dietitian consult
training filter placement prior
B-blockers Oral hygiene to surgery Stent Polypharmacy evaluation
Statins Feeding tube Psychiatric/neurologic
evaluation
Relative Uncorrected myocardial ischemia FEV1 < 70 % Childs A or worse Inability to
contraindication VO2 max < 11 ml/kg/min FVC < 80 % Portal maintain
hypertension weight
W.A. Papenfuss and T.L. Demmy
5 Optimization of Patients for Esophageal Cancer Surgery 53

Cardiac already on these drugs should have them

continued in the perioperative setting [12]. Statin
Arrhythmia is common after esophagectomy, use may reduce the risk of myocardial infarction
occurring in 20–30 % of patients [5]. The most as well as the risk of postoperative atrial flutter in
common arrhythmia is atrial fibrillation occur- patients undergoing noncardiac surgery [13].
ring in 10–20 % of cases [6, 7] It is associated Their use should be considered in patients
with an increased hospital length of stay and deemed to be at elevated risk of postoperative
increased risk of postoperative death. A retro- cardiovascular events.
spective review of esophagectomy found an asso-
ciation between male gender, age greater than
65 years, history of COPD, history of cardiac dis- Pulmonary
ease, and neoadjuvant therapy with an increased
risk of postoperative atrial fibrillation [7, 8]. Pulmonary complications are the most frequent
Therefore, patients at increased risk of arrhyth- complication following esophagectomy occurring
mia should be evaluated and appropriate optimi- in 20–40 % of patients. Despite the advances in
zation considered preoperatively. Patients on beta minimally invasive surgery, all patients should be
blockade should have this continued, electrolytes prepared for possible thoracotomy. Pulmonary
repleted, and statin therapy considered. testing should be considered in patients who have a
There continues to be debate over postopera- history of smoking, pulmonary disease, or signs
tive arrhythmia prevention [9]. A prospective and symptoms suggestive of an underlying pulmo-
randomized trial of amiodarone initiated at the nary disorder. Patients with compromised pulmo-
time of general anesthesia induction found a sig- nary function as evidenced by FVC < 80 % and
nificant decrease in post-esophagectomy atrial FEV1 < 70 % predicted have been shown to have
fibrillation (15 % versus 40 %), with no increased an increased risk of pulmonary complications [14].
adverse effects [9]. However, amiodarone use can Patients who are identified to have an increased
cause pneumonitis in 10–15 % of patients lead- risk of pulmonary complications may benefit from
ing to death in 10 % of those affected. For this preoperative rehab or training [15, 16].
reason, amiodarone use should be reserved for Preoperative interventions can prevent post-
patients at high risk for postoperative atrial fibril- operative pneumonia. Smoking cessation at least
lation (e.g., age > 70 or cardiac disease) and low 1 month prior to surgery has been associated with
risk for pulmonary injury. decreased incidence of pneumonia following tho-
Myocardial infarction is uncommon after racic surgery. Smoking cessation, respiratory
esophagectomy (1–2 %). However, patients training (incentive spirometry, respiratory muscle
should be assessed for their risk for postoperative stretching, deep diaphragm breathing, and effec-
MI according to the ACC AHA guidelines [10]. tive cough), and attention to oral hygiene/plaque
Esophageal surgery is considered intermediate removal decrease pulmonary complications
risk with a 1–5 % chance of perioperative following esophagectomy [17]. Minimally inva-
MI. Preoperative assessment may include EKG, sive esophagectomy causes fewer pulmonary
echocardiography, exercise stress testing, and complications. In an open-label, randomized
coronary angiography as appropriate. For patients trial, minimally invasive esophagectomy greatly
with limited fitness at centers with cardiopulmo- decreased inpatient pulmonary complications
nary exercise testing, a VO2 max less than 11 ml/ (29 versus 9 % open) [18].
kg/min predicts complications following esopha-
gectomy [11].
The use of perioperative beta-blockers and Hepatic Disease
statins deserves mention. While the addition of
beta-blockers as prophylaxis against postopera- Patients with liver disease are at an increased risk
tive events is controversial, patients who are of mortality following surgery. Alcohol use
54 W.A. Papenfuss and T.L. Demmy

contributing to squamous cell esophageal cancer Table 5.2 Frailty in the surgical patient
risk factors may also induce cirrhosis and liver Functional factors Medical factors
dysfunction. While varices may be seen on pre- Difficulty with activities Diabetes
operative imaging, liver dysfunction may be of daily living
occult until the perioperative setting. Mortality Weight loss Pulmonary disease (COPD,
pneumonia)
approaches 100 % in patients with Childs C crite-
Body mass index Cardiac disease (CHF, MI,
ria. Even Childs A patients have mortality as high hypertension)
as 10 % following esophagectomy [19]. In a Grip strength Peripheral vascular disease
review of 18 known cirrhotics undergoing esoph- Gait speed Cerebral vascular disease
agectomy, Tachibana et al. found an overall (TIA, CVA)
16.7 % mortality (versus 5.7 % in noncirrhotics). History of falls Delirium
One-year and 3-year survivals were also signifi- History of depression
cantly less [20]. The presence of cirrhosis should
be considered in all patients who have a history
of liver disease, overt physical signs on examina- of postoperative complications [25]. In a recent
tion, irregularities on liver function tests or imag- study of esophagectomy patients in the NSQIP
ing, or known risk factors. Liver biopsy may be database, both morbidity and mortality increased
necessary to confirm the diagnosis. with the presence of 1 of 11 NSQIP-measured
preoperative variables as determined by a modi-
fied frailty index. As the number of items present
Age in the frailty index increased from zero to five, the
rate of a serious complication requiring ICU
Using a specific age exclusion for esophagectomy admission increased from 18 to 61 %. Mortality
is controversial [21]. Age-related comorbidities rate increased from 1.8 to 23.1 % [26].
foster complications which are tolerated poorly In summary, age alone should not preclude
because of concomitant reductions in organ esophagectomy but should be considered in the
reserve. There are recent reports in the literature context of the patients overall functional status,
regarding the safety of esophagectomy performed frailty index, and associated comorbid conditions.
in elderly patients. Pultrum et al. report their
experience performing extended esophageal
resection via thoracolaparotomy at a high volume Obesity
center. While there was no difference in overall
survival, perioperative morbidity was predictably Obesity is an epidemic problem causing an
higher in patients greater than or equal to 70 years, increased incidence of distal esophageal cancer.
particularly in regard to pulmonary, cardiac, and Therefore, surgeons can expect to encounter
infectious complications [22]. This report has more obese patients with esophageal cancer.
been criticized as potentially difficult to repro- Obese patients have higher rates of diabetes and
duce because few centers could achieve the underlying cardiac and pulmonary diseases.
authors’ case volumes [21]. A recent pooled anal- Preoperative evaluation of the obese patient may
ysis of 25 studies revealed that elderly patients require echocardiography, cardiopulmonary
were less likely to receive neoadjuvant therapy exercise testing, pulmonary function testing
and more likely to experience inhospital mortality (with special attention to functional residual
and pulmonary and cardiac complications [23]. capacity), evaluation for obstructive sleep apnea,
More important than age is overall patient risk modification for venous thromboembolism
frailty. Multiple factors have been described and (VTE), and optimizing glycemic control for
shown to be associated with postoperative out- patients with a HgA1c > 8 % [27–29].
comes (Table 5.2) [24]. A prospective study found The incremental contribution of obesity to
the degree of frailty to be associated with the rate perioperative morbidity and mortality is
5 Optimization of Patients for Esophageal Cancer Surgery 55

controversial. Obesity itself has not been related thoracotomies, upper abdominal (e.g., anti-reflux
to increased morbidity and mortality in patients or ulcer) surgeries, and prior head and neck pro-
undergoing surgery for intra-abdominal cancer cedures deserve mention.
[30]. However, anastomotic and wound compli- Orringer et al. reported their experience per-
cations increase in obese patients with diabetes forming transhiatal esophagectomy for benign
[31–33]. In addition, several studies report no disease in patients having had prior operation for
detrimental effect on survival in the obese esoph- GERD or hiatal hernia. Thoracotomy was neces-
ageal cancer patient [33–35]. Minimally invasive sary in 16.6 % and a colonic conduit was required
esophagectomy in the obese patient is also feasi- in 10.6 % of patients [40]. MIE has also been
ble with similar morbidity and mortality but lon- reported in patients after thoracotomy for end-
ger operative times [36]. Like age, obesity, per se, stage achalasia [41].
should not preclude open or minimally invasive Esophageal cancer after bariatric surgery is
esophagectomy; however, care must be taken uncommon. However, with the increased use of
when managing coexistent comorbidities. bariatric surgery, we can expect reports to
increase. A recent series describes an experi-
ence of five minimally invasive esophagecto-
Venous Thromboembolism mies following gastric bypass. Four had
undergone laparoscopic Roux-en-y gastric
Thromboembolic events occur in 14–32 % of bypass and one patient had open bypass. One
patients undergoing neoadjuvant therapy for patient required colonic interposition for recon-
esophageal cancer [37, 38]. Such patients require struction after esophagectomy. There was no
extended anticoagulation therapy for treatment mortality in their series. The previously
and prevention of end-organ damage, which may bypassed stomach is utilized as the new gastric
delay time to surgery. Decisions regarding the conduit, while the Roux limb is utilized for jeju-
timing of surgery, the role of perioperative anti- nostomy tube placement [42].
coagulation, and IVC filter placement need to be Prior head and neck surgery can complicate
made on a case-by-case basis. Current guidelines esophagectomy depending on the planned surgi-
suggest the use of inferior vena cava filters in cal approach. A cervical anastomosis may prove
patients with residual DVT and a contraindica- challenging given prior dissection or radiation
tion to anticoagulation, recurrent DVT or PE within the operative field. For this reason, a tho-
despite anticoagulation, and patients undergoing racic dissection and anastomosis should be con-
major surgery within 2 months of a thromboem- sidered in these patients.
bolic event [39]. Removal of the filter should be
considered once the patient is deemed appropri-
ate to resume anticoagulation and is easiest Nutritional Assessment
performed within 10–14 days of placement [39]. and Optimization
Inferior vena cava filter placement in patients
with recent DVT/PE before planned esophagec- Patients with esophageal cancer frequently pres-
tomy may decrease the risk of fatal perioperative ent with dysphagia and variable degrees of weight
pulmonary embolism. loss prior to diagnosis. For this reason, nutritional
assessment before any treatment is imperative.
Assorted methods can assess the nutritional
Prior Surgical History status of cancer patients. Clinical parameters
include weight loss, dietary change as a marker
Minimally invasive esophagectomy requires for dysphagia, and gastrointestinal symptoms
operating in both the abdominal and thoracic including nausea, vomiting, diarrhea, and
cavities and is made more complex by previous anorexia. Physical exam findings suggestive of
surgical procedures in these regions. Previous malnutrition include loss of subcutaneous fat,
56 W.A. Papenfuss and T.L. Demmy

Table 5.3 Factors associated with malnutrition Table 5.4 Advantages and disadvantages of different
methods of enteral support
Weight loss > 10 %
BMI < 20 kg/m2 Advantages Disadvantages
Albumin < 3.5 g/dL Gastrostomy Ease of placement Potential injury to
Prealbumin < 10 mg/dL Bolus feeds conduit
Degree of dysphagia Jejunostomy Evaluate for Unable to bolus
Gastrointestinal symptoms metastatic disease (requires pump)
Muscle wasting Able to use Usually surgically
post-resection placed
Loss of subcutaneous fat
Esophageal Immediate relief Retrosternal pain
Ascites
stent of dysphagia
Edema
Improved quality Requires removal
of life before resection
muscle wasting, edema, and ascites as signs of Migration/
perforation
protein calorie malnutrition (see Table 5.3).
Laboratory evaluations include assessments of
rapid turnover proteins including albumin (half- accomplished in a number of ways: esophageal
life 20 days), prealbumin (half-life 2–3 days), stenting, gastrostomy, or jejunostomy. Each of
and transferrin (half-life 8–10 days) [43]. these methods has its own advantages and disad-
Weight loss greater than 10 % over 3–6 months vantages (see Table 5.4).
and greater than 5 % over 1 month suggests sig- Gastrostomy can be achieved endoscopically
nificant malnutrition [44]. Preoperative nutri- and by interventional radiology techniques or
tional supplementation, provided as TPN, was surgical placement. The use of gastrostomy
found to be beneficial only in the most malnour- before esophagectomy is somewhat controversial
ished [45]. Immuno-enhanced enteral supple- due to the risks of injuring the future gastric con-
mentation has been studied with the hope of duit or its blood supply. In general, percutaneous
decreasing morbidity and mortality following endoscopic gastrostomy tubes have low compli-
major surgery for gastrointestinal cancer. A ran- cation rates, and esophagectomies following
domized controlled trial that utilized preopera- placements have not been associated with
tive immunotherapy (supplementation of omega increased conduit-related complications [52].
3 fatty acids) failed to demonstrate a significant Transoral placement poses its own difficulties
difference in length of stay or morbidity in esoph- due to an obstructing tumor. Additionally, a
agectomy patients [46]. However, a meta-analysis recent study identified g-tube site metastasis in
of studies using immunonutrition in the periop- 9.4 % of patients undergoing endoscopic place-
erative setting for patients undergoing elective ment in esophageal cancer [53].
gastrointestinal cancer operations showed shorter Percutaneous radiologic gastrostomy (PRG) is
length of stay and fewer postoperative infectious a radiologic technique whereby the stomach is
complications [47]. At this time, the use of accessed under radiologic guidance. PRG has the
immunonutrition in the perioperative setting theoretical advantage of avoiding the primary
remains controversial. Severely malnourished malignancy during placement of the feeding
patients may benefit but should be treated for tube. PRG was placed successfully in 96.3 % of
approximately 2 weeks preoperatively [48, 49]. patients, and there were no conduit-related com-
Often, esophageal cancer patients need addi- plications attributable to the procedure in all
tional nutritional support. Enteral is preferred resected patients [54]. Open or laparoscopic gas-
over parenteral nutrition to avoid infectious com- trostomy tube placement allows for direct visual-
plications. This is especially important when ization of tube placement and avoids injury to
multimodality therapy is considered. The ability other organs or the future conduit vasculature.
to maintain nutritional status fosters completion Jejunostomy placement has been described by
of multimodality regimens [50, 51]. This can be percutaneous [55] and endoscopic means but is
5 Optimization of Patients for Esophageal Cancer Surgery 57

usually accomplished surgically. The advantage operative approach. In the event of metastatic
of a laparoscopic approach is that it allows for an disease, a gastrostomy can be placed at the same
assessment of undetected peritoneal surface setting to facilitate definitive chemotherapy.
metastasis while avoiding manipulation of the
future gastric conduit. If metastatic disease is Conclusion
encountered at the time of laparoscopy, then a Surgery of the esophagus for malignant dis-
permanent gastrostomy tube can be placed at that ease continues to be challenging despite
time. Laparoscopic placement has been shown to advances in surgical technique and periopera-
be feasible and safe without significant postop- tive management. Appropriate patient evalua-
erative sequelae [56]. Choice of jejunal tube loca- tion, selection, and optimization in the setting
tion requires careful consideration as not to of multimodality therapy are critical to
hinder future surgical therapy. decreasing the overall morbidity and mortality
Esophageal stent placement for preoperative of esophageal surgery for malignant disease.
nutritional optimization of the esophageal cancer
patient is another option. Several recent trials
have demonstrated their use in the near obstructed References
patient destined for neoadjuvant chemoradiother-
apy [57–59]. In addition to providing nutrition, 1. Wouters MW, Wijnhoven BP, Karim-Kos HE, et al.
High-volume versus low-volume for esophageal
stenting palliated obstructive symptoms leading
resections for cancer: the essential role of case-mix
to an improved quality of life [60]. The choice of adjustments based on clinical data. Ann Surg Oncol.
stent is very important. Self-expanding metal 2008;15(1):80–7. Accessed 19 Oct 2013.
stents are difficult to remove and are generally 2. Whooley BP, Law S, Murthy SC, Alexandrou A,
Wong J. Analysis of reduced death and complication
used in patients ineligible for definitive esopha-
rates after esophageal resection. Ann Surg.
geal resection. Self-expanding silicone stents are 2001;233(3):338–44. Accessed 19 Oct 2013.
preferred for patients treated with operative/cura- 3. Ajani JA, Barthel JS, Bentrem DJ, et al. Esophageal
tive intent. They provide immediate relief of and esophagogastric junction cancers. J Natl Compr
Canc Netw. 2011;9(8):830–87. Accessed 1 Nov 2013.
obstruction and are easier to remove before or at
4. Schröder W, Zähringer M, Stippel D, et al. Does celiac
the time of planned surgical resection. Stent trunk stenosis correlate with anastomotic leakage of esoph-
migration is uncommon but managed easily endo- agogastrostomy after esophagectomy? Dis Esophagus.
scopically. Gastrointestinal perforation is rare. 2002;15(3):232–6. doi:10.1046/j.1442-2050.2002.00252.x.
Accessed 24 Oct 2013.
Deciding between modalities requires consid-
5. Murthy SC, Law S, Whooley BP, Alexandrou A, Chu K,
ering specific case factors. When compared to Wong J. Atrial fibrillation after esophagectomy is a marker
jejunostomy, stent placement was associated for postoperative morbidity and mortality. J Thorac
with improved dysphagia scores, improved Cardiovasc Surg. 2003;126(4):1162–7. doi:10.1016/
S0022-5223(03)00974-7. Accessed 14 Aug 2013.
weight gain, fewer interruptions in neoadjuvant
6. Ojima T, Iwahashi M, Nakamori M, et al. Atrial fibril-
therapy, improvement in albumin levels, and less lation after esophageal cancer surgery: an analysis
percentage of body weight lost [43, 60]. of 207 consecutive patients. Surg Today. 2014;44:
Gastrostomy tubes may be more convenient for 839–47. doi:10.1007/s00595-013-0616-3. Accessed
19 Oct 2013.
the patient but have the unlikely possibility of
7. Rao VP, Addae-Boateng E, Barua A, Martin-Ucar AE,
damaging the gastric conduit precluding its use. Duffy JP. Age and neo-adjuvant chemotherapy
Our preferred approach to the patient requir- increase the risk of atrial fibrillation following oesoph-
ing nutritional support for planned neoadjuvant agectomy. Eur J Cardiothorac Surg. 2012;42(3):438–
43. doi:10.1093/ejcts/ezs085. Accessed 19 Oct 2013.
therapy is laparoscopic jejunostomy placement.
8. Ma J, Wang Y, Zhao Y, et al. Atrial fibrillation after
This allows for evaluation of the future gastric surgery for esophageal carcinoma: clinical and prog-
conduit, direct anatomic assessment of GE junc- nostic significance. World J Gastroenterol. 2006;12(3):
tion tumors, and the ability to rule out metastatic 449–52. Accessed 19 Oct 2013.
9. Tisdale JE, Wroblewski HA, Wall DS, et al. A random-
peritoneal disease. Thoughtful placement of the
ized, controlled study of amiodarone for prevention of
jejunostomy will not complicate the future atrial fibrillation after transthoracic esophagectomy.
58 W.A. Papenfuss and T.L. Demmy

J Thorac Cardiovasc Surg. 2010;140(1):45–51. esophagectomy in 18 consecutive patients. Ann Surg

doi:10.1016/j.jtcvs.2010.01.026. Accessed 19 Oct 2013. Oncol. 2000;7(10):758–63. Accessed 22 Oct 2013.
10. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 21. Orringer MB. Editorial: Age does not preclude an
2007 guidelines on perioperative cardiovascular evalua- esophagectomy⋯ if only it were that simple. Ann
tion and care for noncardiac surgery: executive sum- Surg Oncol. 2010;17(6):1487–9. doi:10.1245/s10434-
mary – a report of the American College of Cardiology/ 010-0989-0. Accessed 20 Aug 2013.
American Heart Association task force on practice 22. Pultrum BB, Bosch DJ, Nijsten MW, et al. Extended
guidelines (Writing Committee to revise the 2002 guide- esophagectomy in elderly patients with esophageal
lines on perioperative cardiovascular evaluation for non- cancer: minor effect of age alone in determining the
cardiac surgery). Circulation. 2007;116(17):1971–96. postoperative course and survival. Ann Surg Oncol.
doi: 10.1161/CIRCULATIONAHA.107.185700 . 2010;17(6):1572–80. doi:10.1245/s10434-010-0966-7.
Accessed 26 Oct 2013. Accessed 20 Aug 2013.
11. Forshaw MJ, Strauss DC, Davies AR, et al. Is cardio- 23. Markar SR, Karthikesalingam A, Thrumurthy S, Ho A,
pulmonary exercise testing a useful test before esoph- Muallem G, Low DE. Systematic review and pooled
agectomy? Ann Thorac Surg. 2008;85(1):294–9. analysis assessing the association between elderly age
Accessed 19 Oct 2013. and outcome following surgical resection of esopha-
12. London MJ, Hur K, Schwartz GG, Henderson WG. geal malignancy. Dis Esophagus. 2013;26(3):250–62.
Association of perioperative ß-blockade with mortal- doi:10.1111/j.1442-2050.2012.01353.x. Accessed 2
ity and cardiovascular morbidity following major Nov 2013.
noncardiac surgery. JAMA. 2013;309(16):1704–13. 24. Partridge JSL, Harari D, Dhesi JK. Frailty in the older sur-
doi:10.1001/jama.2013.4135. Accessed 19 Oct 2013. gical patient: a review. Age Ageing. 2012;41(2):142–7.
13. Chopra V, Wesorick DH, Sussman JB, et al. Effect of doi:10.1093/ageing/afr182. Accessed 2 Nov 2013.
perioperative statins on death, myocardial infarction, 25. Revenig LM, Canter DJ, Taylor MD, et al. Too frail
atrial fibrillation, and length of stay: a systematic for surgery? Initial results of a large multidisciplinary
review and meta-analysis. Arch Surg. 2012;147(2): prospective study examining preoperative variables
181–9. Accessed 19 Oct 2013. predictive of poor surgical outcomes. J Am Coll
14. Kuwano H, Sumiyoshi K, Sonoda K, et al. Relationship Surg. 2013;217(4):665–70.e1. doi:10.1016/j.jamcoll-
between preoperative assessment of organ func- surg.2013.06.012. Accessed 4 Nov 2013.
tion and postoperative morbidity in patients with 26. Hodari A, Hammoud ZT, Borgi JF, Tsiouris A,
oesophageal cancer. Eur J Surg. 1998;164(8):581–6. Rubinfeld IS. Assessment of morbidity and mortality
doi:10.1080/110241598750005679. Accessed 14 Aug after esophagectomy using a modified frailty index.
2013. Ann Thorac Surg. 2013;96:1240–5. doi:10.1016/j.
15. Agrelli TF, De Carvalho Ramos M, Guglielminetti R, athoracsur.2013.05.051. Accessed 17 Oct 2013.
Silva AA, Crema E. Preoperative ambulatory inspira- 27. Owers CE, Abbas Y, Ackroyd R, Barron N, Khan
tory muscle training in patients undergoing esopha- M. Perioperative optimization of patients undergoing
gectomy. A pilot study. Int Surg. 2012;97(3):198–202. bariatric surgery. J Obes. 2012;2012:781546.
doi:10.9738/CC136.1. Accessed 19 Oct 2013. doi:10.1155/2012/781546. Accessed 4 Nov 2013.
16. Inoue J, Ono R, Makiura D, et al. Prevention of post- 28. Donohoe CL, Feeney C, Carey MF, Reynolds
operative pulmonary complications through intensive JV. Perioperative evaluation of the obese patient.
preoperative respiratory rehabilitation in patients with J Clin Anesth. 2011;23(7):575–86. doi:10.1016/j.
esophageal cancer. Dis Esophagus. 2013;26(1):68– jclinane.2011.06.005. Accessed 12 Nov 2013.
74. doi:10.1111/j.1442-2050.2012.01336.x. Accessed 29. Underwood P, Askari R, Hurwitz S, Chamarthi B,
24 Oct 2013. Garg R. Preoperative A1C and clinical outcomes in
17. Akutsu Y, Matsubara H. Perioperative management patients with diabetes mellitus undergoing major non-
for the prevention of postoperative pneumonia with cardiac surgical procedures. Diabetes Care.
esophageal surgery. Ann Thorac Cardiovasc Surg. 2014;37(3):611–6. doi:10.2337/dc13-1929.
2009;15(5):280–5. Accessed 26 Oct 2013. 30. Mullen JT, Davenport DL, Hutter MM, et al. Impact of
18. Biere SS, Van Berge Henegouwen MI, Maas KW, body mass index on perioperative outcomes in patients
et al. Minimally invasive versus open oesopha- undergoing major intra-abdominal cancer surgery.
gectomy for patients with oesophageal cancer: Ann Surg Oncol. 2008;15(8):2164–72. doi:10.1245/
a multicentre, open-label, randomised controlled s10434-008-9990-2. Accessed 11 Oct 2013.
trial. Lancet. 2012;379(9829):1887–92. doi:10.1016/ 31. Kassis ES, Kosinski AS, Ross Jr P, Koppes KE, Donahue
S0140-6736(12)60516-9. Accessed 21 Aug 2013. JM, Daniel VC. Predictors of anastomotic leak after
19. Lu M, Liu Y, Wu Y, Kao C, Liu H, Hsieh M. Is it safe esophagectomy: an analysis of the society of thoracic
to perform esophagectomy in esophageal cancer surgeons general thoracic database. Ann Thorac Surg.
patients combined with liver cirrhosis? Interact 2013. doi:10.1016/j.athoracsur.2013.07.119. Accessed
Cardiovasc Thorac Surg. 2005;4(5):423–5. Accessed 27 Oct 2013.
19 Oct 2013. 32. Bhayani NH, Gupta A, Dunst CM, Kurian AA, Halpin VJ,
20. Tachibana M, Kotoh T, Kinugasa S, et al. Esophageal Swanström LL. Does morbid obesity worsen outcomes
cancer with cirrhosis of the liver: results of after esophagectomy? Ann Thorac Surg. 2013;95(5):
5 Optimization of Patients for Esophageal Cancer Surgery 59

1756–61. doi:10.1016/j.athoracsur.2013.01.015. Accessed 45. Perioperative total parenteral nutrition in surgical

27 Oct 2013. patients. N Engl J Med. 1991;325(8):525–32.
33. Kayani B, Okabayashi K, Ashrafian H, et al. Does obe- Accessed 2 Nov 2013.
sity affect outcomes in patients undergoing esophagec- 46. Sultan J, Griffin SM, Di Franco F, et al. Randomized
tomy for cancer? A meta-analysis. World J Surg. clinical trial of omega-3 fatty acid-supplemented
2012;36(8):1785–95. doi:10.1007/s00268-012-1582-4. enteral nutrition versus standard enteral nutrition in
Accessed 27 Oct 2013. patients undergoing oesophagogastric cancer surgery.
34. Melis M, Weber J, Shridhar R, et al. Body mass index Br J Surg. 2012;99(3):346–55. doi:10.1002/bjs.7799.
and perioperative complications after oesophagec- Accessed 26 Oct 2013.
tomy for adenocarcinoma: a systematic database 47. Zhang Y, Gu Y, Guo T, Li Y, Cai H. Perioperative
review. BMJ Open. 2013;3(5). doi: 10.1136/bmjopen- immunonutrition for gastrointestinal cancer: a system-
2012-001336. Accessed 11 Oct 2013. atic review of randomized controlled trials. Surg Oncol.
35. Scarpa M, Cagol M, Bettini S, et al. Overweight 2012;21(2):e87–95. doi:10.1016/j.suronc.2012.01.002.
patients operated on for cancer of the esopha- Accessed 26 Oct 2013.
gus survive longer than normal-weight patients. 48. Nespoli L, Coppola S, Gianotti L. The role of the
J Gastrointest Surg. 2013;17(2):218–27. doi:10.1007/ enteral route and the composition of feeds in the nutri-
s11605-012-2023-2. Accessed 27 Oct 2013. tional support of malnourished surgical patients.
36. Kilic A, Schuchert MJ, Pennathur A, et al. Impact of Nutrients. 2012;4(9):1230–6. doi:10.3390/nu4091230.
obesity on perioperative outcomes of minimally invasive Accessed 26 Oct 2013.
esophagectomy. Ann Thorac Surg. 2009;87(2):412–5. 49. Abunnaja S, Cuviello A, Sanchez JA. Enteral and par-
doi:10.1016/j.athoracsur.2008.10.072. Accessed 11 Oct enteral nutrition in the perioperative period: state of
2013. the art. Nutrients. 2013;5(2):608–23. doi:10.3390/
37. Verhage RJJ, Van Der Horst S, Van Der Sluis PC, nu5020608. Accessed 26 Oct 2013.
Lolkema MP, Van Hillegersberg R. Risk of thrombo- 50. Miyata H, Yano M, Yasuda T, et al. Randomized
embolic events after perioperative chemotherapy ver- study of clinical effect of enteral nutrition support
sus surgery alone for esophageal adenocarcinoma. during neoadjuvant chemotherapy on chemotherapy-
Ann Surg Oncol. 2012;19(2):684–92. doi:10.1245/ related toxicity in patients with esophageal cancer.
s10434-011-2005-8. Accessed 20 Aug 2013. Clin Nutr. 2012;31(3):330–6. doi:10.1016/j.
38. Teman NR, Silski L, Zhao L, et al. Thromboembolic clnu.2011.11.002. Accessed 12 Nov 2013.
events before esophagectomy for esophageal cancer do not 51. Schattner M. Enteral nutritional support of the patient
result in worse outcomes. Ann Thorac Surg. 2012;94(4): with cancer: route and role. J Clin Gastroenterol.
1118–25. doi:10.1016/j.athoracsur.2012.05.109. Accessed 2003;36(4):297–302. doi:10.1097/00004836-
20 Aug 2013. 200304000-00004. Accessed 12 Nov 2013.
39. Bakshi P, Partridge J. Indications for and use of infe- 52. Margolis M, Alexander P, Trachiotis GD, et al.
rior vena cava filters in the preoperative phase. BMJ Percutaneous endoscopic gastrostomy before multi-
(Online). 2013;347(7928). doi: 10.1136/bmj.f5807. modality therapy in patients with esophageal cancer.
Accessed 4 Nov 2013. Ann Thorac Surg. 2003;76(5):1694–8. doi:10.1016/
40. Chang AC, Lee JS, Sawicki KT, Pickens A, Orringer S0003-4975(02)04890-7. Accessed 14 Aug 2013.
MB. Outcomes after esophagectomy in patients with 53. Ellrichmann M, Sergeev P, Bethge J, et al. Prospective
prior antireflux or hiatal hernia surgery. Ann Thorac evaluation of malignant cell seeding after percutaneous
Surg. 2010;89(4):1015–23. doi:10.1016/j.athorac- endoscopic gastrostomy in patients with oropharyngeal/
sur.2009.10.052. Accessed 20 Aug 2013. esophageal cancers. Endoscopy. 2013;45(7):526–31.
41. Carter YM, Bond CD, Benjamin S, Marshall doi:10.1055/s-0033-1344023. Accessed 27 Oct 2013.
MB. Minimally invasive transhiatal esophagectomy 54. Tessier W, Piessen G, Briez N, Boschetto A, Sergent
after thoracotomy. Ann Thorac Surg. 2013;95(2):e41– G, Mariette C. Percutaneous radiological gastrostomy
3. doi:10.1016/j.athoracsur.2012.07.084. Accessed 21 in esophageal cancer patients: a feasible and safe
Aug 2013. access for nutritional support during multimodal ther-
42. Rossidis G, Browning R, Hochwald SN, Abbas H, Kim apy. Surg Endosc. 2013;27(2):633–41. doi:10.1007/
T, Ben-David K. Minimally invasive esophagectomy s00464-012-2506-y. Accessed 4 Oct 2013.
is safe in patients with previous gastric bypass. Surg 55. Van Overhagen H, Schipper J. Percutaneous jejunos-
Obes Relat Dis. 2014;10(1):95–100. doi:10.1016/j. tomy. Semin Intervent Radiol. 2004;21(3):199–204.
soard.2013.03.015. Accessed 11 Oct 2013. doi:10.1055/s-2004-860878. Accessed 5 Nov 2013.
43. Bower MR, Martin II RC. Nutritional management 56. Ben-David K, Kim T, Caban AM, Rossidis G, Rodriguez
during neoadjuvant therapy for esophageal cancer. J SS, Hochwald SN. Pre-therapy laparoscopic feeding
Surg Oncol. 2009;100(1):82–7. doi:10.1002/jso.21289. jejunostomy is safe and effective in patients undergoing
Accessed 15 Aug 2013. minimally invasive esophagectomy for cancer. J
44. Nitenberg G, Raynard B. Nutritional support of the cancer Gastrointest Surg. 2013;17(8):1352–8. doi:10.1007/
patient: issues and dilemmas. Crit Rev Oncol Hematol. s11605-013-2231-4. Accessed 27 Oct 2013.
2000;34(3):137–68. doi:10.1016/S1040-8428(00)00048-2. 57. Krokidis M, Burke C, Spiliopoulos S, et al. The use of
Accessed 12 Nov 2013. biodegradable stents in malignant oesophageal strictures
60 W.A. Papenfuss and T.L. Demmy

for the treatment of dysphagia before neoadjuvant treat- 59. Webb M, Griffin SM, Shenfine J. Self-expanding
ment or radical radiotherapy: a feasibility study. esophageal stents and neoadjuvant therapy. Ann Surg
Cardiovasc Intervent Radiol. 2013;36(4):1047–54. Oncol. 2011;18(1):286. doi:10.1245/s10434-010-1096-y.
doi:10.1007/s00270-012-0503-0. Accessed 27 Oct 2013. Accessed 5 Oct 2013.
58. Pellen MG, Sabri S, Razack A, Gilani SQ, Jain 60. Siddiqui AA, Sarkar A, Beltz S, et al. Placement
PK. Safety and efficacy of self-expanding remov- of fully covered self-expandable metal stents in
able metal esophageal stents during neoadjuvant patients with locally advanced esophageal cancer
chemotherapy for resectable esophageal cancer. Dis before neoadjuvant therapy. Gastrointest Endosc.
Esophagus. 2012;25(1):48–53. doi:10.1111/j.1442- 2012;76(1):44–51. doi:10.1016/j.gie.2012.02.036.
2050.2011.01206.x. Accessed 4 Oct 2013. Accessed 15 Aug 2013.
 Pathogenesis of Gastric Cancer
6
Fátima Carneiro and Heike I. Grabsch

Introduction to females and show prominent geographical

variation ranging from 3.9 in Northern Africa to
The vast majority of malignant neoplasms of the 42.4 in Eastern Asia per 100,000 males [4].
stomach are adenocarcinomas. Non-epithelial Over the past 50 years, incidence and mortality
tumors of the stomach include lymphomas, neuro- rates of the non-cardia GC have been decreasing in
endocrine and soft tissue tumors. In this chapter, almost all countries, whereas incidence rates of
we will focus on gastric (adeno)carcinoma (GC). GC at the cardia have been stable or increasing [3].
GC represents a morphological, biologically
and genetically heterogeneous group of tumors
with multifactorial etiologies [1]. Etiology and Risk Factors of Gastric
Most GCs are sporadic. However, familial Adenocarcinoma
clustering is observed in about 10 % of sporadic
GC. Hereditary GC accounts for 1–3 % of cases Helicobacter pylori (H. pylori) infection plays a
and two hereditary syndromes have been major role in contributing to an increased risk of
described – hereditary diffuse type gastric cancer GC. Most non-cardia GCs develop from a back-
(HDGC) and gastric adenocarcinoma and proxi- ground of H. pylori infected mucosa [5]. Factors
mal polyposis of the stomach (GAPPS). associated with the pathogenicity of H. pylori
include virulence factors such as cagA in the cag
pathogenicity island and vacA, the vacuolating
Epidemiology cytotoxin [6]. Interestingly, strains producing the
cagA protein are associated with a greater risk of
Despite a steady decline in GC incidence at a rate developing cancer of the distal stomach [7, 6].
of approximately 5 % per year since the 1950s Although the risk of GC has been related to the
[2], GC is still the fifth most common cancer presence of a vacA genotype in some European
worldwide. In 2012, almost three quarters of new countries and North America, such a relationship
GCs occurred in Asia, and more than two fifths has not been observed in East Asia suggesting
occurred in China [3]. Age-standardized GC inci- that consequences of the vacA genotype may be
dence rates are twice as high in males compared dependent on the geographical region.

F. Carneiro, MD, PhD (*) H.I. Grabsch, MD, PhD, FRCPath (*)
Department of Pathology, IPATIMUP and Medical Department of Pathology, Maastricht University
Faculty, Centro Hospitalar de São João, Medical Center, P. Debyelaan 25, 6202 AZ Maastricht,
Rua Dr. Roberto Frias S/N, Porto 4200-465, Portugal The Netherlands
e-mail: [email protected] e-mail: [email protected]

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 61
DOI 10.1007/978-3-319-09342-0_6, © Springer International Publishing Switzerland 2015
62 F. Carneiro and H.I. Grabsch

The development of GC after H. pylori infec- appears to potentiate the carcinogenic effect of
tion has been considered a multistep process pro- infection with cagA-positive H. pylori.
gressing from chronic active pangastritis or Other clinicopathological conditions which
corpus-predominant gastritis to increasing loss of have been associated with an increased risk of
gastric glands (atrophy), replacement of the normal GC are autoimmune gastritis, peptic ulcer dis-
mucosa by intestinal metaplasia, and malignant ease, hypertrophic gastropathies, gastric stump
transformation [8]. However, corpus-predominant (operated stomach), and gastric polyps.
gastritis with multifocal gastric atrophy and hypo-
chlorhydria or achlorhydria is only seen in approxi-
mately 1 % of subjects infected with H. pylori, and Precursor Lesions of Gastric
most H. pylori infected individuals will remain Carcinoma
asymptomatic. Only 1–5 % of the H. pylori infected
population will develop GC indicating a role for Gastric dysplasia (synonym: intraepithelial
other causative agents and/or host factors. neoplasia (IEN)) is considered the precursor
Individual genetic GC susceptibility appears to lesion of the so-called intestinal type of gastric
involve a relatively large number of different genes carcinoma and can have a flat, slightly depressed,
including those involved in the protection of the or polypoid growth pattern. The prevalence of
gastric mucosa against damaging agents and in gastric dysplasia varies between 20 % in high-
inflammatory response, such as polymorphisms of risk areas and 4 % in Western countries where
the interleukin-1β (IL1B) gene [9, 10]. gastric carcinoma is less common [15]. Dysplasia
It has been estimated that 10 % of GC are asso- is more frequent in males, patients over 70 years
ciated with Epstein Barr virus (EBV) infection of age and most commonly affects the lesser
[11]. Considering the worldwide GC incidence, curve and the antrum.
EBV-associated GC is the largest group of carci- Gastric dysplasia is characterized by cellular
nomas within all EBV-associated malignancies. atypia and disorganized glandular architecture.
Certain dietary habits have been associated Recognition of gastric dysplasia and determina-
with an increased risk of GC [12]. These include tion of its grade is critical because it predicts
high intakes of salt-preserved and/or smoked both, the risk of malignant transformation and the
foods as well as low intakes of fresh fruits and veg- risk of metachronous gastric cancer.
etables. A recent meta-analysis suggested a poten- In an attempt to standardize the terminology
tial 50 % higher risk of GC associated with intake used to describe the grade of dysplasia and distin-
of pickled vegetables but interestingly indicated a guish it from adenocarcinoma, several proposals
potential stronger association between GC and including the Padova and Vienna classifications
intake of pickled vegetables for patients in Korea have been made [16–18].
and China [13]. Meat consumption, specifically According to the most recent WHO classifica-
red meat and processed meat, has been associated tion [1], dysplasia is graded as high or low grade.
with an increased risk of GC in the distal stomach, Low-grade dysplasia/IEN shows minimal archi-
whereas a high consumption of fruits, vegetables, tectural disarray and only mild to moderate cyto-
cereals, nuts and seeds, seafood, and olive oil was logical atypia. The nuclei are elongated and basally
shown to be associated with a significant reduction located, and mitotic activity is mild to moderate
in the risk of developing GC. There is currently no (Fig. 6.1a). High-grade dysplasia/IEN shows pro-
conclusive evidence for an association between nounced architectural disarray and severe cyto-
alcohol consumption and GC. logical atypia with numerous mitoses, which can
Smoking has also been associated with an be atypical (Fig. 6.1b). The cell nuclei in high-
increased risk of GC depending on the number of grade dysplasia are typically no longer basally
cigarettes and the duration of smoking; the epide- located and may contain prominent nucleoli.
miological association is not explicable by bias Low-grade dysplasia progresses to adenocar-
or confounding factors [14]. Smoking also cinoma in up 23 % of cases within 10 months to
6 Pathogenesis of Gastric Cancer 63

a b

Fig. 6.1 Precursor lesion of intestinal type gastric cancer – nuclei. Nuclei are larger and rounder and vary more in size
dysplasia. (a) Low-grade dysplasia. Pseudostratification of and shape. Loss of basal orientation of nuclei in many cells.
nuclei. Nuclei are elongated and mostly basally orientated. Basal membrane around individual glands still intact
Few mitotic figures. (b) High-grade dysplasia. Crowding of

4 years, whereas malignant transformation of EGCs are classified into three types based on
high-grade dysplasia has been reported to occur the endoscopic appearance according to the Paris
in 60–80 % of cases. classification: type I (protruded), polypoid growth
It is noteworthy that precursor lesions of diffuse (subcategorized into Ip (pedunculated) and Is
type GC are not well characterized, except for (sessile)); type II (superficial), non-polypoid
hereditary diffuse type GC (see below under the growth (subcategorized into type IIa (slightly ele-
section on genetic predisposition and hereditary vated), type IIb (flat), and type IIc (slightly
syndromes). depressed)); and type III, excavated growth [21]
(Figs. 6.2 and 6.3).
The macroscopic appearance of advanced GC
Pathology of Gastric is classified using the Borrmann classification
Adenocarcinoma [22] which divides GC into four distinct types:
type I, polypoid type; type II, fungating; type III,
Macroscopy ulcerated; and type IV, diffusely infiltrative
(Fig. 6.4).
GC can present at an early or advanced disease
stage. “Early gastric carcinoma” (EGC) is defined
as a carcinoma which has infiltrated the mucosa Microscopy
or submucosa regardless of the presence or
absence of lymph node metastases [19, 20]. While the macroscopic appearances are different
Conversely, GCs infiltrating into the muscularis between early and advanced GC, the histological
propria and beyond are defined as “advanced.” appearances are similar. Two major histological
64 F. Carneiro and H.I. Grabsch

Fig. 6.2 Paris classification

of early gastric cancer: type I
(protruding), Ip pedunculated
and Is sessile; type II Mucosa
(superficial), IIa elevated, IIb Ip Is Submucosa
flat, and IIc slightly
depressed; type III (ulcerated)

IIa IIb IIc

III

b
a

Mucosa
c

Submucosa

Muscularis propria

Fig. 6.3 Endoscopic resection (ESD) of a well- Deep resection margin located in the muscularis propria
differentiated early gastric cancer. (a) Macroscopy of the ensuring complete (curative) resection of the tumor.
endoscopic resection specimen after fixation with a super- (c) Microscopy of the well-differentiated adenocarcinoma
ficially elevated lesion (Paris type IIa). (b) Macroscopy of infiltrating the submucosa (pT1b, red arrow) and adjacent
the serial cross sections through the lesion showing a intramucosal adenocarcinoma (Images courtesy of
tumor which is infiltrating the submucosa (red arrows). Dr. T. Arai, Tokyo)
6 Pathogenesis of Gastric Cancer 65

I II III IV

b1 c1 d1

b2 c2 d2

Fig. 6.4 Macroscopy of advanced gastric cancer. (type III) – (c1) deep ulceration visible macroscopically from
(a) Borrmann classification. I polypoid type, II fungating the mucosal surface and (c2) infiltration into the attached
type, III ulcerated type, and IV diffusely infiltrative. lesser omentum visible macroscopically on cross sectioning.
(b) Polypoid gastric cancer (type I) – (b1) macroscopy of the (d) Diffusely infiltrative gastric cancer (type IV) – (d1) dif-
mucosal surface showing a large polypoid lesion and (b2) fuse thickening of the gastric folds visible on macroscopy of
cross section showing tumor infiltrating into the superficial the mucosal surface and (d2) diffuse infiltration of the whole
layer of the muscularis propria. (c) Ulcerated gastric cancer depth of the wall into the perigastric fat on cross section

GC subtypes (intestinal type GC and diffuse type classifications based on tumor morphology, GC
GC) have been described by Laurén [23] which can be classified on the basis of the presence or
have different clinicopathological profiles and absence of cell differentiation markers – MUC5AC
molecular pathogenesis and often occur in dis- and trefoil peptide TFF1 (markers of surface gas-
tinct epidemiologic settings. tric epithelium (foveolar cells)), MUC6 and trefoil
According to the World Health Organization peptide TFF2 (markers of mucus neck cell, pyloric
(WHO) [1], GCs are classified as tubular, papil- gland, and Brunner’s gland cells), and MUC2,
lary, mucinous, poorly cohesive (with or without CDX-2, and CD10 (markers of intestinal goblet
signet ring cells), and mixed (Fig. 6.5). Tubular cells) – into four phenotypes: (1) gastric, (2) mixed
and papillary carcinomas roughly correspond to gastric and intestinal, (3) intestinal, and (4) unclas-
the intestinal type and poorly cohesive carcino- sifiable or null phenotype which does not express
mas correspond to the diffuse type according to any of these markers [27–29].
Laurén’s classification (Table 6.1). The Laurén
and WHO classifications are the ones most com-
monly used outside of Japan. In Japan, the rec- Staging and Prognosis of Advanced
ommended histological typing is similar but not Gastric Cancer
100 % identical to the WHO classification [24].
Nakamura’s classification into differentiated Staging
and undifferentiated subtype is used together with
the size of the lesion and presence or absence The staging for carcinoma of the stomach
of ulceration to decide whether a lesion can be was substantially modified in 2009 as detailed
treated endoscopically [25, 26]. Apart from the in Table 6.2. Major changes included the
66 F. Carneiro and H.I. Grabsch

a b c

d e f

Fig. 6.5 Histological subtypes of gastric cancer. (a) Tubular type (moderately differentiated); (b) diffuse type;
(c) papillary type; (d) mucinous type; (e) undifferentiated, solid type; and (f) poorly cohesive type with signet ring cells

Table 6.1 Classification of GC

Laurén Nakamura
classification World Health Organization 2010 Japanese classification 2011 classification
Papillary Papillary Differentiated type
Tubular Tubular 1
Intestinal type
Tubular 2
Mucinous Mucinous
Poorly cohesive, including Signet ring cell Undifferentiated type
Diffuse type signet ring cell carcinoma Poorly differentiated,
and other variants non-solid type
Mixed (intestinal Mixed type (tubular/papillary – –
and diffuse type) and poorly cohesive/signet ring)
Undifferentiated Poorly differentiated, solid type
Adenosquamous
Indeterminate type Undifferentiated type
Medullary
Hepatoid

subdivision of T1 cancers into T1a (mucosa) adjacent structures). Consequently, the categori-
and T1b (submucosa), the renaming of T2a zation of the T (depth of invasion) is now uni-
(muscularis propria) as T2 and T2b (subserosa) form throughout the gastrointestinal tract,
as T3, and the subdivision of T4 (serosa) whereas differences remain for the categoriza-
into T4a (penetrates serosa) and T4b (invades tion of the N (presence or absence of regional
6 Pathogenesis of Gastric Cancer 67

Table 6.2 TNM classification of gastric carcinoma A recent meta-analysis comparing survival
T – Primary tumor rates after gastrectomy between GC patients from
TX Primary tumor cannot be assessed the West and the East from patients recruited into
T0 No evidence of primary tumor large randomized controlled clinical trials showed
Tis Carcinoma in situ: intraepithelial tumor without an association between type of surgical resection
invasion of the lamina propria, high-grade dysplasia performed in the East and improved survival [30].
T1 Tumor invades lamina propria, muscularis mucosae,
The known difference in surgical techniques
or submucosa
T1a Tumor invades lamina propria or muscularis
between the East and the West is one potential
mucosae variable that may be responsible for discrepancy in
T1b Tumor invades submucosa outcomes. Noguchi et al. [31] reported a survival
T2 Tumor invades muscularis propria difference between high-volume centers in the
T3 Tumor invades subserosa USA and Japan which was no longer apparent
T4 Tumor perforates serosa or invades adjacent after adjusting for tumor location. Verdecchia
structures et al. [32] demonstrated that the survival of Italian
T4a Tumor perforates serosa GC patients was inferior to that of Japanese GC
T4b Tumor invades adjacent structures
patients and that this survival difference disap-
N – Regional lymph nodes
peared after adjusting for stage. Bollschweiler
NX Regional lymph nodes cannot be assessed
et al. [33] compared the survival of Japanese and
N0 No regional lymph node metastasis
German GC patients and concluded that the coun-
N1 Metastasis in 1–2 regional lymph nodes
N2 Metastasis in 3–6 regional lymph nodes
try itself was a prognostic factor. Higher frequency
N3 Metastasis in 7 or more regional lymph nodes of early stage GC and more accurate staging have
N3a Metastasis in 7–15 regional lymph nodes also been associated with improved survival in
N3b Metastasis in 16 or more regional lymph nodes Japan compared with Western nations [34].
M – Distant metastasis Early and advanced GCs differ in prognosis.
M0 No distant metastasis Japanese patients with EGC have an excellent
M1 Distant metastasis prognosis with a 5-year survival rate exceeding
From Edge et al. [20] with permission 90 % after surgical treatment. Nevertheless,
approximately 2 % of EGC recur after curative
resection and lymph node metastases occur in
lymph node metastases). The N categories for 2–3 % of intramucosal carcinomas [35, 36] and
GC are N0 (no regional lymph node metastasis), 20–30 % of submucosal carcinomas [37]. Risk
N1 (1 to 2 lymph node metastases), N2 (3 to 6 factors for lymph node metastasis in EGC include
lymph node metastases), N3a (7 to 15 lymph age at time of diagnosis (the younger, the more
node metastases), and N3b (metastases in 16 or frequent the lymph node metastases), size greater
more regional lymph nodes) [19, 20]. 20 mm, depressed macroscopic type, grade of
differentiation, presence of an ulcer or scar, lym-
phatic channel invasion, and submucosal inva-
Spreading and Prognosis sion by more than 500 μm [35, 37].
Five-year survival rate of advanced GC, the
Gastric carcinomas can spread by (i) direct exten- most frequent type in the West, is around 23 %
sion to adjacent organs, (ii) lymphatic invasion, when treated by surgery alone and around 36 %
(iii) blood vessel invasion, and/or (iv) peritoneal when treatment includes perioperative chemo-
dissemination. Intestinal type GCs preferentially therapy [38]. For advanced GC, depth of infiltra-
metastasize hematogenously to the liver, whereas tion into the wall (T category of the TNM
diffuse type GCs preferentially metastasize to classification), number of lymph node metastases
peritoneal surfaces [1]. GCs with mixed histo- (N category of the TNM classification), and
logical phenotype exhibit the metastatic patterns presence of distant metastases (M category of
of both types. the TNM classification) remain the strongest
68 F. Carneiro and H.I. Grabsch

a b c

Fig. 6.6 Development model for diffuse type GC in CDH1 germline mutation carriers encompassing: (a) in situ
carcinoma, (b) pagetoid spread of signet ring cells, and (c) early intramucosal carcinoma. The arrow heads highlight a
gland that shows in situ carcinoma

prognostic indicators [19, 20]. Lymphatic and families with HDGC syndrome as families
venous invasion are also predictors of poor sur- meeting one of two criteria: (i) two or more docu-
vival in GC. Perineural invasion correlates with mented cases of diffuse type GC in first- or sec-
T stage and tumor size and may serve as a marker ond-degree relatives with at least one of them
of advanced disease [39]. diagnosed before the age of 50 years or (ii) three
or more documented cases of diffuse GC in first-
or second-degree relatives independent of the age
Genetic Predisposition at diagnosis [41]. Women in these families have
and Hereditary Syndromes an elevated risk of lobular breast cancer. The cri-
teria for genetic testing were updated in 2010
First-degree relatives of patients with GC are [42]. In several HDGC families, a higher inci-
almost three times more likely to develop GC dence of orofacial clefts has been noted [43, 44].
themselves compared to the general population Alterations of the CDH1 gene, which encodes
which has been partially attributed to H. pylori E-cadherin, constitute the genetic causal event in
infection and to the potential role of IL-1 gene HDGC patients [45]. In clinically defined HDGC
polymorphisms. patients, CDH1 germline mutations are detected
Genome-wide association studies have impli- in 30–40 % of cases [42]. Seventy-five to eighty
cated the prostate stem cell antigen (PSCA) gene percent of CDH1 mutations are truncating muta-
and the mucin 1 (MUC1) gene as GC susceptibility tions, and the remaining are missense mutations.
factors. Approximately 95 % of the Japanese popu- In addition, large germline deletions have also
lation have at least one of the two risk genotypes, been found in HDGC families which tested nega-
and approximately 56 % of the population have tive for point mutations [46].
both risk genotypes [40]. Hereditary GC accounts Another rare but so far the only reported alterna-
for 1–3 % of GC, and two hereditary syndromes tive to CDH1 inactivation in HDGC is the presence
have been described – hereditary diffuse gastric of germline α-E-catenin mutations [47]. Since α-E-
cancer (HDGC) and gastric adenocarcinoma and catenin functions in the same complex as E-cadherin,
proximal polyposis of the stomach (GAPPS). these results call attention to the broader signaling
network surrounding these proteins in HDGC.
A development model has been proposed for
Hereditary Diffuse Gastric Cancer diffuse type GC in CDH1 germline mutation carri-
(HDGC) ers encompassing foveolar hyperplasia, precursor
(intraepithelial) lesions (in situ carcinoma and pag-
On the basis of clinical criteria, the International etoid spread of signet ring cells), early intramuco-
Gastric Cancer Linkage Consortium defined sal carcinoma, and advanced cancer [48] (Fig. 6.6).
6 Pathogenesis of Gastric Cancer 69

Genetic counseling is an essential Gastric Cancer in Other Hereditary

component of the evaluation and management Cancer Syndromes
of HDGC patients and informed consent for
genetic testing is required [ 49]. The recom- The risk of GC is also increased in dominantly
mended youngest age at which to offer testing inherited cancer predisposition syndromes such
to relatives at risk is not well established. as familial adenomatous polyposis and Lynch
Rare cases of clinically significant diffuse syndrome, as well as in patients with Li–Fraumeni
type GC have been reported in affected fami- syndrome with germline mutations of TP53 [52].
lies before the age of 18, but the overall risk
of diffuse type GC before the age of 20 is very
low [49]. Molecular Pathology of Gastric
No therapies other than prophylactic total Carcinoma
gastrectomy are currently available for HDGC
patients. Since the penetrance of HDGC is Like most cancers, GC is the result of accumulated
>80 %, and analysis of gastrectomy speci- genomic changes affecting a number of cellular
mens suggests that microscopic foci of signet functions essential for cancer development: self-
ring cells are almost universally present in sufficiency in growth signals, escape from anti-
CDH1 germline mutation carriers even if the growth signals, resistance to apoptosis, sustained
endoscopic examination was unremarkable, replicative potential, angiogenesis induction, and
prophylactic gastrectomy should be strongly invasive or metastatic potential. These genomic
considered [42]. changes arise through three major pathways: mic-
rosatellite instability, chromosomal instability, and
a CpG island methylator phenotype.
Gastric Adenocarcinoma Furthermore, genetic and epigenetic changes
and Proximal Polyposis may affect oncogenes and tumor suppressor
of the Stomach (GAPPS) genes [53]. It seems that some oncogenes and
some tumor suppressor genes are preferentially
Recently, a new hereditary syndrome has been altered in a specific histological subtype of GC,
identified: gastric adenocarcinoma and proximal such as HER2, KRAS, APC, and DCC in intesti-
polyposis of the stomach, which is characterized nal type GC and BCL2, FGFR2 (formerly K-
by the autosomal dominant transmission of fundic SAM), CDH1, and RB1 in diffuse type GC. Other
gland polyposis including areas of dysplasia or oncogenes and tumor suppressor genes such as
intestinal type gastric adenocarcinoma restricted CTNNB1, MET, MYC, PTEN, and TP53 are
to the proximal stomach with no evidence of altered in both histological subtypes (for a review,
colorectal or duodenal polyposis or other herita- see Lauwers et al. 2010 [1]).
ble gastrointestinal cancer syndromes. This syn-
drome was originally identified in Australian and
North American families [50] but has also been MicroRNA
reported in Japanese families [51]. The genetic
defect behind this syndrome has not yet been Several miRNAs have been shown to be related to
identified. certain GC subtypes, GC progression, and potential
The clinical management of GAPPS families treatment targets, albeit with inconsistent results
must balance the limitations of endoscopic sur- probably related to small sample sizes [54, 55].
veillance, the patient-specific risk of morbidity
associated with prophylactic surgery, and the risk
of GC within the specific family. All first-degree Whole Genome Studies
relatives of affected patients should be advised to
have an upper gastrointestinal endoscopy and Modern high-throughput molecular methods are
colonoscopy [50]. being used with the aim to complement traditional
70 F. Carneiro and H.I. Grabsch

histopathological diagnosis and prognosis 5. Fukase K, Kato M, Kikuchi S, et al. Effect of eradica-
tion of Helicobacter pylori on incidence of metachro-
prediction in GC and to contribute to a better
nous gastric carcinoma after endoscopic resection of
understanding of the biology of GC at a molecu- early gastric cancer: an open-label, randomized con-
lar level. trolled trial. Lancet. 2008;372:392–7.
Zang et al. found an average of 50 mutations/ 6. Basso D, Zambon CF, Letley DP, et al. Clinical rele-
vance of Helicobacter pylori cagA and vacA gene
GC mostly affecting genes involved in cell adhe-
polymorphisms. Gastroenterology. 2008;135:91–9.
sion and chromatin remodeling and identified 7. Figueiredo C, Machado JC, Pharoah P, et al.
two new putative tumor suppressor genes, FAT4 Helicobacter pylori and interleukin 1 genotyping: an
and ARID1A [56]. opportunity to identify high-risk individuals for gas-
tric carcinoma. J Natl Cancer Inst. 2002;94:1680–7.
Tan et al. investigated the gene expression
8. Correa P. Helicobacter pylori and gastric carcinogen-
profile in a large number of GC cell lines and esis. Am J Surg Pathol. 1995;19 Suppl 1:S37–43.
identified two major genomic subtypes [57]. The 9. El-Omar EM, Carrington M, Chow W-H, et al.
genomic subtype classification was concordant Interleukin-1 polymorphisms associated with increased
risk of gastric cancer. Nature. 2000;404:398–402.
with the phenotypic classification according to
10. Machado JC, Figueiredo C, Canedo P, et al. A proin-
Laurén in 64 % of GC, suggesting that gene flammatory genetic profile increases the risk for
expression studies can lead to the identification chronic atrophic gastritis and gastric carcinoma.
of distinct GC subtypes which are not distin- Gastroenterology. 2003;125:364–71.
11. Fukayama M, Ushiku T. Epstein-Barr virus-associated
guishable based on histology.
gastric carcinoma. Pathol Res Pract. 2011;207:529–37.
Two very recent genome-wide copy number 12. World Cancer Research Fund/American Institute for
profiling studies using high-resolution SNP Cancer Research. Food, nutrition, physical activity,
arrays have demonstrated the power of modern and the prevention of cancer: a global perspective.
Washington, DC: American Institute for Cancer
molecular technology in identifying new clini-
Research; 2007.
cally relevant subtypes of GC [58, 59]. Both stud- 13. Ren JS, Kamangar F, Forman D, Islami F. Pickled food
ies identified independently that up to 37 % of and risk of gastric cancer – a systematic review and
GC show high amplifications of genes encoding meta-analysis of English and Chinese literature.
Cancer Epidemiol Biomarkers Prev. 2012;21:905–15.
druggable tyrosine kinase receptor proteins such
14. Gonzalez CA, Pera G, Agudo A, et al. Smoking and
as FGFR2, HER2, EGFR, and MET. Furthermore, the risk of gastric cancer in the European Prospective
these gene amplifications were almost always Investigation into Cancer and Nutrition (EPIC). Int
exclusive emphasizing the molecular heterogene- J Cancer. 2003;107:629–34.
15. Farinati F, Rugge M, Di Mario F, et al. Early and
ity of GC and the need to select patient’s treat-
advanced gastric cancer in the follow-up of moderate
ment based on molecular profiles. and severe gastric dysplasia patients. A prospective
study. I.G.G.E.D.–Interdisciplinary Group on Gastric
Epithelial Dysplasia. Endoscopy. 1993;25:261–4.
16. Rugge M, Correa P, Dixon MF, et al. Gastric dyspla-
References sia: the Padova international classification. Am J Surg
Pathol. 2000;24:167–76.
1. Lauwers GY, Carneiro F, Graham DY, Curado MP, 17. Schlemper RJ, Riddell RH, Kato Y, et al. The Vienna
Franceschi S, Montgomery E, Tatematsu M, Hattori classification of gastrointestinal epithelial neoplasia.
T. Gastric carcinoma. In: Bosman FT, Carneiro F, Gut. 2000;47:251–5.
Hruban RH, Theise ND, editors. WHO Classification 18. Dixon MF. Gastrointestinal epithelial neoplasia:
of Tumours of the Digestive System. 4th ed. Lyon: Vienna revisited. Gut. 2002;51:130–1.
IARC Press; 2010. p. 48–58. 19. Sobin LH, Gospodarowicz MK, Wittekind C. TNM
2. Howson CP, Hiyama T, Wynder EL. The decline in classification of malignant tumors. Oxford: Wiley-
gastric cancer: epidemiology of an unplanned tri- Blackwell; 2009.
umph. Epidemiol Rev. 1986;8:1–27. 20. Edge SB, Byrd DR, Compton CC, Fritz AG, Greene
3. Carneiro F. Stomach cancer. In: Stewart BW, Wild CP, FL, Trotti A. AJCC cancer staging manual. New York:
editors. World cancer report 2014. Lyon: IARC Press; Springer; 2009.
2014. p. 545–57. 21. The Paris endoscopic classification of superficial neo-
4. Ferlay J, Shin HR, Bray F, et al. Estimates of world- plastic lesions: esophagus, stomach, and colon:
wide burden of cancer in 2008: GLOBOCAN 2008. November 30 to December 1, 2002. Gastrointest Endosc.
Int J Cancer. 2010;127:2893–917. 2003;58:S3–43.
6 Pathogenesis of Gastric Cancer 71

22. Borrmann R. Handbuch der speziellen pathologischen endoscopic resection for undifferentiated-type early
Anatomie und Histologie. In: von Henke F, Lubarch O, gastric cancer. Gastric Cancer. 2009;12:148–52.
editors. IV/erster Teil. Berlin, Julius: Springer; 1926. 37. Tajima Y, Murakami M, Yamazaki K, et al. Risk fac-
p. 864–71. tors for lymph node metastasis from gastric cancers
23. Lauren P. The two histological main types of gastric with submucosal invasion. Ann Surg Oncol. 2010;17:
carcinoma: diffuse and so-called intestinal-type carci- 1597–604.
noma. An attempt at a histo-clinical classification. 38. Cunningham D, Allum WH, Stenning SP, et al.
Acta Pathol Microbiol Scand. 1965;64:31–49. Perioperative chemotherapy versus surgery alone for
24. JGCA. Japanese classification of gastric carcinoma, resectable gastroesophageal cancer. N Engl J Med.
3rd English edition. Gastric Cancer. 2011;14:101–12. 2006;355:11–20.
25. Nakamura K, Sugano H, Takagi K. Carcinoma of the 39. Kooby D, Suriawinata A, Klimstra DS, et al. Biologic
stomach in incipient phase: its histogenesis and histo- predictors of survival in node-negative gastric cancer.
logical appearances. Gann. 1968;59:251–8. Ann Surg. 2003;237:828–37.
26. Ono H. Early gastric cancer: diagnosis, pathology, 40. Saeki N, Ono H, Sakamoto H, Yoshida T. Genetic fac-
treatment techniques and treatment outcomes. Eur J tors related to gastric cancer susceptibility identified
Gastroenterol Hepatol. 2006;18:863–6. using a genome-wide association study. Cancer Sci.
27. Machado JC, Nogueira AM, Carneiro F, Reis CA, 2013;104:1–8.
Sobrinho-Simoes M. Gastric carcinoma exhibits dis- 41. Caldas C, Carneiro F, Lynch HT, et al. Familial gastric
tinct types of cell differentiation: an immunohisto- cancer: overview and guidelines for management.
chemical study of trefoil peptides (TFF1 and TFF2) J Med Genet. 1999;36:873–80.
and mucins (MUC1, MUC2, MUC5AC, and MUC6). 42. Fitzgerald RC, Hardwick R, Huntsman D, et al.
J Pathol. 2000;190:437–43. International Gastric Cancer Linkage Consortium.
28. Kushima R, Vieth M, Borchard F, Stolte M, Mukaisho Hereditary diffuse gastric cancer: updated consensus
K, Hattori T. Gastric-type well-differentiated adeno- guidelines for clinical management and directions for
carcinoma and pyloric gland adenoma of the stomach. future research. J Med Genet. 2010;47:436–44.
Gastric Cancer. 2006;9:177–84. 43. Frebourg T, Oliveira C, Hochain P, et al. Cleft lip/pal-
29. Shiroshita H, Watanabe H, Ajioka Y, Watanabe G, ate and CDH1/E-cadherin mutations in families with
Nishikura K, Kitano S. Re-evaluation of mucin phe- hereditary diffuse gastric cancer. J Med Genet. 2006;
notypes of gastric minute well-differentiated-type 43:138–42.
adenocarcinomas using a series of HGM, MUC5AC, 44. Kluijt I, Siemerink EJ, Ausems MG, et al. CDH1-
MUC6, M-GGMC, MUC2 and CD10 stains. Pathol related hereditary diffuse gastric cancer syndrome:
Int. 2004;54:311–21. clinical variations and implications for counseling. Int
30. Markar SR, Karthikesalingam A, Jackson D, Hanna J Cancer. 2012;131:367–76.
GB. Long-term survival after gastrectomy for cancer 45. Guilford P, Hopkins J, Harraway J, et al. E-cadherin
in randomized, controlled oncological trials: compari- germline mutations in familial gastric cancer. Nature.
son between West and East. Ann Surg Oncol. 1998;392:402–5.
2013;20:2328–38. 46. Oliveira C, Senz J, Kaurah P, et al. Germline CDH1
31. Noguchi Y, Yoshikawa T, Tsuburaya A, Motohashi H, deletions in hereditary diffuse gastric cancer families.
Karpeh MS, Brennan MF. Is gastric carcinoma differ- Hum Mol Genet. 2009;18:1545–55.
ent between Japan and the United States? Cancer. 47. Majewski IJ, Kluijt I, Cats A, et al. An α-E-catenin
2000;89:2237–46. (CTNNA1) mutation in hereditary diffuse gastric can-
32. Verdecchia A, Mariotto A, Gatta G, Bustamante- cer. J Pathol. 2013;229:621–9.
Teixeira MT, Ajiki W. Comparison of stomach cancer 48. Carneiro F, Huntsman DG, Smyrk TC, et al. Model of
incidence and survival in four continents. Eur J Cancer. the early development of diffuse gastric cancer in
2003;39:1603–9. E-cadherin mutation carriers and its implications for
33. Bollschweiler E, Boettcher K, Hoelscher AH, Sasako patient screening. J Pathol. 2004;203:681–7.
M, Kinoshita T, Maruyama K, Siewert JR. Is the prog- 49. Oliveira C, Seruca R, Hoogerbrugge N, Ligtenberg
nosis for Japanese and German patients with gastric M, Carneiro F. Clinical utility gene card for:
cancer really different? Cancer. 1993;71:2918–25. Hereditary diffuse gastric cancer (HDGC). Eur J Hum
34. Reid-Lombardo KM, Gay G, Patel-Parekh L, Ajani Genet. 2013;21(8), doi:10.1038/ejhg.2012.247.
JA, Donohue JH. Treatment of gastric adenocarci- 50. Worthley DL, Phillips KD, Wayte N, et al. Gastric
noma may differ among hospital types in the United adenocarcinoma and proximal polyposis of the stom-
States, a report from the National Cancer Data Base. ach (GAPPS): a new autosomal dominant syndrome.
J Gastrointest Surg. 2007;11:410–20. Gut. 2012;61:774–9.
35. Yamao T, Shirao K, Ono H, et al. Risk factors for 51. Yanaru-Fujisawa R, Nakamura S, Moriyama T, et al.
lymph node metastasis from intramucosal gastric car- Familial fundic gland polyposis with gastric cancer.
cinoma. Cancer. 1996;77:602–6. Gut. 2012;61:1103–4.
36. Hirasawa T, Gotoda T, Miyata S, et al. Incidence of 52. Sereno M, Aguayo C, Guillén Ponce C, et al. Gastric
lymph node metastasis and the feasibility of tumours in hereditary cancer syndromes: clinical
72 F. Carneiro and H.I. Grabsch

features, molecular biology and strategies for preven- 57. Tan IB, Ivanova T, Lim KH, et al. Intrinsic subtypes
tion. Clin Transl Oncol. 2011;13:599–610. of gastric cancer, based on gene expression pattern,
53. Carneiro F, Oliveira C, Leite M, Seruca R. Molecular predict survival and respond differently to chemo-
targets and biological modifiers in gastric cancer. therapy. Gastroenterology. 2011;141:476–85. 485.
Semin Diagn Pathol. 2008;25:274–87. e1–11.
54. Wang F, Sun GP, Zou YF, Hao JQ, Zhong F, Ren WJ. 58. Deng N, Goh LK, Wang H, et al. A comprehensive
MicroRNAs as promising biomarkers for gastric survey of genomic alterations in gastric cancer reveals
cancer. Cancer Biomark. 2012;11:259–67. systematic patterns of molecular exclusivity and co-
55. Yin Y, Li J, Chen S, Zhou T, Si J. MicroRNAs as diag- occurrence among distinct therapeutic targets. Gut.
nostic biomarkers in gastric cancer. Int J Mol Sci. 2012;61:673–84.
2012;13:12544–55. 59. Dulak AM, Schumacher S, van Lieshout J, et al.
56. Zang ZJ, Cutcutache I, Poon SL, et al. Exome Gastrointestinal adenocarcinomas of the esophagus,
sequencing of gastric adenocarcinoma identifies recur- stomach and colon exhibit distinct patterns of genome
rent somatic mutations in cell adhesion and chromatin instability and oncogenesis. Cancer Res. 2012;72:
remodeling genes. Nat Genet. 2012;44:570–4. 4383–93.
 Standards for Surgical Therapy
of Gastric Cancer 7
Roderich E. Schwarz

Gastric cancer (GC) continues to represent a Surgical Objectives

formidable health challenge worldwide based on
its unaltered high incidence in certain geographic General objectives of operative therapy for gas-
areas such as East Asia or South America, its tric cancer are more easily compiled than suc-
commonly advanced stage at diagnosis, and its cessfully accomplished on a consistent basis in
limited curability for disease in intermediate and clinical practice [4]. They include surgical
advanced stages [1]. Even in the United States, removal of tumor tissue, the provision of local
where gastric adenocarcinoma mortality has grad- and regional disease control, the optimization of
ually decreased from the most common form of curative potential, the provision of intraoperative
cancer-related deaths in the 1940s and gastric and pathologic staging information which occa-
cancer incidence is among the lowest in the world, sionally includes the confirmation of a gastric
curability remains a significant problem [2]. cancer diagnosis, the restoration of function lost
Before the onset of surgical therapy, gastric can- or limited through a resection such as reestablish-
cer was incurable. Since the first groundbreaking ing gastrointestinal continuity after gastrectomy,
accomplishments with partial gastrectomy by and minimizing any resulting postoperative mor-
Billroth in 1881, Y-jejunostomy reconstruction by bidity. The latter objective, for example, would
Roux in 1893, and total gastrectomy by Schlatter include strategies to avoid splenectomy or distal
in 1897, operative therapy of gastric malignancy pancreatectomy if possible to reduce infectious
has gone through more than a century of contin- morbidity or to furbish gastric rather than esoph-
ued refinement and ever-improving accomplish- ageal anastomoses when feasible to minimize
ments (Table 7.1) [3]. This chapter intends to anastomotic leaks. While all listed objectives
highlight the critical objectives, indications, and appear equally valid in a minimally invasive sur-
standard techniques for operative procedures in gery (MIS) context, improved recovery potential
gastric cancer treatment and to describe the posi- and minimized morbidity obviously carry special
tion of surgical therapy within the context of mul- appeal for a MIS rationale and approach to gas-
tidisciplinary approaches for mid-stage and tric cancer treatment.
metastatic stomach malignancy.

Operative Intent
R.E. Schwarz, MD, PhD, FACS (*)
Department of Surgical Oncology, Indiana University
The intent to conduct an operation for gastric
Health Goshen Center for Cancer Care,
200 High Park Ave, Goshen, IN 46526, USA cancer can be highly variable. In most cases, a
e-mail: [email protected] procedure is justified to cure the underlying

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 73
DOI 10.1007/978-3-319-09342-0_7, © Springer International Publishing Switzerland 2015
74 R.E. Schwarz

Table 7.1 Important steps in the historic development of a resection is able to accomplish [5]. In any
operative gastric cancer therapy
circumstance, a curative outcome after resection
Year of cannot be expected unless all known locoregional
procedure or Operative disease is completely removed, usually en bloc,
Surgeon(s) publication accomplishment
and is generally not possible if diffuse extrare-
Pean 1879 Unsuccessful pyloric
resection gional metastatic disease does exist. Even if the
Rydygier 1880 Unsuccessful pyloric complete removal of all gross disease with nega-
resection tive margins (R0 resection) has been performed,
Billroth 1881 First successful pyloric subsequent recurrence remains common for gas-
resection (Billroth I) tric cancers of mid-stage due to the presence of
Kocher 1893 Posterior nonvisualized micrometastases at the time of
gastroduodenostomy
operation [6]. This mechanism and the fact that
Billroth 1885 Antrectomy after loop
gastrojejunostomy previously undetected metastatic disease is iden-
(2-stage, Billroth II) tified intraoperatively are the most common
Krönlein 1887 End-to-side reasons if a preoperative curative intent cannot be
gastrojejunostomy achieved [4]. Macroscopically visible residual
Woelfler 1881 Y-gastroenterostomy disease and positive peritoneal cytology are
Roux 1893 Retrocolic virtual guarantees for symptomatic disease recur-
Y-gastrojejunostomy
rence to develop [7]. Microscopic positive
Connor 1884 Unsuccessful total
gastrectomy margins (R1 status) impart an increased local
Schlatter 1897 First successful total recurrence risk, but are in addition a surrogate for
gastrectomy higher-risk disease and a greater failure rate in
Brigham, 1898 Three successful total extraregional sites [8]. In addition to the curative
Richardson gastrectomies intentions, a diagnostic component or the provi-
Hoffmeister 1908 Greater curvature
sion of tumor tissue for specific purposes can
gastrojejunostomy
Reichel–Polya 1911 Full length
provide the rationale to operate on a patient with
gastrojejunostomy gastric cancer, specifically if the diagnosis is sus-
McNeer 1951 Radical gastrectomy, pected but remains unconfirmed through endo-
extended scopic biopsy means, or if more advanced
lymphadenectomy intra-abdominal disease extent is suspected but
Appleby 1953 Radical en bloc not confirmed through imaging modalities.
gastrectomy with
resection of celiac Another common preoperative intent is the palli-
artery ation of symptoms that cannot be alleviated
Hunt 1952 Pouch through lesser invasive means such as endoscopy
esophagojejunostomy or interventional radiologic techniques [9].
reconstruction
Examples for this approach are obstruction
Merendino 1955 Small bowel
interposition
symptoms not relieved through stent placement
reconstruction or resection needs for tumor-related bleeding not
Kitano 1992 Laparoscopically amenable to palliative radiation or interventional
assisted distal vascular manipulation. In this context it is impor-
gastrectomy tant not to confuse the terms “palliative” and
Azagra 1996 Laparoscopic total
“noncurative”; a noncurative operative procedure
gastrectomy
is hardly ever justifiable in a patient who does not
suffer from symptoms that require a specific sur-
malignancy. Curability criteria for surgical gical intervention, while an operation with pallia-
therapy depend on the underlying disease extent tive intent is primarily driven by the patient’s
and biologic behavior and are set by the relatively symptoms irrespective of whether potential
limited scope of local and regional tumor control curability is still given or not [10]. Therefore,
7 Standards for Surgical Therapy of Gastric Cancer 75

Table 7.2 Preoperative intents to provide operative therapy for gastric cancer
Intent Examples Comments on requirements or conditions
Diagnostic Diagnostic laparoscopy Enhances clinical staging either prior to
induction therapy or at beginning of planned
resection; rarely required to prove and treat
suspected gastric cancer that failed
endoscopic biopsy confirmation attempts
Curative Gastrectomy with D2 lymphadenectomy Requires absence of extraregional
metastases; all multimodality options
considered; goal not achieved through R2
and most R1 resections; need for symptom
control may affect timing of resection
Palliative Intestinal bypass for malignant peritoneal Nonoperative or less invasive options always
bowel obstruction, gastrojejunostomy bypass preferred if feasible; possible benefits to
resection of tumor reported only in low
tumor burden settings; although most often
in noncurative setting, palliative-intent
gastrectomy can result in curative procedure
if disease extent is smaller than expected
Noncurative Gastrectomy for asymptomatic stage IV Cannot be supported or justified without
tumor other compelling intents documented
Preemptive Resection of gastric tumor to prevent Hardly ever indicated; should not prompt a
obstruction in setting of metastatic disease separate planned operation
Supportive Surgical feeding jejunostomy tube placement Nonoperative or less invasive means
prior to preoperative therapy preferred if possible
Tissue provision Resection of gastric cancer tissue for Hardly ever indicated; less invasive
on-protocol vaccine generation nonoperative means preferred

preoperative intents for operative therapy for overall survival (OS) compared to gastrectomy
gastric cancer can exist in single or in combined alone [11]. Adjuvant therapy options with OS
form (Table 7.2). The surgeon is advised to benefits and particular relevance to practice
clearly define preoperative intents, for guidance within the United States include postoperative
of the informed consenting process with patient chemotherapy with chemoradiation according
and family members, for appropriate positioning to the Intergroup 0116 trial [12], perioperative
of the operative step within the sequence of mul- chemotherapy analogous to the MAGIC or
tidisciplinary treatment options, as well as for ACCORD07 trials [13, 14], or preoperative che-
enabling correct interpretation of outcomes. moradiation analogous to the CROSS trial [15].
Preoperatively clearly defined palliative or diag- Details on these multimodality treatment options
nostic intents for operations have a greater chance exceed the scope of this chapter. As there is cur-
to be successfully achieved compared to proce- rently no single, evidence-based approach to
dures performed with curative intent [4]. multimodal GC therapy, various regimens are in
use based on local centers’ expertise and pre-
ference. In general, preoperative (neoadjuvant)
Operative Therapy as Part approaches are preferred, as tolerance to
of a Multidisciplinary Strategy treatment is greater, delivery is more likely com-
plete, and as clinical or pathologic response to
Due to the high risk for recurrence after resection such treatment may represent an important prog-
of mid-stage GC, additional treatment options nostic surrogate for disease behavior and future
have been increasingly applied. In numerous recurrence risk [16, 17]. Perioperative chemo-
phase 3 randomized controlled trials, adjuvant therapy appears to be most useful for mid- and
therapy has been demonstrated to lead to superior distal third gastric tumors, while preoperative
76 R.E. Schwarz

chemoradiation may be preferred for proximal metastases [22, 23]; laparoscopic ultrasound may
gastric or GE junction lesions. Importantly, any slightly increase metastasis detection rates [24].
operation plans would have to be balanced In addition, peritoneal washing cytology may be
against these important strategies, especially for considered if subsequent treatment steps are
curative goals, and formal multidisciplinary eval- affected by positive results [25]. Timing or fre-
uation of appropriate treatment options prior to quency of staging laparoscopy around preopera-
initiation of therapy should be mandatory. tive therapy is being debated [26]. Patients with
“Surgical” therapy of GC therefore includes persisting positive washing cytology findings
knowledge of and support for multimodality invariably have poor OS outlook, while those
treatment and the insight to adapt to effects of with positive peritoneal cytology status that
other treatments, especially regarding assessment turned negative have shown longer survival [7].
of tumor response to preoperative therapy and
delineation of an appropriate resection extent.
Standards for Curative Mid-Stage
Gastric Cancer Operative Therapy
Preoperative Aspects
Technical Aspects of Resection
Most patients will present to the surgeon with
biopsy-proven adenocarcinoma through endo- State-of-the-art curative-intent gastrectomy
scopic means. Accurate clinical staging includes requires R0 resection and should be accompanied
computed tomography imaging and endoscopic by an extended lymphadenectomy (D2 dissec-
ultrasound (EUS) evaluation. It is important to tion) [27, 28]. Whether open or minimally inva-
have precise documentation in regard to primary sive surgical (MIS) techniques are utilized
tumor location and extent prior to initiation of appears to be of lesser consequence oncologi-
preoperative therapy, as responses to this treat- cally, as long as principles of complete local
ment may render intraoperative localization resection and regional dissection are adhered to
attempts difficult. PET scans do not appear man- [29–32]. The following operative components
datory for GC staging, but may have a more reli- are based on open gastrectomy standards, but
able role in proximal or GE junction primaries or seem to be equally relevant for a MIS approach.
to guide preoperative chemotherapy on protocol; For early GC (T1N0), endoscopic mucosal resec-
diffuse-type GCs tend to be less well imaged on tion (EMR) or submucosal dissection (ESD) may
PET scans [18, 19]. Resectable tumors are best suffice as definitive therapy [33]; both require
approached in terms of resection extent based on proper specialty skills and currently appear to be
their pretreatment extent and stage, irrespective limited to few centers within the United States
of restaging findings. Even major clinical with appropriate technical and clinical expertise.
responses are often incomplete on pathologic EMR and ESD techniques will not be described
examination, supporting this more “radical” in further detail within this chapter. For all more
approach [20, 21]. An exception would be the advanced stages of nonmetastatic gastric adeno-
rare scenario of an unresectable tumor being ren- carcinoma, complete locoregional resection is
dered resectable due to a response to initial che- the central component for curative-intent therapy.
motherapy or radiation. The intraoperative In the operating room, general endotracheal anes-
specifics are thus best delineated preoperatively, thesia is introduced, and the patient is usually
including planned operative approach (open ver- placed in a supine position for a planed open celi-
sus laparoscopic), placement of incision(s), otomy; planned laparoscopic resection may favor
resection extent, and preferred reconstruction. different positions based on the operating sur-
Staging laparoscopy is strongly recommended as geon’s preference. It may be helpful to consider a
an operative complement to preoperative imag- short repeat upper gastrointestinal endoscopy
ing, as it is most sensitive in detecting after induction of anesthesia prior to resection or
small-volume peritoneal or visceral surface later for anastomotic assessment [34]; the author
7 Standards for Surgical Therapy of Gastric Cancer 77

has used this liberally to verify tumor location adequate. For lesions in the middle third of the
and extent and to assess the mucosal appearance stomach, the decision between total or near-total
of gastric or esophageal components to be used in gastrectomy depends on the proximal margin sta-
the reconstruction or for anastomotic sites. In tus and considerations for possibly safer recon-
addition, laparoscopy should be performed now struction (gastrojejunostomy leak rates have been
unless already done in a separate setting. In up to described as occurring half as often as those after
20 % of cases, laparoscopic confirmation of esophagojejunostomy [42]). Proximal third lesions
intra-abdominal metastases will still provide the will essentially always require either total gastrec-
opportunity to avoid an otherwise noncurative tomy or proximal gastrectomy with a special
gastrectomy in this setting. reconstruction such as small bowel interposition
A transabdominal approach will be sufficient [43]. Proximal gastrectomy with subsequent
for most complete resections, but incision place- esophagogastrostomy is not recommended, espe-
ment for open gastrectomy is not standardized and cially after pyloroplasty, for concerns of signifi-
follows personal preferences. While many sur- cant reflux. Avoiding any pyloromyotomy or
geons choose upper midline incisions, the author pyloroplasty in this setting is recommended, but
prefers a bilateral subcostal margin incision does not completely preempt reflux-related prob-
approach. Rarely is there benefit to a combined lems; distal gastric emptying problems that require
left thoracoabdominal incision, but for high, large endoscopic or even operative management may
gastric tumors in obese patients, this can generate occur in 5–15 % of cases. Lesions at the GE junc-
much superior exposure if needed. A routine tho- tion require special operative planning based on
racoabdominal approach for GC resection is not the lesions’ epicenter and, more importantly, the
beneficial compared to the transabdominal-only proximal disease extent. Siewert type I lesions
access and thus not recommended [35, 36]. With require a transthoracic or transhiatal esophagec-
proper exposure and resectability established, the tomy and should not be approached with an
main resective objectives are R0 resection and attempt to perform a gastrectomy [44]. Siewert
lymphadenectomy. Total gastrectomy out of prin- type II lesions are located at the gastric cardia;
ciple is not necessary; lesser extent resections, these can either be approached via esophagogas-
especially for distally located tumors, have shown trectomy with retrogastric LND analogous to type
comparable survival results, with fewer morbidity I lesions or through an extended gastrectomy as
and functional challenges [37, 38]. Appropriate long as not more than 3 cm of distal esophageal
macro- and microscopically negative margins involvement exists and proximal negative margins
should be obtained as feasible at duodenal and (of 2 cm or greater) can be obtained [45]. Siewert
esophageal resection sites. In challenging scenar- type III lesions are in biologic terms proximal gas-
ios of advanced disease burden, it can be accept- tric cancers, and a transabdominal approach
able to leave a positive margin at these sites, as should be fully sufficient as long as no more exten-
long as parameters such as serosal involvement or sive submucosal esophageal involvement exists
significant nodal burden imply a minimal curative [44, 45]. It appears permissible to decide upon the
potential. Intragastric margins of 5 cm are tradi- best resection extent for proximal gastric cancers
tionally recommended for subtotal gastrectomy, at close to the GEJ intraoperatively through esopha-
least for intestinal-type disease [39, 40]; diffuse- geal transection and frozen section analysis, as
type lesions may require wider margins. A healthy long as the surgeon is experienced with perform-
tissue esophageal margin length of 2 cm seems to ing an esophagectomy in this setting and prepared
be sufficient for resection of Siewert type II and III to do so if necessary and as long as right gastric
lesions treated with gastrectomy [41]. The choice and gastroepiploic vasculature is initially pre-
of gastrectomy extent (and of lymphatic dissection served for a gastric tube reconstruction in case an
extent) will not only depend on location and extent esophageal resection becomes necessary.
of the primary tumor but also on potential recon- Total or near-total omentectomy is frequently
struction needs and options (Fig. 7.1). In general, performed en bloc with a gastrectomy for cancer
for distal lesions a subtotal gastrectomy is and represents a good way to initiate the
78 R.E. Schwarz

dissection. Omental bursectomy has been widely oncologic standpoint, especially for transmural
applied as a means to accomplish more complete tumors with serosal involvement and progression
resection of posterior wall lesions; it includes risks [46, 47], it nevertheless appears to be a use-
removal of the anterior peritoneal leaf of the ful technique to identify the relevant retroperito-
mesocolon in an attempt to not enter the lesser neal plains above the pancreas for identifying
sac and completely remove this retrogastric lymph nodes at hepatic and splenic arteries.
structure. While it appears less sensible from an Careful attention is applied to not injure the

a Tumor location Resection option

b Tumor location

Resection options

Fig. 7.1 Schematic

representation of gastric
resection extent based on the
location of the primary
adenocarcinoma.
(a) Resection extent for distal
third tumors; (b) resection
extent options for middle
third tumors; (c) resection
extent options for proximal
third tumors including
Siewert type III lesions;
(d) resection extent options
for Siewert type II lesions
7 Standards for Surgical Therapy of Gastric Cancer 79

Tumor location
c

Resection options

Tumor location
d

Resection options

Fig. 7.1 (continued)

80 R.E. Schwarz

pancreas parenchyma in the process. For all necessary and indicated when resulting in a R0
gastric tumors except those close to the GE resection that still offers curative potential. In this
junction, the proximal duodenum is freed and situation, the surgeon ought to be prepared to per-
prepared for transection; in this process, dissec- form an en bloc segmental hepatectomy, dia-
tion of gastroepiploic LNs off the underlying phragmatic resection, pancreatosplenectomy, left
pancreas and deep ligation and transection of adrenalectomy, or colectomy as required.
gastroepiploic vessels will keep the inferior para-
pyloric and gastroepiploic (level 6) LNs on the
specimen and will allow for easy access to the Additional Aspects of Lymph
duodenum. The dissection is now carried from Node Dissection
distal to proximal, with division of lesser omen-
tum and mobilization of paragastric tissues at the The propensity of gastric adenocarcinomas to
lesser curvature up to the diaphragmatic crus. If involve lymph nodes (LNs) is high. Although
the extended LND is to be performed en bloc, actively debated over the past decade, lymphade-
common hepatic artery LNs are now mobilized nectomy at the time of curative-intent gastrectomy
and kept with the specimen. The origin of the left has shown benefits to staging accuracy and to can-
gastric artery should always be identified and cer control and has thus become standard of care
divided for cancer resections; splenic artery [28, 52]. Resection of the appropriate paragastric
nodes are dissected away from pancreas and and of second echelon (left gastric, common
artery, and short gastric vessels are divided close hepatic, splenic, celiac artery) LNs (D2 dissection)
to the spleen. The spleen can most frequently be is generally sufficient; wider dissections have not
preserved unless direct tumor involvement or a shown superior results [35]. This procedure should
large hilar LN burden requires splenectomy. yield at least 15 or more LNs for the pathologic
Splenic hilar LN involvement is rare for tumors evaluation, but greater total LN counts have been
not located at the fundus or proximal two thirds associated with better survival outcomes [53–55].
of the greater curvature. Even when splenic hilar A long-term survival or disease-specific control
dissection is desired in fundus or greater curva- benefit to extended LN dissection (ELND) has
ture primaries, spleen-preserving hilar LN dis- now been demonstrated in at least two randomized
section has been applied, since spleen preservation controlled trials, despite a greater early morbidity
may have important benefits for reduced postop- and mortality in the Dutch trial after D2 dissection
erative morbidity [48–51]. The proximal transec- [28, 52, 56]. These were related to an increased
tion is now determined based on anticipated rate of pancreatosplenectomy with D2 dissection
margin needs. This is either at the level of the [57], but this survival hazard has been superseded
distal esophagus or transgastric with preservation by a long-term overall survival benefit due to
of the proximal stomach if feasible. In the latter greater disease control. As discussed earlier, sple-
scenario, the lesser curvature transection should nectomy and distal pancreatectomy are strongly
extend close to the GE junction without narrow- discouraged unless deemed necessary based on
ing the esophagogastric passage, primarily to tumor involvement [58, 59].
support a complete left gastric artery LND, while ELND can be performed en bloc with the gas-
more length can be preserved toward the greater trectomy as described above, or in a separate
curvature if possible. This then shall allow for an specimen. The paragastric nodes (i.e., paracar-
easier reconstruction, with a subsequent anasto- dial, lesser and greater curvature, right gastric
mosis close to the greater curvature transection artery, and gastroepiploic artery LNs) are always
site. Completion of the retroperitoneal dissection best removed with the adjacent stomach portion.
with celiac lymphadenectomy and clearance of Since the LN group to be removed is variable
tissues to the diaphragmatic crural tissue com- based on the tumor location, a good strategy is to
pletes the gastrectomy. For locally advanced remove any paragastric LNs adjacent to stomach
tumors, multivisceral resections are occasionally that is also to be removed. Dissection of the
7 Standards for Surgical Therapy of Gastric Cancer 81

a b

d e

Fig. 7.2 Intraoperative images of a 2-step extended artery; (c) Completion of retroperitoneal lymphadenec-
lymphadenectomy and subsequent reconstruction (a) tomy at celiac, hepatic, and splenic arteries. CHA com-
Appearance of the left gastric artery pedicle during resec- mon hepatic artery, SA splenic artery, SV splenic vein; (d)
tion of a proximal gastric cancer; (b) Appearance after Completed esophagojejunostomy; (e) Completed jejuno-
transection of the left gastric artery pedicle and proximal gastrostomy between small bowel (Merendino) interposi-
gastrectomy. CHA common hepatic artery, SA splenic tion and distal remnant stomach

named artery LNs will then complete a sensible tomy as initial step, to be followed by the
D2 dissection. If these left gastric, common retroperitoneal dissection of these structures as
hepatic, splenic, and celiac artery LNs do not second step (Fig. 7.2). This allows not only for
appear grossly abnormal, the author has divided better exposure but also improved pathologic
the left gastric artery pedicle to facilitate gastrec- identification of relevant retrogastric LN involve-
82 R.E. Schwarz

ment. The left gastric artery should generally be between the esophagus and distal gastric reser-
divided in cancer resections, in part for better voir and the avoidance of pyloric manipulation
nodal clearance; occasionally, an accessory left if possible present acceptable options, as shown
hepatic artery is encountered that can be pre- in (Fig. 7.2) [43]. As a general important aspect,
served, as LNs can be dissected around the proxi- reconstruction preferences should not compro-
mal left gastric artery, and the gastric branch can mise the resection extent. Pouch reconstructions
be divided after separating from the hepatic are rarely performed in the United States as
branch. In most Western patients, it is not possi- there has been no convincing evidence of post-
ble to identify all LNs of interest visually during operative nutritional superiority; some reports
the dissection. The goal is therefore to free the describe a potential long-term quality of life
relevant and named arterial vasculature of all sur- benefit [67, 68].
rounding lympho-areolar and adipose tissue,
rather than obtain specific LNs or a certain total
number of LNs. LN counts are determined by the Additional Intra- and Postoperative
pathologist and do not only reflect radicality of Considerations
dissection, but also quality of the specimen
pathologic examination, and other clinicopatho- Considerable variability and different preferences
logic factors including preoperative therapy exist regarding technical details of operative
effects and nutritional implications. A median aspects during gastrectomy. This applies to
total LN count between 20 and 30 appears to be anastomotic techniques, duodenal stump clo-
an acceptable standard [27, 53, 54]. In some sure, dissection techniques using sharp tools,
Asian centers, limiting the LND in patients with traditional electrocoagulation, or newer energy
low likelihood for LN involvement is being devices and extends to details of incision clo-
explored, such as through laparoscopic sentinel sure and others. In general, no specific tech-
LN biopsy for early GCs [60, 61], but these tech- nique has demonstrated clear and universally
niques are not yet accepted as proven standards. accepted evidence of superiority over others,
despite numerous trial or meta-analysis-based
efforts. The author prefers hand-sewn inversion
Technical Aspects of Reconstruction of the duodenal staple line closure, hand-sewn
dual-layer anastomoses between the esophagus
Most gastric resections are followed by Roux- or stomach remnant and jejunum, and intraop-
en-Y jejunal reconstruction, either as esophago- erative integrity testing of proximal anastomoses
jejunostomy or gastrojejunostomy (Fig. 7.3). through orogastric/orojejunal tube instillation of
The jejunal limb is best created with a length of methylene blue-containing saline solution. After
around 45 cm to achieve the lowest degree of total gastrectomy, postoperative nasojejunal
both Roux-stasis and of dumping problems decompression is unnecessary [69, 70]; with a
postoperatively [62]. Billroth 1 and 2 recon- significant-size gastric remnant, temporary naso-
structions have been described after distal gas- gastric decompression may be considered. There
trectomy, but appear acceptable regarding appears to be no benefit to routinely placed drains
appropriate oncologic dissection extent and despite some divergent clinical results, but spe-
functional outcomes only for very distally cial indications for intraoperative drainage may
located tumors [63, 64]. A potentially challeng- exist such as after partial pancreatic resection or
ing scenario for either reconstruction technique in case of a transhiatal high esophageal anasto-
is that of a small proximal gastric reservoir with mosis in a setting of having entered the pleural
uncontrolled access of biliary small bowel con- space during the dissection [71–73]. Placement
tents and the related bile reflux risk [65, 66]. of feeding tubes for postoperative nutrition
Similarly, after proximal gastrectomy, a small support is equally debatable [74, 75]. It is the
distal reservoir too and biliary reflux have to be author’s practice to always provide jejunal feed-
avoided. A Merendino small bowel interposition ing access to patients undergoing esophagectomy
7 Standards for Surgical Therapy of Gastric Cancer 83

or total or near-total gastrectomy, but to use them the surgical planning for best postoperative
selectively in the rare cases of distal gastrectomy recovery.
based on the patient’s nutritional risk status [76].
While most patients do not require postoperative
enteral nutrition support, any failure of sufficient Surgical Palliation Aspects
oral food intake within 1–2 weeks and severe
preoperative malnutrition render the initiation of Surgeons frequently are called upon to decide on
tube feeding unproblematic with a feeding tube the most appropriate way to palliate symptoms of
available. Other means of standardized postop- GC. For mid-stage and potentially curable dis-
erative management including venous thrombo- ease, obstructive symptoms caused by the pri-
embolic prophylaxis, incentive spirometry, early mary tumor may influence the therapy sequence,
activation, cardioprotective therapy, etc. complete with the resection performed up front to address

a
Post-resection status Reconstruction options

Post-resection status Reconstruction

Fig. 7.3 Reconstruction options after subtotal gastrectomy, reconstruction option after total gastrectomy; (d) preferred
total gastrectomy, or proximal gastrectomy. (a) Billroth II reconstruction option after proximal gastrectomy; (e) gastric
reconstruction options after distal gastrectomy; (b) recon- pull-up reconstruction after esophagogastrectomy
struction option after near-total gastrectomy; (c) preferred
84 R.E. Schwarz

Fig. 7.3 (continued) Post-resection status Reconstruction

Post-resection status Reconstruction

Post-resection status Reconstruction

e
7 Standards for Surgical Therapy of Gastric Cancer 85

symptom control needs and complete resection.

In cases of bleeding from the primary tumor, 13.8 Locoregional Peritoneal 13.1
short course radiation has been an effective
7.1 3.0 8.4
palliative option [77–79], and even chemother-
apy may control the low continuous blood loss 0.6
associated with larger, ulcerated tumors [80]. For
3.1 1.1
symptom control in settings of incurable disease,
operations should generally be avoided if possi-
ble. Nonoperative treatment can also improve Distant, hematogenous 14.6
mild symptoms, and specific nonoperative inter- 9.8
ventions such as endoscopic stenting, tumor
reduction, or bleeding management may result in
Fig. 7.4 Failure pattern after gastrectomy for gastric can-
the desired control [81, 82]. Obstruction due to
cer. Graphic representation of failure patterns after gas-
large intragastric tumor burden in the setting of trectomy and D2 dissection for gastric cancer, without
metastatic disease provides great challenges. A routine use of adjuvant therapy. Data pooled from three
palliative-intent gastrectomy under these circum- series [91–93]. All numbers represent % values based on
total patient n = 2,753; recurrences: n = 909 (33 %)
stances may be indicated, but treatment goals are
rarely reported well [9], outcomes are frequently
disappointing, and benefits above available sys- remains a significant challenge, as recurrence
temic and supportive therapies are unproven rates are high. Peritoneal recurrence is common
despite some retrospective reports of more effec- among patients with T4 primaries, and significant
tive palliation and longer survival in highly LN involvement correlates with hematogenous
selected patients [83–85]. Success after palliative- metastasis and recurrence in distant sites [58];
intent gastrectomy also depends strongly on the isolated local recurrences appear rare (Fig. 7.4)
overall disease burden and pattern [86]. In cases [91, 92]. Overall survival after resections alone
of distal gastric obstruction, gastrojejunostomy appears to be primarily dependent on whether
may succeed and allow for avoiding a more com- serosal invasion of the primary tumor or nodal
plex resection. Malignant bowel obstruction due involvement is present (Fig. 7.5). Thus, TNM
to peritoneal carcinomatosis presents another staging criteria remain the dominant prognostic
scenario for which a palliative operation may be components after gastrectomy alone within
required, but where outcomes frequently fall nomograms for disease-specific survival [94, 95].
short of the desired goal [87]. In this case, bypass However, response to preoperative therapy is
or drainage procedures may be more feasible another powerful prognostic parameter and likely
than resection. a surrogate for favorable biologic behavior, as
metabolic and pathologic responses are linked to
best survival outcomes [16, 17]. It is possible that
Postoperative Outcomes the recently observed improvement in postgas-
trectomy survival is due to increased use of adju-
Postoperative morbidity after gastrectomy for vant therapy options (Fig. 7.5b). Postoperative
GC remains formidable but manageable for most chemotherapy with radiation has led to a survival
patients. Anastomotic leaks are linked to most benefit [12], and contemporary perioperative che-
deep site infections, and no specific reconstruc- motherapy such as in the MAGIC trial has
tion technique has emerged as superior in pre- improved long-term survival by roughly 10 %
venting leaks. Postoperative infections have been [13]. For proximal cancers including those of the
linked to inferior survival outcomes [88]. There GE junction, preoperative chemoradiation (as in
are well-established volume-outcome relation- the CROSS trial) has demonstrated survival ben-
ships for postoperative mortality as well as over- efits over surgical resection alone [15], with a sig-
all long-term survival [89, 90]. Disease control nificant reduction in locoregional and peritoneal
86 R.E. Schwarz

Fig. 7.5 Survival outcomes a

after gastrectomy for gastric
1 p < 0.0001
cancer. (a) OS after
curative-intent gastrectomy.
MSKCC data from the era .8 T1-3N0 (n = 201)
prior to widespread adjuvant

Cum. survival
therapy use, data from .6
Schwarz et al. [58]. (b) OS
after gastrectomy, by time T1-3Npos (n = 157)
.4
period. SEER data (Courtesy
of R. Nelson, Ph.D., 2014).
T4N0 (n = 100)
(c) Survival outcomes in three .2
key trials of adjuvant therapy
(Adjuv.) in addition to surgical T4Npos (n = 530)
0
resection alone (Surg. only)
of gastric or GE junction
0 24 48 72 96 120 (months)
cancer. The bars represent b
5-year overall survival data
1.0 1992−1997 (n = 6031) MS=2.3 years, 3-years=43 %, 5-years=34 %
(in %) after gastrectomy with 1998−2003 (n = 9028) MS=2.4 years, 3-years=45 %, 5-years=36 %
and without perioperative ECF 0.9 2004−2010 (n = 10967) MS=3.0 years, 3-years=50 %, 5-years=40 %
chemotherapy from the
0.8
MAGIC trial [13], 3-year
Survival probability

recurrence-free survival (in %) 0.7

after gastrectomy with and
0.6
without postoperative
5FU-LV chemotherapy and 0.5
chemoradiation from the
0.4
Intergroup 0116 trial [12],
and median overall survival 0.3
(in months) after
0.2
esophagogastrectomy with
and without preoperative 0.1
chemoradiation for GE Log-rank p-value=<.0001
0.0
junction and esophageal
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
cancer from the CROSS
Overall survival (years)
trial [15] # at Risk
1992–1997 6,031 2,609 1,835 1,423 1,106 568 108
1998–2003 9,028 4,053 2,851 1,275 195 11 11
2004–2010 10,967 3,078 577 39 39 39 39

Magic

INT 0116 Adjuv.

Surg. only

Cross

0 10 20 30 40 50 60
7 Standards for Surgical Therapy of Gastric Cancer 87

recurrences [96] (Fig. 7.5c). The only trial to recurrence rates are high for large lesions, high
compare preoperative chemoradiation with pre- mitotic counts, and ruptured lesions or cases with
operative chemotherapy alone for resected GE intraoperative spillage of liquid contents. Modified
junction cancers failed to show a statistically sig- NIH criteria have been validated to delineate well
nificant difference due to small numbers of high- versus low-risk constellations [103]. Patients
enrolled patients, but also indicated a lower haz- with resected high risk GISTs have been shown to
ard ratio in favor of chemoradiation [97]. benefit from postoperative targeted adjuvant ther-
apy with the c-kit kinase inhibitor imatinib based
on 2 RCTs [104, 105]. Longer therapy in this set-
Prophylactic Gastrectomy ting for 3 years or possibly more appears to have
survival benefits compared to 1-year treatment.
Hereditary diffuse-type GC based on germline
CDH1 (E-cadherin) gene mutations can be
effectively addressed through prophylactic References
gastrectomy prior to onset of invasive carcinoma
[98]. Individuals from an affected kindred require 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E,
genetic counseling and testing [99]. Total gastrec- Forman D. Global cancer statistics. CA Cancer J Clin.
2011;61(2):69–90.
tomy is the only sensible procedure and is usually 2. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics,
performed during late adolescence or early adult- 2014. CA Cancer J Clin. 2014;64(1):9–29.
hood, as endoscopic surveillance has shown chal- 3. Robinson JO. The history of gastric surgery. Postgrad
lenges in identifying histologic alterations reliably Med J. 1960;36:706–13.
4. Schwarz RE. Factors influencing change of preopera-
[100]. Prophylactic gastrectomy specimens from tive treatment intent in a gastrointestinal cancer prac-
gene carriers harbor occult microscopic cancer in tice. World J Surg Oncol. 2007;5:32.
as many as 80 % of cases. Long-term results after 5. Karpeh Jr MS, Brennan MF. Gastric carcinoma. Ann
prophylactic gastrectomy show overall good Surg Oncol. 1998;5(7):650–6.
6. Karpeh MS, Leon L, Klimstra D, Brennan MF. Lymph
functional recovery and adaptation [101]. Another node staging in gastric cancer: is location more impor-
autosomal dominant syndrome, in this case of tant than Number? An analysis of 1,038 patients. Ann
intestinal-type gastric adenocarcinoma and proxi- Surg. 2000;232(3):362–71.
mal polyposis of the stomach (GAPPS), has 7. Mezhir JJ, Shah MA, Jacks LM, Brennan MF, Coit
DG, Strong VE. Positive peritoneal cytology in
recently been described [102]. patients with gastric cancer: natural history and out-
come of 291 patients. Ann Surg Oncol. 2010;17(12):
3173–80.
Special Considerations 8. Kim SH, Karpeh MS, Klimstra DS, Leung D, Brennan
MF. Effect of microscopic resection line disease on
for Gastrointestinal Stromal gastric cancer survival. J Gastrointest Surg. 1999;3(1):
Tumors (GISTs) 24–33.
9. Miner TJ, Jaques DP, Karpeh MS, Brennan
GISTs are rare but well-defined mesenchymal MF. Defining palliative surgery in patients receiving
noncurative resections for gastric cancer. J Am Coll
tumors of the GI tract and most commonly occur Surg. 2004;198(6):1013–21.
in the stomach. They are characterized by unique 10. Schwarz RE. Defining palliation in patients undergo-
biologic behavior and specific progression pat- ing gastrectomy for gastric cancer. J Am Coll Surg.
terns and thus represent a small yet unique subset 2004;199(6):1001–2.
11. Sehdev A, Catenacci DV. Perioperative therapy for
of “gastric cancers.” Definitive therapy of GISTs locally advanced gastroesophageal cancer: current
greater than 2 cm in size is surgical resection, for controversies and consensus of care. J Hematol Oncol.
which full-thickness local excision with negative 2013;6:66.
margins is sufficient. Laparoscopic resection tech- 12. Smalley SR, Benedetti JK, Haller DG, et al. Updated
analysis of SWOG-directed intergroup study 0116: a
niques are often applied. LN involvement is phase III trial of adjuvant radiochemotherapy versus
extremely rare, and ELND is not required. Cure observation after curative gastric cancer resection.
rates are high for small, low-grade tumors, but J Clin Oncol. 2012;30(19):2327–33.
88 R.E. Schwarz

13. Cunningham D, Allum WH, Stenning SP, et al. 26. Cardona K, Zhou Q, Gonen M, et al. Role of repeat
Perioperative chemotherapy versus surgery alone for staging laparoscopy in locoregionally advanced gas-
resectable gastroesophageal cancer. N Engl J Med. tric or gastroesophageal cancer after neoadjuvant
2006;355(1):11–20. therapy. Ann Surg Oncol. 2013;20(2):548–54.
14. Ychou M, Boige V, Pignon JP, et al. Perioperative 27. Siewert JR, Bottcher K, Stein HJ, Roder JD. Relevant
chemotherapy compared with surgery alone for prognostic factors in gastric cancer: ten-year results of
resectable gastroesophageal adenocarcinoma: an the German Gastric Cancer Study. Ann Surg. 1998;
FNCLCC and FFCD multicenter phase III trial. J Clin 228(4):449–61.
Oncol. 2011;29(13):1715–21. 28. Songun I, Putter H, Kranenbarg EM, Sasako M, van
15. van Hagen P, Hulshof MC, van Lanschot JJ, et al. de Velde CJ. Surgical treatment of gastric cancer:
Preoperative chemoradiotherapy for esophageal or 15-year follow-up results of the randomised nation-
junctional cancer. N Engl J Med. 2012;366(22): wide Dutch D1D2 trial. Lancet Oncol. 2010;11(5):
2074–84. 439–49.
16. Lowy AM, Mansfield PF, Leach SD, Pazdur R, 29. Vinuela EF, Gonen M, Brennan MF, Coit DG, Strong
Dumas P, Ajani JA. Response to neoadjuvant chemo- VE. Laparoscopic versus open distal gastrectomy for
therapy best predicts survival after curative resection gastric cancer: a meta-analysis of randomized con-
of gastric cancer. Ann Surg. 1999;229(3):303–8. trolled trials and high-quality nonrandomized studies.
17. Lordick F, Ott K, Krause BJ, et al. PET to assess early Ann Surg. 2012;255(3):446–56.
metabolic response and to guide treatment of adeno- 30. Choi YY, Bae JM, An JY, Hyung WJ, Noh SH.
carcinoma of the oesophagogastric junction: the Laparoscopic gastrectomy for advanced gastric
MUNICON phase II trial. Lancet Oncol. 2007;8(9): cancer: are the long-term results comparable with
797–805. conventional open gastrectomy? A systematic review
18. Ott K, Fink U, Becker K, et al. Prediction of response and meta-analysis. J Surg Oncol. 2013;108(8):550–6.
to preoperative chemotherapy in gastric carcinoma by 31. Kim HH, Han SU, Kim MC, et al. Long-term results
metabolic imaging: results of a prospective trial. of laparoscopic gastrectomy for gastric cancer: a
J Clin Oncol. 2003;21(24):4604–10. large-scale case–control and case-matched Korean
19. Ott K, Herrmann K, Lordick F, et al. Early metabolic multicenter study. J Clin Oncol. 2014;32(7):627–33.
response evaluation by fluorine-18 fluorodeoxyglu- 32. Oh SY, Kwon S, Lee KG, et al. Outcomes of mini-
cose positron emission tomography allows in vivo mally invasive surgery for early gastric cancer are
testing of chemosensitivity in gastric cancer: long- comparable with those for open surgery: analysis of
term results of a prospective study. Clin Cancer Res. 1,013 minimally invasive surgeries at a single institu-
2008;14(7):2012–8. tion. Surg Endosc. 2014;28(3):789–95.
20. Cheedella NK, Suzuki A, Xiao L, et al. Association 33. Tanabe S, Ishido K, Higuchi K, et al. Long-term out-
between clinical complete response and pathological comes of endoscopic submucosal dissection for early
complete response after preoperative chemoradiation gastric cancer: a retrospective comparison with con-
in patients with gastroesophageal cancer: analysis in a ventional endoscopic resection in a single center.
large cohort. Ann Oncol. 2013;24(5):1262–6. Gastric Cancer. 2014;17(1):130–6.
21. Yoshikawa T, Tanabe K, Nishikawa K, et al. Accuracy 34. Nishikawa K, Yanaga K, Kashiwagi H, Hanyuu N,
of CT staging of locally advanced gastric cancer after Iwabuchi S. Significance of intraoperative endoscopy
neoadjuvant chemotherapy: cohort evaluation within a in total gastrectomy for gastric cancer. Surg Endosc.
randomized phase II study. Ann Surg Oncol. 2014;21: 2010;24(10):2633–6.
S385–9. 35. Sasako M, Sano T, Yamamoto S, et al. D2 lymphade-
22. Conlon KC, Karpeh Jr MS. Laparoscopy and laparo- nectomy alone or with para-aortic nodal dissection for
scopic ultrasound in the staging of gastric cancer. gastric cancer. N Engl J Med. 2008;359(5):453–62.
Semin Oncol. 1996;23(3):347–51. 36. Kurokawa Y, Sasako M, Sano T, et al. Functional out-
23. Badgwell B, Cormier JN, Krishnan S, et al. Does neo- comes after extended surgery for gastric cancer. Br J
adjuvant treatment for gastric cancer patients with posi- Surg. 2011;98(2):239–45.
tive peritoneal cytology at staging laparoscopy improve 37. Gouzi JL, Huguier M, Fagniez PL, et al. Total versus
survival? Ann Surg Oncol. 2008;15(10):2684–91. subtotal gastrectomy for adenocarcinoma of the gas-
24. Samee A, Moorthy K, Jaipersad T, et al. Evaluation of tric antrum. A French prospective controlled study.
the role of laparoscopic ultrasonography in the stag- Ann Surg. 1989;209(2):162–6.
ing of oesophagogastric cancers. Surg Endosc. 2009; 38. Bozzetti F, Marubini E, Bonfanti G, Miceli R, Piano
23(9):2061–5. C, Gennari L. Subtotal versus total gastrectomy for
25. Lorenzen S, Panzram B, Rosenberg R, et al. Prognostic gastric cancer: five-year survival rates in a multicenter
significance of free peritoneal tumor cells in the peri- randomized Italian trial. Italian Gastrointestinal
toneal cavity before and after neoadjuvant chemother- Tumor Study Group. Ann Surg. 1999;230(2):170–8.
apy in patients with gastric carcinoma undergoing 39. Pugliese R, Maggioni D, Sansonna F, et al. Subtotal
potentially curative resection. Ann Surg Oncol. 2010; gastrectomy with D2 dissection by minimally inva-
17(10):2733–9. sive surgery for distal adenocarcinoma of the stomach:
7 Standards for Surgical Therapy of Gastric Cancer 89

results and 5-year survival. Surg Endosc. 2010;24(10): 54. Schwarz RE, Smith DD. Clinical impact of lymphad-
2594–602. enectomy extent in resectable gastric cancer of
40. Shin D, Park SS. Clinical importance and surgical advanced stage. Ann Surg Oncol. 2007;14(2):317–28.
decision-making regarding proximal resection margin 55. Smith DD, Nelson RA, Schwarz RE. A comparison of
for gastric cancer. World J Gastrointest Oncol. 2013; five competing lymph node staging schemes in a
5(1):4–11. cohort of resectable gastric cancer patients. Ann Surg
41. Mine S, Sano T, Hiki N, et al. Proximal margin length Oncol. 2014;21(3):875–82.
with transhiatal gastrectomy for Siewert type II and 56. Bonenkamp JJ, Hermans J, Sasako M, et al. Extended
III adenocarcinomas of the oesophagogastric junc- lymph-node dissection for gastric cancer. N Engl J
tion. Br J Surg. 2013;100(8):1050–4. Med. 1999;340(12):908–14.
42. Bozzetti F, Marubini E, Bonfanti G, et al. Total versus 57. Hartgrink HH, Van De Velde CJ, Putter H, et al.
subtotal gastrectomy: surgical morbidity and mortal- Extended lymph node dissection for gastric cancer:
ity rates in a multicenter Italian randomized trial. The who may benefit? Final results of the randomized
Italian Gastrointestinal Tumor Study Group. Ann Dutch gastric cancer group trial. J Clin Oncol. 2004;
Surg. 1997;226(5):613–20. 22(11):2069–77.
43. Stein HJ, Feith M, Mueller J, Werner M, Siewert 58. Schwarz RE, Karpeh MS, Brennan MF. Surgical man-
JR. Limited resection for early adenocarcinoma in agement of gastric cancer: the Western experience. In:
Barrett’s esophagus. Ann Surg. 2000;232(6):733–42. Daly JM, Hennessy TPJ, Reynolds JV, editors.
44. Siewert JR, Stein HJ. Classification of adenocarci- Management of upper gastrointestinal cancer.
noma of the oesophagogastric junction. Br J Surg. London: W.B. Saunders; 1999. p. 83–106.
1998;85(11):1457–9. 59. Kodera Y, Schwarz RE, Nakao A. Extended lymph
45. Siewert JR, Feith M, Werner M, Stein HJ. node dissection in gastric carcinoma: where do we stand
Adenocarcinoma of the esophagogastric junction: after the Dutch and British randomized trials? J Am
results of surgical therapy based on anatomical/topo- Coll Surg. 2002;195(6):855–64.
graphic classification in 1,002 consecutive patients. 60. Kitagawa Y, Takeuchi H, Takagi Y, et al. Sentinel node
Ann Surg. 2000;232(3):353–61. mapping for gastric cancer: a prospective multicenter
46. Yoshikawa T, Tsuburaya A, Kobayashi O, et al. Is bur- trial in Japan. J Clin Oncol. 2013;31(29):3704–10.
sectomy necessary for patients with gastric cancer 61. Mayanagi S, Takeuchi H, Kamiya S, et al. Suitability
invading the serosa? Hepatogastroenterology. 2004; of sentinel node mapping as an index of metastasis in
51(59):1524–6. early gastric cancer following endoscopic resection.
47. Yamamura Y, Ito S, Mochizuki Y, Nakanishi H, Ann Surg Oncol. 2014;21:2987–93.
Tatematsu M, Kodera Y. Distribution of free cancer 62. Behrns KE, Sarr MG. Diagnosis and management of
cells in the abdominal cavity suggests limitations of gastric emptying disorders. Adv Surg. 1994;27:233–55.
bursectomy as an essential component of radical sur- 63. Zong L, Chen P. Billroth I. vs. Billroth II vs. Roux-en-Y
gery for gastric carcinoma. Gastric Cancer. 2007; following distal gastrectomy: a meta-analysis based on
10(1):24–8. 15 studies. Hepatogastroenterology. 2011;58(109):
48. Schwarz RE. Spleen-preserving splenic hilar lymph- 1413–24.
adenectomy at the time of gastrectomy for cancer: 64. Hirao M, Takiguchi S, Imamura H, et al. Comparison
technical feasibility and early results. J Surg Oncol. of Billroth I and Roux-en-Y reconstruction after distal
2002;79(1):73–6. gastrectomy for gastric cancer: one-year postopera-
49. Schwarz RE, Zagala-Nevarez K. Gastrectomy cir- tive effects assessed by a multi-institutional RCT. Ann
cumstances that influence early postoperative out- Surg Oncol. 2013;20(5):1591–7.
come. Hepatogastroenterology. 2002;49(48):1742–6. 65. Ishikawa M, Kitayama J, Kaizaki S, et al. Prospective
50. Hyung WJ, Lim JS, Song J, Choi SH, Noh SH. randomized trial comparing Billroth I and Roux-en-Y
Laparoscopic spleen-preserving splenic hilar lymph procedures after distal gastrectomy for gastric
node dissection during total gastrectomy for gastric carcinoma. World J Surg. 2005;29(11):1415–20; dis-
cancer. J Am Coll Surg. 2008;207(2):e6–11. cussion 1421.
51. Huang CM, Chen QY, Lin JX, et al. Laparoscopic 66. Kojima K, Yamada H, Inokuchi M, Kawano T,
spleen-preserving no. 10 lymph node dissection for Sugihara K. A comparison of Roux-en-Y and Billroth-I
advanced proximal gastric cancer using a left reconstruction after laparoscopy-assisted distal
approach. Ann Surg Oncol. 2014;21:2051. gastrectomy. Ann Surg. 2008;247(6):962–7.
52. Wu CW, Hsiung CA, Lo SS, et al. Nodal dissection 67. Fein M, Fuchs KH, Thalheimer A, Freys SM,
for patients with gastric cancer: a randomised con- Heimbucher J, Thiede A. Long-term benefits of Roux-
trolled trial. Lancet Oncol. 2006;7(4):309–15. en-Y pouch reconstruction after total gastrectomy: a
53. Smith DD, Schwarz RR, Schwarz RE. Impact of total randomized trial. Ann Surg. 2008;247(5):759–65.
lymph node count on staging and survival after gas- 68. Zong L, Chen P, Chen Y, Shi G. Pouch Roux-en-Y vs
trectomy for gastric cancer: data from a large No Pouch Roux-en-Y following total gastrectomy: a
US-population database. J Clin Oncol. 2005;23(28): meta-analysis based on 12 studies. J Biomed Res.
7114–24. 2011;25(2):90–9.
90 R.E. Schwarz

69. Carrere N, Seulin P, Julio CH, Bloom E, Gouzi JL, gastric cancer: a systematic review. Gastric Cancer.
Pradere B. Is nasogastric or nasojejunal decompres- 2012;15 Suppl 1:S138–45.
sion necessary after gastrectomy? A prospective ran- 83. Chang YR, Han DS, Kong SH, et al. The value of
domized trial. World J Surg. 2007;31(1):122–7. palliative gastrectomy in gastric cancer with distant
70. Pacelli F, Rosa F, Marrelli D, et al. Naso-gastric or metastasis. Ann Surg Oncol. 2012;19(4):1231–9.
naso-jejunal decompression after partial distal gastrec- 84. Kokkola A, Louhimo J, Puolakkainen P. Does non-curative
tomy for gastric cancer. Final results of a multicenter gastrectomy improve survival in patients with metastatic
prospective randomized trial. Gastric Cancer 2013. gastric cancer? J Surg Oncol. 2012;106(2):193–6.
[Epub ahead of print]. 85. Karpeh Jr MS. Palliative treatment and the role of sur-
71. Petrowsky H, Demartines N, Rousson V, Clavien gical resection in gastric cancer. Dig Surg. 2013;30(2):
PA. Evidence-based value of prophylactic drainage in 174–80.
gastrointestinal surgery: a systematic review and 86. Bonenkamp JJ, Sasako M, Hermans J, van de Velde
meta-analyses. Ann Surg. 2004;240(6):1074–84; dis- CJ. Tumor load and surgical palliation in gastric can-
cussion 1084–1075. cer. Hepatogastroenterology. 2001;48(41):1219–21.
72. Kumar M, Yang SB, Jaiswal VK, Shah JN, Shreshtha 87. Blair SL, Chu DZ, Schwarz RE. Outcome of palliative
M, Gongal R. Is prophylactic placement of drains nec- operations for malignant bowel obstruction in patients
essary after subtotal gastrectomy? World J Gastroenterol. with peritoneal carcinomatosis from nongynecologi-
2007;13(27):3738–41. cal cancer. Ann Surg Oncol. 2001;8(8):632–7.
73. Ishikawa K, Matsumata T, Kishihara F, Fukuyama Y, 88. Tokunaga M, Tanizawa Y, Bando E, Kawamura T,
Masuda H. Laparoscopy-assisted distal gastrectomy Terashima M. Poor survival rate in patients with post-
for early gastric cancer with versus without prophy- operative intra-abdominal infectious complications
lactic drainage. Surg Today. 2011;41(8):1049–53. following curative gastrectomy for gastric cancer.
74. McCarter MD, Gomez ME, Daly JM. Early postopera- Ann Surg Oncol. 2013;20(5):1575–83.
tive enteral feeding following major upper gastrointes- 89. Birkmeyer JD, Sun Y, Goldfaden A, Birkmeyer NJ,
tinal surgery. J Gastrointest Surg. 1997;1(3):278–85; Stukel TA. Volume and process of care in high-risk
discussion 285. cancer surgery. Cancer. 2006;106(11):2476–81.
75. Patel SH, Kooby DA, Staley 3rd CA, Maithel SK. An 90. Birkmeyer JD, Sun Y, Wong SL, Stukel TA. Hospital
assessment of feeding jejunostomy tube placement at volume and late survival after cancer surgery. Ann
the time of resection for gastric adenocarcinoma. Surg. 2007;245(5):777–83.
J Surg Oncol. 2013;107(7):728–34. 91. Schwarz RE, Zagala-Nevarez K. Recurrence patterns
76. Schwarz RE. Simple feeding jejunostomy technique after radical gastrectomy for gastric cancer: prognos-
for postoperative nutrition after major upper gastroin- tic factors and implications for postoperative adjuvant
testinal resections. J Surg Oncol. 2002;79(2):126–30. therapy. Ann Surg Oncol. 2002;9(4):394–400.
77. Kim MM, Rana V, Janjan NA, et al. Clinical benefit of 92. D’Angelica M, Gonen M, Brennan MF, Turnbull AD,
palliative radiation therapy in advanced gastric can- Bains M, Karpeh MS. Patterns of initial recurrence in
cer. Acta Oncol. 2008;47(3):421–7. completely resected gastric adenocarcinoma. Ann
78. Hashimoto K, Mayahara H, Takashima A, et al. Surg. 2004;240(5):808–16.
Palliative radiation therapy for hemorrhage of unre- 93. Yoo CH, Noh SH, Shin DW, Choi SH, Min
sectable gastric cancer: a single institute experience. JS. Recurrence following curative resection for gas-
J Cancer Res Clin Oncol. 2009;135(8):1117–23. tric carcinoma. Br J Surg. 2000;87(2):236–42.
79. Chaw CL, Niblock PG, Chaw CS, Adamson DJ. The 94. Dikken JL, Baser RE, Gonen M, et al. Conditional
role of palliative radiotherapy for haemostasis in unre- probability of survival nomogram for 1-, 2-, and
sectable gastric cancer: a single-institution experi- 3-year survivors after an R0 resection for gastric can-
ence. Ecancermedicalscience. 2014;8:384. cer. Ann Surg Oncol. 2013;20(5):1623–30.
80. Koucky K, Wein A, Konturek PC, et al. Palliative 95. Hirabayashi S, Kosugi S, Isobe Y, et al. Development
first-line therapy with weekly high-dose 5-fluorouracil and external validation of a nomogram for overall sur-
and sodium folinic acid as a 24-hour infusion (AIO vival after curative resection in serosa-negative, locally
regimen) combined with weekly irinotecan in patients advanced gastric cancer. Ann Oncol. 2014;25:1179–84.
with metastatic adenocarcinoma of the stomach or 96. Oppedijk V, van der Gaast A, van Lanschot JJ, et al.
esophagogastric junction followed by secondary met- Patterns of recurrence after surgery alone versus pre-
astatic resection after downsizing. Med Sci Monit. operative chemoradiotherapy and surgery in the
2011;17(5):CR248–58. CROSS trials. J Clin Oncol. 2014;32(5):385–91.
81. Boskoski I, Tringali A, Familiari P, Mutignani M, 97. Stahl M, Walz MK, Stuschke M, et al. Phase III com-
Costamagna G. Self-expandable metallic stents for parison of preoperative chemotherapy compared with
malignant gastric outlet obstruction. Adv Ther. 2010; chemoradiotherapy in patients with locally advanced
27(10):691–703. adenocarcinoma of the esophagogastric junction. J Clin
82. Mahar AL, Coburn NG, Karanicolas PJ, Viola R, Oncol. 2009;27(6):851–6.
Helyer LK. Effective palliation and quality of life out- 98. Chen Y, Kingham K, Ford JM, et al. A prospective
comes in studies of surgery for advanced, non-curative study of total gastrectomy for CDH1-positive heredi-
7 Standards for Surgical Therapy of Gastric Cancer 91

tary diffuse gastric cancer. Ann Surg Oncol. 2011; 102. Worthley DL, Phillips KD, Wayte N, et al. Gastric
18(9):2594–8. adenocarcinoma and proximal polyposis of the
99. Seevaratnam R, Coburn N, Cardoso R, Dixon M, stomach (GAPPS): a new autosomal dominant
Bocicariu A, Helyer L. A systematic review of the indi- syndrome. Gut. 2012;61(5):774–9.
cations for genetic testing and prophylactic gastrec- 103. Joensuu H, Vehtari A, Riihimaki J, et al. Risk of
tomy among patients with hereditary diffuse gastric recurrence of gastrointestinal stromal tumour after
cancer. Gastric Cancer. 2012;15 Suppl 1:S153–63. surgery: an analysis of pooled population-based
100. Fujita H, Lennerz JK, Chung DC, et al. Endoscopic cohorts. Lancet Oncol. 2012;13(3):265–74.
surveillance of patients with hereditary diffuse gas- 104. Dematteo RP, Ballman KV, Antonescu CR, et al.
tric cancer: biopsy recommendations after topo- Adjuvant imatinib mesylate after resection of local-
graphic distribution of cancer foci in a series of 10 ised, primary gastrointestinal stromal tumour: a ran-
CDH1-mutated gastrectomies. Am J Surg Pathol. domised, double-blind, placebo-controlled trial.
2012;36(11):1709–17. Lancet. 2009;373(9669):1097–104.
101. Worster E, Liu X, Richardson S, et al. The impact of 105. Joensuu H, Eriksson M, Sundby Hall K, et al. One vs
prophylactic total gastrectomy on health-related three years of adjuvant imatinib for operable gastro-
quality of life: a prospective cohort study. Ann Surg. intestinal stromal tumor: a randomized trial. JAMA.
2014;260:87–93. 2012;307(12):1265–72.
 Endoscopic Submucosal
Dissection for Gastric Cancer: 8
Its Indication, Technique,
and Our Experience

Hiroki Sato and Haruhiro Inoue

Introduction submucosal dissection (ESD) expanded the

criteria to include tumors more than 2 cm with
Implementation of screening and surveillance or without ulcers to be successfully removed
upper GI endoscopy and the availability of endo- endoscopically in an en bloc fashion [7, 8].
scopes allowed for the identification and diagno- As an extended indication, submucosal inva-
sis of early gastric carcinoma (EGC) which can sion (sm1, infiltration depth less than 500 μm
still be resected endoscopically. with no vessel permeation) is also accepted.
EGCs are defined as those in which invasion is Several retrospective studies in EGC tumors
limited to either the mucosa or submucosa irre- with a low risk of lymph node metastasis fur-
spective of lymph node involvement [1]. In those ther support the extension of the indication
lesions, at present, widely accepted techniques for ESD [3 , 9].
for endoscopic resection (ER), particularly endo- Moreover, patients with EGC who are poor
scopic mucosal resection (EMR), are indicated surgical candidates due to underlying comorbidi-
only for differentiated adenocarcinoma smaller ties may be managed safely and less invasively
than 2 cm and confined to the mucosa [2, 3] so as through ESD [10, 11].
to assure en bloc resection of the tumor. EMR is In this chapter, we present the principles and
an endoscopic resection technique which utilizes practice of ESD for EGC and describe our expe-
snare wire [4, 5]. In 1993, Cap-EMR method was rience and clinical outcomes.
developed, which allows easy resection of muco-
sal lesions [6]. This technique is currently modi-
fied and further popularized as EMR-C. Indication for Cure Rates
Compared to these techniques, however, with Endoscopic Resection
advancements in the technique of endoscopic
The Japanese EMR/ESD guidelines have been
described [2, 12].

H. Sato, MD (*) • H. Inoue, MD, PhD

Digestive Disease Center, Showa University, Indication for Endoscopic Resection
Northern Yokohama Hospital,
35-1 Chigasakichuo, Tsuzuki-ku,
Yokohoma 224-8503, Japan Contrary to the conventional criteria for ER
e-mail: [email protected] which limits EMR/ESD to differentiated-type

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 93
DOI 10.1007/978-3-319-09342-0_8, © Springer International Publishing Switzerland 2015
94 H. Sato and H. Inoue

adenocarcinomas without ulcerative findings Therapeutic Efficacy of Endoscopic

(UL(−)), of which the depth of invasion is clini- Resection
cally diagnosed as T1a and the diameter is ≦2 cm,
the expanded criteria for ESD encompass tumors The resected specimen should be handled
clinically diagnosed as T1a which includes the according to the rules described in the Japanese
following: classification [2] (Table 8.1).
(i) Differentiated type, UL(−), but >2 cm in Curative tumor resection was previously
diameter defined as en bloc resection of a differentiated-
(ii) Differentiated type, UL(+), and ≦3 cm in type tumor, ≦2 cm in size, pT1a, negative hori-
diameter zontal margin (HM0), negative vertical margin
(iii) Undifferentiated type, UL(−), and ≦2 cm in (VM0), and no lymphovascular infiltration
diameter (ly(−), v(−)).
Data regarding the “therapeutic efficacy” of Following the expanded criteria for ESD in
ESD when applying the expanded criteria is lim- EGC, resection is considered curative when all of
ited; hence, the decision to do ESD should be on the following conditions are fulfilled:
a case-to-case basis and the procedure offered • En bloc resection, HM0, VM0, ly(−), v(−)
with caution (Fig. 8.1). (a) Tumor size ≧2 cm, histologically of a dif-
ferentiated type, pT1a, UL(−)
(b) Tumor size ≦3 cm, histologically of a dif-
ferentiated type, pT1a, UL(+)
(c) Tumor size ≦2 cm, histologically of
undifferentiated type, pT1a, UL(−)
(d) Tumor size ≦3 cm, histologically of dif-
ferentiated type, pT1b (SM1,<500 μm
from the muscularis mucosae)
Follow-up abdominal ultrasonography or
computed tomography (CT) scan and annual or
biannual endoscopy are still recommended in all
cases of curative resection meeting the expanded
criteria.

Fig. 8.1 This is a case of a 93-year-old female who was

referred to our hospital for the treatment of a gastric tumor
that by biopsy was group 3. Conventional endoscopy
Summary
revealed a circumferential, slightly elevated lesion in the
gastric antrum, which was resected by ESD as an The expanded criteria for ESD are now being
expanded indication. Pathological diagnosis was employed as more endoscopists become skillful
110 × 65 mm, type 0–IIa, adenocarcinoma, T1a, ly0, v0,
HM0, and VM0
in performing ESD. As the procedure is offered

Table 8.1 Indication for T1a T1b

ESD: current and expanded
UL(−) UL(+) SM1 SM1<
indications
20 mm 20 mm< 30 mm 30 mm< 30 mm Any size
Differentiated-type
Undifferentiated-type

:current indication
:expanded indication
8 Endoscopic Submucosal Dissection for Gastric Cancer: Its Indication, Technique, and Our Experience 95

Table 8.2 High-frequency knives for ESD Tokyo Japan), which has a long sharp non-covered
Needle Flush knife (DK2618JB/DK2618JN, needle, was originally used for ESD). This knife
Knife type Fujifilm Co.) enables the endoscopist to perform ESD with
Flex knife (KD-630 L, Olympus Co.) direct visualization of the area to be cut. Both
Triangle Tip Knife (KD-640 L, incision and dissection can be done using this
Olympus Co)
device. Among the short-needle type, Flush knife
Dual Knife (KD-650 L/KD-650Q,
Olympus Co) (Fujifilm Co., Tokyo, Japan) is used routinely in
Hook Knife (KD-620LR/KD-620QR, our hospital. The Flush knife has a water-jet
Olympus Co.) function that makes additional submucosal
IT-knife IT-knife, IT-knife-2, IT-knife-nano injection possible. It also has two kinds of tip
type (KD-610 L/KD-611 L/KD-612, types (needle or ball-chip), and four kinds of
Olympus Co.)
knife lengths are available at 5 mm intervals
Non-IT- Mucosectom (DP-2518, PENTAX Co.)
knife type (1.0–3.0 mm). The endoscopist should take into
SAFE knife (DK2518DV1, Fujifilm Co.)
consideration both which organ (i.e., esophagus,
Swanblade (DC-D2618, PENTAX,
Tokyo) stomach, colon) and the location of the lesion in
Scissors Clutch Cutter (DP2618DT, Fujifilm Co.) choosing the suitable tip type and protruding
forceps type SB knife and SB knife Jr. (MD-47706/ length to be used for treatment. In our hospital,
MD-47704 and MD-47703, Sumitomo the Flush knife with a needle type tip, 2.0 mm in
Bakelite Co.) length, is employed for gastric ESD.
The other category is the hook type, repre-
to more patients, further studies on clinical sented by the Hook Knife (Olympus Co., Tokyo,
efficacy and safety following the expanded crite- Japan). The distal L-shaped hook has a rotatory
ria are warranted [13]. function allowing for incision and dissection in
longitudinal and lateral directions. This is done
by simply turning the handle to point the tip of
Device for ESD [14–16] the hook in the desired direction. Moreover, the
Hook Knife also enables the operator to “hook”
Knives for ESD tissue and pull away from the muscle layer, as its
name suggests. This method minimizes the risk
Knives used for ESD have been conventionally of perforation especially in cases of severe fibro-
classified into needle type or insulated tip type sis as cutting is done away from the muscle layer.
based on the design, shape, and method of use. At The drawback of using this knife, however, is that
present, various knives are available including the amount of tissue that can be hooked at a time
those designed for use in areas in which the is limited prolonging procedure time.
approach is difficult or knives which increase the
safety intraoperatively. However, use of this wide IT-Knife Type
range of knives requires an adequate understand- IT-knife and IT-knife-2 (Olympus Co., Tokyo,
ing of the properties of each to allow their appli- Japan) are the main knives used for gastric
cation under appropriate conditions. ESD. The ceramic insulator attached at the tip of
The ESD knives currently available in Japan the needle-shaped knife does not conduct elec-
are described and classified below into (1) needle tricity thereby minimizing invasiveness and
type, (2) IT-knife type (insulation-tipped type), reduces the risk of perforation. It allows lateral
(3) non-IT-knife type, and (4) scissors forceps cutting from a vertical approach. To perform
type (Table 8.2, Fig. 8.2). steady submucosal dissection, the ceramic glob-
ule and the sheath should be positioned properly
Needle Type on the surface of the incision or area to be dis-
Short needle is the representative type in this sected. It is said that procedure time is shorter
category (although Needle Knife (Olympus Co., when using IT-knife effectively because the
96 H. Sato and H. Inoue

Fig. 8.2 Knives for ESD. Upper left, Flush knife; Upper right, Hook Knife; Center left, IT-knife-2; Center right,
Mucosectom; Lower left, Clutch Cutter; Lower right, Flex knife

contact area between the blade and surface is of this knife is its high coagulation capacity
increased due to “line” touch (understandably, causing tissue carbonization or charring which
needle type attach in a point), and tension can be eventually prevents proper tissue contact and
applied during the dissection due to non- adequate depth dissection (see below “high-
electrical conductivity of the tip. The downside frequency generator”).
8 Endoscopic Submucosal Dissection for Gastric Cancer: Its Indication, Technique, and Our Experience 97

Recently, the IT-knife-nano was developed, The internal diameter of scope is a minimum
with a compact ceramic tip and small disklike of 2.8 mm considering the external diameter of
structure of the backside electrical blade. It is each device. For the purpose of removing the
expected to work effectively in difficult cases smoke or mucus during the procedure, a 3.2 mm
particularly with severe fibrosis. diameter channel is more desirable. A two-
channel scope enables us to perform efficient
Non-IT-Knife Type (Insulator suction (one channel for the device, another for
Processing) suction) and effective dissection (one channel
Mucosectom (HOYA PENTAX Co., Tokyo, for the device, another for injection).
Japan) is a knife wherein the tip and the lateral 3. Flexure point
sides are covered by an insulator. In addition, the The maximum angle of the scope is an
blade has a rotatable function enabling adjust- important factor in approaching the lesion.
ments in the direction of dissection. This knife The multi-bending function means that the
can be used as a secondary knife rather than a second flexure point is set at the posterior side
primary knife, reserving its use in cases where of the first, which enables an approach to any
the position of the knife is perpendicular to the lesion.
muscle layer and area of dissection especially in 4. Outside diameter
cases of severe fibrosis. In general, big external diameter means
multi-function (water-jet function, 2-channel,
Scissors Forceps Type etc.). However, for lightening patients’ pain,
This device was developed combining the design using the small scope is better.
concepts of the conventional knife like Needle Small scope also has a small turning circle,
Knife and IT-knife type. The Clutch Cutter which is useful when dissection is performed
(Fujifilm Co., Tokyo, Japan) is representative of by handling the endoscopic arm, particularly
this category. These knives have the capability to in the curve.
grasp and cut tissue in direct view. Moreover the In our hospital, a 9.9 mm endoscope with
Clutch Cutter can also be used to perform water-jet function (Olympus GIF Q260J) is used
hemostasis. as our standard. Then if a close approach is diffi-
A large variety of ESD devices have now been cult, the multi-bending scope (Olympus GIF
developed and launched by several companies. 2TQ260M) is used.
Each knife has some unique characteristics in
terms of sharpness for incision/dissection and
capacity to do hemostasis. Therefore, the endos- Distal Attachment
copist should be cognizant of the knife features
which will guide choosing the tool or device to be A transparent tip hood is necessary for manipula-
used in a particular situation. tion in ESD. It exerts tension on the submucosal
layer and aids in easy entry into the submucosa.
In addition, stable knife operation is possible dur-
Endoscope ing the procedure with good visibility even under
conditions of body motion, breath movement,
The endoscope for ESD should have the and heartbeat by holding down the front mucosa
following: or holding up the lesion.
1. Flush function During the procedure with distal attachment,
The water-jet function is important to clear frequent use of an anti-fouling composition is
the mucus from the lesion or to find the bleed- important. Particularly when ST hood is used,
ing point. Clear water with small amount of because of its narrow vision, it is effective.
dimethicone is used in our hospital. In our hospital, cylindrical hood (Olympus Co
2. Channel (size, number) D-201-11804) and ST hood (Fujifilm Co
98 H. Sato and H. Inoue

DH-15GR)/Short ST hood (Fujifilm Co Endoscopic Resection for EGC

DH-15GR) are used depending on the situation. with Gastric Ulcer

Before Endoscopic Resection

High-Frequency Generator (HFG)
In principle, open gastric ulcer with EGC should be
A high-frequency generator (HFG) is an appli- cured by antisecretory medication such as proton
ance designed for incision or coagulation of tis- pump inhibitor (PPI) or H2 receptor antagonist (H2
sue. It uses heat-generated high-frequency blocker) if ESD is kept in mind as the first choice.
electrical current applied to the tissue. Incision (In this sentence, open gastric ulcer is defined as
results from a vapor explosion of cellular mem- active (A) and healing (H) stages, not including the
brane caused by continuous delivery of low- scar (S) stage according to the classification pro-
voltage current resulting in rapid generation of posed by Sakita and Fukutomi [18].) One of the
heat, whereas coagulation results from shrinkage reasons is, particularly in the acute phase, it is dif-
of tissue and evaporation of moisture using inter- ficult, almost impossible to assess the details of
mittent, high-voltage current. tumors (invasion depth, the extent, etc.) because of
The details with regard to the energy setting of modifying factors like inflammatory reaction and
the HFG were discussed previously [14, 17]; edema. Another reason is that the specimen is eas-
however, in general, the configuration mode ily torn by submucosal dissection because the tis-
should be arranged per organ (i.e., esophagus, sue connection is too weak, particularly in the acute
stomach, colon) and per knife or forceps devices. ulcer phase. Antisecretory medication is also effec-
Regarding the current density, if it is high, tive to prevent ulcer relapse of EGC with scar.
large amount of heat is made, and the incision In cases of gastric ESD at Northern Yokohama
capacity increases. If it is low, coagulation capac- Hospital from July 2007 through March 2013, 38
ity increases. Meanwhile, current density is influ- differentiated adenocarcinomas had endoscopic
enced by contact area. For example, a point attach ulcer findings on the first endoscopy (recurrent can-
(Flush knife, Flex knife) has high current density; cers were excluded). Thirteen carcinomas were
on the contrary, a line attach (IT-knife) has low found with the open ulcer. Every patient received
current density. Therefore, if the same output medication therapy of PPI or H2 blocker and fol-
waveforms are used, the needle type point has low-up endoscopy was performed to assess the
more incision capacity, and IT-knife has more ulcer stage (improvement/no change/exacerbation).
coagulation capacity (the tissue carbonizes eas- Regarding cases with open gastric ulcer (n = 13), as
ily). Thus, if the same device is used, current den- the median interval between the first and follow-up
sity would also change depending on the method endoscopic examinations was 81.6 days (range
of attaching the tissue. 28–152), a change in ulcer stage was observed with
In our hospital, the HFG VIO 300D (ERBE improvement for 12 patients (92.3 %) and exacer-
Elektromedizin GmbH, Tübingen, Germany) is bation for 1 patient (7.7 %). The case with unhealed
being used during ESD. The following are the ulcer is described separately below.
HFG settings we use for gastric ESD during the Additionally, regarding cases of scar stage, no
different procedural steps: mucosal marking, exacerbation was found with a mean of 60.4 fol-
forced coagulation mode, 30 w, and effect 4; low-up days (range 15–150) (Fig. 8.3).
mucosal incision, Endocut I mode, cut duration
3, cut interval 2, and effect 2; and submucosal
dissection, forced coagulation mode, 45 w, and ESD Procedure
effect 4.
In the next section, among the expanded indi- (i) Marking
cations, ESD for EGC with gastric ulcer and for Marking should be performed 3–5 mm
undifferentiated adenocarcinoma is described. away from the margin of the tumor.
8 Endoscopic Submucosal Dissection for Gastric Cancer: Its Indication, Technique, and Our Experience 99

Fig. 8.3 Every patient who Mucosal gastric cancer

Follow up endoscopy
was diagnosed with a with ulcer findings Intervention
differentiated mucosal cancer (n = 38)
PPI or H2 blocker
with ulcer received antisecre-
tory medication. Out of Improvement
13 cases with active ulcer, (n = 12)
improvement was seen in
With open ulcer
12 patients (92.3 %) and (n = 13)
exacerbation in 1 patient
(7.7 %). In cases with ulcer Exacerbation
scarring (n = 25), no (n = 1)
exacerbation was found. We
may infer antisecretory
medication is useful to With ulcer scar
prevent ulcer relapse (n = 25)

No change

(ii) Injection and incision

Mucosal distention is accomplished by
injection of a solution in to the submucosa
through the mucosa. Injection is done prior
to incision with the objective of lifting the
mucosa away from the muscle layer. In cases
where submucosal injection does not pro-
duce an adequate bleb, fibrosis should be
suspected and the incision line reconsidered.
In our hospital, glycerine (glycerol; Chugai
Pharm Co., Tokyo, Japan) is routinely used
as an injection solution. An alternative to this
is normal saline with indigo carmine. In the
presence of severe fibrosis, sodium hyaluro-
Fig. 8.4 If fibrosis is suspected in the marginal area of
nate solution (MucoUp; Johnson & Johnson EGC, the incision line should be 1 cm away from the mar-
Co., Tokyo, Japan) is effective to keep gin of the scar to make enough of a flap allowing access
enough distance between the mucosal and into the fibrotic submucosal layer
muscle layer and prevent perforation.
Incision is made with reference to the
location of the tumor and scar. In cases when dissection is accomplished over the muscle
the scar is located at the margins of the tumor, layer with repeated injection.
the incision should be made 1 cm away In cases where the muscle layer and sub-
(Fig. 8.4) [19]. This allows creation of an mucosal fibrosis are adherent, the clear layer
adequate flap making dissection of the sub- is absent. Submucosal dissection is per-
mucosal layer in the fibrotic area possible. formed starting from the periphery going to
(iii) Submucosal dissection the center of the lesion. Dissection is per-
EGC with ulceration has a very thin, formed connecting bilateral submucosal
clear layer of fibrotic submucosa separating layers paying careful attention not to dam-
the mucosa and muscle layer. Submucosal age the muscle layer.
100 H. Sato and H. Inoue

Fig. 8.5 ME with NBI shows fine-network superficial Fig. 8.6 Exacerbation of the gastric ulcer was seen
pattern indicative of well-differentiated adenocarcinoma. despite PPI and H. pylori eradication therapy. However,
This allows identification of demarcation for cancer the endoscopic appearance shows no evidence of invasion
(arrows point to the demarcation line for cancer) into the submucosa

In addition, CO2 insufflation should be Figure 8.6 shows the state of the gastric ulcer
used throughout the procedure in case a per- before treatment. Unfortunately at this time, the
foration occurs. endoscopic appearance of the ulcer was worse
compared to its condition on the previous endos-
copy. However, a decision was made to perform
A Case of Severe Fibrosis with Ulcer ESD due to the following reasons as follows:
(1) for definitive diagnosis (biopsy has a high
This is a case of a 65-year-old male with history bleeding risk due to antiplatelet therapy and not a
of unstable angina. He had coronary angioplasty reliable enough diagnostic tool, so in this case,
and was maintained on antiplatelet medications. resection biopsy was chosen) and (2) according
Six months prior to ESD, he presented at our to the endoscopic appearance, the depth of the
institution with 2-week history of abdominal pain. relapsed ulcer was estimated as less deep than the
Endoscopic examination revealed a 2 cm ulcer at layer dissected by the ESD technique; hence en
the posterior wall of the gastric angle. The patient bloc resection is possible.
was prescribed PPI and triple therapy for H. pylori Submucosal dissection was started from the
eradication because of the positive serologic test. periphery (1 cm away from the scar) going to the
A repeat endoscopy was performed at a later date center of the lesion taking all the precautions by
with a week of antiplatelet cessation and biopsies assessing dissection depth. Then, as anticipated,
were taken from margin of the ulcer. The histo- dissection was technically difficult because of
pathologic results were indefinite for neoplasia; paucity of submucosal space and much fibrosis
hence endoscopically, no cancer was identified. below the ulcer (Fig. 8.7). With repeated
A follow-up endoscopy 3 months after initial injection, as the safe layer on each side was con-
endoscopy showed a healing ulcer; however, nected, submucosal dissection was performed.
magnifying endoscopy with narrow band imag- Nonetheless, a minor perforation was encoun-
ing (ME-NBI) showed a fine-network pattern tered at the bottom of the ulcer, which was suc-
with demarcation line suggesting a well- cessfully closed by placing endoscopic clips
differentiated adenocarcinoma [20] (Fig. 8.5). (Olympus Co., Tokyo, Japan). However, after
Thus, ESD was contemplated for diagnostic and dissection was further advanced before clipping
therapeutic purposes. because instant clipping interferes with
8 Endoscopic Submucosal Dissection for Gastric Cancer: Its Indication, Technique, and Our Experience 101

dissection. Finally, ESD was accomplished in an malignant nature. Malignant ulceration usually
en bloc fashion. arises at the margin of cancer in the presence of
The patient was started on clear liquids 24 h acid and pepsin. The repetitive cycle of inflam-
post procedure and diet was progressed subse- mation and repair in the epithelial cells triggers
quently without untoward events. the formation of fibrosis. Ultimately, cancer cells
Histological examination of the resected spec- spread superficially or malignant invasion may
imen showed well-differentiated adenocarci- occur along the fibrosis [18].
noma, pT1a-M, 30 × 19 mm, UL(+), ly0, v0, Jong Pil Im et al. reported that the use of anti-
pHM0, and pVM0, satisfying the currently secretory medication in mucosal cancer and a
expanded indication for endoscopic treatment longer interval between the first and follow-up
(Fig. 8.8). Although careful follow-up is needed, endoscopy were independently associated with
ESD was considered to have accomplished the healing of malignant ulcers [21]. Cancer should
same objective as radical therapy. be resected at the most appropriate time when the
This case shows that ulcerations in EGC may ulcer is healed.
relapse even after PPI treatment due to its It has been recognized that ESD for EGC
should be delayed until after ulcer healing has
occurred; however, it is difficult to ascertain the
time interval when this would occur even with PPI
therapy. Ulcers in EGC behave in a different man-
ner as peptic ulcer and healing is dependent on fac-
tors such as ulcer size and depth. Moreover,
documentation of healing on endoscopy does not
guarantee that the ulcer will not recur. A malignant
ulcer may relapse as the cancer cells invade into
the submucosa due to the malignant cycle [22].
Jae IK Lee reported that endoscopic resection
should be restricted to cases showing significant
improvement in the size and depth of ulcer at fol-
low-up endoscopy [7]. Although this proposition
is ideal, the fact that it is difficult to determine
Fig. 8.7 Severe fibrosis was encountered below the ulcer when significant improvement in ulcer size and
which made identification and dissection of the submu- depth occurs makes this problematic. In the
cosa difficult untoward event of perforation, endoscopic closure

Fig. 8.8 Pathological diagnosis was well-differentiated adenocarcinoma, pT1a-M, 30 × 19 mm, UL(+), ly0, v0, pHM0,
and pVM0, satisfying the criteria for expanded indication for ESD, Arrow demarcation line of mucosal cancer
102 H. Sato and H. Inoue

using clips is effective [23]. An experienced intramucosal EGC (0.4 %) [3]. Thus, the currently
endoscopist can successfully remove EGCs lim- accepted definitive treatment worldwide is surgi-
ited to the mucosa with non-healing ulcers via cal resection.
ESD and also manage its complications (i.e., per- However, in a study by Gotoda et al., none of
foration) endoscopically when it occurs. Of the 141 undifferentiated lesions without ulcer-
course, such high-quality ESD should be per- ation, less than 20 mm in size, were associated
formed only by an experienced endoscopist. with positive lymph nodes [3]. Recently, the
In our opinion, it is of utmost importance to Japanese Gastric Cancer Treatment Guidelines
ascertain tumor depth prior to ESD. The presence expanded the indication for ESD to include
of gastric ulcer makes endoscopic and even path- undifferentiated EGC without ulceration.
ological diagnosis by biopsy challenging due to
factors like inflammation, edema, and superficial
regenerated epithelial cell infiltration [24–27]. Procedure
Biopsy and ME-NBI are complementary diag-
nostic tools. In circumstances when biopsy is Undifferentiated carcinomas sometimes have dif-
contraindicated or inconclusive, ME-NBI proves fuse invasion; hence, submucosal dissection
to be a useful tool to aid in the assessment of should be done in the deep submucosal layer to
invasion depth and extent of cancer and also achieve adequate tumor-free vertical margins. In
determine pathologic type based on the examina- general, the technique used to perform ESD on
tion of surface patterns [20, 28–33]. undifferentiated carcinoma follows the same
principles as in differentiated cancer.
Signet ring cell carcinoma usually expands
ESD for Undifferentiated superficially with 0–IIb or 0–IIc macroscopic
Adenocarcinoma type, with white color or same color of adjacent
normal epithelium (Fig. 8.9). The non-exposed
Principle expanding subepithelium of the tumor margin is
often very difficult to identify. A 1 cm tumor-free
Undifferentiated intramucosal EGC demon- margin confirmed by biopsy in 4 directions
strates a relatively higher probability of lymph around the lesion should be made in order to
node metastasis (LNM) (4.2 %) than differentiated achieve complete resection (Fig. 8.10).

a b

Fig. 8.9 Conventional endoscopy reveals a faded color, NBI, the area appears slightly brownish (c). With
flat lesion in the gastric antrum (a), and gastric area has ME-NBI, microsurface architecture begins to disintegrate,
irregular pattern with indigo carmine stain (b). Under and corkscrew-like vessels are observed (d)
8 Endoscopic Submucosal Dissection for Gastric Cancer: Its Indication, Technique, and Our Experience 103

c d

Fig. 8.9 (continued)

Fig. 8.10 Negative biopsies (4 points around the lesion)

were performed to determine the margin histopathologi-
cally. ESD including the 4 negative biopsies is a good
method for complete resection (arrows point to the biopsy
points)

Conclusion The outcome of expanded criteria for ESD

We summarized the technique used for ESD in EGC appears promising. However, data is
in EGC. With the improvements in equipment, limited and studies to validate the “new” crite-
technique, and endoscopic skills, the indica- ria are warranted.
tion for ESD has now expanded to include
undifferentiated cancers. However, due dili-
gence is needed in determining which patients References
are good candidates for this procedure.
Performing ER is only half of the equation in 1. Japanese Research Society for Gastric Cancer. The
the treatment of EGC. We cannot overempha- general rules for the gastric cancer study in surgery
and pathology. Jpn J Surg. 1981;11:127–39.
size the importance of proper recognition of 2. Japanese Gastric Cancer Association. Japanese gas-
lesions and this can be achieved by tric cancer treatment guidelines 2010 (ver. 3). Gastric
ME-NBI. Identification of lesions amenable Cancer. 2011;14:113–23.
for endoscopic treatment and endoscopic 3. Gotoda T, Yanagisawa A, Sasako M, et al. Incidence
of lymph node metastasis from early gastric cancer:
skills is equally important to accomplish a estimation with a large number of cases at two large
safe and curative treatment. centers. Gastric Cancer. 2000;3:219–25.
104 H. Sato and H. Inoue

4. Tada M, Murakami A, Karita M, et al. Endoscopic resec- 20. Yokoyama A, Inoue H, Minami H, et al. Novel narrow-
tion of early gastric cancer. Endoscopy. 1993;25:445–50. band imaging magnifying endoscopic classification for
5. Deyhle P, Largiader F, Jenny S, et al. A method for early gastric cancer. Dig Liver Dis. 2010;42:704–8.
endoscopic electroresection of sessile colonic polyps. 21. Im JP, Kim SG, Kim JS, et al. Time-dependent mor-
Endoscopy. 1973;5:38–40. phologic change in depressed-type early gastric can-
6. Inoue H, Takeshita K, Hori H, et al. Endoscopic cer. Surg Endosc. 2009;23:2509–14.
mucosal resection with a cap-fitted panendoscope for 22. Shimizu S, Tada M, Kawai K. Early gastric cancer: its
esophagus, stomach, and colon mucosal lesions. surveillance and natural course. Endoscopy. 1995;27:
Gastrointest Endosc. 1993;39:58–62. 27–31.
7. Lee JI, Kim JH, Choi BJ, et al. Indication for endo- 23. Minami S, Gotoda T, Ono H, et al. Complete endo-
scopic treatment of ulcerative early gastric cancer scopic closure of gastric perforation induced by endo-
according to depth of ulcer and morphological change. scopic resection of early gastric cancer using endoclips
J Gastroenterol Hepatol. 2012;27:1718–25. can prevent surgery (with video). Gastrointest Endosc.
8. Gotoda T, Yamamoto H, Soetikno RM. Endoscopic 2006;63:596–601.
submucosal dissection of early gastric cancer. 24. Sano T, Okuyama Y, Kobori O, et al. Early gastric
J Gastroenterol. 2006;41:929–42. cancer. Endoscopic diagnosis of depth of invasion.
9. Lee H, Yun WK, Min BH, et al. A feasibility study on Dig Dis Sci. 1990;35:1340–4.
the expanded indication for endoscopic submucosal dis- 25. Namieno T, Koito K, Hiigashi T, et al. Endoscopic
section of early gastric cancer. Surg Endosc. 2011;25: prediction of tumor depth of gastric carcinoma for
1985–93. assessing the indication of its limited resection. Oncol
10. Shimura T, Joh T, Sasaki M, et al. Endoscopic submu- Rep. 2000;7:57–61.
cosal dissection is useful and safe for intramucosal 26. Lv SX, Gan JH, Ma XG, et al. Biopsy from the base and
gastric neoplasms in the elderly. Acta Gastroenterol edge of gastric ulcer healing or complete healing may
Belg. 2007;70:323–30. lead to detection of gastric cancer earlier: an 8 years
11. Isomoto H, Ohnita K, Yamaguchi N, et al. Clinical endoscopic follow-up study. Hepatogastroenterology.
outcomes of endoscopic submucosal dissection in 2012;59:947–50.
elderly patients with early gastric cancer. Eur J 27. Banerjee S, Cash BD, Dominitz JA, et al. The role of
Gastroenterol Hepatol. 2010;22:311–7. endoscopy in the management of patients with pep-
12. Ishikawa S, Togashi A, Inoue M, et al. Indications for tic ulcer disease. Gastrointest Endosc. 2010;71:
EMR/ESD in cases of early gastric cancer: relationship 663–8.
between histological type, depth of wall invasion, and 28. Nagahama T, Yao K, Maki S, et al. Usefulness of
lymph node metastasis. Gastric Cancer. 2007;10:35–8. magnifying endoscopy with narrow-band imaging for
13. Seiichiro A, Ichiro O, Haruhisa S, et al. Short- and determining the horizontal extent of early gastric can-
long-term outcomes of endoscopic submucosal dis- cer when there is an unclear margin by chromoendos-
section for undifferentiated early gastric cancer. copy (with video). Gastrointest Endosc. 2011;74:
Endoscopy. 2013;45(09):703–7. 1259–67.
14. Matsui N, Akahoshi K, Nakamura K, et al. Endoscopic 29. Kiyotoki S, Nishikawa J, Satake M, et al. Usefulness
submucosal dissection for removal of superficial gas- of magnifying endoscopy with narrow-band imaging
trointestinal neoplasms: a technical review. World J for determining gastric tumor margin. J Gastroenterol
Gastrointest Endosc. 2012;4:123–36. Hepatol. 2010;25:1636–41.
15. Oyama T. Basic technique of endoscopic submucosal 30. Nakayoshi T, Tajiri H, Matsuda K, et al. Magnifying
dissection for esophageal and gastric tumor-knack, pit endoscopy combined with narrow band imaging system
fall, and indications. Medical View Co., Ltd, Tokyo; for early gastric cancer: correlation of vascular pattern
2007. with histopathology (including video). Endoscopy.
16. Kodashima S, Fujishiro M, Koike K. The features of 2004;36:1080–4.
the high-frequency Knives for ESD of colorectal neo- 31. Hayee B, Inoue H, Sato H, et al. Magnification narrow-
plasms. Stomach Intestine. 2013;48:193–6. band imaging for the diagnosis of early gastric cancer:
17. Rey JF, Beilenhoff U, Neumann CS, et al. European a review of the Japanese literature for the Western
Society of Gastrointestinal Endoscopy (ESGE) guide- endoscopist. Gastrointest Endosc. 2013;78:452–61.
line: the use of electrosurgical units. Endoscopy. 32. Kobara H, Mori H, Fujihara S, et al. Prediction of
2010;42:764–72. invasion depth for submucosal differentiated gastric
18. Sakita T, Oguro Y, Takasu S, et al. Observations on cancer by magnifying endoscopy with narrow-band
the healing of ulcerations in early gastric cancer. The imaging. Oncol Rep. 2012;28:841–7.
life cycle of the malignant ulcer. Gastroenterology. 33. Kikuchi D, Iizuka T, Hoteya S, et al. Usefulness of
1971;60:835–9. passim. magnifying endoscopy with narrow-band imaging for
19. Takahashi A, Oyama O. ESD for early gastric cancer determining tumor invasion depth in early gastric can-
with ulcer scar. Stomach Intestine. 2013;48:63–71. cer. Gastroenterol Res Pract. 2013;2013:217695.
 Multimodality Therapy
in Gastric Cancer 9
Usha Malhotra and Mei Ka Fong

Introduction high failure rates with surgical resection alone,

optimal management of locoregional gastric
Gastric cancer is common worldwide with cancer involves utilization of chemotherapy and/
extensive variations between continents. Though or radiation in conjunction with surgery. For
the incidence of gastric cancer overall has been patients with surgically unresectable advanced
declining steadily since the 1970s, it still remains stage or metastatic disease, treatment algorithms
an enormous health issue. Also notable is the are palliative involving utilization of cytotoxic
increase in proximal gastric cancer in the western chemotherapy in combination with other modal-
hemisphere over the past few decades. Worldwide ities for symptomatic management.
989,600 new cases of gastric cancer were esti- In this chapter, pivotal studies supporting the
mated in 2008 with 738,000 deaths accounting role of chemotherapy and radiation in manage-
for 10 % of total cancer-related deaths [1]. In the ment of both localized and metastatic gastric can-
United States, 21,600 new cases of gastric cancer cer will be reviewed. Surgical approaches are
were estimated in 2013 with 10,990 deaths [2]. described in detail in other sections.
Based on SEER data, 5-year survival rate still
remains a dismal 25 % which is a marginal
improvement over 14 % in 1970s. Management of Localized
For patients diagnosed with localized or and Locoregional Gastric Cancer
locally advanced disease defined as disease con-
fined to the stomach and/or regional lymph Based on patterns and risk of recurrence, the ben-
nodes, the intent of therapy is curative. His- efit of additional chemotherapy and/or radiation
torically, for this subset of patients, the mainstay is most substantial in stages IB–III where the
of treatment has been surgery. However, due to cancer is resectable and potentially curable but
there is a high risk of recurrence. There is cur-
rently a wide variation in utilization of these
U. Malhotra, MD (*) modalities in the adjuvant (after surgical resec-
Department of Medicine, tion) and neoadjuvant (prior to surgical resection)
Roswell Park Cancer Institute,
settings with no established standard of care.
Elm and Carlton Street, Buffalo, NY 41263, USA
e-mail: [email protected] This is due to a multitude of factors including
postulated regional variations in tumor biology
M.K. Fong, PharmD
Department of Pharmacy, between eastern and western hemisphere dictat-
Roswell Park Cancer Institute, Buffalo, NY, USA ing a preference for certain treatment paradigms

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 105
DOI 10.1007/978-3-319-09342-0_9, © Springer International Publishing Switzerland 2015
106 U. Malhotra and M.K. Fong

Table 9.1 Perioperative chemotherapy trials

Treatment arms Primary end point Secondary end points
MAGIC [3]
Surgery alone Epirubicin 50 mg/m2 day 1 5-year OS 5-year PFS
Cisplatin 60 mg/m2 day 1 HR 0.75 HR 0.66
5-FU 200 mg/m2 daily 95 % CI 0.6–0.93 95 % CI 0.53–0.81
Every 21 days ×3 cycles P < 0.001
Surgery Rate of resection
Epirubicin 50 mg/m2 day 1 79.3 % vs. 70.3 %
Cisplatin 60 mg/m2 day 1 (p = 0.03)
5-FU 200 mg/m2 daily
Every 21 days ×3 cycles
French FNCLCC/FFCD [5]
Surgery alone 5-FU 800 mg/m2 days 1–5 5-year OS Disease-free survival
Cisplatin 100 mg/m2 day 1 38 % vs. 24 % (HR 0.69; 34 % vs. 19 % (5-year rate:
95 % CI 0.5–0.95; p = 0.02) 34 % v 19 %; HR, 0.65;
95 % CI, 0.48–0.89;
P = 0.003)
Every 28 days ×2–3 cycles R0 resection
Surgery 84 % vs. 73 %
5-FU 800 mg/m2 days 1–5
Cisplatin 100 mg/m2 day 1 or 2
Every 28 days ×3–4 cycles
EORTC 40954 [7]
Surgery alone Cisplatin 50 mg/m2 on days 1, OS Rate of resection
15, 29
5-FU 2,000 mg/m2 continuous Trial terminated 81.9 % vs. 66.7 %
intravenous infusion over 24 h on prematurely (p = 0.036)
days 1, 8, 15, 22, 29, 36
Leucovorin 500 mg/m2 over 2 h PFS
every 48 days ×2 cycles
Surgery Trial terminated
prematurely

including surgical techniques and change in Role of Chemotherapy

classification of gastric and gastroesophageal
junction (GEJ) cancers over time. For instance, in Perioperative Neoadjuvant
the most recent TNM classification, tumors aris- Chemotherapy (Table 9.1)
ing in the gastric cardia that are within 5 cm of A number of studies have evaluated the benefit of
GEJ and extend to the GEJ are staged and treated chemotherapy both in the neoadjuvant/periopera-
as esophageal cancer rather than gastric cancer. tive as well as postoperative settings. The ratio-
Additionally, the majority of clinical trials have nale behind adding chemotherapy in the
included broad-category patients with esopha- neoadjuvant setting is potential downstaging of
geal, GEJ, and gastric cancers, hence making the cancer by early exposure to chemotherapy in che-
interpretation of data for any of the locations mosensitive cases and improving patient selec-
challenging. There is a lack of phase III trials tion by sparing surgery in patients that are at high
addressing these controversies and hence lack of risk of metastasis as micrometastatic disease may
a consensus for establishing uniform guidelines become evident after chemotherapy but prior to
worldwide. surgery. A possible disadvantage is delaying
9 Multimodality Therapy in Gastric Cancer 107

Fig. 9.1 Kaplan-Meier 1.0

curve showing overall Log-rank P = 0.02
survival from date of random Hazard ratio = 0.69
assignment (From Ychou (95 % CI, 0.50−0.95)
0.8
et al. [5] with permission)

Overall survival
0.6

0.4

0.2
Surgery
Chemotherapy + surgery

0 1 2 3 4 5 6 7
Time (months)

No. at risk
Surgery 111 79 53 38 27 16 13 7
Chemotherapy
+ surgery 113 93 65 53 41 27 17 14

surgery and hence resectable disease may become arm, p = 0.03). Hematologic toxicity with grade 3
unresectable in the interim. and 4 neutropenia was reported in 23 %, and the
A pivotal trial, MAGIC, conducted by Medical incidence of non-hematologic grade 3 and 4 tox-
Research Council (MRC) in the United Kingdom, icities was not very high (12 %) demonstrating an
evaluated the role of perioperative chemotherapy acceptable toxicity profile, but only 42 % of the
in combination with surgery. A total of 503 patients patients assigned to the combined modality arm
were randomly assigned to either surgery alone or were able to complete all therapy. This highlights
surgery with three preoperative and three postop- the decreased tolerance to chemotherapy in the
erative 21-day cycles of chemotherapy consisting postoperative setting. This trial established periop-
of epirubicin (50 mg/m2 day 1), cisplatin (60 mg/ erative chemotherapy as standard of care for oper-
m2 day 1), and 5-fluorouracil, 5-FU (200 mg/m2 able gastric cancer in Europe.
daily) also known as ECF [3]. Eligibility criteria Another trial demonstrating the benefit of
required the presence of T2 or more advanced perioperative chemotherapy is the French
biopsy-proven adenocarcinoma with good perfor- FNLCC/FFCD multicenter trial [5]. A total of
mance status. Seventy-four percent had gastric, 224 patients were randomized to surgery alone or
11 % distal esophageal, and 15 % had GEJ cancer. surgery with perioperative chemotherapy con-
The trial demonstrated benefit of adding chemo- sisting of infusional 5-FU (800 mg/m daily for
therapy with significant improvement in 5-year 5 days) and cisplatin (100 mg/m2 on day 1 or 2)
survival (36 % vs. 23 % hazard ratio [4], 0.75; every 28 days with two or three cycles delivered
95 % confidence interval (CI), 0.6–0.93 in favor of preoperatively and three or four cycles given
combined modality arm) as well as progression- postoperatively for a total of six cycles. Of the
free survival (HR 0.66, 95 % CI 0.53–0.81, 224 patients with stage II or higher resectable
p < 0.001 in favor of combined modality arm). disease enrolled in this trial, 55 had gastric 144
Additionally, increased rate of curative resection GEJ and 25 had distal esophageal cancer. This
was seen in the combined modality arm (79.3 % in trial also demonstrated an improvement in 5-year
combined modality arm and 70.3 % in surgery-only survival (38 % vs. 24 %, Fig. 9.1) and disease-free
108 U. Malhotra and M.K. Fong

survival (34 % vs. 19 %) with the addition of Table 9.2 Adjuvant chemotherapy
chemotherapy. Additionally, the rate of R0 resec- Primary Secondary
tion also improved with addition of perioperative Treatment arms end point end points
chemotherapy (84 % vs. 73 %). Like the prior ACTS-GS [16]
study, only 50 % of patients were able to receive Surgery Surgery Overall Relapse-free
alone survival at survival at
therapy postoperatively.
5 years: 5 years
In contrast, an EORTC randomized trial failed 72 % vs. 65.4 % vs.
to show any benefit of adding preoperative che- 61 % 53.1 % (HR
motherapy to surgery. In the EORTC 40954 trial, (HR 0.68; 0.653;
95 % CI 95 % CI
a total of 144 of the planned 360 patients with
0.52–0.87; 0.537–0.793)
stage III and IV gastric and GEJ adenocarcinoma p = 0.003)
were randomized to surgery or preoperative S1 80–120 mg
chemotherapy consisting of two 48-day cycles of daily for 4 weeks
cisplatin (50 mg/m2 on days 1, 15, and 29) and every 6 weeks
for 1 year
leucovorin with FU (leucovorin 500 mg/m2 over
CLASSIC [13]
2 h followed by FU 2,000 mg/m2 continuous infu-
Surgery Surgery Disease- Overall
sion over 24 h on days 1, 8, 15, 22, 29, and 36) alone Capecitabine free survival,
[6, 7]. The trial was stopped early due to poor 1,000 mg/m2 survival, 3 years:
accrual and failed to show a survival benefit with twice daily in 3 years: 78 % vs.
days 1–14 74 % vs. 69 %, HR
the addition of chemotherapy. Analysis of the
Oxaliplatin 59 % (HR 0.66, 95 %
accrued patients demonstrated an improvement in 0.56, 95 % CI 0.51–0.85
130 mg/m2 on
the R0 resection rate (81.9 % vs. 66.7 %, p = .036) day 1 every CI
and a higher incidence of postoperative complica- 21 days for 0.44–0.72,
p < 0.0001)
tions in the chemotherapy arm (27 % vs. 16 %). 8 cycles

Adjuvant Chemotherapy (Table 9.2)

A number of trials have been conducted to assess significantly better at 65.4 % in the S1 group
role of chemotherapy in the adjuvant setting for compared to 53.1 % in the surgery-only group
gastric cancer and have failed to show any sur- (HR 0.653; 95 % CI, 0.537–0.793).
vival benefit [6, 8–12]. Two large phase III stud- In the multicenter CLASSIC trial, the adju-
ies, ACTS-GS and CLASSIC conducted in Japan vant chemotherapy regimen consisted of
and East Asia (South Korea, Taiwan and China), capecitabine (1,000 mg/m2 twice daily in days
respectively, demonstrated a significant survival 1–14) plus oxaliplatin (130 mg/m2 on day 1)
benefit with adjuvant chemotherapy establishing given every 21 days for 8 cycles. A total of 1,035
adjuvant chemotherapy after adequate surgery as patients with stage II, IIIA, or IIIB gastric cancer
standard of care in these regions. In the ACTS-GS patients were randomly assigned to surgery with
trial, 1,059 patients with stage II and III gastric D2 lymphadenectomy alone vs. surgery followed
cancer were randomized to surgery (included a by adjuvant chemotherapy. At a median follow-
D2 lymphadenectomy in both arms) with and up of 34 months, there was a significant improve-
without adjuvant therapy with S1 administered at ment in 3-year disease-free survival in the
a dose of 80–120 mg daily for 4 out of 6 weeks chemotherapy arm (74 vs. 59 % in chemotherapy
for one year. S1 is an oral fluoropyrimidine that vs. surgery-alone arm, HR for death 0.56, 95 %
consists of three components: ftorafur (tegafur), CI 0.44–0.72, p < 0.0001) as well as a marginally
gimeracil (5-chloro-2,4 dihydropyridine), and significant improvement in OS at the time of ini-
oteracil (potassium oxonate). The addition of tial report in 2012 (83 vs. 78 %, HR 0.72, 95 %
adjuvant S1 led to an improvement in 5-year CI 0.52–1.00) with more robust improvement in
overall survival from 61 to 72 %. Relapse-free OS with longer follow-up (78 vs. 69 %, HR for
survival at 5 years was also found to be death 0.66 %, 95 % CI 0.51–0.85) [13, 14].
9 Multimodality Therapy in Gastric Cancer 109

On evaluation of tolerance and toxicity, grade 3 addition of adjuvant radiation (27 % vs. 10 % in
and 4 adverse events were reported in 56 % of favor of radiation), but no significant difference
patients in the combined modality group and in was observed in OS between the three arms.
only 6 % of patients in the surgery-only group. In another study conducted by the European
Only 67 % of patients were able to complete all 8 Organization for Research and Treatment of
planned cycles of chemotherapy with 90 % of Cancer (EORTC), 115 patients underwent sur-
patients requiring dose modifications. gery and then were randomly assigned to four
High survival rates even in the surgery-alone different groups in the adjuvant setting [17]. The
arms in both these trials have led to a debate first group received 55.5 Gy of postoperative
about the pertinence of this data to western popu- radiation only, while the other three groups
lation. Epidemiological and clinical variations in received radiation in combination with short-
gastric cancer between eastern and western popu- term 5-FU, long-term 5-FU, and both short-term
lations have led to a hypothesis that there is a dif- and long-term 5-FU. Unadjusted analysis showed
ference in biology of gastric cancer and hence a significant difference in OS between the four
variable response to therapies in different parts of groups, but when other pertinent prognostic fac-
the world. tors were added to the model, there was no sig-
Additionally a recent meta-analysis also sup- nificant difference in survival.
ported the role of adjuvant chemotherapy for A number of trials have evaluated the role of
resectable gastric cancer [15]. Based on these radiation in combination with chemotherapy in
studies, adjuvant chemotherapy only is the stan- the adjuvant setting. In the Intergroup 0116 study,
dard of care in East Asia. 556 patients with stage IB through IV gastric or
gastroesophageal cancer were randomized to
observation vs. adjuvant chemoradiation after
Role of Radiation surgery [18]. Chemoradiation consisted of an ini-
tial 28-day cycle of 5-FU and leucovorin given
Radiation in most cancers has been shown to on days 1–5, followed by 5-FU based concurrent
have a role in improving local disease control. chemoradiation for 5 weeks (radiation dosage
Based on the natural history of gastric cancer, was 45 Gy at 1.8 Gy per day, given 5 days per
local recurrence has been reported in a high pro- week along with 5-FU on first 4 and last 3 days of
portion of cases, which led evaluation of radia- radiation), break for 1 month, and then two addi-
tion with or without chemotherapy in addition to tional cycles of chemotherapy. At a 4-year
surgical resection for patients with potentially median follow-up, there was a significant differ-
curable disease. ence in median survival (36 vs. 27 months),
In one of the earlier studies conducted by the 3-year disease-free survival (48 % vs. 31 %), OS
British Stomach Cancer Group (Table 9.3), (50 % vs. 41 %), and local failures (29 % vs.
patients were randomly assigned to surgery 19 %) in favor of the tri-modality therapy arm.
alone, surgery followed by 45–50 Gy of radia- With a longer 10-year median follow-up, OS
tion, and surgery followed by chemotherapy con- continued to be significantly better in the com-
sisting of eight courses of 5-FU, doxorubicin, bined modality arm (43 % vs. 28 %, HR 1.32,
and mitomycin [9]. This trial demonstrated an 95 % CI 1.10–1.60, p = 0.0046) [19]. This study
improvement in the local control rate with the established the role of concurrent chemoradiation

Table 9.3 Postoperative radiation (RT)

British Stomach Cancer Group [9, 17]
Surgery alone Surgery Surgery Overall survival Local control
45–50 Gy RT 5FU, doxorubicin, No difference in 17 % increase in
mitomycin ×8 cycles OS among the improvement in local
3 treatment arms control with adjuvant RT
110 U. Malhotra and M.K. Fong

as an effective adjuvant regimen but has been a paclitaxel/cisplatin (PC) for two cycles followed
focus of considerable criticism as more than half by concurrent chemoradiation with paclitaxel and
of the patients enrolled in this study underwent cisplatin. The PCF arm was closed early due to
inadequate D0 lymph node dissection and only excessive gastrointestinal toxicity and the trial
10 % underwent D2 lymph node dissection. failed to achieve its primary end point of improve-
In a CALGB 80101 study, adjuvant combina- ment in 2-year DFS and, hence, further evaluation
tion chemotherapy with chemoradiation based on in a phase III study was not recommended [22].
the INT 0116 regimen was compared with a more A recent phase III ARTIST trial conducted in
intense postoperative regimen consisting of one Korea provided a direct comparison of chemother-
cycle of ECF followed by concurrent chemora- apy and chemoradiation in the adjuvant setting
diation and 2 more cycles of dose-reduced after surgery with D2 lymph node dissection [23].
ECF. The rationale was that more intensive sys- Four hundred and fifty-eight patients were ran-
temic chemotherapy may translate to better domly assigned postoperatively to either chemo-
OS. As reported in the American Society of therapy arm consisting of capecitabine and
Clinical Oncology meeting in 2011, there was no cisplatin (capecitabine 2,000 mg/m2/day 1–14 and
difference in survival between the two arms [20]. cisplatin 60 mg/m2 on day 1, repeated every 3
To evaluate an alternative chemotherapy back- weeks) for 6 cycles or the chemoradiation arm
bone with concurrent radiation, a trial conducted consisting of two cycles of chemotherapy with
at MD Anderson Cancer Center evaluated a neo- capecitabine and cisplatin as above followed by
adjuvant regimen consisting of induction chemo- concurrent chemoradiation for 5 weeks
therapy for 2 cycles (5-FU 750 mg/m2/day days (capecitabine 1,650 mg/m2/day with radiation,
1–5, cisplatin 15 mg/m2/days 1–5, and paclitaxel 1.8 Gy/day for 5 days/week for a total of 45 Gy)
200 mg/m2 day 1) followed by concurrent chemo- followed by two additional cycles of chemother-
radiation (45 Gy over 5 weeks, 5-FU 300 mg/m2/ apy. Though DFS was not significantly prolonged
day 5 days/week, and paclitaxel 45 mg/m2 on with addition of radiation for the entire study
days 1, 8, 15, 22, and 29) and then surgery. Of the group (p = 0.0862), a subgroup of patients with
41 patients enrolled, the majority had proximal surgical pathological lymph node involvement
gastric cancer (83 %), 40 patients underwent sur- experienced superior DFS in the chemoradiation
gery, and 78 % had an R0 resection. Pathological arm (p = 0.0365). Based on these results a subse-
complete and partial response (defined as less quent trial ARTIST II will evaluate the role of
than 10 % residual cancer cells) was seen in 20 chemoradiation in node-positive disease.
and 15 % of patients, respectively. At a median Table 9.4 summarizes the abovementioned
follow-up of 36 months, OS was found to be sig- trials.
nificantly associated with pathological response
(both complete and partial, p = 0.006) in addition
to R0 resection, postsurgical nodal positivity, N Management of Metastatic
stage, and T stage [21]. Gastric Cancer
In the Radiation Therapy Oncology Group
(RTOG) 0114 randomized phase II study, two Unlike localized and locoregional gastric cancer,
postoperative adjuvant regimens consisting of the predominant method of treatment for meta-
induction chemotherapy followed by concurrent static gastric cancer is chemotherapy. Best support-
chemoradiation were evaluated. A total of 87 ive care for metastatic gastric cancer has a median
patients were randomly assigned to receive two survival of 3 months [24]. With the advent of newer
cycles of chemotherapy consisting of paclitaxel/ chemotherapy treatment options in advanced gas-
cisplatin/5-FU (PCF) followed by concurrent tric cancer, survival has improved by 60 % (HR
chemoradiation with paclitaxel and 5-FU or 0.39) with minimal impact on quality of life [24].
Table 9.4 Postoperative chemoradiation 9

Intergroup 0116 [19]

Surgery only Surgery 4-year OS 3-year DFS
5-FU plus leucovorin days 1–5 in a 28-day cycle 36 months vs. 27 months 48 % vs. 31 %
×1 cycle followed by RT 45 Gy given over
5 days per week for 5 weeks with 5-FU plus
leucovorin on days 1–5 and days 28–32 followed
by 1 month off treatment then 5-FU plus
leucovorin days 1–5 in a 28-day cycle ×2 cycles
10-year OS
43 % vs. 28 % HR 1.32,
95 % CI 1.10–1.60,
p = 0.0046
EORTC [17]
Surgery Surgery Surgery Surgery No difference in OS among the 4 arms
Multimodality Therapy in Gastric Cancer

55.5 Gy RT 55.5 Gy RT + 5-FU short term 55 Gy RT + 5-FU long term 55.5 Gy RT + 5-FU short after adjusting for appropriate prognostic
term and long term factors
CALGB 80101 [20]
Surgery Surgery No difference in OS
ECF ×1 cycle followed by
concurrent chemoradiation
followed by reduced-dose
ECF ×2 cycles
RTOG 0114 [22]
Paclitaxel/cisplatin/5-FU Paclitaxel/cisplatin ×2 cycles Disease-free survival,
×2 cycles 2 years
Chemoradiation with paclitaxel and cisplatin Failed to reach 2-year
×2 cycles disease-free survival
PCF arm terminated early
due to toxicity
ARTIST [23]
Capecitabine and Capecitabine and cisplatin ever 21 days ×2 Disease-free survival
cisplatin every 21 days cycles followed by chemoradiation 45 Gy over
×6 cycles 5 days for 5 weeks
Not statistically
significant; however DFS
improved significantly in
111

patients with pathological

lymph node involvement
112 U. Malhotra and M.K. Fong

First-Line Treatment of Metastatic survival and overall response rates as secondary

Gastric Cancer outcomes. The DCF arm significantly improved
TTP by 1.9 months compared to CF (5.6 months
First-line treatment of metastatic gastric cancer vs. 3.7 months, p < 0.001). In addition to TTP,
began with single-agent chemotherapy. Many overall survival significantly improved, from
different classes of chemotherapeutic agents have 8.6 months in the CF arm to 9.2 months in
been used, including anthracyclines, platinums, the DCF arm with a median follow-up of
taxanes, and fluoropyrimidines. However, single- 23.4 months. The two-year survival was doubled
agent chemotherapy had an overall response rate in the DCF arm, from 9 to 18 %.
of approximately 20 %[25]. Wagner et al. con- The frequency of neutropenia was higher in
ducted a meta-analysis regarding the effects of the DCF arm with 82 % developing grade 3/4
chemotherapy on advanced gastric cancer and neutropenia and 29 % with reported febrile neu-
showed that the combination of chemotherapy tropenia or neutropenia infection. However, the
agents improved overall survival by 17 % [24]. incidence of febrile neutropenia or neutropenia
The combination cisplatin and fluorouracil (CF) infection was reduced by more than half (12 %)
became the standard of comparison, with a when secondary prophylaxis with granulocyte
median survival of approximately 7 months [26]. colony-stimulating growth factors was adminis-
The combination fluorouracil and irinotecan, tered. The frequency of the non-hematologic tox-
which had been used with success in metastatic icities was similar in both arms with the exception
colorectal cancer, was studied in the V306 nonin- of diarrhea where DCF had a significantly higher
feriority trial comparing it to CF [27]. The V306 incidence. The CF arm had a higher incidence of
study demonstrated noninferiority of fluorouracil stomatitis, although this did not prove to be sta-
and irinotecan, with a median overall survival of tistically significant. The frequency of dose
9 months in the irinotecan arm vs. 8.7 months in reductions and treatment discontinuation was
the control arm. Similarly, the JCOG9912 sup- similar in both arms. In addition, follow-up study
ported the efficacy of irinotecan with fluorouracil assessing quality of life was prospectively per-
in gastric cancer with a median overall survival formed using a validated quality of life question-
of 12.3 months [16]. naire [29]. The quality of life study showed time
The doublet therapy with platinum and a to 5 % definitive deterioration was significantly
pyrimidine analogue was then subsequently higher in the DCF arm at 6.5 months compared to
paired with other chemotherapy classes and 4.2 months in the CF arm, indicating that quality
studied for improved survival compared to that of life is preserved in patients receiving DCF.
of the doublet therapy. The TAX325 trial studied Due to the high risk of neutropenia in the DCF
the effect of docetaxel, cisplatin, and fluoroura- arm, several modified versions of the DCF regi-
cil (DCF) compared to CF in untreated advanced men were made and studied. A phase II study by
gastric cancer [28]. Ninety-seven percent of the Shah et al. [30] involving 60 patients had two
patients enrolled in this study had metastatic arms that investigated the effects of a modifica-
gastric cancer with the remaining population tion, using the results of the historical DCF regi-
having locally advanced or recurrent gastric can- men for comparison. The modified DCF regimen
cer. The study included patients who had was as such: docetaxel 40 mg/m2, fluorouracil
received prior treatment with radiation, surgery, 400 mg/m2 bolus, leucovorin 400 mg/m2, fluoro-
and chemotherapy. However, this study essen- uracil 1,000 mg/m2 daily for 2 days starting on
tially assessed the effects of DCF in the chemo- day 1, and followed by cisplatin 40 mg/m2 on day
therapy-naïve population, as the percentage of 3 of a 14-day cycle. The second arm of the
patients who had received prior treatment with study kept the original DCF regimen (docetaxel
chemotherapy was 3 %. The study was powered 75 mg/m2, cisplatin 75 mg/m2, fluorouracil
to measure time to progression (TTP) with supe- 1,000 mg/m2 daily for 5 days starting on day 1
riority of DCF as compared to CF in overall of a 28-day cycle) but with the addition of
9 Multimodality Therapy in Gastric Cancer 113

granulocyte colony-stimulating factor (GCSF). significant difference in any of the arms regard-
Thirty-eight percent of patients who received ing progression-free survival and overall response
mDCF had grade 3/4 neutropenia, with 4 % devel- rate. Each regimen had its own unique set of tox-
oping febrile neutropenia. Forty-three percent of icities. There was a higher incidence of grade 3/4
those who received standard DCF with GCSF had hand-foot syndrome and neutropenia in ECX,
neutropenia, and there was incidence of 14 % compared to ECF. Conversely, both the EOF and
febrile neutropenia. The 6-month PFS was 90 and EOX arms held significantly lower rates of grade
78 % in the mDCF and standard DCF arms, 3/4 neutropenia when compared to ECF. However,
respectively. This small study demonstrates a the frequency of febrile neutropenia was similar
reduction in neutropenia and febrile neutropenia among all groups. Other grade 3/4 toxicities that
with mDCF, without compromising on efficacy. were significantly higher in the EOF arm included
There are several toxicities and inconve- anemia, diarrhea, stomatitis, and peripheral neu-
niences associated with the CF regimen that ropathy when compared to ECF. EOX also had a
could be improved upon. While cisplatin is significantly higher rate of grade 3/4 diarrhea,
shown to be active in gastric cancer, some of its peripheral neuropathy, and lethargy. The REAL-2
toxicities such as neuropathy, nephropathy, and study supports the use of capecitabine and oxali-
ototoxicity may limit the use of cisplatin after platin in triple therapy with epirubicin.
toxicities have set in. Oxaliplatin is a third-
generation platinum with an oxalate leaving
group, replacing the chlorine leaving groups that Targeted Therapies
are found in cisplatin. The oxalate binding to the for Metastatic Disease
DNA adducts results in a bulky side group inhib-
iting DNA base excision [31]. This mechanism Targeted therapies have more recently made their
has proved to be effective in gastrointestinal way into cancer treatments. The first targeted
malignancies, such as colorectal cancer [32]. therapy approved in metastatic gastric cancer was
Fluorouracil is administered as a continuous bevacizumab, a vascular endothelial growth fac-
intravenous infusion, requiring patients to carry tor (VEGF) inhibitor. At the time of the study,
an infusion pump or hospital admission for che- median survival for metastatic gastric cancer with
motherapy. Capecitabine is an oral fluoropyrimi- cisplatin-based treatment capped at 10 months
dine that was shown to be noninferior to its [34]. The phase II study of bevacizumab in gastric
intravenous counterpart in the treatment of cancer in combination with cisplatin and irinote-
colorectal cancer. The REAL-2 study evaluated can improved TTP to 8.3 months and showed
the potential replacement of cisplatin with oxali- an overall survival of 12.3 months [35]. VEGF
platin and fluoropyrimidine with capecitabine inhibition-related toxicities were seen including
through a two-by-two study design, powered to grade 3 hypertension in 28 % of patients. Notably,
determine noninferiority [33]. In this study, 25 % of patients developed thromboembolism.
patients were randomized to either epirubicin Similarly, Shah et al. utilized bevacizumab in
with cisplatin or oxaliplatin. Each group was then combination with modified DCF in a phase II trial
further randomized to receive either fluorouracil showing improvement in median progression-free
or capecitabine. The majority of the study partici- survival of 12 months [36].
pants had metastatic disease, but all participants The human epidermal growth factor receptor
were chemotherapy naïve. The median follow-up 2 (HER2) inhibitor, trastuzumab, was originally
was similar among all groups, with a range of developed for HER2-positive breast cancer and
17.5–19.3 months. The study found that both has shown to improve outcomes in HER2 protein
substitutions met their prespecified margin for expressing gastric cancer. The ToGA trial ran-
noninferiority. The survival data for EOX showed domized patients to receive cisplatin with a fluo-
an improvement over ECF by 1.3 months with a ropyrimidine with or without trastuzumab [37].
9 % increase in 1-year survival. There was no Participants in this study could not have received
114 U. Malhotra and M.K. Fong

Fig. 9.2 (a) Median a Median

overall survival and overall
survival
(b) progression-free 1.0
Events (months) HR (95 % CI) p value
survival in the primary 0.9
analysis population. Trastuzumab 167 13.8 0.74 (0.60–0.91) 0.0046
0.8 plus chemotherapy
HR hazard ratio Chemotherapy alone 182 11.1
0.7

Survival probability
(From Bang et al. [37]
with permission) 0.6

0.5

0.4

0.3

0.2

0.1
11.1 13.8
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36
Time (months)
Number ar risk
Trastuzumab plus 294 277 246 209 173 147 113 90 71 56 43 30 21 13 12 6 4 1 0
chemotherapy
Chemotherapy 290 266 223 185 143 117 90 64 47 32 24 16 14 7 6 5 0 0 0
alone

b Median
1.0 progression-free
survival
0.9 Events (months) HR (95 % CI) p value

0.8 Trastuzumab 226 6.7 0.71 (0.59–0.85) 0.0002

plus chemotherapy
0.7
Survival probability

Chemotherapy alone 235 5.5

0.6

0.5

0.4

0.3

0.2

0.1
5.5 6.7
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
Time (months)
Number ar risk
Trastuzumab plus 294 258 201 141 95 60 41 28 21 13 9 8 6 6 6 4 2 0
chemotherapy
Chemotherapy 290 238 182 99 62 33 17 7 5 3 3 2 2 1 1 0 0 0
alone

prior chemotherapy treatment for their metastatic patients developed cardiac complications and
disease and had to have adequate cardiac func- there was no significant difference between the
tion, as measured by ejection fraction, blood two arms.
pressure, and medical history. The study was pri-
marily designed to detect overall survival with
secondary measures of progression-free survival, Salvage Therapy in Metastatic
TTP, and overall tumor response rate. Median Gastric Cancer
overall survival in the trastuzumab arm was
13.8 months, compared to 11.1 months in the Several studies have been conducted regarding
chemotherapy alone arm (Fig. 9.2). There were salvage therapy in metastatic gastric cancer. For
no significant differences in grade 3/4 toxicities the most part, all agents and combinations studied
with the exception of diarrhea, which was 5 % in first-line treatment may be successfully used as
higher in the trastuzumab arm. Less than 1 % of second-line treatment and salvage therapy.
9 Multimodality Therapy in Gastric Cancer 115

However, residual side effects from previous practices around the world. In North America,
treatments, such as neuropathy from cisplatin, postoperative chemoradiation remains popu-
may limit the effectiveness of future combination lar based on INT 0116, perioperative chemo-
therapies employing similar agents. Kang et al. therapy based on MAGIC trial is preferred in
studied the benefits of salvage chemotherapy in Europe and also employed in North America,
patients with metastatic gastric cancer who failed while in East Asia the trend is more toward
first-line therapy [38]. Study participants were postoperative chemotherapy after surgical
randomized in a 2:1 ratio of salvage chemother- resection with standard of care being D2
apy to best supportive care. Salvage chemother- lymph node dissection. Role of targeted thera-
apy involved single-agent docetaxel 60 mg/m2 pies in this setting is still under clinical evalu-
every 3 weeks or single-agent irinotecan 150 mg/ ation and recommended only in the setting of
m2 every 2 weeks. The study was powered to a clinical trial.
detect an improvement in overall survival. There While metastatic gastric cancer is chemo-
was a 34 % reduced risk of death in the chemo- therapy sensitive, the relapse rate is high with
therapy arms compared to best supportive care a low 2-year survival. Over the years, combi-
(HR 0.657, p = 0.007); however, there was no dif- nation chemotherapy has improved median
ference between the two chemotherapy arms. OS with minimal added toxicities. The devel-
While there are no published studies regarding opment of targeted therapies has improved the
trastuzumab in the second-line setting, a newer landscape of cancer treatment outcomes in
agent was recently studied as second line in a general, but its role in gastric cancer is limited.
phase III trial. Ramucirumab is a fully human- With the success of targeted therapies such as
ized IgG1 monoclonal antibody that targets the trastuzumab and ramucirumab, there is great
VEGF2 receptor [39]. The REGARD trial ran- potential for further improvements in survival
domized patients with metastatic gastric cancer in patients with metastatic gastric cancer.
who failed first-line therapy to either ramuci-
rumab 8 mg/kg every 14 days plus best support-
ive care or placebo plus best supportive care [40]. References
The primary objective was overall survival with
secondary outcomes in progression-free survival, 1. Jemal A, et al. Global cancer statistics. CA Cancer J
Clin. 2011;61(2):69–90.
overall response rate, duration of response, and
2. National Cancer Institute: PDQ® Gastric Cancer
quality of life. The results of the study showed a Treatment. Bethesda: National Cancer Institute. Date
median OS advantage of 1.4 months in the ramu- last modified 15 Feb 2013. Available at. http://cancer.
cirumab arm (p = 0.042). As expected with VEGF gov/cancertopics/pdq/treatment/gastric/Health
Professional. Accessed 20 Nov 2013.
inhibition, the frequency of hypertension was
3. Cunningham D, et al. Perioperative chemotherapy
higher in the ramucirumab arm (16 % vs. 8 %), versus surgery alone for resectable gastroesophageal
but with no other significant differences in toxici- cancer. N Engl J Med. 2006;355(1):11–20.
ties between the two arms. 4. du Bois A, et al. Addition of epirubicin as a third drug to
carboplatin-paclitaxel in first-line treatment of advanced
ovarian cancer: a prospectively randomized gyneco-
Conclusion logic cancer intergroup trial by the Arbeitsgemeinschaft
Poor overall survival and high recurrence rates Gynaekologische Onkologie Ovarian Cancer Study
after surgery for potentially curable localized Group and the Groupe d’Investigateurs Nationaux pour
l’Etude des Cancers Ovariens. J Clin Oncol. 2006;
and locoregional gastric cancer have sup-
24(7):1127–35.
ported the need for additional therapy. Based 5. Ychou M, et al. Perioperative chemotherapy compared
on pivotal phase III trials, perioperative che- with surgery alone for resectable gastroesophageal
motherapy, postoperative chemoradiation, and adenocarcinoma: an FNCLCC and FFCD multicenter
phase III trial. J Clin Oncol. 2011;29(13):1715–21.
postoperative chemotherapy are acceptable
6. Adjuvant treatments following curative resection for
options, but the choice of regimen varies gastric cancer. The Italian Gastrointestinal Tumor
widely based on institutional and regional Study Group. Br J Surg. 1988;75(11):1100–4.
116 U. Malhotra and M.K. Fong

7. Schuhmacher C, et al. Neoadjuvant chemotherapy radiotherapy (CRT) compared with bolus 5-FU/LV
compared with surgery alone for locally advanced can- before and after CRT: Intergroup trial CALGB 80101.
cer of the stomach and cardia: European Organisation ASCO Meet Abstr. 2011;29(15_suppl):4003.
for Research and Treatment of Cancer randomized 21. Ajani JA, et al. Paclitaxel-based chemoradiotherapy
trial 40954. J Clin Oncol. 2010;28(35):5210–8. in localized gastric carcinoma: degree of pathologic
8. Engstrom PF, et al. Postoperative adjuvant response and not clinical parameters dictated patient
5-fluorouracil plus methyl-CCNU therapy for gastric outcome. J Clin Oncol. 2005;23(6):1237–44.
cancer patients. Eastern Cooperative Oncology Group 22. Schwartz GK, et al. Randomized Phase II Trial
study (EST 3275). Cancer. 1985;55(9):1868–73. Evaluating Two Paclitaxel and Cisplatin‚ ÄìContaining
9. Hallissey MT, et al. The second British Stomach Chemoradiation Regimens As Adjuvant Therapy in
Cancer Group trial of adjuvant radiotherapy or che- Resected Gastric Cancer (RTOG-0114). J Clin Oncol.
motherapy in resectable gastric cancer: five-year fol- 2009;27(12):1956–62.
low-up. Lancet. 1994;343(8909):1309–12. 23. Lee J, et al. Phase III trial comparing capecitabine plus
10. Coombes RC, et al. A randomized trial comparing adju- cisplatin versus capecitabine plus cisplatin with con-
vant fluorouracil, doxorubicin, and mitomycin with no current capecitabine radiotherapy in completely
treatment in operable gastric cancer. International resected gastric cancer with D2 lymph node dissection:
Collaborative Cancer Group. J Clin Oncol. 1990;8(8): the ARTIST trial. J Clin Oncol. 2012;30(3):268–73.
1362–9. 24. Wagner AD, et al. Chemotherapy in advanced gastric
11. Nitti D, et al. Randomized phase III trials of adjuvant cancer: a systematic review and meta-analysis based
FAMTX or FEMTX compared with surgery alone in on aggregate data. J Clin Oncol. 2006;24(18):2903–9.
resected gastric cancer. A combined analysis of the 25. Shah MA, Kelsen DP. Gastric cancer: a primer on the
EORTC GI Group and the ICCG. Ann Oncol. 2006; epidemiology and biology of the disease and an over-
17(2):262–9. view of the medical management of advanced disease.
12. De Vita F, et al. Adjuvant chemotherapy with epirubi- J Natl Compr Canc Netw. 2010;8(4):437–47.
cin, leucovorin, 5-fluorouracil and etoposide regimen 26. Vanhoefer U, et al. Final results of a randomized
in resected gastric cancer patients: a randomized phase phase III trial of sequential high-dose methotrexate,
III trial by the Gruppo Oncologico Italia Meridionale fluorouracil, and doxorubicin versus etoposide, leu-
(GOIM 9602 Study). Ann Oncol. 2007;18(8):1354–8. covorin, and fluorouracil versus infusional fluorouracil
13. Bang YJ, et al. Adjuvant capecitabine and oxaliplatin and cisplatin in advanced gastric cancer: a trial of the
for gastric cancer after D2 gastrectomy (CLASSIC): a European Organization for Research and Treatment
phase 3 open-label, randomised controlled trial. of Cancer Gastrointestinal Tract Cancer Cooperative
Lancet. 2012;379(9813):315–21. Group. J Clin Oncol. 2000;18(14):2648–57.
14. Noh SH, et al. O-0007 adjuvant Capecitabine and 27. Dank M, et al. Randomized phase III study comparing
Oxaliplatin (XELOX) for gastric cancer after D2 gas- irinotecan combined with 5-fluorouracil and folinic
trectomy: final results from the classic trial. Ann acid to cisplatin combined with 5-fluorouracil in che-
Oncol. 2013;24(suppl 4):iv14. motherapy naive patients with advanced adenocarci-
15. Diaz-Nieto R, Orti-Rodriguez R, Winslet M. Post- noma of the stomach or esophagogastric junction.
surgical chemotherapy versus surgery alone for Ann Oncol. 2008;19(8):1450–7.
resectable gastric cancer. Cochrane Database Syst 28. Van Cutsem E, et al. Phase III study of docetaxel and
Rev. 2013;9, CD008415. cisplatin plus fluorouracil compared with cisplatin
16. Boku N, et al. Fluorouracil versus combination of iri- and fluorouracil as first-line therapy for advanced gas-
notecan plus cisplatin versus S-1 in metastatic gastric tric cancer: a report of the V325 Study Group. J Clin
cancer: a randomised phase 3 study. Lancet Oncol. Oncol. 2006;24(31):4991–7.
2009;10(11):1063–9. 29. Ajani JA, et al. Quality of life with docetaxel plus cis-
17. Bleiberg H, et al. Adjuvant radiotherapy and chemo- platin and fluorouracil compared with cisplatin and
therapy in resectable gastric cancer. A randomized trial fluorouracil from a phase III trial for advanced gastric
of the gastro-intestinal tract cancer cooperative group or gastroesophageal adenocarcinoma: the V-325
of the EORTC. Eur J Surg Oncol. 1989;15(6):535–43. Study Group. J Clin Oncol. 2007;25(22):3210–6.
18. Macdonald JS, et al. Chemoradiotherapy after surgery 30. Shah MA, et al. Random assignment multicenter phase
compared with surgery alone for adenocarcinoma of II study of modified docetaxel, cisplatin, fluorouracil
the stomach or gastroesophageal junction. N Engl J (mDCF) versus DCF with growth factor support (GCSF)
Med. 2001;345(10):725–30. in metastatic gastroesophageal adenocarcinoma (GE).
19. Smalley SR, et al. Updated analysis of SWOG-directed ASCO Meet Abstr. 2010;28(15_suppl):4014.
intergroup study 0116: a phase III trial of adjuvant radio- 31. Mathe G, et al. Oxalato-platinum or 1-OHP, a third-
chemotherapy versus observation after curative gastric generation platinum complex: an experimental and
cancer resection. J Clin Oncol. 2012;30(19):2327–33. clinical appraisal and preliminary comparison with cis-
20. Fuchs CS, et al. Postoperative adjuvant chemoradiation platinum and carboplatinum. Biomed Pharmacother.
for gastric or gastroesophageal junction (GEJ) adeno- 1989;43(4):237–50.
carcinoma using epirubicin, cisplatin, and infusional 32. Rixe O, et al. Oxaliplatin, tetraplatin, cisplatin, and
(CI) 5-FU (ECF) before and after CI 5-FU and carboplatin: spectrum of activity in drug-resistant cell
9 Multimodality Therapy in Gastric Cancer 117

lines and in the cell lines of the National Cancer HER2-positive advanced gastric or gastro-oesophageal
Institute’s Anticancer Drug Screen panel. Biochem junction cancer (ToGA): a phase 3, open-label, ran-
Pharmacol. 1996;52(12):1855–65. domised controlled trial. Lancet. 2010;376(9742):
33. Cunningham D, et al. Capecitabine and oxaliplatin for 687–97.
advanced esophagogastric cancer. N Engl J Med. 2008; 38. Kang JH, et al. Salvage chemotherapy for pretreated
358(1):36–46. gastric cancer: a randomized phase III trial compar-
34. Shah MA, Schwartz GK. Treatment of metastatic ing chemotherapy plus best supportive care with best
esophagus and gastric cancer. Semin Oncol. 2004; supportive care alone. J Clin Oncol. 2012;30(13):
31(4):574–87. 1513–8.
35. Shah MA, et al. Multicenter phase II study of irinotecan, 39. Spratlin JL, et al. Phase I pharmacologic and biologic
cisplatin, and bevacizumab in patients with metastatic study of ramucirumab (IMC-1121B), a fully human
gastric or gastroesophageal junction adenocarcinoma. immunoglobulin G1 monoclonal antibody targeting
J Clin Oncol. 2006;24(33):5201–6. the vascular endothelial growth factor receptor-2.
36. Shah MA, et al. Phase II study of modified docetaxel, J Clin Oncol. 2010;28(5):780–7.
cisplatin, and fluorouracil with bevacizumab in 40. Fuchs CS, et al. Ramucirumab monotherapy for pre-
patients with metastatic gastroesophageal adenocarci- viously treated advanced gastric or gastro-oesophageal
noma. J Clin Oncol. 2011;29(7):868–74. junction adenocarcinoma (REGARD): an interna-
37. Bang YJ, et al. Trastuzumab in combination with che- tional, randomised, multicentre, placebo-controlled,
motherapy versus chemotherapy alone for treatment of phase 3 trial. Lancet. 2014 Jan 4;383(9911):31–9.
 Laparoscopic Transhiatal
Esophagectomy for Esophageal 10
Cancer

Dido Franceschi, Elizabeth Paulus,

and Danny Yakoub

Introduction Laparoscopic techniques were first adapted

into the field of esophageal disease in 1991 with
The incidence of esophageal cancer has increased laparoscopic fundoplication, performed by
over the last several decades, and the incidence of Dallemagne et al. [9]. With this, the shift toward
adenocarcinoma now surpasses that of squamous minimally invasive esophageal surgery began.
cell carcinoma [1]. Esophagectomy is the best Traditional approaches via open transhiatal or
curative option for the treatment of resectable transthoracic (Ivor Lewis) resections were first
esophageal cancer but is a complex operation “hybridized” with minimally invasive techniques,
with significant morbidity and mortality. While where parts of the procedure were performed in a
the overall morbidity and mortality in those who minimally invasive fashion and other parts via
are surgically treated has declined, approaching standard incisions. In 1993, Collard and col-
40–50 and 8–11 %, respectively, it is still signifi- leagues [10] published their initial experience
cant [2, 3]. with thoracoscopic mobilization of the esopha-
Over the past decade, minimally invasive gus. The first esophagectomy performed com-
esophagectomy (MIE) has been gaining favor as pletely via laparoscopy through a transhiatal
an attractive alternative to open resection with the approach was in 1995 by DePaula et al. [11]. In
potential to reduce surgical trauma, decrease 1999, Watson et al. first described a completely
morbidity, and shorten the length of hospital stay minimally invasive Ivor Lewis technique [12].
[4–8]. However, unlike laparoscopic procedures for
other malignancies such as the colon, stomach,
Electronic supplementary material Supplementary and even liver, laparoscopic esophagectomies
material is available in the online version of this have not become commonplace in the manage-
chapter at 10.1007/978-3-319-09342-0_10. Videos can
also be accessed at http://www.springerimages.com/
ment of esophageal cancer. The procedure is
videos/978-3-319-09341-3. technically demanding requiring expertise in
advanced laparoscopy and esophageal surgery.
D. Franceschi, MD (*)
Department of Surgery, University of Miami Hospital There is also the perceived lack of traditional
and Jackson Memorial Hospital, benefits from laparoscopic approach such as
1120 NW 14th Street, Clinical Research decreased hospital stay, ICU stay, or morbidity.
Building – 4th Floor, Miami, FL 33136, USA
Some centers that initially embraced it aban-
e-mail: [email protected]
doned the routine use of MIE [13].
E. Paulus, MD • D. Yakoub, MD, PhD
Just like the open approach, there are several
Division of Surgical Oncology,
University of Miami – Miller School of Medicine/ variations of minimally invasive esophagectomy.
Jackson Memorial Hospital, Miami, FL, USA They include thoracoscopy combined with

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 119
DOI 10.1007/978-3-319-09342-0_10, © Springer International Publishing Switzerland 2015
120 D. Franceschi et al.

laparotomy, thoracoscopy combined with any patient with a lesion that bridges the
laparoscopy, hand-assisted thoracotomy, hand- gastroesophageal (GE) junction may not be con-
assisted laparotomy or minilaparotomy, and sidered a candidate for this approach unless the
laparoscopic transhiatal or hand-assisted laparo- gastric margin can be cleared and an esophago-
scopic transhiatal [14]. Most experience has been gastrectomy can be done either via open approach
gained with a combined thoracoscopic and lapa- or minimally invasively. As with other laparo-
roscopic approach [4, 5]. scopic procedures, patients with extensive adhe-
Herein, we describe a completely laparoscopic sions and scar tissue over the abdomen or chest
approach with a cervical esophagogastric anasto- wall, particularly in areas where the thoracoscope
mosis for tumors located mainly in the gastro- or laparoscope would be placed, are a higher-risk
esophageal junction. group for treatment with MIE. Older patients and
those with comorbid conditions are not candi-
dates for surgery due to the high morbidity with
Indications either a MIE or standard procedure, but they may
benefit more from nonsurgical therapy [20].
Minimally invasive approaches to treatment of
benign esophageal diseases have been met with
widespread acceptance. This includes diseases Surgical Technique (Video 10.1)
such as achalasia, paraesophageal hernia, and
other complex esophageal disorders [15–18]. Positioning
This has not been the case with malignant disease
of the esophagus. Currently, no criteria define The patient is positioned in a supine position,
when a minimally invasive procedure should be with the left arm tucked to the side. The patient is
performed over an open procedure [19]. However, secured to the laparoscopic table and a footboard
an increasing trend exists for many high-volume is used. The abdomen, chest, and neck are
institutions to use minimally invasive esophagec- prepped under sterile condition.
tomy (MIE) in treatment of Barrett’s disease with
high-grade dysplasia and in patients with small
resectable lesions that have limited nodal involve- Abdominal Dissection
ment (N0-1). This includes T1 (invasion of the
lamina propria or submucosa), T2 (invasion of Positioning of the ports is a modification of that
the muscularis propria), and some instances of described by Hochwald and Ben-David [4]; as
T3 lesions (invasion of the adventitia). for our approach, the ports are placed closer to
Neoadjuvant chemoradiation is not a contraindi- the costal margin. Pneumoperitoneum is estab-
cation to a minimally invasive approach [7]. lished either through a 5 mm Optiview trocar
inserted under direct vision into the lateral aspect
of the left subcostal region or alternatively with a
Contraindications traditional Hasson technique above the umbili-
cus. A 30° scope is inserted, and the abdomen is
Currently, no standardized contraindications explored for the presence of metastases. Once the
exist regarding the use of minimally invasive decision is made to proceed, the remaining ports
esophagectomy. However, T4 lesions (invasion can now be placed under laparoscopic visualiza-
of surrounding tissues) are generally not amena- tion to avoid intra-abdominal injury. A 5-mm port
ble to any form of surgical resection. Extensive is inserted at the subxiphoid area and replaced
nodal disease and metastatic disease are also with a Nathanson liver retractor and secured to
advanced stages that may require an open surgi- the right side of the table for liver retraction. Two
cal approach or even endoscopic stenting for pal- 12-mm ports are placed, one on the right midcla-
liation instead of an attempt at MIE. Furthermore, vicular line and the other on the left midclavicu-
10 Laparoscopic Transhiatal Esophagectomy for Esophageal Cancer 121

½-inch Penrose around the GE junction and

secure it with an endo-loop.
Mobilization of the upper greater curvature is
done by the surgeon on the right side of the
patient. The stomach is retracted superiorly and
to the right. The assistant surgeon provides coun-
tertraction of the omentum, and the lesser sac is
entered at a point generally halfway up the greater
curvature of the stomach. We utilize the
Articulating Tissue Sealer (ENSEAL® G2,
Ethicon, USA) to divide the gastrocolic omentum
in a standard fashion. Care is taken to preserve
the right gastroepiploic vessels. Dissection is car-
ried out toward the gastrosplenic ligament, and
the short gastric vessels are divided. The fundus
is further released by dissection from the superior
splenic pole and division of the pancreaticogas-
tric attachments and the posterior vagus nerve.
Dissection progresses until the left crus is identi-
fied. The fascia over it is incised sharply, and the
posteroinferior mediastinum is entered. If suc-
cessfully completed initially, the surgeon should
Fig. 10.1 Port placement for a laparoscopic transhiatal be able to visualize the Penrose drain previously
esophagectomy
placed around the GE junction. If not previously
placed, the Penrose can be secured at this time.
lar line a few centimeters below the costal margin We now proceed to mobilize the lower half of
(Fig. 10.1). This varies depending on the patient’s the greater curvature of the stomach in a similar
body habitus. The final 5-mm port is inserted at fashion. This is performed from the left side of
the right lateral subcostal region for additional the table. All adhesions of the gastric posterior
retraction. wall to the pancreas are dissected until full mobi-
To start, the left lobe of the liver is retracted lization of the stomach is achieved. A Kocher
and the lesser omentum is incised and entered. maneuver is performed to ensure optimal gastric
The fascia covering the right crus of the dia- mobilization to the thorax. Adhesions of the duo-
phragm is incised sharply and an attempt is made denum to the liver, gallbladder, or porta hepatis
to create a retrogastric tunnel exiting to the left are divided.
and superior to the gastric cardia. Generally, the Attention is now directed toward isolating the
connective alveolar tissue in this area is loose and left gastric artery and vein. The stomach is
permits passage of a blunt grasper to the left of retracted superiorly, and surrounding lymph
the gastroesophageal junction. The stomach is nodes and fatty tissues are dissected to adequately
retracted inferiorly during this maneuver to visualize the celiac axis and origin of the com-
enhance visualization. Occasionally, especially mon hepatic artery and splenic artery.
with bulky tumors of the GE junction, posterior Retropancreatic lymph nodes along the proximal
visualization is not optimal and we prefer to com- splenic artery may be included in the dissection.
plete the mobilization of the upper half of the Lymphatic and fatty tissue is cleared up to the
greater curvature of the stomach prior to com- crus which should have been previously dis-
pleting the retrogastric tunnel. When we are able sected, guaranteeing the stomach is completely
to visualize the grasper as it exits the left side of free except for the left gastric vessels. The left
the GE junction, above the left crus, we pass a gastric artery and vein are divided and ligated
122 D. Franceschi et al.

can be carried up to the thoracic inlet and distal

cervical esophagus.

Cervical Component

A left cervical incision is performed along the

anterior sternocleidomastoid muscle. We rou-
tinely incise the strap muscles transversely and
expose the middle thyroid vein which is com-
monly divided and ligated with sutures. The jug-
ular vein and carotid artery are retracted laterally
as the thyroid is retracted superiorly and laterally.
Fig. 10.2 The use of open grasper on esophagus to facili-
tate mediastinal dissection This exposes the cervical esophagus. Care is
taken to identify the recurrent laryngeal nerve.
The esophagus is circumferentially dissected and
with the ECHELON FLEX™ Powered a Penrose is secured around it. Using mild trac-
ENDOPATH® Stapler (Ethicon, USA) utilizing a tion of the proximal cervical esophagus, blunt
vascular load (1.5 mm staples). dissection is used to free the cervical esophagus
The gastric conduit is created with a series of to the thoracic inlet. The degree of dissection will
firings of the 6 cm ECHELON FLEX™ Powered depend on the degree of success of our mediasti-
ENDOPATH® Stapler (Ethicon, USA) along the nal dissection. Once the esophagus is completely
lesser curvature of the stomach. Care is taken to free from adjacent tissues, the esophagus and
pull the nasogastric tube back into the GE junc- gastric tube can be pulled through the cervicot-
tion prior to stapler application. The stapler is omy. The nasogastric tube is pulled into the cer-
applied from the right 12 mm port. Stapling is vical esophagus, and after completing transection
completed except for the final centimeter, leaving of the stomach, an end-to-end esophagogastric
the esophagus attached to the conduit so it can be anastomosis is performed. Care is taken to bring
used to transfer the stomach into the posterior the gastric tube oriented correctly, with the staple
mediastinum. line toward the right mediastinum. Laparoscopic
Attention is returned to the hiatus where the visualization as well as assistance with the trans-
distal esophagus is further mobilized circumfer- fer is performed through the abdominal ports.
entially while utilizing retraction on the Penrose For bulky tumors that may not be able to be
previously placed. Included in this step should be successfully pulled through the mediastinum, we
wide division of the phrenoesophageal ligament. vary the technique by amputating the proximal
After this, the posterior mediastinum can be esophagus and pulling the esophagus into the
entered to continue dissection with mediastinal abdomen. A small upper midline incision is
lymph node dissection up to the level of the made, and the specimen is then removed abdomi-
carina. nally after completing transection of the stomach.
At this level, superior dissection continues The gastric conduit is brought up to the posterior
close to the esophagus with mobilization of the mediastinum by using a large Foley attached to
proximal esophagus away from the trachea and the end of a thin laparoscopy bag that drapes over
prevertebral fascia. Traction of the Penrose the conduit. The Foley is introduced into the neck
located at the GE junction inferiorly and laterally and brought out the hiatus. By applying suction
aides with the dissection. The use of an open and traction to the Foley, the plastic adheres to
grasper on the esophagus (Fig. 10.2) to generate the viscera and allows for gentle traction and cor-
posterior or anterior traction close to the area of rect orientation of the gastric tube as it is brought
dissection is also useful. Generally, dissection out the neck (Fig. 10.3).
10 Laparoscopic Transhiatal Esophagectomy for Esophageal Cancer 123

Fig. 10.4 Side-to-side anastomosis from the cervical

esophagus to the gastric tube, utilizing a 6 cm stapler

Fig. 10.3 Gastric tube is placed in a plastic laparoscopy

bag attached to a Foley previously passed from the neck
through the mediastinum. Suction on the Foley collapses
the plastic on the viscera and allows for gentle traction
into the neck Fig. 10.5 Closure of anterior wall of anastomosis with a
triangulation technique

For the esophagogastric anastomosis, we Outcomes

prefer a modification of the technique described
by Collard et al. [21], utilizing a side-to-side Complications and outcomes are significantly
anastomosis. The back wall is created with an influenced by the volume of patients, because a
application of the 6 mm ECHELON FLEX™ large learning curve exists. High-volume centers
Powered ENDOPATH® Stapler (Ethicon, USA) tend to have more experience and, therefore, bet-
(2.5 mm) (Fig. 10.4). The anterior layer is closed ter outcomes than smaller-volume hospitals [22,
utilizing a triangulation technique with a 23]. Minimally invasive techniques for esopha-
PROXIMATE® Reloadable Staplers (TX) 30 mm geal resection have been reported to have accept-
(Ethicon, USA) (Fig. 10.5). The nasogastric tube ably reduced procedure-related morbidity
is advanced to just proximal to the pylorus. A without compromising disease-free survival
pyloromyotomy is not performed. A 19-round rates. Luketich et al. have the largest reported
Jackson-Pratt drain is placed in each pleural cav- experience to date. Their initial series of 222
ity and brought out through the hiatus into the patients has now grown to more than 1,000
abdomen and out of the abdominal port sites. A patients. In this series, mortality was 1.4 % ver-
jejunal feeding tube is placed, and all incisions sus 5.5 % for an open approach [5]. Furthermore,
are closed. the survival curve at 19 months of follow-up was
124 D. Franceschi et al.

comparable in both groups. In another analysis of Table 10.1 Short-term results for laparoscopic
transhiatal esophagectomy
41 elderly patients over the age of 75 years who
underwent minimally invasive esophagectomy, Parameter Result
no operative deaths occurred, with a survival of Mean operative time 160–390 min
81 % at 20 months of follow-up [24]. A recent Mean blood loss 220–400 cc
meta-analysis of the available literature suggests Conversion rate 0–16.6 %
that patients undergoing MIE had better opera- Anastomotic leak 0–8.3 %
Mean number of retrieved lymph 8–14
tive and postoperative outcomes with no compro-
nodes
mise in oncologic outcomes (as assessed by Mean hospital stay 6.4–12.1 days
lymph node retrieval) [14]. Patients receiving Thirty-day mortality 0–13.6 %
MIE had significantly lower blood loss and
shorter postoperative ICU and hospital stay.
There was a 50 % decrease in total morbidity in 22 [28, 30] with the percentage of cancer patients
the MIE group. Subgroup analysis of comorbidi- ranged from 17 [11] to 100 % [28]. Results com-
ties demonstrated significantly lower incidence pare favorably with the open procedure. Mean
of respiratory complications after MIE; however, operative time reported varies widely between
other postoperative outcomes such as anasto- studies from 160 to 390 min. Anastomotic leak
motic leak, anastomotic stricture, gastric conduit rate varies between 0 and 8.3 % with 30-day mor-
ischemia, chyle leak, vocal cord palsy, and tality ranging from 0 to 13.6 %. Surgical margin
30-day mortality were comparable between the data were satisfactory when stated but were not
two techniques. commented on in detail, and long-term oncologic
The only reported trial of minimally invasive outcomes are not reported in any study.
esophagectomy versus open esophagectomy per-
formed in the Netherlands [25] randomized 56
patients to open esophagectomy and 59 patients Summary
to MIE. 16 (29 %) patients in the open esopha-
gectomy group had pulmonary infection in the Laparoscopic transhiatal esophagectomy was the
first 2 weeks compared with five (9 %) in the first totally minimally invasive approach to
minimally invasive group (relative risk (RR) esophagectomy that did not include a thoracot-
0.30, 95 % CI 0.12–0.76; p = 0.005). 19 (34 %) omy or laparotomy. This technique is similar to
patients in the open esophagectomy group had that of open blunt transhiatal esophagectomy
pulmonary infection in the hospital compared except that the blunt mediastinal esophageal dis-
with seven (12 %) in the minimally invasive section is replaced by a laparoscopic transhiatal
group (0.35, 0.16–0.8; p = 0.005). dissection of the mediastinal esophagus. The
These findings suggest that minimally inva- indications for a total laparoscopic transhiatal
sive esophagectomy can be safely performed in esophagectomy are similar to those of standard
selected patients and even those considered high open transhiatal esophagectomy, and the proce-
risk that might not otherwise be considered for an dure is particularly useful for patients who have
open surgery. Likewise, there seems to be good lower- or middle-third tumors with significant
evidence of short-term benefits when compared proximal involvement or in conjunction with
to open procedures. A recent analysis also sug- long-segment Barrett’s esophagus. The anasto-
gests that MIE is cost-effective compared to open mosis is performed in the neck and allows the
esophagectomy in patients with resectable esoph- surgeon to maximize the proximal margin. The
ageal cancer [26]. main limitations of this technique include a lim-
The short-term results of several series of lapa- ited view of the middle and upper third of the
roscopic transhiatal esophagectomies (11, 27–30) mediastinum; however, that can be improved by
are listed in Table 10.1. The number of surgical the use of long instruments and adequate port
cases in these five studies ranged from 9 [27] to positioning.
10 Laparoscopic Transhiatal Esophagectomy for Esophageal Cancer 125

Review of the literature suggests that the esophageal cancer? A meta-analysis. Surg Endosc.
2010;24:1621–9.
short-term outcomes are superior to the open
15. Luketich JD, Fernando HC, Christie NA, et al.
approach; however, there is no data about the Outcomes after minimally invasive esophagomyot-
long-term survival. The procedure should be omy. Ann Thorac Surg. 2001;72:1909–13.
done in a high-volume center. 16. Pierre A, Luketich JD, Fernando HC, et al. Results of
laparoscopic repair of giant paraesophageal hernia:
200 consecutive patients. Ann Thorac Surg. 2002;
74:1909–15.
References 17. Krasna MJ, Jiao X. Thoracoscopic and laparoscopic
staging for esophageal cancer. Semin Thorac
1. Eslik GD. Epidemiology of esophageal cancer. Cardiovasc Surg. 2000;12:186–94.
Gastroenterol Clin North Am. 2009;38:17–25. 18. Luketich JD, Schauer P, Landreneau R, et al. Minimally
2. Chang AC, Ji H, Birkmeyer NJ, Orringer MB, invasive surgical staging is superior to endoscopic ultra-
Birkmeyer JD. Outcomes after transhiatal and trans- sound in detecting lymph node metastases in esophageal
thoracic esophagectomy for cancer. Ann Thorac Surg. cancer. J Thorac Cardiovasc Surg. 1997;114:817–23.
2008;85(2):424–9. 19. Pennathur A, Zhang J, Chen H, Luketich JD. The
3. Connors RC, Reuben BC, Neumayer LA, et al. ‘Best Operation’ for esophageal cancer. Ann Thorac
Comparing outcomes after transthoracic and tran- Surg. 2010;89:2163–7.
shiatal esophagectomy: a 5-year prospective cohort 20. Luketich JD, Alvelo-Rivera M, Buenaventura PO,
of 17,395 patients. J Am Coll Surg. 2007;205: et al. Minimally invasive esophagectomy: outcomes
735–40. in 222 patients. Ann Surg. 2003;238:486–95.
4. Hochwald SN, Ben-David K. Minimally invasive esoph- 21. Collard JM, Romagnoli R, Goncette L, et al.
agectomy with cervical anastomosis. J Gastrointest Terminalized semimechanical side-to-side suture tech-
Surg. 2012;16:1775–81. nique for cervical esophagogastrostomy. Ann Thorac
5. Luketich JD, Pennathur A, Awais O, et al. Outcomes Surg. 1998;65:814–7.
after minimally invasive esophagectomy: review of 22. Birkmeyer JD, Siewers A, Finlayson E, et al. Hospital
over 1000 patients. Ann Surg. 2012;256(1):95–103. volume and surgical mortality in the United States. N
6. Hoppo T, Jobe BA, Hunter JG. Minimally invasive Engl J Med. 2002;346:1128–37.
esophagectomy: the evolution and technique of mini- 23. Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon
mally invasive surgery for esophageal cancer. World J volume and operative mortality in the United States.
Surg. 2011;35(7):1454–63. N Engl J Med. 2003;349(22):2117–27.
7. Ben-David K, Rossidis G, Zlotecki RA, et al. 24. Perry Y, Fernando HC, Buenaventura PO, et al.
Minimally invasive esophagectomy is safe and effec- Minimally invasive esophagectomy in the elderly.
tive following neoadjuvant chemoradiation therapy. JSLS. 2002;6:299–304.
Ann Surg Oncol. 2011;18(12):3324–9. 25. Biere SS, van Berge Henegouwen MI, Maas KW,
8. Peyre CG, Peters JH. Minimally invasive surgery for et al. Minimally invasive versus open oesophagec-
esophageal cancer. Surg Oncol Clin N Am. 2013;22: tomy for patients with oesophageal cancer: a multi-
15–25. centre, open-label, randomised controlled trial.
9. Dallemagne B, Weerts JM, Jehaes C, et al. Laparoscopic Lancet. 2012;379(9829):1887–92.
Nissen fundoplication: preliminary report. Surg Laparosc 26. Lee L, Sudarshan M, Li C, et al. Cost-effectiveness of
Endosc. 1991;1:138–43. minimally invasive versus open esophagectomy for
10. Collard JM, Lengele B, Otte JB, et al. En bloc and esophageal cancer. Ann Surg Oncol. 2013;20:3732–9.
standard esophagectomies by thoracoscopy. Ann 27. Swanstrom L, Hansen P. Laparoscopic total esopha-
Thorac Surg. 1993;56(3):675–9. gectomy. Arch Surg. 1997;132:943–7.
11. DePaula AL, Hashiba K, Ferreira EA, et al. Laparoscopic 28. Del Genio A, Rossetti G, Napolitano V, et al.
transhiatal esophagectomy with esophagogastroplasty. Laparoscopic esophagectomy in the palliation treat-
Surg Laparosc Endosc. 1995;5:1–5. ment of advanced esophageal cancer after radioche-
12. Watson DI, Davies N, Jamieson GG. Totally endo- motherapy. Surg Endosc. 2004;18:1789–94.
scopic Ivor Lewis esophagectomy. Surg Endosc. 1999; 29. Bonavina L, Incarbone R, Bona D, Peracchia A.
13:293–7. Esophagectomy via laparoscopy and transmediastinal
13. Smithers BM, Gotley DC, Martin I, et al. Comparison endodissection. J Laparoendosc Adv Surg Tech A.
of the outcomes between open and minimally invasive 2004;14:13–6.
esophagectomy. Ann Surg. 2007;245:232–40. 30. Avital S, Zundel N, Szomstein S, Rosenthal R.
14. Nagpal K, Kamran A, Vats A, et al. Is minimally Laparoscopic transhiatal esophagectomy for esophageal
invasive surgery beneficial in the management of cancer. Am J Surg. 2005;190:69–74.
 Laparoscopic and Thoracoscopic
Esophagectomy with EEA 11
Anastomosis

R. Taylor Ripley, David D. Odell,

and James D. Luketich

Introduction mortality, improved pain control, and a decrease

in hospital length of stay.
Minimally invasive esophagectomy (MIE) is Currently, minimally invasive approaches for
now an accepted surgical approach for esopha- esophagectomy include laparoscopic transhiatal,
geal malignancy and the occasional benign con- laparoscopic-thoracoscopic 3-hole (McKeown),
ditions. Meta-analyses evaluating the results of and laparoscopic-thoracoscopic Ivor Lewis
MIE have shown improved perioperative out- esophagectomy. The choice between MIE
comes and similar oncologic outcomes when approaches is often based on surgeon preference,
compared to open esophagectomy with the but tumor location may influence the surgical
advantages of minimally invasive surgery. MIE approach. For example, a mid-esophageal squa-
techniques are now a combination of laparos- mous cell carcinoma is often best treated by a
copy and thoracoscopy for a totally minimally McKeown esophagectomy. The anticipated mor-
invasive esophagectomy. While this approach is bidity of the operation varies with the choice of
technically demanding and associated with a sig- surgical approach. The creation of a cervical anas-
nificant learning curve, it is an excellent option tomosis has a higher incidence of recurrent laryn-
for esophageal resection. In our experience, geal nerve injury, anastomotic leak, stricture, and
MIE is associated with a reduction in blood pharyngoesophageal swallowing dysfunction. In
loss, decreased respiratory complications, lower contrast, transthoracic approaches have a higher
incidence of cardiopulmonary complications and
potentially greater morbidity if an anastomotic
leak occurs.
Electronic supplementary material Supplementary Our initial approach to MIE utilized a modi-
material is available in the online version of this
chapter at 10.1007/978-3-319-09342-0_11. Videos can
fied McKeown (3-hole) technique that proved to
also be accessed at http://www.springerimages.com/ have equivalent oncologic outcome and morbid-
videos/978-3-319-09341-3. ity to the open technique. Secondary to the mor-
R.T. Ripley, MD (*) bidity of the cervical neck dissection, and the
Division of Thoracic Surgery, Memorial Sloan- current predominance of adenocarcinoma with
Kettering Cancer Center, primarily lower third esophageal tumors, our
1275 York Avenue, New York, NY 10065, USA
preferred approach is now a laparoscopic-
e-mail: [email protected]
thoracoscopic (Ivor Lewis) esophagectomy and a
D.D. Odell, MD, MMSc • J.D. Luketich, MD
Department of Cardiothoracic Surgery, University of
Pittsburgh Medical Center, UPMC Presbyterian,
Pittsburgh, PA, USA The authors have nothing to disclose.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 127
DOI 10.1007/978-3-319-09342-0_11, © Springer International Publishing Switzerland 2015
128 R.T. Ripley et al.

two-field lymphadenectomy (celiac, left gastric, Endoscopic Evaluation

splenic, paraesophageal, and subcarinal lymph
nodes). The minimally invasive Ivor Lewis is The operation begins with esophagogastroduode-
appropriate for most distal esophageal cancers, noscopy (EGD). The location of the tumor is con-
GE junction tumors with cardia extension, and firmed and the proximal and distal extents are
short-to-moderate length Barrett’s esophagus assessed. The esophagus is examined for evi-
with high-grade dysplasia. In addition, when the dence of Barrett’s changes proximal to the
length of the gastric conduit is compromised by intended resection margin with four quadrant
either not enough stomach or a close margin, an biopsies in areas of concern. Endoscopic exami-
intrathoracic anastomosis decreases the neces- nation of the stomach is necessary to assess suit-
sary length of the conduit. Total laparoscopic and ability for use as a conduit. Insufflation should be
thoracoscopic Ivor Lewis resections should not a minimum to reduce small bowel distention
be performed for upper third or high mid- which may significantly decrease domain and
esophageal cancers. The following describes our increase the difficulty of laparoscopy.
current technique for laparoscopic-thoracoscopic
Ivor Lewis minimally invasive esophagectomy.
Laparoscopic Phase

Surgical Technique Positioning and Laparoscopic Port

Placement
Anesthetic Considerations The patient is supine with the arms at a 60° angle.
A foot board is placed to allow steep reverse
Anesthetic management during MIE poses spe- Trendelenburg during the hiatal dissection. The
cific challenges. All patients receive an arterial costal margin is identified and a line is drawn
blood pressure monitoring line; a central venous from the xiphoid to the umbilicus. This line is
catheter placement is not routine. A double- then divided into thirds. The first port placed
lumen endotracheal tube is placed initially in using a direct Hassan cutdown approach in the
anticipation of the thoracoscopic phase. In right paramedian position roughly 2 cm lateral to
patients with mid- or upper-thoracic tumors, a the midline just cephalad to the junction of the
single-lumen endotracheal tube is initially placed lower and middle thirds of the described line. For
for preoperative bronchoscopy to evaluate airway patients with a protuberant abdomen or unusually
involvement. long distance from the xiphoid to the umbilicus,
Patients generally require volume loading dur- the “thirds rule” may need to be modified by pay-
ing the laparoscopic phase secondary to the pneu- ing attention to the absolute distance from the
moperitoneum and reverse Trendelenburg. The xiphoid to the middle ports. If this distance is too
patient can develop significant hypercarbia and great, adequate visualization in the crural area
acidosis secondary to CO2 insufflation. The sur- will be difficult. Five abdominal ports are used
geon must communicate with the anesthesiolo- for gastric mobilization (10 mm right and 5 or 10
gists about vasopressors because they can directly left paramedian, 5 mm right and left subcostal,
impact the viability of the gastric conduit. Simple and a second 5 mm right lateral subcostal port for
measures to correct these problems include lower- liver retraction (Fig. 11.1)). After the Hassan cut-
ing the insufflation pressure, decreasing the down, the remaining ports are placed under direct
degree of the reverse Trendelenburg, and increas- laparoscopic vision. A sixth port is placed in the
ing volume. In addition to changing ventilator set- right paraumbilical region to assist in placement
tings, hypercarbia can be corrected by reversing of the feeding jejunostomy tube. All ports should
the pneumoperitoneum. Communication through- be at least a hand’s breadth apart to avoid inter-
out the procedure between the surgeon and the ference between instruments. Additionally, skin
anesthesiologist is imperative. and fascial incisions should be small to avoid
11 Laparoscopic and Thoracoscopic Esophagectomy with EEA Anastomosis 129

vascular pedicle identified. A complete lymph

node dissection is performed by dissection of the
left gastric and celiac lymph nodes toward the
specimen. This dissection is continued laterally
along the splenic artery, the superior border of the
pancreas, and superiorly toward the crura along
the preaortic plane. If the nodes appear bulky or
otherwise malignant, they are sent for frozen sec-
tion evaluation. Once assured of the resectability
of the tumor and nodal disease, the crural dissec-
tion and complete mobilization of the lower
5 mm 5 mm
esophagus are performed. The right crus is dis-
sected first; this dissection is continued anterior
by transecting the phrenoesophageal ligaments to
5 mm expose the anterior hiatus. The left crus is often
5 mm exposed either by a combination of the anterior
10 mm dissection along the medial crural border and by
mobilizing the fundus of the stomach. This com-
bined dissection exposes the retroesophageal
window which ensures complete mobilization of
Fig. 11.1 Laparoscopic port placement. The 10 mm port the superior portion of the lesser curve complet-
in the right paramedian position is placed first via a direct ing 360° exposure of the gastroesophageal junc-
cutdown technique. The left paramedian port may be con- tion. During this dissection, the exposure is
verted from a 5 mm port (as shown) to a 10 mm port if a maximized by dividing the left gastric artery and
larger camera is desired (© Heart, Lung and Esophageal
Surgery Institute University of Pittsburgh Medical Center) vein with endovascular GIA stapler.
After identifying the gastrocolic omentum, the
antrum of the stomach is retracted, and a window
subcutaneous emphysema. The liver retractor is is created in the greater omentum, allowing
brought in through the right lateral subcostal port access to the lesser sac. The remaining short gas-
and positioned to elevate the left lobe of the liver tric vessels are divided while ensuring the dissec-
and expose the hiatus. This retractor is most tion is above the gastroepiploic arcade. The
effective if the port is as posterior and cephalad fundus is retracted to the right to dissect the
as possible in Morrison’s pouch. remaining retro-gastric short gastric arteries
The camera is placed in the left paramedian while exposing the left gastric artery and vein.
port position for the majority of the laparoscopic Care should be taken to ensure that all nodes are
phase. The surgeon works from the right side of swept toward the specimen side and to avoid nar-
the table using the right paramedian and subcos- rowing of the splenic or hepatic arteries. This dis-
tal ports. From the left, the assistant controls the section should complete the mobilization of the
camera and a second grasper for retraction fundus and proximal stomach. Gastric mobiliza-
(through the left subcostal port). tion is carried inferiorly to the pyloro-antral
region. Meticulous attention must be paid during
Gastric Mobilization this phase of the dissection to avoid injury to the
Inspection of the abdomen is performed to evalu- gastroepiploic arcade. Mistakenly transecting the
ate injuries occurring during port placement and arcade often renders the conduit unusable. Direct
to check for liver, omental, or other intraperito- handling and instrumentation of the conduit por-
neal metastasis. Biopsies of suspicious lesions tion of the stomach should be avoided. Adequate
are sent for frozen section evaluation. The gastro- mobilization has been achieved when the pylorus
hepatic ligament is opened and the left gastric reaches the caudate lobe. Depending on prior
130 R.T. Ripley et al.

location for gastric transection. Starting at the

antrum, the staple line is then directed superiorly,
toward the fundus, parallel to the line of the short
gastric vessels along the greater curvature. A
conduit width of 3–4 cm is preferred (Fig. 11.2).
An unusually thick antrum may require that one
chooses a greater staple height (e.g., the black
Endo GIA loads) to get an adequate staple line
integrity. The length of the conduit and margin of
resected stomach should be assessed and modi-
fied if there is concern for extension of the tumor
onto the gastric cardia. Sutures may be placed to
reinforce the staple line if there is concern about
its integrity though usually not necessary.
Fig. 11.2 Anatomy of the completed gastric conduit. The
right gastroepiploic arcade forms the primary blood sup-
ply. The right gastric artery is also preserved and contrib-
Pyloroplasty
utes some blood supply to the gastric antrum (© Jennifer The pylorus is identified and 2-0 Surgidac
Dallal, James D. Luketich, MD) (Covidien, Mansfield, MA) stay sutures are
placed on the superior and inferior aspects using
the Endostitch device (US Surgical, Norwalk,
operations and adhesions, enough mobility may CT) (Fig. 11.3). The anterior wall of the pylorus
require a lysis of adhesions and a partial or a is transected with an ultrasonic shears. The pylo-
complete Kocher maneuver. romyotomy is closed transversely in a Heineke-
Mikulicz fashion using simple, interrupted 2.0
Creation of Gastric Tube Surgidac sutures. An omental patch is placed
The gastric tube is created prior to the pyloro- over the pyloroplasty and sutured in place.
plasty and placement of the feeding jejunostomy
tube to allow time for assessment of conduit via- Feeding Jejunostomy Tube Placement
bility prior to transition to the thoracoscopic A 12 Fr jejunostomy catheter is placed in the left
phase. The gastric tube follows the arc of the lower quadrant using a percutaneous technique.
greater curve of the stomach and is based on the The transverse colon is retracted superiorly to
right gastroepiploic artery (Fig. 11.2). Prior to expose the ligament of Treitz, and a position on
creating the gastric conduit, the nasogastric tube, the jejunum 30–40 cm downstream is identified.
if previously placed, is pulled back to the mid- The antimesenteric border of the bowel is sutured
esophagus. An endovascular stapling technique to the abdominal wall with a 2-0 Surgidac suture.
allows for a controlled creation of the gastric tube The 12 mm right paraumbilical port is used with
conduit. The first staple load is a vascular (gold) the camera positioned in the right paramedian
load for the adipose tissue and vessels along the location. A Seldinger technique is used to intro-
lesser curve above the level of the right gastric duce the catheter into the jejunum under direct
artery. No stomach is divided with this firing. The laparoscopic vision. Air insufflation is used to
remainder of the firings divides the stomach with verify luminal placement. The jejunum is tacked
45 mm purple loads (Endo GIA Reloads with Tri- circumferentially to the abdominal wall. An addi-
staple Technology, Covidien, Mansfield, MA). tional suture is placed in the distal limb of the
The course of the greater curvature is precisely jejunum to prevent volvulus and obstruction.
followed by applying traction to the fundus and
antrum by the assistants and by traction on the Preparation for Thoracoscopic Phase
specimen side with the surgeon’s left hand. This The gastric conduit is assessed for viability. Once
three-point traction provides a clear view of the viability of the conduit is assured, the most
11 Laparoscopic and Thoracoscopic Esophagectomy with EEA Anastomosis 131

a b

Pyloroplasty
incision
Identification
of pylorus
muscle

Pyloroplasty
Pyloroplasty closed
transversely with
auto suture device

Fig. 11.3 Laparoscopic creation of a pyloroplasty (a) with vertical closure in a Heineke-Mikulicz fashion (b) (© Heart,
Lung and Esophageal Surgery Institute University of Pittsburgh Medical Center)

superior portion of the gastric tube is stitched to on the right side and the assistant stands on the
the specimen (Fig. 11.4). Maintaining the align- left side of the table. Five thoracoscopic ports are
ment of the conduit to avoid twisting as the stom- used (Fig. 11.5). A 10-mm camera port is placed
ach is brought into the chest is imperative. The in the 8th or 9th intercostal space slightly anterior
greater curvature along the short gastric vessels is to the midaxillary line. A 10-mm working port is
sutured to the staple line of the proximal gastric placed in the 8th or 9th intercostal space posterior
remnant. If an omental flap has been created, the to the posterior axillary line. Another 10-mm port
distal end is sutured to the conduit tip. If hemo- is placed in the anterior axillary line at the
stasis of the staple line is needed, clips are 4th intercostal space for a fan-shaped lung retrac-
applied. The specimen and gastric conduit are tor aids in retracting the lung to expose the
placed in the lower mediastinum while preserv- esophagus. A 5-mm port is placed just inferior to
ing the proper orientation. If the hiatal opening is the tip of the scapula.
large, the crura are reapproximated with a stitch
to prevent delayed thoracic herniation of the dis- Thoracoscopic Dissection
tal conduit. This step requires considerable judg- and Resection of the Esophagogastric
ment by an experienced surgeon because a tight Specimen
hiatus may compromise the venous drainage of Retraction of the diaphragm is essential to the
the conduit. A nasogastric tube (if not previously thoracoscopic phase of the dissection. A 48 in., 0
placed) is placed in the esophagus prior to tho- Surgidac suture is placed through the central ten-
racic positioning. don of the diaphragm using the Endostitch. The
suture is brought out through the lateral chest
wall at the level of the insertion of the diaphragm
Thoracoscopic Phase through a small stab incision, retracting the dia-
phragm inferiorly and exposing the distal esoph-
Positioning and Port Placement agus. The inferior pulmonary ligament is divided
The patient is turned to the left lateral decubitus to the inferior pulmonary vein to maximize
position, and location of the double-lumen endo- retraction of the lung. The esophageal dissection
tracheal tube is reconfirmed. The surgeon stands is started on the avascular plane along the surface
132 R.T. Ripley et al.

Fig. 11.4 The gastric

conduit is sutured to the Mobilized
distal portion of the at hiatus
specimen. The lesser
curvature staple line faces
toward the right so as to
preserve orientation
(© Heart, Lung and
Esophageal Surgery Institute
University of Pittsburgh Attaching
Medical Center) gastric tube
to specimen

Fig. 11.5 Thoracoscopic 10–mm port 5–mm port 10–mm port

port placement

5–mm port

of the pericardium. This dissection is carried from collapsing and can aid in visualization
superiorly to the subcarinal space ensuring that while removing subcarinal lymph nodes. The
the lymph nodes are dissected with the esopha- lung is retracted anteriorly and the pleura incised
gus (Fig. 11.6). Care must be taken to identify the along the anterior border of the esophagus to the
membranous wall of the right mainstem bron- level of the azygous vein. The pleura above the
chus because it is at risk of injury during this azygous vein is opened to facilitate the exposure
phase of the dissection. Removing suction from of the vein with division with the endo-GIA vas-
the right lung will prevent the membranous wall cular (gold) load. Above the level of the azygous
11 Laparoscopic and Thoracoscopic Esophagectomy with EEA Anastomosis 133

facilitated by lifting the specimen into the chest,

cutting the suture between the conduit and the
specimen, and tacking the proximal conduit to
the diaphragm. As the specimen is rolled toward
the apex, the remaining adhesions to the deep
margin are transected. The left pleural space may
be entered if needed to remove a bulky tumor.
Once mobilization of the esophagus has been
completed, a 4–5 cm mini-thoracotomy without
rib spreading is created between the surgeon’s
working port and the tip of the scapula. A wound
protector (Applied Medical, Rancho Santa
Margarita, CA) is placed to protect the skin and
chest wall. The esophagus is sharply transected
using laparoscopic scissors at or above the level
of the azygous vein determined by the proximal
extent of tumor. The nasogastric tube is pulled
back into the proximal esophagus under direct
vision during transection. The esophagogastrec-
tomy specimen is then withdrawn through the
wound protector, opened, grossly examined, and
sent for frozen section evaluation of the margins.
Fig. 11.6 Thoracoscopic mobilization of the esophagus.
The pleura is incised longitudinally in line with the esopha- Creation of Gastroesophageal
gus. The lung is anteriorly retracted to provide adequate Anastomosis (Video 11.1)
exposure. All nodal tissue is excised en bloc with the esoph- Next, the esophagogastric anastomosis is created.
ageal tissue (© Jennifer Dallal, James D. Luketich, MD) An EEA stapling device is utilized (Fig. 11.7).
The anvil of the stapler is placed in the proximal
vein, the dissection no longer includes periesoph- end of the esophagus and sutured with two purse-
ageal tissue to avoid injury to the recurrent laryn- string 2-0 Surgidac. All layers of the esophagus
geal nerve and the airway. The vagus is cut at this must be included to ensure a competent anasto-
level to prevent traction injury to the recurrent mosis. The ideal size is a 28 mm EEA stapler
laryngeal nerve. The extent of superior dissection which will help to minimize stricture formation
and mobilization depends upon the location of and to reduce postoperative dilation. If the proxi-
the tumor and the intended site of resection. Next, mal esophagus is not large enough to accommo-
the posterior mobilization is begun. The pleura is date the 28 mm anvil, a Foley catheter with a
divided in the groove posterior to the esophagus 30 cc balloon can be used to gently dilate the
near the diaphragm. This dissection is kept super- esophageal lumen in an attempt to facilitate
ficial to avoid injury to the thoracic duct and placement of the anvil. If dilatation fails, the
underlying thoracic aorta. Bridging lymphatics 25 mm EEA may be necessary. The gastric con-
and aortoesophageal vessels are controlled with duit is pulled further into the chest and angled
endoclips. Thoracic duct ligation should be con- toward the mini-thoracotomy. The orientation of
sidered if there is concern for trauma to the duct the gastric tube requires that the staple is facing
or tumor extension into the duct. This lateral dis- the lateral chest wall. The tip of the gastric con-
section is carried along the length of the esopha- duit is opened using ultrasonic shears to the right
gus from the gastroesophageal junction to above side of the staple line. The EEA stapler is placed
the azygous vein. The contralateral pleura is the through the wound protector and inserted into the
deep margin of the dissection. This dissection is conduit using the atraumatic graspers to pull the
134 R.T. Ripley et al.

on slight tension. Once the proximal esophagus is

taught, the EEA is turned and advanced toward
the esophagus to prevent additional tension on
the esophagus. An estimate of the amount of con-
duit that will lie in the chest is performed at this
point. The stapler is then fired and withdrawn.
The tissue rings are inspected grossly to insure
that they are complete and sent for permanent
pathology.
After creating the anastomosis, the remaining
gastric tip with the gastrotomy is resected with
2–3 loads of the endovascular GIA stapler
(Fig. 11.8). The anastomosis and conduit resec-
tion need to be sufficiently separated to prevent
ischemia of the intervening tissue. If an omental
flap was created during the abdominal dissection,
it is wrapped around the anastomosis by placing
between the airway and the anastomosis and
suturing into place. The chest is then thoroughly
irrigated and inspected for hemostasis.
Fig. 11.7 Creation of the esophagogastric anastomosis
using the EEA stapler. The anvil is sewn into the esopha- Drain Placement and Closure
gus with 2 concentric sutures and the stapler introduced
through a gastrotomy in the gastric conduit. Conduit ori-
Adequate drainage of the mediastinum
entation is maintained by keeping the lesser curvature surrounding the anastomosis is imperative to
staple line facing the camera while the stapler is docked minimize complications in the event of an anas-
(© Heart, Lung and Esophageal Surgery Institute tomotic leak. A 10 mm Jackson-Pratt drain is
University of Pittsburgh Medical Center)
placed posteriorly along the anastomosis and
a 28-French chest tube is placed in the pleural
conduit over the EEA stapler while gently space. The nasogastric tube is advanced past
pushing the stapler toward the mediastinum. The the anastomosis under thoracoscopic visualiza-
EEA is advanced to a position on the conduit tion. The gastric conduit is sutured to the right
appropriate for an anastomosis. While maintain- crus and the diaphragm edge with a single 2-0
ing traction with the graspers on the gastrotomy Endostitch to prevent delayed herniation. A long
edges, the EEA is shifted toward the proximal aspirating needle is used to instill a multilevel
anastomosis to gently slide additional conduit intercostal nerve block. The access incision is
into the mediastinum without lifting redundant closed. The Jackson-Pratt drain is secured with
conduit into the pleural space. Bringing excess multiple sutures to the skin to prevent dislodge-
stomach into the chest with the intent of minimiz- ment. Once all the incisions are closed, the
ing tension on the anastomosis is a mistake. A patient is turned to the supine position and the
redundant conduit above the diaphragm can lead oropharynx and nasopharynx are suctioned of all
to significant problems with conduit emptying. secretions. The patient is reintubated with a sin-
Once the appropriate length has determined, the gle-lumen endotracheal tube. If a tube exchange
stapler spike is brought out along the greater catheter is necessary, it should be used with cau-
curve of the gastric conduit. The spike is care- tion because it may injure the right mainstem
fully docked with the anvil. Prior to creating the bronchus. A toilet bronchoscopy is performed
anastomosis, the EEA spike is turned without while examining the right and left mainstem
moving the stapler to lift the esophagus until it is bronchi for injury.
11 Laparoscopic and Thoracoscopic Esophagectomy with EEA Anastomosis 135

Fig. 11.8 Closure of the

gastrotomy is performed
using a reticulating Endo Esophagus
GIA stapler. This portion of
resected stomach represents
the final gastric margin
(© Heart, Lung and
Esophageal Surgery Institute
University of Pittsburgh
Medical Center)
Gastric
tube
Excess stomach
trimmed and
closed

Postoperative Care drainage is acceptable (typically 200 cc/24 h).

The Jackson-Pratt drain is pulled back 3–5 cm on
Patients are transferred to the intensive care unit postoperative day 5 so as to prevent delayed
and usually transition to the ward on postopera- fistulization to the anastomosis and resecured.
tive day 1. The typical hospital stay is 7 days in The drain is removed at the first postoperative
patients with an uncomplicated postoperative clinic visit in about 2 weeks.
course. The nasogastric tube may be removed on
postoperative day 2 if the patient is awake and
alert and nasogastric drainage is minimal. Enteral Outcomes
nutrition, in the form of “trickle” (30 cc/h) jeju-
nostomy tube feeds, is started typically on post- Our group has refined the minimally invasive
operative day 2. On day 3, tube feeds are advanced approach to esophageal resection over a period of
to goal nutritional intake and cycled over an 18 h several years in an effort to decrease the morbid-
period to facilitate ambulation during the daytime ity and mortality of open esophagectomy [1–5].
hours. A contrast esophagram is obtained on day Our minimally invasive approach to esophageal
3–4 if the patient has adequate pulmonary toilet resection was first described in 1998 [3], fol-
and a good cough. If there is no evidence of leak, lowed by the description of the initial experience
oral intake is initiated as 1–2 oz of clear liquids in 77 patients undergoing minimally invasive
per hour. This amount is advanced over 2 days to esophagectomy (MIE) in 2000 [5]. We described
full liquids with no more than 3–4 oz/h while the further experience with 222 patients in 2003
continuing cycled tube feeds. The chest tube on [2]. In 2011, we reported a large series of over
the operative side is removed once the volume of 1,000 MIEs [4], and currently are approaching
136 R.T. Ripley et al.

the 2,000 mark, with mortality rates in the range 2. Luketich JD, Alvelo-Rivera M, Buenaventura PO,
et al. Minimally invasive esophagectomy: outcomes in
of 1 % [4]. In a recent prospective study of 17
222 patients. Ann Surg. 2003;238:486–94; discussion
centers experienced in minimally invasive esoph- 494–5.
ageal surgery, minimally invasive esophagectomy 3. Luketich JD, Nguyen NT, Weigel T, et al. Minimally inva-
was associated with a 2 % mortality rate and sive approach to esophagectomy. JSLS. 1998;2:243–7.
4. Luketich JD, Pennathur A, Awais O, et al. Outcomes
offered a safe and oncologically equivalent alter-
after minimally invasive esophagectomy: review of
native to open esophagectomy [6]. over 1000 patients. Ann Surg. 2012;256:95–103.
5. Luketich JD, Schauer PR, Christie NA, et al. Minimally
invasive esophagectomy. Ann Thorac Surg.
2000;70:906–11; discussion 911–2.
References 6. Pennathur A, Luketich JD, Landreneau RJ, et al. Long-
term results of a phase II trial of neoadjuvant chemo-
1. Bizekis C, Kent MS, Luketich JD, et al. Initial experience therapy followed by esophagectomy for locally
with minimally invasive Ivor Lewis esophagectomy. Ann advanced esophageal neoplasm. Ann Thorac Surg.
Thorac Surg. 2006;82:402–6; discussion 406–7. 2008;85:1930–6; discussion 1936–7.
 Laparoscopic and Thoracoscopic
Esophagectomy with Side-Side 12
Thoracic Anastomosis

Kfir Ben-David and Isaac P. Motamarry

Over the past few decades, there has been a major difficulties for patients with esophageal
constant increase in the number of patients diag- cancer is accurate preoperative staging.
nosed with esophageal cancer in the United States. Noninvasive staging modalities include com-
There were 17,990 newly diagnosed patients with puted tomography (CT) of the chest, abdomen,
esophageal cancer in 2013, and 15,210 patients and pelvis and endoscopic ultrasound (EUS).
died from this malignancy. Although squamous EUS has become more sensitive with greater than
cell carcinoma is the most common malignancy of 93 % accuracy in differentiating mucosal versus
the esophagus worldwide, adenocarcinoma is con- submucosal lesions. However, there are multiple
siderably more prevalent in the United States. limiting factors including the location, type of
Regardless of histologic character, this malig- lesion, method and frequency of EUS probe, and
nancy has a reported overall 5-year survival rate of the experience of the endosonographer [2]. EUS-
13–18 % since most patients have advance disease guided fine-needle aspiration (FNA) for lymph
at initial presentation [1]. node staging has been compared to PET/CT in
Esophageal carcinomas are generally asymp- recent studies, and PET/CT has consistently pre-
tomatic, with patients typically complaining of dicted nodal status as well as response to neoad-
dysphagia or odynophagia. These symptoms are juvant therapy [3].
generally considered late manifestations of the The initial workup includes: a barium swallow
disease process. The esophagus lacks a serosa, to assess anatomy and esophageal function, fol-
which gives way to dilation, and patients are usu- lowed by an EGD for tissue biopsy. An EUS can
ally not symptomatic until 60 % of the circumfer- also be used for biopsy but is more frequently
ence is obstructed. Consequently, one of the used to assess depth of malignant penetration. CT
and PET-CT are used to evaluate for metastatic
disease, and a PET-CT is obtained post neoadju-
Electronic supplementary material Supplementary
material is available in the online version of this vant chemoradiation to assess response to treat-
chapter at 10.1007/978-3-319-09342-0_12. Videos can ment at our institution. Neoadjuvant therapy is
also be accessed at http://www.springerimages.com/ performed for T2–T4 and/or node-positive, M0
videos/978-3-319-09341-3.
malignancy [4]. Following the completion of
K. Ben-David, MD (*) neoadjuvant therapy, patients are restaged with
Department of Surgery, University of Florida Health, radiographic CT/PET imaging, and surgery is
Gainesville, FL, USA
e-mail: [email protected] offered to medically fit patients who do not have
metastatic disease.
I.P. Motamarry, MD
Department of General Surgery, Minimally invasive esophagectomy was first
University of Florida Shands, Gainesville, FL, USA described by DePaula and was a primarily

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 137
DOI 10.1007/978-3-319-09342-0_12, © Springer International Publishing Switzerland 2015
138 K. Ben-David and I.P. Motamarry

laparoscopic transhiatal approach with a cervical

anastomosis [5]. Since then, multiple publica-
tions have surfaced regarding various techniques
and outcomes shifting from open esophagectomy
to a minimally invasive approach which has
greatly decreased the overall morbidity and mor-
tality of the operation [6]. In fact, Luketich has
recently reported the largest MIE series with over
1,000 patients with a surgical mortality rate less
than 2 % for the first 500 cases and 0.9 % for the
latter 500 [7]. Similarly, our institution reported a
mortality rate of 1 %, pneumonia 9 %, anasto-
motic leak 4 %, and median length of stay of
7.5 days, utilizing the three-field esophagectomy
approach [8]. Consequently, there have been
many controversies regarding the optimal mini-
mally invasive surgical treatment of esophageal Fig. 12.1 Abdominal trocar placement
cancer. More specifically, a three-field esopha-
gectomy is superior to a two-field with regards to
lymphadenectomy, 5-year survival rate, periop- 12 mm port is placed 6 cm superior and lateral to
erative morbidities and mortality [9]. Although this port on the right side. These will serve as the
many prefer an Ivor Lewis minimally invasive surgeon’s working ports. The patient is placed in
esophagectomy with circular anastomosis [7], we steep reverse Trendelenburg position. A 5-mm
prefer the side-side thoracic anastomosis [10] incision is created inferior and to the left of the
since we have noticed a reduced anastomotic xiphoid process to allow for the placement of a
stricture formation. It also provides an excellent liver retractor (Fig. 12.1). The surgeon stands on
option for patient with previous neck dissections, the patient’s right side while the assistant is on
foregut surgery, or gastroesophageal tumors the contralateral side.
extending into the proximal stomach.

Abdominal Dissection
Operative Description
The gastrohepatic ligament is divided. The dis-
The patient is intubated with a double-lumen section is continued to the right crus up toward
endotracheal tube to achieve isolated left lung the phrenoesophageal ligament and across the
ventilation. An 18-gauge nasogastric tube is diaphragm to the left crus. A retrogastric tunnel
placed to help with gastric decompression. The is created by incising the tissue along the right
abdominal cavity is entered with a 5-mm trocar crus of the diaphragm. Care is taken to visualize
under direct vision into the lateral aspect of the the left crus posteriorly and place a blunt grasper
left subcostal region. The abdomen is evaluated behind the gastroesophageal junction (GEJ) from
for evidence of metastatic disease. If there is no right to left just inferior to the crura. The blunt
evidence of metastatic disease, additional trocars grasper is visualized as it exits the loose connec-
are placed under direct visualization. These tive alveolar tissue at the angle of His. A Penrose
include a 5-mm camera port, 2 cm to the left and drain is placed around the gastroesophageal junc-
superior to the umbilicus to be controlled by the tion. The Penrose is secured around the GE junc-
operative assistant. A 12-mm port is placed at the tion with an endo-loop.
same level as the previous port just lateral to the Following this portion, the operating surgeon
rectus muscle on the right side. An additional grasps the anterior aspect of the stomach, and the
12 Laparoscopic and Thoracoscopic Esophagectomy with Side-Side Thoracic Anastomosis 139

lesser sac is entered about halfway up the greater are divided inferiorly and the stomach is mobilized
curvature of the stomach. A tissue sealing device away from the pancreas until the gastroduodenal
is utilized to divide the gastrocolic omentum and artery is visualized. The surgeon on the right side
short gastric vessels. Great care is taken to pre- of the table mobilizes the first and second portions
serve the right gastroepiploic vessels (Fig. 12.2). of the duodenum along the superior aspect of the
The stomach is mobilized all the way until the left duodenum until the common bile duct is reached.
crus and Penrose drain are visualized at the angle Adhesions between the first portion of the duode-
of His. Mobilization of the lower half of the num to the liver, gallbladder, or porta hepatis are
greater curvature of the stomach is created carefully divided. A formal Kocher maneuver is
between the right transverse colon and the right often not necessary for an intrathoracic esophago-
gastroepiploic vessels. The gastropancreatic folds gastric anastomosis.
The right gastric artery along the superior
aspect of the lesser curvature of the stomach is
divided with a laparoscopic sealing device. The
lesser curvature of the stomach is elevated. The
surgeon skeletonizes the left gastric artery and
vein at their base to assure an extensive lymphad-
enectomy is achieved. Subsequently, the left gas-
tric artery and vein are stapled and divided with a
vascular load. The lymphadenectomy dissection
is continued along the common hepatic artery,
splenic artery, and superior portion of the pan-
creas toward the left crus (Fig. 12.3).
The gastric conduit is created by a series of
laparoscopic stapler firings along the lesser cur-
vature of the stomach. The nasogastric tube is
pulled back above the gastroesophageal junction.
The first firing is done via the right upper abdom-
Fig. 12.2 Division of the gastrocolic omentum inal 12 mm port. The stapler is introduced onto

Fig. 12.3 Skeletonization and division of left gastric artery and vein (From Hochwald and Ben-David [16] with
permission)
140 K. Ben-David and I.P. Motamarry

Fig. 12.4 Creation of the gastric conduit (From Hochwald and Ben-David [16] with permission)

the stomach 4 cm proximal to the pylorus along previously described [11]. The port site incisions
the lesser curvature of the stomach just proximal and liver retractor incision are sutured closed and
to the divided right gastric vessels. The stapler dressed appropriately after expelling the
firings continue along the body and fundus of the pneumoperitoneum.
stomach. The final division of the stomach is not
done until after each staple line is reinforced with
a single interrupted inverting suture of 2-0 silk. Thoracic Dissection
These sutures are placed at the junction of the
staple lines and are used as handles for subse- The patient is placed in the left lateral decubitus
quent transfer of the stomach to the posterior position ensuring that all of their bony promi-
mediastinum into the right chest (Fig. 12.4). nences are well padded. The right lung is deflated
Following the final application of the stapler and a 5-mm trocar is placed under direct vision
and division of the gastric conduit from the using a 5-mm 0° scope just inferior to the tip of
proximal stomach and GEJ, the esophagus is the right scapula. This serves as the camera port
further mobilized. The Penrose drain is pulled for the duration of the case, and the scope is
laterally and medially enabling mobilization of switched to a 5 mm 30° scope. The right chest
the distal esophagus through the hiatus into the cavity is insufflated with 8 mmHg of carbon
posterior mediastinum. The lower esophagus is dioxide (CO2) pressure. This allows for further
widely dissected incorporating all lymphatic tis- lung collapse during the thoracic portion of the
sue. Subsequently, the gastric tube is sutured to procedure. A 5-mm port is placed in the seventh
the lesser curvature side of the upper divided intercostal space along the posterior axillary line.
stomach utilizing two interrupted 2-0 silk A 12-mm trocar is placed in the tenth intercostal
sutures. space just above the diaphragmatic insertion
The Penrose is placed through the hiatus into slightly anterior to the vertebral bodies. A 12-mm
the posterior mediastinum while maintaining port is placed anteriorly in the seventh intercostal
appropriate orientation of the gastric conduit to space and is utilized for the lung retractor
prevent organoaxial rotation of the gastric tube (Fig. 12.5).
when it is being pulled into the right chest cavity. The lung is retracted anteriorly. The inferior
This is done by aligning the sutures placed along pulmonary ligament is divided. The lower esoph-
the lesser curvature of the stomach and straight- agus is widely dissected with an ENSEAL® tis-
ening the gastric tube. A 16-French feeding tube sue sealing device (Ethicon Endo-Surgery, Inc.,
is inserted into the proximal jejunum as we have Cincinnati, OH), and the Penrose drain is
12 Laparoscopic and Thoracoscopic Esophagectomy with Side-Side Thoracic Anastomosis 141

Fig. 12.5 Thoracic port placement (From Hochwald and Ben-David [16] with permission)

Fig. 12.6 Dissection of the distal esophagus (From Hochwald and Ben-David [16] with permission)

identified in the posterior mediastinum from our included with the specimen. If the thoracic duct
previous abdominal dissection. Care is taken not is identified, it is suture ligated or clipped. The
to enter into the left pleural space during this azygous vein is divided with a 45- or 60-mm
portion of the dissection. The esophagus is mobi- vascular load stapler (Fig. 12.7). The dissection
lized from its distal end to the level of the azygos continues with mobilization of the proximal
vein. The Penrose is advanced along the esophagus away from the trachea. It is important
esophagus during this dissection (Fig. 12.6). to continue the esophageal mobilization just
Periesophageal and subcarinal lymph nodes are distal to the thoracic inlet.
142 K. Ben-David and I.P. Motamarry

Fig. 12.7 Division of azygous vein (From Hochwald and Ben-David [16] with permission)

Thoracic Anastomosis (Video 12.1)

The NGT is removed
The gastroesophageal junction and gastric from the native
esophagus
conduit are pulled into the chest cavity. The sur-
geon keeps the orientation of the gastric tube so
that it does not twist. This is done by pulling on
the sutures placed along the lesser curvature of
the stomach and straightening the gastric tube as
it is pulled through the posterior mediastinum.
The proximal stomach specimen is separated
from the gastric conduit by dividing the previ-
ously placed 2-0 silk sutures that were tethering
them together. The posterior aspect of the gastric
tube is placed alongside the anterior aspect of the
esophagus with gentle tension superiorly. Cautery
Fig. 12.8 Esophagotomy at the native esophagus
is utilized to make an opening in the medial
aspect of the esophagus, 4–5 cm above the
divided azygous vein. The tip of the nasogastric line. The stapler is fired and removed (Fig. 12.11).
tube is pulled out from the esophagotomy The nasogastric tube is then advanced through
(Fig. 12.8). Similarly, electrocautery is used to the anastomosis and the tip left in the lower
create a gastrotomy on the posterior aspect of the aspect of the gastric conduit (Fig. 12.12). The
gastric conduit (Fig. 12.9). The anvil of a 6-cm common openings in the stomach and esophagus
staple load is introduced alongside the nasogas- are aligned with the aid of 2-0 silk stay sutures
tric tube into the esophagus, and the staple car- (Fig. 12.13), and the esophagogastrostomy is
tridge is placed in the stomach (Fig. 12.10). The sealed with the firings of the 6-cm linear stapler
stapler is closed, and care is taken to make sure (Fig. 12.14). The specimen is transected with
the nasogastric tube is not caught in the staple these same stapler firings. The omentum left on
12 Laparoscopic and Thoracoscopic Esophagectomy with Side-Side Thoracic Anastomosis 143

Gastrotomy is made on the

posterior wall of the 60 mm side-side
Gastric conduit anastomosis is
created

Fig. 12.9 Gastric conduit gastrotomy Fig. 12.11 Side-to-side linear esophagogastrostomy
anastomosis

The NGT is advanced

into the conduit
Anvil advanced into
the Esophagus

Fig. 12.10 The anvil of a 6-cm staple load is introduced Fig. 12.12 Advancement of the nasogastric tube through
alongside the nasogastric tube in the esophagus, and the the anastomosis into the gastric conduit
staple cartridge is placed in the stomach

the gastric conduit is brought in a circumferential advantages. All components of the operation are
fashion around the anastomosis and sutured back done under direct vision with minimal blunt
to the gastric conduit as an additional buttressing dissection. Appropriate lymphadenectomy can
layer (Fig. 12.15). A 24-French chest tube is be easily accomplished as we have previously
placed through the inferior 12 mm port and posi- described [8, 10]. The intrathoracic anastomosis
tioned along the posterior mediastinum. The is performed utilizing a 6-cm stapler, without
ports are removed and the incisions are closed concern regarding the functional lumen size.
with absorbable sutures. Potential advantages of a long side-to-side sta-
Minimally invasive esophagectomy utilizing pled anastomosis include lower leak rates due to
thoracoscopic and laparoscopic techniques with less tension and lower stenosis rates. In fact, a
thoracic esophagogastric anastomosis has several recent meta-analysis illustrated that anastomotic
144 K. Ben-David and I.P. Motamarry

Stay sutures are placed to

provide tension

The omentum and the

conduit are used to
buttress the staple line

Fig. 12.13 Alignment of the esophagogastrostomy Fig. 12.15 Buttressing the anastomosis with omental
opening pedicle

hospital stay, decreasing morbidity and lower

Side to side anastamosis respiratory complications when compared with
created over the NGT.
open esophagectomy [13, 14]. Despite these
superior outcomes, patients undergoing a laparo-
scopic/thoracoscopic Ivor Lewis resection with a
circular esophagogastric anastomosis have been
shown to have a 26 % anastomotic stricture rate
[15]. Hence, we have refined our technique for
construction of a thoracic esophagogastrostomy
using a 6-cm side-to-side linear stapled anasto-
mosis. Although this method is applicable to the
majority of patients undergoing minimally inva-
sive resection of the esophagus for esophageal or
GE junction cancer, it has become our procedure
of choice for patients with previous neck dissec-
Fig. 12.14 Closure of the esophagogastrostomy opening tions, foregut surgery, or gastroesophageal
with a linear stapler tumors extending into the proximal stomach.

leak was seen more commonly in the cervical References

group (13.64 %) than in the thoracic group
(2.96 %) [12]. This group is also significantly 1. Worni M, Castleberry AW, Gloor B, et al. Trends and
less likely to experience vocal cord paresis/paral- outcomes in the use of surgery and radiation for the
treatment of locally advanced esophageal cancer: a
ysis as noted by Luketich and colleagues [7]. In propensity score adjusted analysis of the surveillance,
addition, although not common, anastomotic epidemiology, and end results registry from 1998 to
leaks can be managed with minimal intervention 2008. Dis Esophagus. 2014;27(7):662–9.
to the patient. 2. Thosani N, Singh H, Kapadia A, et al. Diagnostic
accuracy of EUS in differentiating mucosal versus
Minimally invasive esophagectomy is a safe submucosal invasion of superficial esophageal can-
alternative to the open technique. Patients under- cers: a systematic review and meta-analysis.
going these operations benefit from shorter Gastrointest Endosc. 2012;75:242–53.
12 Laparoscopic and Thoracoscopic Esophagectomy with Side-Side Thoracic Anastomosis 145

3. Cerfolio RJ, Bryant AS, Ohja B, et al. The accuracy of 10. Ben-David K, Sarosi GA, Cendan JC, et al. Technique
endoscopic ultrasonography with fine-needle aspira- of minimally invasive Ivor Lewis esophagogastrec-
tion, integrated positron emission tomography with tomy with intrathoracic stapled side-to-side anasto-
computed tomography, and computed tomography in mosis. J Gastrointest Surg. 2010;14:1613–8.
restaging patients with esophageal cancer after neoad- 11. Ben-David K, Kim T, Caban AM, et al. Pre-therapy
juvant chemoradiotherapy. J Thorac Cardiovasc Surg. laparoscopic feeding jejunostomy is safe and effective
2005;129:1232–41. in patients undergoing minimally invasive esophagec-
4. Ben-David K, Rossidis G, Zlotecki RA, et al. Minimally tomy for cancer. J Gastrointest Surg. 2013;17:1352–8.
invasive esophagectomy is safe and effective following 12. Markar SR, Arya S, Karthikesalingam A, et al.
neoadjuvant chemoradiation therapy. Ann Surg Oncol. Technical factors that affect anastomotic integrity fol-
2011;18:3324–9. lowing esophagectomy: systematic review and meta-
5. DePaula AL, Hashiba K, Ferreira EA, et al. analysis. Ann Surg Oncol. 2013;20:4274–81.
Laparoscopic transhiatal esophagectomy with esoph- 13. Biere SS, van Berge Henegouwen MI, Maas KW,
agogastroplasty. Surg Laparosc Endosc. 1995;5:1–5. et al. Minimally invasive versus open oesophagec-
6. Kim T, Hochwald SN, Sarosi GA, et al. Review of tomy for patients with oesophageal cancer: a multi-
minimally invasive esophagectomy and current con- centre, open-label, randomised controlled trial.
troversies. Gastroenterol Res Pract. 2012;2012:683213. Lancet. 2012;379:1887–92.
7. Luketich JD, Pennathur A, Awais O, et al. Outcomes 14. Biere SS, Maas KW, Bonavina L, et al. Traditional inva-
after minimally invasive esophagectomy: review of sive vs. minimally invasive esophagectomy: a multi-
over 1000 patients. Ann Surg. 2012;256:95–103. center, randomized trial (TIME-trial). BMC Surg.
8. Ben-David K, Sarosi GA, Cendan JC, et al. Decreasing 2011;11:2.
morbidity and mortality in 100 consecutive minimally 15. Nguyen NT, Hinojosa MW, Smith BR, et al.
invasive esophagectomies. Surg Endosc. 2012;26: Minimally invasive esophagectomy: lessons learned
162–7. from 104 operations. Ann Surg. 2008;248:1081–91.
9. Ye T, Sun Y, Zhang Y, et al. Three-field or two-field 16. Hochwald SN, Ben-David K. Minimally invasive
resection for thoracic esophageal cancer: a meta- esophagectomy with cervical esophagogastric anasto-
analysis. Ann Thorac Surg. 2013;96:1933–41. mosis. J Gastrointest Surg. 2012;16(9):1775–81.
 Laparoscopic and Thoracoscopic
Transhiatal Esophagectomy 13
with Cervical Anastomosis

Moshim Kukar and Steven N. Hochwald

In this chapter, we will outline our technique for lymph node dissection has been previously
laparoscopic and thoracoscopic transhiatal esopha- reviewed elsewhere in this book and in previ-
gectomy with cervical anastomosis [1]. We utilize ous publications [4].
this technique routinely in patients with esophageal • Review the details of endoscopy, extent of
and Siewert’s types 1 and 2 gastroesophageal junc- stomach involvement, and location of tumor
tion cancer [2]. This technique has been shown to be in reference to the gastroesophageal (GE)
safe in the setting of neoadjuvant chemoradiation junction. The surgeon should be prepared to
since most patients in the western world present with perform an intraoperative endoscopy if
locally advanced carcinoma and receive multimo- needed.
dality treatment [3]. This technique is best reserved • Patients are instructed to drink 6–8 oz of
for those patients who have no history of previous whole milk or cream 6 h prior to the start of
gastric surgery such as a Nissen fundoplication. In the procedure. In those patients who are not
such patients, the amount of gastric conduit available able to tolerate this amount of liquid, the
to reach the neck may be limited and an intrathoracic cream can be given through a gastrostomy
esophagogastric anastomosis may be preferable as or jejunostomy feeding tube. In our experi-
described in other portions of this book. ence and others, it has been demonstrated to
significantly reduce postoperative chyle
leaks [5].
Preoperative Preparation

• The goals of surgery are to obtain a R0 resec- Anesthetic/Induction Phase

tion and remove appropriate lymph nodes
while minimizing morbidity. The extent of • Patient is intubated with double-lumen endo-
tracheal tube. Under bronchoscopic guidance,
Electronic supplementary material Supplementary the tube is confirmed in position so that patient
material is available in the online version of this can be maintained on single ventilation during
chapter at 10.1007/978-3-319-09342-0_13. Videos can
thoracic dissection.
also be accessed at http://www.springerimages.com/
videos/978-3-319-09341-3. • Arterial line is placed.
• We do not routinely use central venous access/
M. Kukar, MD (*) • S.N. Hochwald, MD, FACS monitoring unless otherwise indicated.
Department of Surgical Oncology,
• An 18 F nasogastric tube is placed carefully
Roswell Park Cancer Institute,
Elm & Carlton Streets, Buffalo, NY 14263, USA especially in patients with bulky lesions, and
e-mail: [email protected] location is confirmed in the stomach. It is

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 147
DOI 10.1007/978-3-319-09342-0_13, © Springer International Publishing Switzerland 2015
148 M. Kukar and S.N. Hochwald

Fig. 13.1 Patient positioning

important that this tube is placed while the • The bean bag is desufflated with patient in this
patient is in a supine position. position, making sure that the patient is not
• Foley catheter is placed. rotated.
• After positioning, repeat bronchoscopy is per-
formed to confirm the location of endotracheal
Patient Positioning (Fig. 13.1) tube. The right lung is clamped and single-
lung ventilation is begun.
• A bean bag and an overlying gel pad are prep-
ositioned on the operating table.
• The patient is positioned on the operating Thoracoscopic Dissection
table so that the anterior superior iliac spine (Video 13.1)
lies just inferior to the break of the table.
• Patient is positioned in left lateral decubitus Port Placement (Fig. 13.2)
position. The trough supporting the right arm
is positioned so as to allow the right arm to fall • Access: In most patients, two five and two
forward. An axillary roll is placed and all twelve mm ports are necessary to perform the
areas are appropriately padded and secured. thoracic esophageal dissection. After desuf-
• The table is raised and flexed so as to open the flating the right lung, a 5 mm Optiview trocar
intercostal spaces and the subcostal spaces with a 5 mm, 0° laparoscope is used to gain
and rotate the hips down and away from the access and inserted under direct vision just
horizontal position. inferior to the tip of the scapula. It is important
13 Laparoscopic and Thoracoscopic Transhiatal Esophagectomy with Cervical Anastomosis 149

• Dissection is begun by dividing the inferior

pulmonary ligament to the level of inferior
pulmonary vein.
• The goal is to encircle the lower esophagus
with a Penrose drain and to widely dissect the
periesophageal tissue free from the posterior
mediastinum. To accomplish this, the peri-
esophageal tissue is widely opened anteriorly
and posteriorly to the lower esophagus using
the ENSEAL® tissue sealing device (Ethicon
Endo-Surgery, Inc., Cincinnati, OH), carefully
Fig. 13.2 Thoracoscopic port placement (From Hochwald preserving the inferior pulmonary vein and
and Ben-David [1] with permission) avoiding entry into the pericardium and left
pleural space.
• The esophagus is encircled carefully using a
that this port is placed about equidistant RealizeTM (Ethicon Endo-Surgery, Inc.,
between the sternum and the vertebral bodies. Cincinnati, OH) dissector and a ¼-in. penrose
This serves as the camera port. The chest is is placed around the esophagus, and the ends
insufflated to a pressure of 8 mmHg. The cam- are secured with an endoloop tie as shown in
era is subsequently switched to a 5 mm, 30° Fig. 13.3. It is critical to have a 2-0 silk tie
scope after safe entry into the right pleural sutured to one end of the penrose and a small
space is confirmed. loop fashioned in the suture that can be
• A 5 mm port is then placed anteriorly in the attached to the RealizeTM and subsequently
seventh intercostal space. This port is utilized pulled around the esophagus. The penrose is
to retract the lung anteriorly during the initial left slightly loose so as to allow its movement
dissection. If a 5 mm fan retractor is not avail- along the length of the esophagus.
able, a 12 mm port can be placed for a larger • Both limbs of the penrose drain are grasped,
fan retractor. During the latter parts of the dis- and the esophagus is widely dissected from
section, this port can be used by the assistant distal to proximal in the right chest.
to help with the high thoracic dissection. Periesophageal lymph nodes are left on the
• A 12 mm port is inserted in the tenth intercos- specimen. The vagus nerves are divided. The
tal space just above the diaphragmatic inser- dissection is carried superiorly toward the
tion in a straight line beneath the camera port. azygous vein. The subcarinal lymph node
• A 5 mm port is inserted in the seventh inter- packet is carefully dissected while preserving
costal space midway between the inferior the membranous portions of the mainstem
12 mm and 5 mm camera ports. This along bronchi.
with the 12 mm port is used as the surgeon’s • During dissection of the lower esophagus
working ports. It is critical that these two ports from the inferior pulmonary ligament to the
are not placed too posteriorly so that interfer- level of the azygous vein, it is critical to apply
ence with the vertebral bodies is encountered. a generous amount of clips to the tissue
between the aorta and the vertebral bodies and
the esophagus. If the thoracic duct is visual-
Dissection ized, it is ligated, otherwise, multiple clips are
applied in this area to prevent injury to the
• 5 mm fan retractor is used to retract the lung thoracic duct or its branches.
anteriorly, exposing the esophagus. If a 5 mm • A window is made superior to the azygous
fan is not available or adequate, this can be vein and it is divided using a 45- or 60-mm
enlarged to a 12 mm port. vascular cartridge of the powered ECHELON
150 M. Kukar and S.N. Hochwald

Fig. 13.3 Thoracoscopic mobilization of the esophagus (From Hochwald and Ben-David [1] with permission)

Fig. 13.4 Thoracoscopic division of azygous vein (From Hochwald and Ben-David [1] with permission)

FLEXTM ENDOPATH® stapler (Ethicon the esophagus above the azygous vein to avoid
Endo-Surgery, Inc., Cincinnati, OH). Fre- injury to the recurrent laryngeal nerve (RLN).
quently the staple line on the azygous vein • We make an attempt to complete the cervical
should be reinforced with titanium clips if esophageal mobilization through the thoraco-
there is any evidence of bleeding (Fig. 13.4). scopic phase of the operation. Excellent visi-
• Dissection is carried superiorly to the azygous bility afforded by high-definition imaging and
vein, and an attempt is made to stay close to long instruments helps us accomplish this
13 Laparoscopic and Thoracoscopic Transhiatal Esophagectomy with Cervical Anastomosis 151

routinely. However, we try to minimize the

use of energy devices on the tissue between
the esophagus and the trachea to avoid injury
to the RLN. The dissection is carried under the
subclavian vessels up to the level of the
clavicle.
• After the cervical esophagus is mobilized at a
point superior to the thoracic inlet, we proceed
with dissection toward the GE junction and
care is taken to ensure that the esophagus is
mobilized down to the diaphragm. However, it
is important not to dissect into the abdominal
cavity.
• After the esophageal mobilization is com- Fig. 13.5 Penrose easily identified after opening the pre-
plete, the ¼-in. penrose is left in the superior vertebral fascia
apex of our dissection. The penrose should
be above the subclavian vessels – in close
proximity to the clavicle to ensure that the
cervical esophageal dissection has been
completed.
• Hemostasis is ensured and a 24 F Blake
drain is left posteriorly in the thoracic cavity
and brought out through the 12 mm port site,
and a purse-string suture is used to secure it
to the skin. This drain is connected to a
pleuravac.
• The right lung is insufflated under direct
vision and the port sites are closed with 4-0
Monocryl and Dermabond is applied.
Fig. 13.6 Left cervical esophageal mobilization

Repositioning Cervical Dissection (Video 13.2)

• The chest tube is connected to −20 mmHg of • A 6 cm skin incision is made along the anterior
wall suction. border of sternocleidomastoid starting from
• To recruit atelectatic right lung segments, the the suprasternal notch.
patient is put on dual lung ventilation with a • The platysma is divided.
PEEP set to 8 mmHg. • Sternocleidomastoid muscle is identified and
• The patient is placed supine and the left arm is moved laterally, carefully ligating and divid-
tucked. A shoulder roll is inserted under the ing any crossing jugular vein branches.
shoulder blades to optimize cervical exposure. • A self-retaining retractor is used to facilitate
• The head is extended and tilted slightly to the further dissection.
right. • The inferior belly of omohyoid is divided,
• The patient is placed supine on the operating exposing the prevertebral fascia.
table. Split-leg tables or stirrups are not used. • Keeping the jugular vein and carotid artery
A footboard is placed to facilitate steep reverse laterally, prevertebral fascia is opened and the
Trendelenburg position during the abdominal penrose is identified (Fig. 13.5) and secured
dissection. with a Kelly clamp (Fig. 13.6).
152 M. Kukar and S.N. Hochwald

• Another 12 mm trocar is inserted in the right

upper quadrant 6 cm superior and lateral to the
other 12 mm port.
• A 5 mm trocar is inserted just below and
slightly to the left of the xiphoid process, the
track is dilated with a hemostat clamp, and a
Nathanson retractor is inserted to retract the
left lobe of the liver anteriorly. The retractor is
secured to the right side of the bed.

Dissection

• The patient is placed in steep reverse

Trendelenburg position. The operating sur-
geon stands on the right side of the table and
the assistant stands on the left side of the table.
• The abdomen is thoroughly inspected for any
Fig. 13.7 Abdominal port placement evidence of metastatic disease.
• Initial dissection is begun by opening the pars
flaccida and the tissue along the right crus of
• Further dissection is done superiorly and infe- the diaphragm. A retrogastric tunnel is made,
riorly to ensure an adequate opening for easy carefully identifying the left crus, and a blunt
retrieval of the specimen and the gastric con- grasper is passed from right to left and is
duit through this incision. visualized below the left crus by carefully
• Since most of the cervical dissection is com- grasping the gastric fundus and pulling inferi-
pleted thoracoscopically, blunt cervical dis- orly and to the right. Care is taken to avoid
section is minimized during this phase to entry into the chest cavity and avoiding injury
avoid injury to the RLN, and this portion of to the spleen. A penrose drain is encircled
the procedure takes only 5–10 min. around the GE junction and secured with an
endoloop tie.
• The lesser sac is entered halfway up along the
Abdominal Dissection (Video 13.3) greater curvature of the stomach, carefully
preserving the right gastroepiploic arcade.
Port Placement (Fig. 13.7) After making an adequate window, the poste-
rior wall of the stomach is grasped by the
• 5 mm Optiview trocar with a 3-piece 5 mm, 0° operating surgeon’s left hand which allows the
laparoscope is inserted under direct vision in right gastroepiploic arcade to flip anteriorly.
the left upper quadrant, just lateral to Palmer’s Care is taken to grasp the stomach to provide
point. retraction. Minimal retraction of the omentum
• A 5 mm trocar is inserted 22 cm below the is performed.
xiphoid just to the left of the midline. This • The dissection is continued along the greater
serves as the camera port and the camera is curvature, dividing the short gastric vessels
switched to a 5 mm, 30° scope. with the vessel sealing device until the left
• A 12 mm trocar is inserted at the same level of crus and the Penrose drains are visible. During
the camera port just to the right of the right this dissection, the left gastroepiploic vessels
rectus abdominis muscle. are ligated near their origin. In addition, the
13 Laparoscopic and Thoracoscopic Transhiatal Esophagectomy with Cervical Anastomosis 153

Fig. 13.8 Skeletonization of the celiac trunk and division of the left gastric artery (from Hochwald and Ben-David [1]
with permission)

short gastric vessels are taken close to the the specimen. After skeletonizing the
spleen to assist in capturing splenic nodes. vessels, the pedicle is transected using a
• Further mobilization of the greater curvature vascular staple load on the powered Endo
toward the pylorus is performed by the assis- GIA (Fig. 13.8 ). Sometimes the left gastric
tant surgeon on the left side of the table. The vein and artery are taken separately to
transverse mesocolon is carefully mobilized facilitate a better nodal dissection. At this
off the right gastroepiploic arcade and the point, the stomach should be completely
head of the pancreas. mobile.
• To facilitate this dissection, gastropancreatic • The gastric conduit is created using multiple
folds are divided until the gastroduodenal 6 cm firings of 3.5 mm or 4.8 mm staple
artery is identified. Once the location of the loads, depending on the thickness of the
gastroduodenal artery is known, even in obese stomach. The operating surgeon’s left-hand
patients, the location of the right gastroepi- port is utilized to fire the first staple load,
ploic vessels can be determined. 4 cm proximal to the pylorus. Additional fir-
• The operating surgeon performs a Kocher’s ings are done using the surgeon’s right-hand
maneuver to mobilize the duodenum so as to port, following the curve of the stomach. We
allow the pylorus to reach the GE junction routinely use 5–6 staple loads (6 cm each),
with no tension. Most times, a full Kocher and care is taken to keep the width of the con-
maneuver is not required. During this portion duit around 5–6 cm. The stomach is not com-
of the dissection, the assistant grasps the pylo- pletely divided until sutures are used to
rus and retracts the stomach and duodenum to reinforce the junction of the staple lines
the patient’s left. (Fig. 13.9).
• The right gastric artery is divided 4 cm proxi- • A 2-0 silk Endostitch is used to reinforce the
mal to the pylorus. Using a Maryland dissec- intersecting staple lines, and the tails are left
tor, a window is made along the lesser long to facilitate passage of the conduit
curvature, and overlying tissue is divided with through the mediastinum and out the cervical
a sealing device. incision. After all the sutures are placed, an
• Nodal tissue along the left gastric vein and additional staple load is used to transect the
artery are dissected and swept up toward upper fundus of the stomach.
154 M. Kukar and S.N. Hochwald

Fig. 13.9 Creation of gastric conduit (From Hochwald and Ben-David [1] with permission)

• Using the penrose as a handle, the GE junction Reconstruction (Video 13.4)

is completely dissected free by widely divid-
ing the phrenoesophageal membrane and con- • The conduit is delivered out the neck incision,
necting the abdominal with the thoracic and the sutures holding it to the specimen are
dissection. The dissection is performed widely cut.
so that all the tissue between the left and right • The esophagus and the conduit are aligned so
crura is left on the specimen. It is important to that a side-side anastomosis can be created
communicate with the anesthesia team as the between the posterior wall of the stomach and
patient may have some hemodynamic instabil- anteromedial aspect of the cervical esophagus.
ity once the dissection is connected between Judicious care is taken to ensure the correct
the chest and abdomen. orientation of the conduit at all times.
• Two 2-0 silk Endostitches are used to anchor • An esophagotomy is made, and the NG tube is
the tip of conduit to the most inferior and right pulled out. A gastrostomy is made on the pos-
part of the transected specimen side of the terior wall of the stomach 4–5 cm proximal to
stomach so as to keep the correct orientation the tip of the conduit.
while pulling the gastric conduit through the • Using a 60, 3.5 mm load on the Endo GIA, a
mediastinum. 6 cm side-to-side stapled anastomosis is cre-
• Botulin toxin (100 units dissolved in 10 ml) ated. After the stapler is closed, the NG tube is
is used to inject into the pylorus. A total of moved to ensure free mobility. The stapler is
five to six ml is injected intramuscularly at fired. The NG tube is advanced through the
2–3 different areas in the anterior pyloric anastomosis and the tip left in the lower aspect
ring. We do not routinely perform a of the gastric conduit (Fig. 13.10).
pyloroplasty. • The common channel is closed with a 60,
• To facilitate the passage of the conduit to the 3.5 mm load of the TA stapler excising the tip
posterior mediastinum, the operating surgeon of the conduit. After the stapler is fired, the TA
grabs the silk tail ends of the sutures placed stapler is left in place and serves as a handle to
on the gastric conduit while the assistant pulls place two 3-0 silk sutures at the crotch of the
on the cervical esophagus. The specimen and staple line to decrease any tension.
gastric conduit are pulled out the cervical • The TA staple line is suture inverted with a
incision while the surgeon preserves the running 3-0 PDS suture. The anastomosis is
proper orientation at all times via visualiza- carefully pushed back into the posterior-
tion of the conduit both in the abdomen and superior mediastinum. A 7 F Jackson-Pratt is
the mediastinum. placed along the anastomosis. The platysma is
13 Laparoscopic and Thoracoscopic Transhiatal Esophagectomy with Cervical Anastomosis 155

Fig. 13.10 Cervical linear-stapled esophagogastrostomy (From Hochwald and Ben-David [1] with permission)

approximated with interrupted 3-0 Vicryl Table 13.1 Pearls and pitfalls
sutures and skin closed with 4-0 Monocryl. Preoperative
• The conduit is gently pulled down to ensure 1. Patient’s anterior superior iliac spine at the level
that redundant conduit is not left in the thoracic of the break of the table
cavity. The gastric conduit is sutured to the left 2. Right shoulder is slightly depressed and should
crus of the diaphragm with 2, 2-0 silk sutures to fall forward
avoid herniation of intra-abdominal contents. Thoracic dissection
• A prefashioned 16 F T tube (back wall is cut 1. Key anatomical structures to identify/preserve:
Inferior pulmonary vein, thoracic duct, membranous
and a portion is removed) is inserted in the portion of the trachea, recurrent laryngeal nerve
proximal jejunum 15–20 cm from the liga- 2. If thoracic duct is not visualized, multiple clips
ment of Treitz. It is anchored to the abdominal should be applied on the lymphatic tissue between
wall with multiple transfacial sutures [6]. the aorta and esophagus
• The two 12 mm ports are closed with 0 Vicryl 3. Mobilize the esophagus past the level of thoracic inlet
using a Carter Thompson device. All incisions 4. Minimize the use of energy device during
esophageal mobilization on the tracheal side to
are infiltrated with lidocaine and Marcaine avoid thermal injury to the recurrent laryngeal nerve
and closed with 4-0 Monocryl and Dermabond Cervical dissection
applied. 1. Minimize blunt dissection
• Table 13.1 details the pearls and pitfalls of 2. Ensure adequate opening so that the specimen and
each phase of dissection. conduit can be delivered easily
Abdominal dissection
1. Holding the posterior wall of the stomach during
Postoperative Care mobilization of greater curvature prevents injury
to right gastroepiploic vessels
2. Adequate mobilization of the first and second
Results utilizing this anastomotic technique have portions of the duodenum to allow the pylorus to
been previously published [2]. Anastomotic leak reach the gastroesophageal junction
rates are less than 5 % with a low stricture rate. For 3. Pull the NG tube back into the esophagus during
postoperative care, we follow an esophagectomy the creation of conduit
pathway at our institution. 4. Conduit width should be approximately 5–6 cm,
and a minimum of 5–6 staple load fires (6 cm
• Patients are transferred to a monitored setting loads) are needed for a conduit to reach the neck
for overnight observation and transferred to 5. Care is maintained to keep the right orientation
the floor on postoperative day 1 with telemetry while delivering the conduit into the mediastinum
monitoring. and during the anastomosis
156 M. Kukar and S.N. Hochwald

• Day 2, they are started on trickle tube feeds 2. Ben-David K, Sarosi GA, Cendan JC, Howard D,
Rossidis G, Hochwald SN. Decreasing morbidity
and the Foley catheter is discontinued.
and mortality in 100 consecutive minimally inva-
• Day 3, NG tube is removed if the chest x-ray sive esophagectomies. Surg Endosc. 2012;26(1):
shows a decompressed conduit. 162–7.
• Day 4, they are given a trial of colored clears 3. Ben-David K, Rossidis G, Zlotecki RA, Grobmyer
SR, Cendan JC, Sarosi GA, Hochwald SN. Minimally
and the neck JP is removed.
invasive esophagectomy is safe and effective follow-
• Day 5, they are advanced to full liquids and ing neoadjuvant chemoradiation therapy. Ann Surg
the right chest Blake drain is removed. Oncol. 2011;18(12):3324–9.
• Day 6–7, patients are advanced to goal tube 4. Kukar M, Hochwald SN. Operative and multimodal
aspects of esophago-gastric junction (EGJ) cancer
feeds when they have full return of bowel
care: western viewpoints. Textbook of complex gen-
function and usually discharged home on eral surgical oncology (in press).
postoperative day 7 with tube feeds for 16 h 5. Shen Y, Feng M, Khan MA, Wang H, Tan L, Wang
and also maintaining a full liquid diet. Q. A simple method minimizes chylothorax after
minimally invasive esophagectomy. J Am Coll Surg.
2014;218(1):108–12.
6. Ben-David K, Kim T, Caban AM, Rossidis G,
References Rodriguez SS, Hochwald SN. Pre-therapy laparo-
scopic feeding jejunostomy is safe and effective in
1. Hochwald SN, Ben-David K. Minimally invasive patients undergoing minimally invasive esophagec-
esophagectomy with cervical esophagogastric anasto- tomy for cancer. J Gastrointest Surg. 2013;17(8):
mosis. J Gastrointest Surg. 2012;16(9):1775–81. 1352–8.
 Laparoscopic and Thoracoscopic
Esophagectomy with Colonic 14
Interposition

Christopher Armstrong, Monica T. Young,

and Ninh T. Nguyen

Introduction stomach or a previous gastric resection rendering

the stomach unusable as a conduit. In these situa-
Surgical management of cancer of the thoracic tions, the colon or small bowel can be used as an
esophagus and proximal stomach is complex. alternative option. Advantages of colonic interpo-
Various surgical approaches have been utilized for sition include lack of acid reflux, preservation of
resection of these lesions [1]. The choice of the gastric reservoir (if the stomach is preserved),
approach is highly dependent on the location and long length, and that it is outside the radiation field
extent of the tumor. Patients with a gastric cardia for those patients receiving neoadjuvant chemora-
cancer without involvement of the esophagus but diotherapy [3]. These advantages are tempered by
with significant involvement of the gastric body increased technical complexity, construction of
may be candidates for total gastrectomy with three anastomoses, and a higher potential for anas-
Roux-en-Y esophageal-jejunal reconstruction. tomotic leak [4].
Patients with isolated gastric cardia cancer may be Although esophageal resection has been tradi-
candidates for an Ivor Lewis esophagogastrec- tionally managed with open surgery, minimally
tomy, transhiatal esophagectomy, or a three-hole invasive techniques have continued to evolve and
McKeown esophagectomy. In most cases, the now are increasingly utilized among specialized
stomach is the preferred conduit for reconstruction centers [2, 5, 6]. Colonic interposition adds fur-
due to its robust blood supply and the technical ther technical complexity to esophageal resection
advantages of a gastric pull-up with a single intra- and is typically undertaken using an open
thoracic or cervical anastomosis [2]. Occasionally, approach. Currently only one case report authored
patients with proximal gastric or distal esophageal by Nguyen et al. describes the surgical steps of a
cancers will have extensive involvement of the minimally invasive Ivor Lewis esophagectomy
with colonic interposition [7]. In this chapter, we
describe our technique of a laparoscopic and
Electronic supplementary material Supplementary
material is available in the online version of this
chapter at 10.1007/978-3-319-09342-0_14. Videos can M.T. Young, MD
also be accessed at http://www.springerimages.com/ Department of General Surgery,
videos/978-3-319-09341-3. University of California Irvine Medical Center,
Orange, CA, USA
C. Armstrong, MD, FRCSC
Department of General Surgery, Rockyview General N.T. Nguyen, MD (*)
Hospital and South Health Campus, Department of Surgery,
University of Calgary, 4448 Front St, SE, University of California Irvine Medical Center,
Calgary, AB T3M 1M4, Canada Orange, CA, USA
e-mail: [email protected] e-mail: [email protected]

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 157
DOI 10.1007/978-3-319-09342-0_14, © Springer International Publishing Switzerland 2015
158 C. Armstrong et al.

thoracoscopic esophagogastrectomy with colonic

interposition.

Technique

Patient Selection/Evaluation

In all cases, preoperative assessment is a crucial

component of the surgical process. Complete
knowledge of the proximal and distal extent of
the tumor is vitally important to planning the cor-
rect operative approach and reconstructive strat-
egy. Patients undergo an extensive endoscopic
evaluation of the proximal stomach and distal
esophagus with biopsies to accurately determine
the proximal and distal extent of the tumor.
Patients also undergo a comprehensive staging
workup which typically includes a CT-PET to
exclude metastatic disease. If the colon is to be
used for reconstruction, it is important that the
patient also undergoes a colonoscopy to exclude
the possibility of a synchronous colonic neo-
plasm. It is our practice that patients undergo CT
angiography to more accurately assess the
colonic blood supply. Patients are given a bowel
preparation preoperatively. The points of proxi-
mal and distal transection of the colonic inter-
Fig. 14.1 Surgical plan for resection and reconstruction
ponat are shown in Fig. 14.1. using the right colon to restore esophageal continuity
(From Nguyen et al. [7] with permission)

Abdominal Phase
and to the right of the umbilicus. These serve as
The patient is positioned supine for the initial the surgeon’s main operating ports. A final 5-mm
abdominal phase of the surgery. We employ a trocar is placed in the left upper quadrant and is
standard port placement for most procedures utilized by the assistant. An initial staging lapa-
involving the stomach or requiring dissection of roscopy is performed to exclude occult metastatic
the diaphragmatic hiatus (Fig. 14.2). This disease. We frequently do an intraoperative upper
involves establishing pneumoperitoneum using a endoscopy as well to ensure accurate assessment
Veress needle placed in the left abdomen lateral of the proximal and distal extent of the tumor.
to the umbilicus at the edge of the rectus abdomi- After staging laparoscopy excludes the pres-
nis. A 12-mm trocar is placed at this site. We then ence of occult metastatic disease, the hepatogas-
insert a 5-mm port in the right subcostal region tric ligament is divided and the left gastric vessels
beneath the inferior edge of the liver at the midax- are exposed. We perform a celiac lymphadenec-
illary line. This port is used for a fixed liver tomy en bloc and then proceed to divide the left
retractor. Another 5-mm port is placed in the gastric artery at the level of the celiac trunk with
right subcostal region at the midclavicular line a single firing of a linear stapler. The stomach is
and a 12-mm port is inserted slightly cephalad further mobilized by dividing the gastrocolic
14 Laparoscopic and Thoracoscopic Esophagectomy with Colonic Interposition 159

flexure. We routinely take down both flexures and

the transverse colon to minimize tension on the
esophagocolonic anastomosis in the chest. Since
the blood supply for the colonic interponat is
based on the middle colic vessels, it is critical to
identify them early and ensure that they are care-
fully preserved. The ileocolic and right colic ves-
sels are divided at the takeoff of the right colic
with a linear stapler. The right colonic mesentery
can usually be divided with bipolar cautery. The
distal aspect of the colon is divided proximal to
5 mm
the splenic flexure. The proximal aspect of the
5 mm
12 mm
divided terminal ileum is anastomosed to the
5 mm
colonic splenic flexure. Our preference is to con-
11 mm
(Camera port) struct a stapled side-to-side anastomosis using a
60-mm linear stapler. The remaining enterotomy
is closed with a two-layer running suture using
the Endo StitchTM (Covidien, CT). The mesen-
teric defect should also be closed to avoid poten-
tial internal herniation postoperatively. The
second anastomosis constructed is the gastrocolic
anastomosis. The distal aspect of the colonic
interponat is anastomosed to the prepyloric gas-
Fig. 14.2 Laparoscopic port placement tric remnant in a side-to-side fashion using a
60-mm linear stapler. Again it is our preference
to close the remaining enterotomy using two-
ligament and short gastric vessels. We routinely layer running suture.
perform a partial omentectomy during this phase Once these two anastomoses have been com-
of the procedure. pleted, we proceed with dividing the upper stom-
After the gastric fundus has been fully dis- ach or distal esophagus in the mediastinum. The
sected, the distal esophagus is mobilized into the cecal pole is sutured to the stomach or esopha-
mediastinum by opening the phrenoesophageal geal stump in preparation for a colonic pull-up
ligament. We routinely try to obtain at least 6 cm into the thorax (Fig. 14.3). We routinely place a
of mediastinal dissection transabdominally. Penrose drain around the distal esophagus in the
Unlike a gastric pull-up, there is no need to pre- mediastinum to aid in identification and retrieval
serve the right gastroepiploic vessels; therefore, of esophagus once in the chest. Our preference
these can be divided. The stomach is then tran- during esophagectomy is to remove the tumor
sected using a linear cutting stapler. In this situa- during the thoracic phase of the procedure via a
tion, it is our preference to leave a small remnant small thoracotomy incision; however, in some
stomach rather than dividing distal to the pylorus instances, a large tumor may be extracted from
as there is a lower risk of anastomotic disruption the abdomen if required. We routinely use a plas-
with a gastrocolic anastomosis compared to a tic wound protector to protect the wound from
duodenocolic anastomosis. In all cases, we rou- direct contact with the tumor. The 12-mm trocar
tinely send a frozen section to ensure that a incision close to the midline is best suited for
microscopically negative distal margin has been tumor extraction and can be enlarged as needed
achieved. to permit tumor extraction. Our protocol during
At this point, the right colon is mobilized any esophagectomy is to place a needle catheter
along the white line of Toldt toward the hepatic jejunostomy in the proximal jejunum to expedite
160 C. Armstrong et al.

point. If the esophagus requires additional

proximal mobilization, the azygous vein can be
divided with a 60-mm linear stapler. The esopha-
gus is then divided proximally using ultrasonic
shears. The remaining distal esophagus and
attached colonic interponat can then be pulled
into the right chest and separated. The surgical
specimen is removed from the chest through the
mini-thoracotomy incision.
When we initially reported our technique of
minimally invasive Ivor Lewis esophagectomy
with colonic interposition, we used a circular
stapled technique (Video 14.1). A 25-mm anvil
was inserted into the esophageal stump and
secured in place with a purse-string suture. The
ileocecal valve was then dilated to permit pas-
sage of the 25-mm circular stapler. The stapler
was inserted transthoracically through the 2.5-cm
trocar site and positioned through the terminal
ileum into the sidewall of the cecum. The esoph-
Fig. 14.3 Gastrocolic anastomosis has been performed agocolic anastomosis is then created by firing the
and the cecum has been anchored to the distal esophagus circular stapler (Fig. 14.4). The anastomosis was
for pull-up into the chest (From Nguyen et al. [7] with
permission) reinforced with a second layer of Lembert
sutures. The remaining enterotomy at the termi-
nal ileum was closed using a linear stapler and
enteral feeding postoperatively while the patient similarly oversewn with a second layer of
is kept nil per os. Lembert sutures. During our experience with
minimally invasive esophagectomy with gastric
pull-up, we subsequently changed our practice to
Thoracic Phase construct an entirely hand-sewn thoracoscopi-
cally performed esophagogastric anastomosis.
The patient is then repositioned in left lateral The esophagocolic anastomosis can also be con-
decubitus position under single lung ventilation. structed thoracoscopically using the Endo
Three trocars and a mini-thoracotomy incision StitchTM. Our preferred technique is a double-
(2–3 cm) are placed in the right chest. A plastic layer closure with interrupted SurgidacTM sutures
wound protector is used in the mini-thoracotomy (Covidien, CT). We routinely position a nasogas-
incision to protect the chest wall from any direct tric tube distal to the proximal anastomosis. We
contact with tumor cells. Dissection is initiated place an apically oriented 28-French chest tube
by mobilizing the inferior pulmonary ligament. as well as a basally oriented 10-French Blake
The mediastinal pleura overlying the distal drain (Johnson & Johnson Gateway, Livingston,
esophagus is incised using bipolar electrocautery, UK) for postoperative drainage.
and the Penrose drain encircling the distal esoph-
agus is retrieved. We have found that the Penrose
drain greatly assists with esophageal retraction Postoperative Care
during intrathoracic esophageal mobilization.
The esophagus is mobilized up to the level of the Following esophageal resection, our patients
azygous vein. We routinely reevaluate the proxi- are routinely sent to the intensive care unit for
mal extent of the tumor endoscopically at this the initial 24–48 h postoperative period.
14 Laparoscopic and Thoracoscopic Esophagectomy with Colonic Interposition 161

Fig. 14.4 A circular stapled

or hand-sewn
esophagocolonic anastomosis
is created in the chest. The
three anastomosis are now
completed: (1) ileocolic
anastomosis, (2) gastrocolic
anastomosis, and (3)
esophagocolic anastomosis
(From Nguyen et al. [7] with
permission)

Patients are given patient-controlled analgesia

with an opioid infusion. Enteral feeds are com-
menced via jejunostomy tube at 48 h and
advanced toward goal caloric intake as the
patient tolerates. The nasogastric tube is left to
gravity drainage and suctioned twice a day by
the surgical team during inpatient rounds. We
routinely perform a water-soluble contrast
study on postoperative day 5 to assess for anas-
tomotic leakage (Fig. 14.5). Patients are typi-
cally discharged from the hospital by
postoperative day 7 provided there is no clinical
evidence of leak. Patients receive enteral feeds
postoperatively and are allowed to start clear
Fig. 14.5 Postoperative water-soluble contrast study dem-
fluids by the second week after surgery. They
onstrating an intact proximal esophagocolonic and distal
are gradually advanced toward a soft diet over gastro-colonic anastomoses without evidence of anasto-
3–4 weeks. motic leak (From Nguyen et al. [7] with permission)
162 C. Armstrong et al.

Table 14.1 Selected series demonstrating morbidity and mortality associated with esophagectomy with colonic
interposition
Graft
Publication Mortality necrosis Anastomotic Anastomotic
Authors year Patients (#) (%) (%) leakage (%) strictures (%)
a
Wilkins [9] 1980 100 9 7 14
a
Isolauri et al. [8] 1987 248 16 3 4
DeMeester et al. 1988 92 5 7.6 4.3 4.3
[10]
Cerfolio et al. 1995 32 9.4 9.4 3.3 24
[11]
Mansour et al. 1997 129 5.9 3.0 14.8 2.3
[12]
Thomas et al. 1997 60 8.3 5.0 10.0 13.5
[13]
a
Kolh et al. [14] 2000 38 2.5 0 0
a
Hagen et al. [15] 2001 72 5.6 5.6 12.5
Popovici [16] 2003 347 4.6 1.2 6.9 6.3
a
Davis et al. [4] 2003 42 16.7 2.4 14.3
a
Knezevic’ et al. 2007 294 4.2 2.4 9.2
[17]
Motoyama et al. 2007 34 0 0 9 6
[18]
Klink et al. [19] 2010 43 14 9 30 19
a
Hamai et al. [20] 2012 40 0 5 17.5
Adapted from Yasuda and Shiozaki [23] with permission
a
Data not given

Discussion of patients and they reported an anastomotic leak

rate of 4 %. The major source of morbidity post-
There is only one published case report to date operatively in their series was pulmonary compli-
describing the technique of laparoscopic esopha- cations, which occurred in 8 % of patients. Other
gectomy with colonic interposition [7]. In this published series of open colonic interposition are
paper, the patient had a large esophagogastric shown in Table 14.1. These series demonstrate a
cancer involving the gastric body. He underwent large variation in overall mortality between 5 and
laparoscopic/thoracoscopic esophagectomy with 16 %. Anastomotic leak was noted to occur in
interposition of a proximal colonic segment 0–14.8 % of patients. The major source of postop-
based on the middle colic vessels. Total operative erative morbidity in most published series of open
time was 4 h for the laparoscopic portion and 2 h esophagectomy with colonic interposition was
for the thoracoscopic portion. Lymphadenectomy pulmonary complications. A minimally invasive
yielded 18 nodes, and microscopically negative approach to esophageal resection appears to
proximal and distal margins were achieved at the lessen the risk of postoperative pulmonary com-
time of resection. The patient tolerated the proce- plications in high-volume centers. This may be
dure well and did not have any evidence of radio- associated with a reduction in overall mortality
logic leakage postoperatively. [21]. The recently published TIME trial that com-
Isolauri and colleagues have published the pared minimally invasive esophagectomy with
largest series of open esophagectomy with recon- traditional open esophagectomy demonstrated a
struction by colonic interposition [8]. In that major reduction in pulmonary complications
series of 248 patients, they describe an overall between the open (29 %) and minimally invasive
mortality of 16 %. Graft necrosis occurred in 3 % groups (9 %) [22]. Patients who underwent
14 Laparoscopic and Thoracoscopic Esophagectomy with Colonic Interposition 163

minimally invasive esophagectomy also had a 7. Nguyen TN, et al. Laparoscopic and thoracoscopic
Ivor Lewis esophagectomy with colonic interposition.
shorter length of hospital stay and improved qual-
Ann Thorac Surg. 2007;84:2120–4.
ity of life postoperatively compared to those 8. Isolauri J, Markkula H, Autio V. Colon interposition
patients undergoing traditional open esophagec- in the treatment of carcinoma of the esophagus and
tomy. It is likely that the advantages observed dur- gastric cardia. Ann Thorac Surg. 1987;43:420–4.
9. Wilkins Jr EW. Long-segment colon substitution for
ing laparoscopic esophagectomy with gastric
the esophagus. Ann Surg. 1980;192:722.
pull-up are transferrable to minimally invasive 10. DeMeester TR, et al. Indications, surgical technique,
colonic interposition, although more published and long-term functional results of colon interposition
series of this technique are needed. or bypass. Ann Surg. 1988;208:460–74.
11. Cerfolio RJ, Allen MS, Deschamps C, Trastek VF,
Pairolero PC. Esophageal replacement by colon inter-
Conclusion position. Ann Thorac Surg. 1995;59:1382–4.
There are certain scenarios where it is not fea- 12. Mansour KA, Bryan FC, Carlson GW. Bowel interpo-
sible to use the stomach as a conduit for sition for esophageal replacement: twenty-five – year
experience. Ann Thorac Surg. 1997;64:752–6.
esophageal reconstruction. Colonic interposi-
13. Thomas P, Fuentes P, Giudicelli R, Reboud E. Colon
tion is a well-established alternative method interposition for esophageal replacement: current
of reconstruction following esophagectomy indications and long-term function. Ann Thorac Surg.
yet adds further complexity to an already 1997;64:757–64.
14. Kolh P, et al. Early stage results after oesophageal
high-risk surgical procedure. Minimally inva-
resection for malignancy—colon interposition vs.
sive techniques for esophageal resection are gastric pull-up. Eur J Cardiothorac Surg. 2000;18:
being increasingly utilized and appear to be 293–300.
advantageous in reducing pulmonary compli- 15. Hagen JA, DeMeester SR, Peters JH, Chandrasoma P,
DeMeester TR. Curative resection for esophageal
cations after esophagectomy. Although lapa-
adenocarcinoma: analysis of 100 en bloc esophagec-
roscopic esophagectomy with colonic tomies. Ann Surg. 2001;234:520–30; discussion
interposition is technically feasible [7], it is 530–1.
best undertaken at high-volume centers by 16. Popovici Z. A new philosophy in esophageal recon-
struction with colon. Thirty-years experience. Dis
surgeons with experience in both laparoscopic
Esophagus. 2003;16:323–7.
esophageal and colonic surgeries. 17. Knez̆ević JD, et al. Colon interposition in the treat-
ment of esophageal caustic strictures: 40 years of
experience. Dis Esophagus. 2007;20:530–4.
18. Motoyama S, et al. Surgical outcome of colon interposi-
tion by the posterior mediastinal route for thoracic
References esophageal cancer. Ann Thorac Surg. 2007;83:1273–8.
19. Klink CD, Binnebösel M, Schneider M, Ophoff K.
1. Kim T, et al. Review of minimally invasive esopha- Operative outcome of colon interposition in the treat-
gectomy and current controversies. Gastroenterol Res ment of esophageal cancer: a 20-year experience.
Pract. 2012;2012:683213. Surgery. 2010. doi:10.1016/j.surg.2009.10.045.
2. Nguyen NT, et al. Minimally invasive esophagec- 20. Hamai Y, Hihara J, Emi M, Aoki Y, Okada M.
tomy: lessons learned from 104 operations. Ann Surg. Esophageal reconstruction using the terminal ileum
2008;248:1081–91. and right colon in esophageal cancer surgery. Surg
3. DeMeester SR. Colon interposition following esopha- Today. 2012;42:342–50.
gectomy. Dis Esophagus. 2001;14:169–72. 21. Biere SS, Cuesta MA, van der Peet DL. Minimally
4. Davis PA, Law S, Wong J. Colonic interposition after invasive versus open esophagectomy for cancer: a
esophagectomy for cancer. Arch Surg. 2003;138: systematic review and meta-analysis. Minerva Chir.
303–8. 2009;64:121–33.
5. Luketich JD, et al. Outcomes after minimally invasive 22. Biere SS, et al. Minimally invasive versus open
esophagectomy: review of over 1000 patients. Ann oesophagectomy for patients with oesophageal can-
Surg. 2012;256:95–103. cer: a multicentre, open-label, randomised controlled
6. Lazzarino AI, et al. Open versus minimally invasive trial. Lancet. 2012;379:1887–92.
esophagectomy: trends of utilization and associated 23. Yasuda T, Shiozaki H. Esophageal reconstruction
outcomes in England. Ann Surg. 2010;252:292–8. with colon tissue. Surg Today. 2011;41(6):745–53.
 Thoracolaparoscopic
Esophagectomy in the Prone 15
Position for Carcinoma
of the Esophagus

C. Palanivelu, Palanivelu Praveen Raj,

Palanisami Senthilnathan, and R. Parthasarathi

In open surgery for intrathoracic esophageal The use of thoracoscopic esophagectomy in

tumors/carcinoma, transhiatal esophagectomy the prone position for esophageal cancer was first
was used for tumors in the lower third [1] of the reported by Cushieri et al. [5] in 1992.
esophagus and a transthoracoscopic McKeown Subsequently, no other group reported using this
three-hole/three-field approach with cervical approach. Many esophageal surgeons have been
anastomosis was used for tumors in the middle interested in performing minimally invasive
and upper thirds of the esophagus [2]. Initially, esophagectomy for cancer. All of the cases
we followed the same principles when using a reported in the literature reported using video-
laparoscopic approach [3]. Our standard approach assisted thoracoscopic (VATS) esophagectomy
for thoracoscopic esophagectomies is with the with the patient in the lateral decubitus position
patient in the prone position [4]. [6–8]. 10 years after the original publication, our
report of 130 cases using the approach with the
patient in the prone position created great enthu-
siasm among many surgeons across the world,
Electronic supplementary material Supplementary including those in Japan, Korea, and Europe [4].
material is available in the online version of this
chapter at 10.1007/978-3-319-09342-0_15. Videos can The author’s video using this approach received
also be accessed at http://www.springerimages.com/ best technique awards in various congresses such
videos/978-3-319-09341-3. as the American College of Surgeons (ACS) in
C. Palanivelu, MS, MCh, FRCS, FACS (*) 2005 [9], the 16th European Congress at
Division of Oesophagastric Surgery, Stockholm [10], and the 10th World Congress of
GEM Hospital and Research Centre, the International Society for Diseases of the
Pankaja Mills Road, Ramanathapuram,
Coimbatore, Tamil Nadu 64105, India
Esophagus [11]. The author performed a live
e-mail: [email protected] thoracolaparoscopic esophagectomy on a patient
P.P. Raj, MS, DNB(GI), DNB(SGI), FALS, FMAS
in the prone position for esophageal cancer dur-
P. Senthilnathan, MS, DNBGen, MRCSEd, ing the Hong Kong Asia Pacific Congress (ELSA)
DNB GISurg, FACS in 2005, which created great enthusiasm among
Department of Surgical Gastroenterology, the Asian group.
GEM Hospital and Research Centre,
Coimbatore, Tamil Nadu, India
Two-field lymphadenectomy was used for sur-
geries up to the infracarinal group of lymph
R. Parthasarathi, MS, FMAS
Division of Advanced Minimally Invasive
nodes. Decisions regarding the type of operation
and GI Surgery, GEM Hospital and Research Centre, were based primarily on the location of the tumor
Coimbatore, Tamil Nadu, India with the goal of low morbidity; radical

165
S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice,
DOI 10.1007/978-3-319-09342-0_15, © Springer International Publishing Switzerland 2015
166 C. Palanivelu et al.

lymphadenectomy was not the surgical goal in • Learning curve is shorter

the early 1990s. • Pneumothorax partially collapses the lung
Thoracoscopic en bloc esophagectomy [4] • Lung falls anteriorly
with the patient in the prone position and three- • Wide exposure is obtained without lung
field radical lymphadenectomy (TLE–3H–3F) retraction
became our standard of practice. We found no
significant change in the incidence of morbidity
and mortality by adopting the three-hole trans- Physiological
thoracic esophagectomy and cervical dissection
approach in comparison with laparoscopic tran- • Single-lumen endotracheal tube
shiatal esophagectomy [12]. • Double-lung ventilation
More and more two-hole esophagectomies • Improved ventilation/perfusion ratio
(TLE–2H) [13–15] with intrathoracic anastomosis • Incidence of postoperative pulmonary compli-
are being performed, limiting the three-hole esoph- cations is lower
agectomy (TLE–3H) approach to removal of upper • Improved postoperative oxygenation
esophageal growths/cancer. Total mediastinal,
extended two-field lymphadenectomy (TLE–2H–
TM) for adenocarcinoma and thoracoscopic modi- Anesthesia Effects
fied three-field lymphadenectomy including the
right cervicothoracic packet of lymph nodes (TLE– The TLE–3 F operative procedure can be divided
2H–3F) along the right recurrent laryngeal group into three steps (1) the thoracoscopic phase: en
are becoming the standard approaches [16–19]. bloc esophagectomy and radical lymphadenec-
Esophagogastrectomy for cancer of the cardia with tomy, (2) the abdominal phase: radical lymph-
shorter gastric tube thoracolaparoscopic esopha- adenectomy and gastric tube formation, and (3)
gectomy with two-field lymphadenectomy (TLE– the cervical phase: specimen extraction, gastric
2H–2F) is also performed. pull up, and cervicogastric anastomosis. Ivor
Lewis esophagogastrectomy (TLE–2F) is per-
formed in two phases (1) the laparoscopic phase:
Advantages of the Prone Approach gastric mobilization and lymph node dissection,
gastric tube formation, and extraction of speci-
Anatomical men and (2) the thoracoscopic phase: esopha-
gectomy, radical lymphadenectomy, and
1. Effects of gravity anastomosis. Rarely there is a second laparo-
– Liver falls caudally scopic phase wherein the specimen extraction is
– Heart falls anteriorly performed through a Pfanneiel incision after
– Mediastinum widens adjusting the patient’s position to supine before
2. No lung ventilation is needed extubation.
In a lateral approach, the esophagus lies in the General anesthesia with a single-lumen endo-
most dependent portion of the chest, where it is tracheal tube and the patient in a semiprone posi-
often obscured by overlying lung. tion is our standard practice, with a specially
made mechanical support (Figs. 15.1 and 15.2).
The semiprone position [11] allows for a lateral
Surgeon thoracotomy to be performed during an emer-
gency without changing the patient’s position.
• Positioning is comfortable for the surgeon The operative field of the right chest is prepared
• Blood does not accumulate near the dissection and draped anteroposteriorly from midline to
field midline.
15 Thoracolaparoscopic Esophagectomy in the Prone Position for Carcinoma of the Esophagus 167

Fig. 15.3 Thoracoscopic ports. (A) Camera 10-mm port.

Fig. 15.1 Operating table side support is fixed to the (B) Left-hand working 5-mm port. (C) Right-hand work-
table side rails. (A) Center knob to adjust the arm. (B) Side ing 5-mm port
knob to pull out the inner rod and adjust the length. (C)
Outer knob to turn the pad in either direction

for the work performed with the right hand is

placed in the third intercostal space. The third
trocar (10 mm) is placed in the seventh intercos-
tal space. The fourth trocar (5 mm) is placed in
the ninth intercostal space. The 10-mm trocar in
the seventh space is useful for applying clips,
vascular clamps, and taking sutures into the tho-
rax; the camera is sometimes used in this trocar
during mobilization of the lower esophagus
(Fig. 15.3). Initially, the right pneumothorax at
10 mmHg partially collapses the right lung,
which lies in the anterior compartment, and then
Fig. 15.2 Right forearm support the pressure is reduced to 6–8 mmHg. In addition
to the pneumothorax, gravity also aids in keeping
the collapsed lung in an anterior position. Two-
lung ventilation is continued throughout the
procedure.
Thoracolaparoscopic The surgeon and camera operator stand at the
Esophagectomy and Three-Field patient’s right side, and the video monitor is posi-
Lymphadenectomy in the Prone tioned directly opposite, on the patient’s left side
Position (Video 15.1) (Fig. 15.4). The surgeon, using a hook, incises
the mediastinal pleura overlying the anterior
TLE–3F: Thoracoscopic Phase aspect of the esophagus, and the inferior pulmo-
nary ligament is released up to the right pulmo-
A right pneumothorax is created either by using a nary vein. Anterior dissection is begun by
closed Veress needle technique or by using mobilizing the esophagus away from the hilum
Visiport. Four trocars are placed into the right and the pericardium [20]. Because of gravity, the
thoracic cavity (Fig. 15.2). The first trocar heart tends to fall down anteriorly; thus, the space
(10 mm) for the camera is placed in the fifth in front of the esophagus is widened. The mobili-
intercostal space corresponding to the level of the zation extends to the level of the azygos vein,
arch of the azygos vein. The second trocar (5 mm) which is skeletonized and divided with double
168 C. Palanivelu et al.

Fig. 15.4 Team setup for

thoracoscopy. The patient is
in the prone position, the
surgeon, camera surgeon,
and assisting surgeon stand
on the right side of the
patient and the monitor is on
the left side of the patient

Fig. 15.5 After pleural incision and dissection of the azy- Fig. 15.7 Umbilical tape is tied around the esophagus
gos vein, the azygos arch is ligated doubly with silk loosely for free sliding and retraction

surgeon. Many surgeons prefer an endovascular

(Endo GIA) stapler to divide the arch of the azy-
gos vein. The parietal pleura posterior to the
esophagus is incised from the level of the azygos
arch vein to the crus. Blunt dissection is used to
identify any potential thoracic duct branches and
perforator vessels from the aorta. The thoracic
duct is identified between the esophagus and
aorta, and is not routinely divided. In case of
tumor infiltration, the thoracic duct is clipped
Fig. 15.6 The azygos arch is divided and the vertebral
caudally at the hiatus and at the thoracic inlet at
side is retracted for exposure
its insertion with the subclavian vein cranially
and is transected. The esophagus is encircled
ligatures on both sides (Fig. 15.5). The posterior with an umbilical tape, and traction by the assis-
end of the thread is kept long and brought out tant through the fourth port provides excellent
through the posterior chest wall. Retraction of the exposure (Figs. 15.7 and 15.8). The surgeon is
thread dorsally separates the divided ends, pro- able to use both hands, simplifying the en bloc
viding a wider view. At the level of the aortic mobilization and lymphadenectomy. Initially, we
arch (Fig. 15.6), the azygos vein is the only struc- used three ports, and have now changed to using
ture that lies between the esophagus and the four ports; the fourth port is used for traction. The
15 Thoracolaparoscopic Esophagectomy in the Prone Position for Carcinoma of the Esophagus 169

Fig. 15.8 Dissection of the esophagus from the trachea Fig. 15.10 Clearance of lymph node stations 7, 8, and 9

and suction of the pleural cavity is performed.

A single 28-F chest tube is placed through the
seventh intercostal space and the lung is
re-expanded.

Abdominal/Laparoscopic Phase

The patient is positioned supine and five ports are

placed. The left lobe of the liver is retracted with
Fig. 15.9 Thoracoscopic view after complete mobiliza-
instruments through the subxiphoid (epigastric)
tion. (A) Trachea. (B) Arch of aorta. (C) Left bronchus.
(D) Right bronchus. (E) Pericardium. (F) Aortopulmonary trocar. A 10-mm port for the camera is placed in
window the epigastrium. Two working ports, a 12-mm
port in the right midclavicular line and a 5-mm
operative time is shorter using the two-handed port in the left midclavicular line are placed. One
technique. The groups of lymph nodes are dis- 5-mm port at the left anterior axillary line for
sected sequentially, the subcarinal, aortobron- gastric traction.
chial, paratracheal, right recurrent laryngeal, and The lesser omentum is incised and the stom-
then left recurrent laryngeal groups (Fig. 15.9) ach retracted to the left and anteriorly with a
are removed in that order [20]. The thoracoscopic grasper. The retroperitoneal lymphadenectomy
approach enables removal of the cervicothoracic (D2) is begun by incising the peritoneum at the
packet of lymph nodes along the right recurrent upper border of the pancreas. The retroperitoneal
laryngeal nerve using a strictly “no touch” tech- lymphatic and areolar tissues are swept superi-
nique [16, 18]. orly by skeletonizing the common hepatic artery,
If tumor is present in the lower esophagus, dissecting cranially along the lateral celiac group
the dissection starts from the upper chest; for (Fig. 15.10). The left gastric vein is divided first
upper growths, the dissection may start from at its insertion with the portal vein, followed by
the lower mediastinum. The entire periesopha- the left gastric artery, and then the celiac axis is
geal tissue and the lymph nodes are removed. If completely cleared. The left gastric artery is
the mediastinal pleura is infiltrated, it is also clipped with Hem-o-lok and divided. The dissec-
excised to obtain an R0 resection. In cases of tion is continued along the splenic artery up to
advanced tumors, where we anticipate excision the splenic hilum. This retroperitoneal dissection
of both pleura, the dissection begins from the extends up to the dissected esophageal hiatus
cranial end. The esophagus may be divided by superiorly, the hilum of the spleen laterally, and
stapling and retracted laterally, which exposes the common hepatic artery and inferior vena cava
the entire mediastinum. Thorough irrigation medially. Finally, the lesser curvature and left
170 C. Palanivelu et al.

Fig. 15.11 Intracorporeal formation of the gastric tube

using an Endo GIA stapler

Fig. 15.12 Divided esophagus, Ryle’s tube with a cover-

gastric nodes are included with the specimen as ing plastic sleeve is being pushed into the posterior
the gastric tube is prepared. mediastinum
The right gastroepiploic artery and the arterial
arcade along the greater curve is carefully place a nasojejunal feeding tube. Only in selected
assessed early to ensure its suitability as a vascu- cases in which the patient develops a leak do we
lar supply to the gastric conduit. The greater perform a feeding jejunostomy. The incidence of
omentum is divided at a safe distance from the developing a leak is very low [13] and feeding
gastroepiploic arcade, and the dissection is con- jejunostomy is not without morbidity.
tinued upward and to the left to divide the gastro-
colic and gastrosplenic ligaments by dividing the
short gastric vessels, keeping the dissection Cervical Phase
closer to the origin of the left gastroepiploic
artery from the splenic origin. On the right side, Specimen Extraction and Gastric
the dissection continues up to the second part of Pull Up
the duodenum. The right gastroepiploic vein is The cervical esophagus is dissected through a left
carefully protected from injury. collar incision and divided. The distal end of the
esophagus in the neck is over sewn and attached
to a long Ryles tube. A long plastic sleeve is used
Gastric Tube Formation as a protective sheath and attached to the esopha-
gus with a separate stitch (Fig. 15.12). The pneu-
A 5-cm-wide gastric conduit is created by means moperitoneum is reestablished and the esophagus
of multiple firings of an Endo GIA 6-cm car- is pulled down until the protective sheath reaches
tridge (Echelon–Ethicon) through the right mid- the peritoneal cavity. The stitch is released and
clavicle port. The stomach is stretched when the lower end of the plastic sheath opened
stapling starts on the lesser curvature, 5 cm away (Fig. 15.13). The Ryles tube in the neck is used to
from the pylorus, and progressing toward the fun- pull the esophagus into the plastic sheath by the
dus of the stomach. A golden cartridge is used to assistant; the surgeon working on the anterior
staple the antrum and blue cartridges are used for wall of the gastric tube pushes the tube carefully,
dividing the body of the stomach (Fig. 15.11). using a hand-over-hand technique and avoiding
A pyloroplasty or pyloromyotomy is per- twisting or spiraling (Fig. 15.14).
formed by incising the pylorus longitudinally and
the closure is performed transversely with inter- Esophagogastric Anastomosis
rupted sutures using 3-0 PDS suture. The place- A small vertical gastrotomy is performed with
ment of the feeding jejunostomy is at the electrocautery. The posterior wall of the esopha-
discretion of the surgeon. Our preference is to gus and the anterior wall of the stomach are then
15 Thoracolaparoscopic Esophagectomy in the Prone Position for Carcinoma of the Esophagus 171

Two-Hole Esophagogastrectomy
and Modified Three-Field
Lymphadenectomy in the Prone
Position

The abdominal dissection is performed first in

the same way as in the three-hole approach. A
gastric tube is formed, leaving adequate proximal
stomach with the specimen. The mediastinal dis-
section is performed beyond the upper limit of
Fig. 15.13 Plastic sleeve in position (from neck wound
connecting the peritoneal cavity) lying in the posterior
the growth. For smaller growths confined to the
mediastinum. (A) Nasogastric tube attached to the divided cardia, the esophagus is divided transhiatally and
end of the cervical esophagus extracted through a Pfannenstiel incision. For
larger growths or if there is greater involvement
of the esophagus, then the division is performed
in the right pleural cavity during the thoraco-
scopic phase.
The patient is moved into a semiprone posi-
tion. A mediastinal lymphadenectomy is per-
formed similar to that described in the three-hole
approach. The esophagus is divided high in the
upper mediastinum, keeping an adequate dis-
tance from the upper limit of the growth, and is
pushed into the peritoneal cavity to be removed
Fig. 15.14 Position of the stomach tube. The pyloro- after completing the thoracic phase. Stapling or a
plasty wound is visible hand-sewn anastomosis is performed. Complete
mediastinal lymphadenectomy for adenocarci-
noma and a modified three-field or extended two-
aligned. A 3-cm long, 3.5-mm Endo GIA stapler field lymphadenectomy by the thoracic approach
is used to perform the posterior anastomosis. The is our preference [22].
anterior anastomosis is performed transversely
using two staplers according to the modified
Collard [19] or Orringer technique [21] and a Postoperative Care
wide stoma is obtained. The anterior wall may
also be approximated with single-layer hand- Generally, the patient is extubated on the operat-
sewn continuous suturing using a 3-0 monofila- ing table and their recovery is good. Because of
ment absorbable suture, beginning at each corner the absence of a thoracotomy, the patients have
and tied in the middle. After completion of the less pain and their breathing is comfortable. On
anastomosis, any redundant stomach is retracted the first day after the surgery, nothing is adminis-
into the abdomen. A nasogastric tube is passed tered by mouth and the nasogastric tube is kept
carefully until it reaches the antrum of the stom- open for decompression of the gastric tube. On
ach and its tip is kept above the pylorus. The gas- the second postoperative day, gastrografin con-
tric tube is then secured to the diaphragmatic trast is administered and gastric emptying is
hiatus anteriorly and laterally using long 2-0 non- assessed. Patients are administered enteral feed-
absorbable sutures to prevent intrathoracic her- ing through a nasojejunal tube, beginning with
niation of the abdominal viscera. A nasojejunal clear fluids. Between the third and fifth day, the
tube is placed across the pylorus into the nasogastric tube is removed and patients are
jejunum. allowed to take oral liquids followed by semi-
172 C. Palanivelu et al.

solid followed soft diet by the end of first week. Demographic characteristics
If dilation of the gastric tube or delayed emptying Number of patients 610
occurs, then the postoperative care changes. Age range 22–87 years
If there is any doubt about the integrity of the Sex (men, women) 67 %, 33 %
anastomosis or delayed emptying of the gastric Period 1997–2002 2002– 2007–
conduit longer than 5 days, endoscopy is per- 2007 2012
Number of patients 45 180 385
formed with the patient under sedation. If there
Type of pathology 45/0 124/56 236/149
is an area of ischemic mucosa or a leak, then (squamous cell
contrast-enhanced computed tomography (CECT) carcinoma/
is performed, looking for collection. Small areas adenocarcinoma)
of mucosal ischemia can heal without additional Tumor location 26 upper; 244 middle; 340
intervention. In this group, placement of a feed- lower + cardia
ing jejunostomy is performed for enteral feeding.
Obvious anastomotic leaks are treated with an
endoscopic stent. If drainage fails and the CT Preoperative comorbidity
scan result shows collection, another thoracos- Number of
copy is performed for complete drainage. patients
Hypertension 47 (8 %)
Diabetes 62 (10 %)
Results Cardiovascular disease 12 (2 %)
Pulmonary disease 27 (4 %)
Neoadjuvant therapy, chemotherapy 135 (22 %)
More than 765 patients with esophageal cancer and/or radiation therapy
were treated by minimally invasive esophagec-
tomy between 1997 and 2013 at GEM Digestive
Cancer Institute, Coimbatore, India. Transhiatal Surgery
esophagectomy was performed in 165 patients
Type of surgery Number of
and thoracolaparoscopic esophagectomy in 610 patients
patients in the prone position. Of these, 132 Ivor Lewis 106 (17 %)
patients received neoadjuvant chemotherapy Two field (2F) 60
and/or radiotherapy for locally advanced disease Modified three field (3F; 46
as determined by staging thoracoscopy. In all 2F + cervicothoracic group)
except 12 patients, esophagectomy was com- Modified McKeown + neck 504 (83 %)
pleted successfully. In 504 patients, TLE–3F anastomosis
with cervical anastomosis was performed and, in
106 patients, two-hole thoracolaparoscopic
esophagectomy with intrathoracic anastomosis Perioperative factors
was performed. Operative time 310 minutes
The anastomotic leak rate was 3 % and the Blood loss 200–600 ml
mortality was 1.1 %. The mean intensive care ICU days 1.5 days
unit (ICU) stay was 2 days and the mean hospital Anastomotic leakages 3%
stay was 7.2 days. Vocal cord palsy was identified Gastric tip necrosis 1.35 %
in 1.5 % of the patients, most recovered in a few Vocal cord paralysis/paresis 1.5 %
days, only one case lasted for 30 days. The Pulmonary complications 2.4 %
median number of lymph nodes identified was Cardiovascular complications 3.75 %
21. No tracheal or bronchial injury was noted. Chylothorax 1%
Two cases had azygos arch venous injury that Overall morbidity 24 %
was managed by a thoracoscopic method. Hospital mortality 1.1 %
15 Thoracolaparoscopic Esophagectomy in the Prone Position for Carcinoma of the Esophagus 173

Pathology experienced anastomotic leaks identified on the

Median tumor 3.9 cm fourth and seventh postoperative days.
size Rethoracoscopies were performed and the
T status T0 8.7 %; T1 6.8 %; T2 25.2 %; anastomoses were revised. Both of these patients
T3 55.3 %; T4 3.9 %
recovered well without further leaks. In one
N status N0 42.7 %; N1 35.9 %; N2 8.7 %;
N3 12.6 % patient after TLE–3F, endoscopy revealed necro-
Margins positive Proximal: 6 cases sis of the proximal part of the gastric tube approx-
Distal: 0 cases imately 3 cm from the tip. Rethoracoscopy was
RO/R1 86 %/14 % performed, the gastric tube was taken back, and a
feeding gastrostomy and a cervicostomy were
performed. After 2 months, coloplasty recon-
Number of lymph nodes harvested struction was performed through the substernal
Mean 24.4 route.
Median 21
Hospital stay 7.2 days
Discussion
Other complications
Thoracoscopic view with the patient in the prone
Pulmonary embolism 2% position is unfamiliar to most surgeons, but easily
Reoperations (thoracoscopic revision) 1.15 %
adaptable because of the excellent ergonomics.
Revision anastomosis 2%
Using a single-lumen endotracheal tube and posi-
Drainage of abscess 3%
tioning the patient in a semiprone lateral approach
Gastric tube pull out 2%
position is easier and takes less time than using a
double balloon and the prone position. The diffi-
culty of an open conversion (posterior thoracot-
Rethoracoscopy for Anastomotic omy) in the case of massive hemorrhage is the
Leak only concern in an emergency with the patient in
the prone position [20]. With the patient in the
After TLE with stapled anastomosis, two patients semiprone position, a lateral thoracotomy can be
had leaks and rethoracoscopies were performed. performed in an emergency without changing the
In the first patient, the leak was diagnosed on the position. However, we never had such an
ninth postoperative day, just a day before their experience.
scheduled discharge. Endoscopy revealed the Thoracoscopic esophagectomy with the
anastomotic leak and collection adjoining the patient in the prone position has several
leak. CT scanning results revealed the collection. advantages, including a wide working space,
Rethoracoscopy was performed and a pneumo- tendency of the blood to collect outside the
thorax was created using Visiport. The collection operative field because of gravity, no need for
was drained, the pleural cavity thoroughly irri- skilled assistance, excellent ergonomic position
gated, and an intercostal drainage tube was for the surgeon, and reduction in lung injury
placed next to the leak. A percutaneous feeding because of the lack of lung handling [23, 24],
jejunostomy was performed. The fistula healed and the two-lung anesthesia with continuous
conservatively in 22 days. In the second patient, perfusion also significantly reduces postopera-
the fistula was a large opening and a stent was tive pulmonary complications. The potential
placed. After 2 weeks, the closure of the fistula advantages of a prone thoracoscopic mobiliza-
was confirmed and the patient was moved to oral tion may also include shortened operative times,
feeding. The stent was removed after 8 weeks. less surgeon fatigue, and shortening of the learn-
After an Ivor Lewis procedure, two patients ing curve [25].
174 C. Palanivelu et al.

Use of a double-balloon endotracheal tube is 3. Dantoc MM, Cox MR, Eslick GD. Does minimally
invasive esophagectomy (MIE) provide for comparable
not only time consuming, but also presents diffi-
oncologic outcomes to open techniques? A systematic
culties in exchanging the tube for a single-lumen review. J Gastrointest Surg. 2012;16(3):486–94.
endotracheal tube at the completion of the tho- 4. Palanivelu C, Senthilnathan P, Parthasarathy R.
racic mobilization and repositioning the patient Minimally invasive esophagectomy: thoracic mobili-
sation of the esophagus and mediastinal lymphade-
to a supine position for the abdominal phase.
nectomy in prone position – experience of 130 cases.
Dissection in front of the trachea and bronchus in J Am Coll Surg. 2006;203(1):7–16.
the presence of a double-balloon endotracheal 5. Cushier A, Shimi S, Banting S. Endoscopic oesopha-
tube may precipitate traumatic injury and delayed gectomy through a right thoracoscopic approach. J R
Coll Surg Edinb. 1992;37:7–11.
leakage. Any untoward incidence, such as injury
6. McAnena OJ, Rogers J, Williams NS. Right thoraco-
to the membranous bronchus or trachea, may be scopically assisted oesophagectomy for cancer. Br J
readily repaired as we do in open surgery [20]. Surg. 1994;81:236–8.
7. Gossot D, Cattan P, Fritsch S, Halimi B, Sarfati E,
Celerier M. Can the morbidity of esophagectomy be
reduced by the thoracoscopic approach? Surg Endosc.
Summary 1995;9:1113–5.
8. Robertson GS, Lloyd DM, Wicks AC, Veitch PS. No
Thoracoscopic esophagectomy with the patient obvious advantages for thoracoscopic two-stage
oesophagectomy. Br J Surg. 1996;83:675–8.
in the prone position is a safe operation, and radi-
9. PraveenRaj P, Palanivelu C, Parthasarathy R. Video pre-
cal en bloc esophagectomy and lymphadenec- sentation international Award session annual conference
tomy may be performed perfectly in a shorter of American College of Surgeons, New Orleans; 2007.
operative time, with less fatigue, reduced blood 10. Palanivelu C, Parthasarthy R, Senthilnathan P. Award
session; EAES best video session: 16th annual confer-
loss, and with a shortened learning curve for the
ence of European Society of Endo Surgeons,
surgeon. There are anatomic and physiologic Stockholm; 2008.
advantages in addition to the ergonomic conve- 11. Palanivelu C, Best abstract technique session; 10th
nience for the surgeon. The thoracolaparoscopic world congress of International Society for Diseases
of Esophagus, Kagoshima; 2010.
esophagectomy with the patient in the prone
12. Hulsher JB, Tijseen JG, Overtop H, et al. Transthoracic
position is likely to be the standard approach for verses transhiatal esophagectomy for carcinoma of
this operation in the future; the two-hole or three- the esophagus: a meta analysis. Ann Thorac Surg.
hole approach depends on the choice of the sur- 2001;72:306–13.
13. Pennathur A, Awais O, Luketich JD. Technique of
geon and the location of the tumor. The prone or
minimally invasive Ivor Lewis esophagectomy. Ann
semiprone approach is an excellent technique for Thorac Surg. 2010;89(6):S2159–62.
extended radical lymphadenectomy and its aim 14. Watson DI, Davies N, Jamieson GG. Totally endo-
should be to improve the survival rate. Modified scopic Ivor Lewis esophagectomy. Surg Endosc.
1999;13(3):293–7.
thoracoscopic extended two-field and modified
15. Nguyen NT, Follette DM, Lemoine PH, et al.
three-field lymphadenectomy and intrathoracic Minimally invasive Ivor Lewis esophagectomy. Ann
anastomosis are currently undergoing clinical tri- Thorac Surg. 2001;72(2):593–6.
als and the early results are encouraging. A mini- 16. Lam KY, Ma LT, Wong J. Measurement of extend of
spread of esophageal carcinoma by serial sectioning.
mally invasive approach may be used with low
J Clin Pathol. 1996;49:124–9.
morbidity and mortality. 17. Alkorki N, Skinner D. Should en bloc esophagectomy
be the standard of care for esophageal carcinoma?
Ann Surg. 2001;234:581–7.
18. Hegan JA, Peters PM, DeMeester TR. Superiority of
References extended enbloc esophagogastrectomy for carcinoma
of the lower esophagus and cardia. J Thorac Cardiovasc
1. Orringer MB, Sloan H. Esophagectomy without Surg. 1993;106:850–8.
thoracotomy. J Thorac Cardiovasc Surg. 1978;76: 19. Collard JM, Romagnoli R, Goncette L, et al.
643–54. Terminalised semi mechanical side to side technique
2. McKeown KC. Total three-stage oesophaectomy for for cervical esophagogastrostomy. Ann Thorac Surg.
cancer of the oesophagus. Br J Surg. 1976;63:259. 1998;65(3):814–7.
15 Thoracolaparoscopic Esophagectomy in the Prone Position for Carcinoma of the Esophagus 175

20. Ozawa S, Ito E, Kazuno A, Chino O, Makuuchi along the left recurrent laryngeal nerve by a minimally
H. Thoracosopic esophagectomy while in a prone invasive esophagectomy in the prone position for tho-
position for esophageal cancer: a preceding anterior racic esophageal cancer. Surg Endosc. 2010;24:
approach method. Surg Endosc. 2013;27:40–7. 2965–73.
21. Orringer MB, Marshall B, Iannettoni MD. Eliminating 24. Akaishi T, Kaneda I, Higuchi N, Kuriya Y, Kuramoto
the cervical esophagogastric leak with a side to side J, Toyoda T, Wakabayashi A. Thoracoscopic en bloc
stapled anastomosis. J Thorac Cardiovasc Surg. total esophagectomy with radical mediastinal lymph-
2000;119(2):277–88. adenectomy. J Thorac Cardiovasc Surg. 1996;112:
22. Nishihira T, Hirayama K, Mori S. A prospective ran- 1533–40. discussion 1540–1.
domised trial of extended cervical and superior medi- 25. Fabian T, Martin J, Katigbak M, McKelvey AA,
astinal lymphadenectomy for esophageal ca of the Federico JA. Thoracoscopic esophageal mobilization
thoracic esophagus. Am J Surg. 1998;175:47–51. during minimally invasive esophagectomy: a head-to-
23. Iwasaki H, Kobayashi K, Uchiyama A, Miyasaka Y, head comparison of prone versus decubitus positions.
Masatsugu T, Koike K, Miyazaki K. Lymphadenectomy Surg Endosc. 2008;22:2485–91.
 Thoracoscopic Enucleation
of Esophageal Benign Tumors 16
Yusuke Kimura, Akira Sasaki, Toru Obuchi,
Takeshi Iwaya, Yuji Akiyama, Masafumi Konosu,
Fumitaka Endo, Koki Otsuka, Hiroyuki Nitta,
Keisuke Koeda, and Go Wakabayashi

Introduction gastrointestinal tract for unrelated reasons [4]. The

treatment strategy for esophageal benign tumors,
Benign esophageal tumors are rare, with a such as leiomyomas, involves continued monitor-
prevalence of 0.005–5.1 %, based on autopsy ing of smaller tumors and surgical resection of
results, and account for <1–10 % of all esophageal larger or symptomatic tumors. Conventional, open
neoplasms [1, 2]. Leiomyomas constitute 70–80 % thoracotomy for enucleation of this tumor type has
of these benign esophageal neoplasms [1–3]. been gradually replaced by less invasive thoraco-
Other benign esophageal tumors, such as granular scopic or laparoscopic approaches [1, 2, 5–7]. In
cell tumors or schwannomas, are extremely rare. the present report, we describe our experience
Esophageal leiomyomas are usually detected in with patients undergoing surgical enucleation of
patients between 20 and 50 years of age, with a esophageal leiomyomas via thoracoscopic or lapa-
twofold male predominance, and most commonly roscopic approaches.
occur in the lower third of the esophagus. At least
50 % of patients with esophageal leiomyomas are
asymptomatic; in symptomatic individuals, dys- Management and Treatment
phagia is the most commonly reported symptom,
followed by chest tightness and pain. These tumors Surgical Indications
are usually discovered, incidentally, during esoph-
agography or endoscopic examination of the upper Surgical indications for thoracoscopic or laparo-
scopic enucleation of esophageal leiomyomas
Electronic supplementary material Supplementary
material is available in the online version of this
include the presence of dysphagia, foreign body
chapter at 10.1007/978-3-319-09342-0_16. Videos can sensations during swallowing, pathological con-
also be accessed at http://www.springerimages.com/ firmation that excludes malignancy, tumors
videos/978-3-319-09341-3. greater than 3 cm in diameter, and tumors that
Y. Kimura, MD, PhD (*) • A. Sasaki, MD, PhD show evidence of growth. Such patients under-
T. Obuchi, MD, PhD • T. Iwaya, MD, PhD went detailed assessment, including esophagog-
Y. Akiyama, MD, PhD • M. Konosu, MD
raphy, endoscopy, endoscopic ultrasound (EUS),
F. Endo, MD, PhD • K. Otsuka, MD, PhD
H. Nitta, MD, PhD • K. Koeda, MD, PhD and computed tomography (CT) evaluations.
G. Wakabayashi, MD, PhD Preoperative EUS-guided fine-needle aspiration
Department of Surgery, (EUS-FNA) was performed only in cases where
Iwate Medical University School of Medicine,
the morphologic appearance did not suggest that
19-1 Uchimaru, Morioka,
Iwate Prefecture 020-8505, Japan malignancy could be excluded with high
e-mail: [email protected] probability.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 177
DOI 10.1007/978-3-319-09342-0_16, © Springer International Publishing Switzerland 2015
178 Y. Kimura et al.

It remains controversial whether esophageal

submucosal tumors (SMT) should be preopera-
tively biopsied. The National Comprehensive
Cancer Network (NCCN) guidelines do not sug-
gest preoperative biopsy of esophageal SMTs
because the biopsy may result in hemorrhaging
and an increased risk of tumor dissemination [8].
Furthermore, surgeons who have performed enu-
cleations have subjectively stated that biopsies
increase the difficulty associated with identifying
the dissection plane. Although we have also oper-
ated on patients who have undergone EUS-
FNAB, we have not observed these complications. B D 4th
Esophageal gastrointestinal stromal tumor 5th
(GIST) resections are essentially limited to either
C 6th
simple enucleation or esophagectomy, but the
specific procedure to be performed is controver- 7th
sial [9]. We believe that complete resection 8th
PAL
remains the standard surgical treatment for local- MAL AAL
ized esophageal GISTs as the procedure reduces
the risk of tumor rupture and the consequent risks
of tumor relapse. In addition, complete resection
also avoids the possibility of microscopic resid-
ual tumors; enucleation methods should not be
indicated for esophageal GISTs.

Surgical Procedures (Video 16.1) Fig. 16.1 Arrangement of ports in thoracoscopic

approach. AAL anterior axillary line, MAL midaxillary
line, PAL posterior axillary line
Thoracoscopic Approach
For tumors of the upper and middle thirds of
the esophagus, we normally select a thoraco- esophagus, which facilitates to confirm the tumor
scopic approach utilizing a balloon-mounted location by intraoperative endoscopy. An upper-
endoscope. In our experience, thoracoscopic GI balloon endoscope is simultaneously inserted
enucleations have been performed under general in to the esophagus so that the tumor could be
anesthesia, administered using a double-lumen shifted toward adventitial layer by balloon pres-
endothoracheal tube for single-lung ventilation, sure in the esophageal lumen (Fig. 16.2). The
with the patient in the left lateral decubitus posi- mediastinal pleura over the tumor is then lon-
tion. We adopt four ports approach: the observa- gitudinally incised, and the tumor is pushed
tion port lies at the midaxillary line in the eighth away from the esophageal wall using a balloon-
intercostal space (A), the main working ports are mounted esophagoscope. The use of this type
located at the posterior axillary line in the fifth of esophagoscope improves the operative speed
(B) and the seventh (C) intercostal space, and the and safety. If necessary, the esophagus is cir-
secondary working port is lying at the anterior cumferentially mobilized in order to expose
axillary line in the fourth intercostal space (D) the tumor, and a Penrose drain is placed around
(Fig. 16.1). Thereafter, a flexible, 10-mm diam- the esophagus. At this stage of the procedure,
eter endoscope is used for the entire procedure. the esophagus can be rotated to some degree to
First, the right lung is retracted to expose the allow visualization of the tumor. The mediastinal
16 Thoracoscopic Enucleation of Esophageal Benign Tumors 179

pleura over the esophageal tumor was divided to combination of laparoscopic coagulating shears
expose the tumor and the adjacent esophagus. (LCS) and Endo Peanut TM(Covidien company)
An esophageal myotomy was performed using a avoiding injury to the mucosa. In order to avoid
wound-healing delay, LCS is not used as much
as possible.
After that blunt dissection was performed sep-
arating the tumor from the mucosa, followed by
applying traction suture to the tumor to aid in
tumor elevation as well as in the dissection which
was done mostly by blunt dissection (Fig. 16.3a).
After tumor enucleation, the specimen was
placed in a retrieval bag introduced through ante-
rior 10-mm trocar and was delivered through this
trocar wound. The dissected area is thoroughly
examined by endoluminal endoscopic inspection
d a after air insufflation of the esophagus. Finally, the
esophageal muscle layer is carefully closed using
b interrupted sutures to prevent the development of
a pseudodiverticulum; a chest tube is also rou-
tinely inserted via a thoracoscopic approach
c (Fig. 16.3b). An alternative method for thoraco-
scopic resection of an esophageal leiomyoma is
demonstrated in Video 16.2.

Laparoscopic Transhiatal Approach

A laparoscopic transhiatal approach is routinely
used for tumors of the lower third of the esopha-
gus. With the patient in a supine position, a pneu-
Fig. 16.2 The esophageal tumor is pushed away from the moperitoneum is established after the placement
esophageal wall using a balloon-mounted esophagoscope. of a 12-mm trocar into the subumbilical area,
(a) esophagoscope, (b) esophageal leiomyoma, (c) esoph-
using an open technique; CO2 insufflation, to a
agus, and (d) balloon

a b

Fig. 16.3 Thoracoscopic enucleation for an esophageal esophageal muscle layer was carefully closed using
leiomyoma. (a) The tumor was removed gently with interrupted sutures to prevent the development of
particular attention for not damaging mucosa. (b) The pseudodiverticulum
180 Y. Kimura et al.

pressure of 10 mmHg, is then maintained. Five procedure should be considered in cases in which
trocars are inserted into the upper abdomen, and technical problems or limitations are noted.
the phrenoesophageal ligament and the short gas-
tric vessels are divided using a Harmonic scalpel Conclusion
(Johnson & Johnson Medical, Cincinnati, OH, A thoracoscopic/laparoscopic approach offers
USA). After dissection of the abdominal esopha- potential advantages, compared with tradi-
gus, a Penrose drain is placed around the esopha- tional thoracotomies. These advantages
gus to aid in esophageal retraction; dissection of include its minimally invasive nature as well
the abdominal esophagus is very similar to a fun- as the lower respiratory morbidity, reduced
doplication dissection. After the esophageal SMT postoperative wound pain, and shorter hospi-
is identified, it is enucleated via a laparoscopic tal stay. In conclusion, thoracoscopic and lap-
transhiatal approach, and a Dor or Toupet fundo- aroscopic transhiatal enucleations for
plication was performed to restore the integrity of esophageal leiomyomas are safe and feasible
the anti-reflux mechanism. procedures. The optimal approach should be
tailored for each patient, based on the location
and size of the tumor.
Clinical and Technical Points

The first successful resection and enucleation of References

a benign esophageal tumor was reported by
Sauerbrach in 1932 [10]. Traditionally, tumors of 1. Jiang G, Zhao H, Yang F, Li J, Li Y, Liu Y, Liu J,
Wang J. Thoracoscopic enucleation of esophageal
the upper and middle thirds of the esophagus
leiomyoma: a retrospective study on 40 cases. Dis
have been approached by a right thoracotomy Esophagus. 2009;22:279–83.
and tumors of the lower third have been 2. Kent M, D’Amato T, Nordman C, Schuchert M,
approached by a left thoracotomy. Surgical enu- Landreneau R, Alvelo-Rivera M, Luketich J. Minimally
invasive resection of benign esophageal tumors.
cleations of esophageal leiomyomas by video-
J Thorac Cardiovasc Surg. 2007;134:176–81.
assisted thoracoscopic surgery have also been 3. Seremetis MG, Lyons WS, deGuzman VC, Peabody
reported and have contributed to the growing Jr JW. Leiomyomata of the esophagus. An analysis of
interest in the use of this approach in recent years 838 cases. Cancer. 1976;38:2166–77.
4. Zaninotto G, Portale G, Costantini M, Rizzetto C,
[5, 11, 12]. There are some clinical and technical
Salvador R, Rampado S, Pennelli G, Ancona E.
points that should also be considered during tho- Minimally invasive enucleation of esophageal leio-
racoscopic/laparoscopic enucleations of esopha- myoma. Surg Endosc. 2006;20:1904–8.
geal leiomyomas. First, the myotomy must be 5. Everitt NJ, Glinatsis M, McMahon MJ. Thoracoscopic
enucleation of leiomyoma of the oesophagus. Br J
performed in the correct direction and at the right
Surg. 1992;79:643.
level, and appropriate traction must be applied on 6. von Rahden BH, Stein HJ, Feussner H, Siewert JR.
the tumor to facilitate its subsequent enucleation. Enucleation of submucosal tumors of the esophagus:
Second, a balloon-mounted esophagoscope minimally invasive versus open approach. Surg
Endosc. 2004;18:924–30.
allows the localization of small tumors and per-
7. Samphire J, Nafteux P, Luketich J. Minimally invasive
mits the confirmation of mucosal integrity and techniques for resection of benign esophageal tumors.
ensures safety. Third, an anti-reflux operation Semin Thorac Cardiovasc Surg. 2003;15:35–43.
needs to be performed to restore the integrity of 8. Jiang P, Jiao Z, Han B, Zhang X, Sun X, Su J, Wang C,
Gao B. Clinical characteristics and surgical treatment
the anti-reflux mechanism, following dissection
of oesophageal gastrointestinal stromal tumours. Eur
and mobilization of the esophagogastric junction J Cardiothorac Surg. 2010;38:223–7.
and as a means of covering the myotomized 9. Demetri GD, Benjamin RS, Blanke CD, Blay JY,
esophageal mucosa. Finally, the difficulty of the Casali P, Choi H, Corless CL, Debiec-Rychter M,
DeMatteo RP, Ettinger DS, Fisher GA, Fletcher CD,
cases, and thus, patient selection should be care-
Gronchi A, Hohenberger P, Hughes M, Joensuu H,
fully considered. Conversion to an open Judson I, Le Cesne A, Maki RG, Morse M, Pappo AS,
16 Thoracoscopic Enucleation of Esophageal Benign Tumors 181

Pisters PW, Raut CP, Reichardt P, Tyler DS, Van den 11. Bardini R, Segalin A, Ruoi A, Pavanello M,
Abbeele AD, von Mehren M, Wayne JD, Zalcberg J, Peracchia A. Videothoracoscopic enucleation of
NCCN Task Force. NCCN Task Force report: manage- esophageal leiomyoma. Ann Thorac Surg. 1992;54:
ment of patients with gastrointestinal stromal tumor 576–7.
(GIST) – update of the NCCN clinical practice guide- 12. Obuchi T, Sasaki A, Nitta H, Koeda K, Ikeda K,
lines. J Natl Compr Canc Netw. 2007;5 suppl 2:S1–29. Wakabayashi G. Minimally invasive surgical enucle-
10. Sauerbrach F. Presentations in the field of thoracic ation for esophageal leiomyoma: report of seven
surgery. Arch Klin Chir. 1932;173:457. cases. Dis Esophagus. 2010;23:E1–4.
 Minimally Invasive Feeding Tube
and Esophageal Stent Placement 17
Erin Schumer and Robert C.G. Martin II

Introduction malnutrition [3]. In multiple reviews, patients have

been shown to have difficulty tolerating these thera-
Nutritional support for esophageal adenocarcinoma pies and report decreased quality of life scores as
and squamous cell carcinoma present a difficult well as worsening of their nutritional status [4].
challenge as approximately 60–80 % of esophageal This data has led to the development of multiple
cancer patients present malnourished, which strategies to improve both the nutritional status and
remains one of the highest rates among those diag- quality of life for patients with esophageal cancer.
nosed with cancer [1]. Poor nutrition has been
shown to negatively affect survival and operative
outcomes [2]. In addition to the well-recognized Evaluation
cachexia resulting from the tumor itself, esophageal
cancers exacerbate this problem through dysphagia A multidisciplinary team should be used to eval-
and obstruction. One current standard therapy for uate the patient presenting with esophageal can-
stage 2 and 3 esophageal adenocarcinomas includes cer. All patients undergo a complete staging
neoadjuvant chemotherapy and/or radiation for workup including imaging studies and an endo-
those planning to undergo surgical resection. This scopic ultrasound. Those patients deemed resect-
regimen can include cisplatin and fluorouracil- able based upon the staging workup and
based chemotherapy, which has side effects of nau- performance status undergo neoadjuvant chemo-
sea, vomiting, and diarrhea in addition to therapy and radiation. As part of the evaluation,
radiation-induced affects, further compounding the patient’s nutritional status is assessed using a
focused history and physical, the subjective
global assessment (SGA), and serum protein
Electronic supplementary material Supplementary
material is available in the online version of this markers such as prealbumin and albumin [5]. The
chapter at 10.1007/978-3-319-09342-0_17. Videos can SGA includes history of the patient’s weight loss,
also be accessed at http://www.springerimages.com/ GI symptoms, functional status, dietary intake, as
videos/978-3-319-09341-3. well as examination focused on the presence of
E. Schumer, MS, MD edema, ascites, muscle wasting, and loss of sub-
Department of General Surgery, University of Louisville, cutaneous fat. In deciding the use of esophageal
550 S. Jackson St., Louisville, KY 40202, USA stenting or laparoscopic or percutaneous jejunos-
e-mail: [email protected]
tomy tube, a number of clinical factors are taken
R.C.G. Martin II, MD, PhD, FACS (*) into account, including patients’ education level,
Division of Surgical Oncology,
University of Louisville, Louisville, KY, USA primary caregivers’ educational level, dietary
e-mail: [email protected] education/understanding, compliance estimation,

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 183
DOI 10.1007/978-3-319-09342-0_17, © Springer International Publishing Switzerland 2015
184 E. Schumer and R.C.G. Martin II

Table 17.1 Outline of steps for placement of endolumi- covered stent (Polyflex, Boston Scientific
nal esophageal stent
Corporation, Natick, MA) is placed using the
Step Action hemoclips as guidance for accurate distal place-
1 Insert endoscope ment. The preferred stent type is based on the
2 Dilate to 16 m with balloon dilator, if necessary type of neoadjuvant therapy that will be per-
3 Place hemoclip 2 cm at planned distal end of formed. If a patient is going to receive neoadju-
stent
vant chemotherapy alone, then we prefer a metal
4 Insert wire and remove endoscope
stent. If a patient is going to receive neoadjuvant
5 Under fluoroscopic guidance, place stent over
wire and advance distal end to hemoclip chemotherapy with radiation therapy, then we
6 Reintroduce endoscope and confirm placement prefer a silicone stent for ease of removal at the
Mm millimeter, Cm centimeter time of esophagogastrectomy [4]. It is essential
that only the 120 mm or 150 mm length stents are
ability to calculate calories during neoadjuvant used with only ≥18 mm OD, in order to reduce
therapy, and the type of neoadjuvant therapy migration rates [4, 6]. The endoscope is then
planned. reintroduced into the esophagus to assess for cor-
rect placement and good apposition of the stent to
the esophageal mucosa. Stents are left in place
until planned resection, death, or the need for
Esophageal Stent reintervention [7].
Stent placement has been shown to improve
Stent placements occur in either endoscopy or in dysphagia scores immediately post-procedure,
combination with staging laparoscopy and infu- thus allowing patients to increase their oral
saport placement in the operating suite by either nutritional intake. Over time, patients demon-
the attending surgeon or gastroenterologist using strate weight gain [8], better tolerance of neoad-
intravenous conscious sedation (Video 17.1 and juvant therapy, and improvement in quality of
Table 17.1). A diagnostic endoscope (Olympus life [9] (Table 17.2). In addition, a significant
America Inc., Center Valley, PA, GIFQ180) is proportion of patients do not proceed with
used with only the need to perform balloon dila- resection following neoadjuvant therapy due to
tion occurring if the diagnostic scope cannot tra- progression of disease. Placement of a stent as
verse the stricture. Using a Controlled Radial opposed to a feeding tube in this population
Expansion Wireguided Balloon Dilator (Boston avoids a more invasive procedure while improv-
Scientific Inc, Microvasive, Natick, MA), the ing overall quality of life [6]. Complications of
stricture is dilated to a maximum of 16 mm prior stent placement include migration, erosion, per-
to stent placement. Care should be taken not to foration, esophageal spasm, and obstruction.
dilate >16 mm, since this will only lead to greater Reintervention rate has been sited to range from
migration of the stent. After the stricture is 20 to 60 % most often for stent migration, but
dilated, a hemoclip is placed 2 cm at the planned this is highly dependent on the type of stent
distal extent of the stent, since all current stents placed [9]. Our recently completed prospective
are distal release (Video 17.2). A guidewire trial demonstrated a reintervention rate of 6 %,
(Jagwire High Performance Guidewire, 0.89 mm, thus proving that with adequate length and outer
Boston Scientific, Natick, MA) is placed across diameter, reintervention rates can be minimal
the lesion and advanced into the distal stomach. [4]. Overall, esophageal stenting is effective,
A wire exchange is performed with removal of more efficient, less invasive, with improved
the endoscope keeping the guidewire in place. quality of life tolerance, and is the optimal way
Using fluoroscopic guidance, either a fully cov- to improve both nutritional measures and over-
ered metal stent (WallFlex, Boston Scientific), all outcomes in the management of esophageal
EndoMaxx, or Cook or a retrievable silicone- cancer.
17 Minimally Invasive Feeding Tube and Esophageal Stent Placement 185

Table 17.2 Outcomes for esophageal stents during neoadjuvant therapy

Date Number of Complication Stent Dysphagia
Author published patients rate (%) migration (%) relief (%) Success of therapy
Langer et al. 2009 38 16 26 97.40 Improved nutritional and
[9] dysphagia results
Bower et al. 2009 25 4 24 100 Improved nutritional
[6] result, tolerance of
neoadjuvant therapy
Siddiqui et al. 2009 12 22 60 100 Improved dysphagia with
[10] similar nutrition
outcomes
Pellen et al. 2012 16 25 44 100 Improved symptoms,
[11] maintenance of nutrition
Brown et al. 2011 32 3 31 100 Improved dysphagia,
[4] maintenance of
performance status
Lopes and 2010 11 27 18 100 Improved dysphagia
Eloubeidi [12]
Adler et al. 2009 13 0 46 100 Improved dysphagia
[13]

Enteral Access the scope and the PEJ tube is advanced over the
wire until the wire can again be snared from the
Enteral access can be gained by placing a jeju- opposite end by the endoscope. The entire appa-
nostomy tube in either a percutaneous, laparo- ratus is again advanced into the pharynx through
scopic, or open technique. the esophagus into the stomach.
Percutaneous endoscopic jejunostomy (PEJ) PEJ tube placement offers the advantage of an
tubes are placed in the endoscopy suite by the endoscopic procedure under conscious sedation,
attending surgeon or gastroenterologist under IV since most patients will require nutritional sup-
conscious sedation. A variable pediatric colono- port during treatment, even if they are able to eat
scope is the best endoscope for this procedure prior to initiating therapy. Potential complica-
given the ability to variably make the scope stiffer tions of this procedure include injury to the bowel
with the adequate length needed to reach the vessels, bleeding, infection, erosion, and bowel
proximal jejunum. The stomach and duodenum injury. In only a small number of patients will the
are traversed and the jejunum is insufflated and esophageal tumors be so obstructive as to not
the site for tube placement is identified using allow passage of the endoscope. Overall, PEJ
transillumination and/or direct finger compres- tubes are an option for enteral support of esopha-
sion. The ideal location for tube placement is geal cancer patients, but should be used with cau-
approximately 4 fingerbreadths below the left tion. The current reported PEJ failure rate based
subcostal margin near the midclavicular line. on the inability to access the proximal jejunum is
This area is then prepped and draped and anes- approximately 15–20 %.
thetized with lidocaine. A 1 cm incision is made A laparoscopic jejunostomy tube is placed as
at the site and a needle is advanced into the jeju- an extension of the diagnostic laparoscopy with
num under direct visualization perpendicular to the use of one umbilical port and 2 additional
the abdominal wall. The guidewire is then intro- 5 mm ports (Fig. 17.1). This procedure is per-
duced through the needle and snared by the endo- formed under general anesthesia in the operat-
scope, which is then pulled back out of the ing room (Table 17.3). Access to the abdomen is
patient’s mouth. The wire is disconnected from gained using a Hassan trocar in the midline with
186 E. Schumer and R.C.G. Martin II

Fig. 17.1 Port sites for laparoscopic jejunostomy: 10 mm

at umbilicus and two right upper quadrant 5 mm ports.
Mm millimeter

Table 17.3 Outline of steps for placement of laparo- Fig. 17.2 Illustration of purse string suture in a laparo-
scopic jejunostomy tube scopic jejunostomy tube located 30–40 cm distal to the
ligament of Treitz
Step Action
1 Place 10 mm umbilical port
2 Place 5 mm RUQ ports ×2 circumferentially around the tube site using a
3 Identify jejunum, adhesiolysis if necessary
laparoscopic suturing device (Fig. 17.2 and
4 Identify jejunum 30–40 cm distal to ligament of
Treitz
Video 17.3). A smaller catheter, 12–16 French,
5 Place purse string at chosen site is passed through the abdominal wall using a
6 Place tube through abdominal wall stab incision and dissection with electrocautery.
7 Make enterotomy and place tube through We prefer to use a T-tube. A jejunotomy is
enterotomy. Secure purse string made, and the feeding tube is inserted into the
8 Fix bowel to intraperitoneal abdominal wall with enterotomy. The bowel is fixed to the abdominal
abdomen desufflated wall at the site of the enterotomy using absorb-
9 Secure tube to skin able suture after the abdomen is desufflated. The
RUQ right upper quadrant, Cm centimeter, Mm millimeter tube is fixed to the skin with Nylon suture and
all ports are closed [14]. An alternative method
two more ports placed in the right upper quad- for laparoscopic feeding tube placement is dem-
rant under direct visualization. Visualization of onstrated in Video 17.4.
the jejunum is achieved by adhesiolysis if The extracorporeal technique can also be per-
needed and with the use of atraumatic bowel formed in appropriate size patients (usually
graspers to the location where the site for the <35BMI), to which the site of the jejunostomy
J-tube placement is chosen. The jejunum is tube is brought out through the umbilical port site
identified and followed backward to the liga- and a direct jejunal tube is placed in the same
ment of Treitz, which commonly requires pull- above technique, but just under direct visualiza-
ing the transverse colon caudally. A site for the tion. Either technique is effective and obtains the
jejunotomy is chosen 30–40 cm distal from the same minimally invasive success of a jejunos-
ligament of Treitz. The actual placement of the tomy tube placement. While laparoscopic feed-
tube can be performed intracorporeally or ing jejunostomy has been more frequently
extracorporeally. Three or four intracorporal described, laparoscopic gastrostomy is also
4-0 Vicryl or PDS sutures are placed feasible.
17 Minimally Invasive Feeding Tube and Esophageal Stent Placement 187

Table 17.4 Outcomes for laparoscopic jejunostomy tube during neoadjuvant therapy
Number of Complication rate Tube exchange
Author Date published patients (%) rate (%) Success of therapy
Ben-David et al. 2013 153 2.60 7.20 Provision of enteral
[15] therapy
Siddiqui et al. 2009 24 4 Improvement in
[10] nutritional status
Jenkinson et al. 2007 43 2.30 20.90 Optimization of
[16] nutrition

Laparoscopic enteral feeding tube placement References

is minimally invasive, although this approach
does subject the patient to general anesthesia. 1. Miller KR, Bozeman MC. Nutrition therapy issues in
Placement is under direct visualization, whether esophageal cancer. Curr Gastroenterol Rep. 2012;14:
356–66.
it is placed into the stomach or jejunum, and 2. Ryan AM, Hearty A, Prichard RS, et al. Association of
allows for avoidance of the greater curvature of hypoalbuminemia on the first postoperative day and
the stomach, thus avoiding the gastroepiploic complications following esophagectomy. J Gastrointest
artery. In addition, laparoscopy can be used for Surg. 2007;11:1355–60.
3. Kleinberg L. Therapy for locally advanced adenocar-
complete visualization of the peritoneal contents, cinoma of the gastroesophageal junction: optimizing
thus allowing for assessment of metastatic dis- outcome. Semin Radiat Oncol. 2013;23:38–50.
ease and completion of staging before neoadju- 4. Brown RE, Abbas AE, Ellis S, et al. A prospective
vant therapy. phase II evaluation of esophageal stenting for neoad-
juvant therapy for esophageal cancer: optimal perfor-
Complications of both laparoscopic gastros- mance and surgical safety. J Am Coll Surg. 2011;
tomy and jejunostomy include bleeding, infec- 212:582–8.
tion, leakage of enteral contents, fistula formation, 5. Poziomyck AK, Weston AC, Lameu EB, et al.
and injury to the small and large bowel. Preoperative nutritional assessment and prognosis in
patients with foregut tumors. Nutr Cancer. 2012;64:
Gastrostomy tubes specifically may cause gastric 1174–81.
outlet obstruction. Jejunostomy tubes may cause 6. Bower M, Jones W, Vessels B, et al. Nutritional sup-
bowel obstruction or intestinal volvulus and are port with endoluminal stenting during neoadjuvant
more likely to become blocked due to the smaller therapy for esophageal malignancy. Ann Surg Oncol.
2009;16:3161–8.
caliber of tube that may safely be placed into the 7. Martin R, Duvall R, Ellis S, Scoggins CR. The use of
jejunum. In a retrospective series, Ben-David self-expanding silicone stents in esophageal cancer
et al. demonstrated that preoperative jejunostomy care: optimal pre-, peri-, and postoperative care. Surg
tube placement is a safe option for enteral access, Endosc. 2009;23:615–21.
8. Bower M, Jones W, Vessels B, et al. Role of esophageal
citing dislodgment or blockage as the most fre- stents in the nutrition support of patients with esopha-
quent complication [15]. Refer to Table 17.4. geal malignancy. Nutr Clin Pract. 2010;25:244–9.
9. Langer FB, Schoppmann SF, Prager G, et al.
Temporary placement of self-expanding oesophageal
Conclusion stents as bridging for neo-adjuvant therapy. Ann Surg
Nutritional status is an important part of the Oncol. 2010;17:470–5.
initial patient assessment and plays a role in 10. Siddiqui AA, Glynn C, Loren D, Kowalski T. Self-
pre- and postoperative outcomes. Patients expanding plastic esophageal stents versus jejunostomy
tubes for the maintenance of nutrition during neoadju-
should be cared for by a multidisciplinary vant chemoradiation therapy in patients with esophageal
team with experience in all types of feeding cancer: a retrospective study. Dis Esophagus. 2009;22:
access. Overall, there are several methods of 216–22.
improving nutrition for esophageal cancer, 11. Pellen MG, Sabri S, Razack A, et al. Safety and effi-
cacy of self-expanding removable metal esophageal
and these methods must be individualized to stents during neoadjuvant chemotherapy for resectable
each patient. esophageal cancer. Dis Esophagus. 2012;25:48–53.
188 E. Schumer and R.C.G. Martin II

12. Lopes TL, Eloubeidi MA. A pilot study of fully 14. Fischer JE. Fischer’s mastery of surgery. Philadelphia:
covered self-expandable metal stents prior to Wolters Kluwer Health/Lippincott Williams &
neoadjuvant therapy for locally advanced Wilkins; 2012.
esophageal cancer. Dis Esophagus. 2010;23: 15. Ben-David K, Kim T, Caban AM, et al. Pre-therapy
309–15. laparoscopic feeding jejunostomy is safe and effective
13. Adler DG, Fang J, Wong R, et al. Placement of in patients undergoing minimally invasive esophagec-
Polyflex stents in patients with locally advanced tomy for cancer. J Gastrointest Surg. 2013;17:1352–8.
esophageal cancer is safe and improves dysphagia dur- 16. Jenkinson AD, Lim J, Agrawal N, Menzies D.
ing neoadjuvant therapy. Gastrointest Endosc. 2009; Laparoscopic feeding jejunostomy in esophagogastric
70:614–9. cancer. Surg Endosc. 2007;21:299–302.
 Robotic Utilization in Esophageal
Cancer Surgery 18
Richard van Hillegersberg, Roy J.J. Verhage,
Pieter C. van der Sluis, Jelle P.H. Ruurda,
and A. Christiaan Kroese

Introduction However, conventional endoscopic surgery has

important limitations, such as a 2-dimensional
Optimal treatment for esophageal cancer consists view, a disturbed hand-eye-coordination, and lim-
of transthoracic en bloc esophagectomy (TTE) ited degrees of freedom. Robotic systems have
with an extensive mediastinal lymph node dissec- been developed to overcome these limitations [4].
tion. This approach through thoracotomy is During esophagectomy, the robotic platform
accompanied by significant morbidity, mainly enables the surgeon to perform an accurate medi-
consisting of cardiopulmonary complications. astinal dissection of the esophagus en bloc with
To reduce surgical trauma and morbidity of surrounding lymphatic tissue and mediastinal fat,
open transthoracic esophagectomy, less invasive often harboring metastatic disease. Robot-assisted
surgical techniques such as transhiatal esopha- thoracoscopic esophagectomy (RAMIE) in con-
gectomy (THE) and minimally invasive esopha- junction with conventional laparoscopy has
gectomy (MIE) have been introduced. shown to be technically feasible. Moreover, it pro-
Recent analyses of the MIE to date have shown vides sufficient oncological resection and is asso-
a decreased operative blood loss, reduced compli- ciated with low blood loss [5, 6]
cation rate, and shorter hospital stay [1–3].

Indications
Electronic supplementary material Supplementary
material is available in the online version of this
chapter at 10.1007/978-3-319-09342-0_18. Videos can Appropriate patient selection is essential to a
also be accessed at http://www.springerimages.com/ successful esophageal surgery program.
videos/978-3-319-09341-3. Approximately 30–40 % of esophageal cancer
R. van Hillegersberg, MD, PhD (*) patients are eligible to undergo an esophagectomy
R.J.J. Verhage, MD, PhD • P.C. van der Sluis, MD, MSc at curative intent, taking into account tumor stage
J.P.H. Ruurda, MD, PhD and comorbidity. The minimally invasive approach
Department of Surgery,
University Medical Center Utrecht, may offer a greater percentage of patients, a poten-
Heidelberglaan 100, Utrecht 3584 CX, tially curative surgical resection. Patients with
The Netherlands stage I–IV disease, i.e., T1–T4a tumors, and no
e-mail: [email protected] evidence of distant metastases are eligible to
A.C. Kroese, MD RAMIE. The ten times magnified 3-dimensional
Division of Anesthesiology, operative field, combined with an excellent manip-
Intensive Care and Emergency Medicine,
University Medical Center Utrecht, ulative freedom, allows radical resection even in
Utrecht, The Netherlands advanced cases [7].

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 189
DOI 10.1007/978-3-319-09342-0_18, © Springer International Publishing Switzerland 2015
190 R. van Hillegersberg et al.

In order to improve oncological outcome, enable selective deflation of the right lung during
multimodality treatment including neoadjuvant the thoracoscopic phase, patients are intubated
chemotherapy or chemoradiotherapy has become with a left-sided double-lumen tube. Patients
the standard of care in recent years [8]. A meta- receive two large-bore peripheral cannulae, a
analysis calculated hazard ratios for all-cause central venous line in the right internal jugular
mortality comparing neoadjuvant chemotherapy vein, an arterial line, a urinary catheter, and a
(0.87 (0.79–0.96); p = 0.005) or chemoradiother- nasogastric tube. Antibiotic prophylaxis is pro-
apy (0.78 (95 % CI 0.70–0.88; p < 0.0001) with vided by i.v. administration of 2,000 mg cefazo-
surgery alone. These data suggest a survival ben- lin and 500 mg metronidazole. Thirty minutes
efit of neoadjuvant chemoradiotherapy or chemo- before incision, 10 mg/kg methylprednisolone is
therapy over surgery alone in patients with administered to minimize postoperative pulmo-
esophageal cancer. However, a clear advantage of nary complications [12].
neoadjuvant chemoradiotherapy over neoadju- Patients receive general anesthesia with either
vant chemotherapy has not been established [9]. propofol or volatile anesthesia. During the thora-
coscopic phase of the operation, patients are posi-
tioned in the left lateral decubitus position, and
Perioperative Management selective ventilation of the left lung is instituted.
Continuous intravenous muscle relaxation is used
Preoperative to facilitate dissection of the esophagus along the
trachea, azygos vein, aorta, and pulmonary veins
All patients planning to undergo RAMIE are seen as sudden, unexpected movements of the patient
by an anaesthesiologist in the preoperative clinic. could have detrimental effects. The patient must
The physical status of the patient is assessed and be protected against inadvertent contact from the
preoperative testing is guided by institutional motions of the robotic arms. After the instruments
guidelines. Patients with the presence of and are connected to the arms of the robot and are
increased degree of perioperative complications placed inside the patient, the body position cannot
(e.g., cardiovascular complications) will be referred be modified unless the instruments are disengaged
for additional specialty care, as necessary, and and removed from the body cavity.
treatment as directed by the anaesthesiologist. When the robotic system is in place, access to
the patient in case of emergency is limited.
Therefore, the surgical team should be capable of
Intraoperative rapidly removing the robot if required.

Thoracic epidural analgesia (TEA) most likely

decreases the risk of postoperative respiratory Management of One-Lung
failure and results in improved pain control [10]. Ventilation in RAMIE
Furthermore, TEA may increase the blood supply
to the esophagogastric anastomosis area after To install one-lung ventilation (OLV), a left-sided
esophagectomy [11]. Although there are no spe- double-lumen tube (DLT) is used. Positioning of
cific publications on the effects of TEA during the DLT is most reliably achieved with a fiberop-
minimally invasive esophagectomy, the advan- tic bronchoscope. It has been shown that left
tages of TEA in the postoperative course of open DLTs, when positioned only by inspection and
esophagectomy can probably be extrapolated to auscultation, were in fact malpositioned in more
thoracoscopic esophagectomy. than 33 % of the cases. After positioning the
Normally, the epidural catheter is placed patient from supine to lateral, the position of the
between the fifth and the eight thoracic vertebrae. DLT is checked again routinely. Cuff pressure is
Usually epidural sufentanil is used intraopera- measured to prevent high intracuff pressures and
tively and a continuous infusion of bupivacaine possible mucosal damage. During OLV both
and morphine is applied postoperatively. To lungs are perfused. Perfusion of the nonventilated
18 Robotic Utilization in Esophageal Cancer Surgery 191

lung inevitably leads to transpulmonary shunting, value in particular in patients with decreased
impairment of oxygenation, and possible hypox- cardiac function. However, at the moment, no
emia. During OLV a protective lung ventilation large-scale randomized trials are available.
(PLV) protocol is applied, consisting of a
pressure-controlled ventilation strategy with a
maximum pressure of 20 cm H2O. Tidal volume Perioperative Complications
is reduced to 6 ml/kg predicted body weight.
Furthermore, 5 cm H2O PEEP is routinely used. The most common complications encountered
Although hypoxemia is a constant threat, the perioperatively include arrhythmias, most often
lowest possible fraction of inspired oxygen seen as the result of manipulation of the heart
(FiO2) is delivered to prevent oxidative damage during the thoracoscopic phase of the operation.
and postoperative acute lung injury. Usually these arrhythmias are self-limited after
In case of hypoxemia, the first treatment is an interruption of the surgical manipulation. Another
increase in FiO2. If no improvement occurs, the complication regularly seen is the development
surgeon is informed and the nonventilated lung is of a pneumomediastinum as a result of the open-
expanded with 100 % oxygen. Our clinical expe- ing of the hiatus during the laparoscopic phase of
rience suggests that dislocation of the DLT, atel- the operation. Hemodynamics may show the
ectasis, and bronchial occlusion of the ventilated characteristics of a tension pneumothorax. Again
lung with blood or secretions are the most occur- the surgeon should be informed immediately and
ring causes of hypoxemia. Therefore, immediate asked to lower the pressure of the pneumoperito-
fiberoptic bronchoscopy is performed to rule out neum. If indicated, thoracic drains are inserted to
or even correct dislocation of the DLT and relieve the pneumomediastinum.
occluded bronchi. Once these are ruled out, a
recruitment maneuver is performed to open pos-
sible atelectasis. Postoperative Care
When hypoxemia persists, the administration
of oxygen with or without CPAP to the nonventi- Postoperatively all patients remain under general
lated lung is a valuable option [13]. Clear commu- anesthesia and are intubated until they are trans-
nication with the surgeon is necessary in these ferred to the intensive care unit. Extubation is
circumstances as both maneuvers may have a neg- aimed for the same day. Although immediate
ative impact on the surgical exposure during thora- extubation in the operating room has been
coscopy. When applying CPAP, the nonventilated described and considered safe, we consider it
lung is first reinflated as CPAP alone does not appropriate to ventilate patients postoperatively
inflate an atelectatic lung. At the end of the thora- until chest X-ray is obtained and information on
coscopic phase, the nonventilated lung is reinflated the actual respiratory status is available. When
under direct vision and extensive recruitment the X-ray shows no significant atelectasis, wean-
maneuvers are performed after which two-lung ing from ventilation is started.
ventilation is restarted and 10 cm H2O PEEP is
added. There is no more need for lung separation
during the rest of the operation and usually the Robot-Assisted Thoracoscopic
DLT is exchanged for a single-lumen tube (SLT). Dissection (Video 18.1)

Robotic Instruments
Fluid Management
• Hook
Fluid strategy during RTE is aimed at a mildly pos- • Cadiere
itive fluid balance of approximately 500–1,000 ml • Needle driver
at the end of the procedure. The use of central • Long tip forceps
venous oxygen saturation may have additional • Hem-o-lok® Ligation clips
192 R. van Hillegersberg et al.

Fig. 18.1 OR setup. The patient is in left lateral position. The robot is docked from the dorsocranial side

Positioning

The patient is positioned in the left lateral decubi-

tus position, tilted 45° toward the prone position.
The operating table is flexed, lowering the legs
and upper thorax (the patient is positioned with
the xiphoid above the pivoting point of the table).
This extends the thorax and widens intercostal
space for introducing trocars. The bedside cart is
brought into the operative field from the dorso-
cranial side of the patient (Fig. 18.1). Before inci-
sion, the right lung is desufflated. A 10-mm
camera port is placed at the sixth intercostal
Fig. 18.2 Port position. Robotic arms 1 (yellow), 2
space, posterior to the posterior axillary line. Two (green) and camera (blue). Two assisting ports (white)
8-mm ports are placed just anterior to the scapu-
lar rim in the fourth intercostal space and more
posterior in the ninth intercostal space. Two tho- tional thoracoscopic assistance such as suction,
racoscopic ports are used in the fifth and seventh traction, and clipping (Fig. 18.2). CO2 insuffla-
intercostal spaces just posterior to the posterior tion of the thoracic cavity with 6 mmHg permits
axillary line. These ports are used for conven- excellent vision, without the need for retracting
18 Robotic Utilization in Esophageal Cancer Surgery 193

a b

Fig. 18.3 (a) Identification of the azygos vein (AV). (b) Division of the azygos vein over the esophagus

a b

Fig. 18.4 (a) The thoracic duct (TD) is identified. (b) The thoracic duct (TD) is identified, clipped at the level of the
diaphragm, and divided

the lung from the operative field. In case of a non- Subsequently, the parietal pleura is dissected at
compliant lung, a retractor can be used. the posterior side of the esophagus cranially to
caudally along the azygos vein, including the
thoracic duct. Paratracheally left, the left recur-
Operative Steps rent nerve is identified and carefully protected.
At the level of the diaphragm, the thoracic duct is
After division of any pulmonary adhesions and a clipped with a 10-mm endoscopic clipping device
proper overview of the operating field is achieved, (Endo ClipTM II; Covidien, Mansfield,
the right pulmonary ligament is divided. The Massachusetts, USA) to prevent postoperative
parietal pleura is dissected at the anterior side of chylous leakage (Fig. 18.4a, b).
the esophagus from the diaphragm up to the azy- At the level of the diaphragm, a Penrose drain
gos arch. The azygos arch is carefully ligated is placed around the esophagus to provide trac-
with robotic hemoloc clips (Fig. 18.3a, b). Then tion, which facilitates esophageal mobilization
dissection of the parietal pleura is continued (Fig. 18.5). The esophagus is then resected en
above the aortic arch for a right paratracheal bloc with the surrounding mediastinal lymph
lymph node dissection. The right vagal nerve is nodes and the thoracic duct from the diaphragm
dissected below the level of the carina. up to the thoracic inlet. Aortoesophageal vessels
194 R. van Hillegersberg et al.

Fig. 18.5 A Penrose drain (PD) is placed around the

esophagus (E) at the level of the pericardium to retract the
esophagus anteriorly

are identified and clipped by the assisting surgeon.

The extensive lymphadenectomy includes the
right- and left-sided paratracheal (lymph node sta-
tion 2R, 2L), tracheobronchial (lymph node sta-
tion 4), aortopulmonary window (station 5),
carinal (station 7), and periesophageal (station 8)
lymph nodes. A 24-Fr chest tube is placed, and
the lung is insufflated under direct vision. Fig. 18.6 Laparoscopic trocar placement

subcostal area, and a 12-mm port is placed

Laparoscopic Dissection pararectally right for the liver retractor (Fig. 18.6).
(Video 18.1) The abdomen is insufflated to a carbon dioxide
pressure level of 15 mmHg.
Instruments

• Harmonic scalpel Operative Steps

• 2× fenestrated bowel clamps
• Endopaddle The hepatogastric ligament is opened. The
• Clipper greater and lesser curvatures are dissected with
• Hem-o-lok® Ligation clips ultrasonic harmonic scalpel (Harmonic Ace®,
Ethicon Endo-Surgery, Johnson & Johnson, New
Brunswick, New Jersey, USA). The hiatus is
Positioning opened, and the distal esophagus is dissected
from the right and left crus. The carbon dioxide
After completion of the robot-assisted thoraco- pressure level is reduced to 6 mmHg to avoid
scopic esophageal mobilization, the patient is put excessive intrathoracic pressure, and a chest tube
in supine position. An 11-mm camera port is is placed in the left pleural sinus. Dissection and
introduced left paraumbilically, and an 11-mm lymphadenectomy then continue around the
working port is placed at the right midclavicular celiac trunk. The left gastric artery and vein then
line at the umbilical level. A 5-mm working port are transected at their origin with Hem-o-lok®
is placed more cranially at the right midclavicular Ligation clips (Teleflex Medical, NC, USA).
line. A 5-mm assisting port is placed in the left Abdominal lymphadenectomy includes lymph
18 Robotic Utilization in Esophageal Cancer Surgery 195

nodes surrounding the left gastric artery and the After 1 day in the ICU, patients are transferred to
lesser omental lymph nodes. a medium care (MC) ward.
The cervical esophagus is mobilized through a Important for postoperative care are a naso-
left-sided longitudinal neck incision along the gastric tube, feeding jejunostomy, and an epi-
sternocleidoid muscle. No formal cervical lymph dural catheter. The nasogastric tube is used for
node dissection is carried out, but cervical lymph gastric decompression and to provide a splinting
nodes are dissected if lymph node metastases are in case of anastomotic dehiscence. Fixation of
suspected macroscopically during the cervical the tube is imperative, as reintroduction can cause
phase of esophagectomy. The esophagus is dis- damage to the anastomosis.
sected and a cord is attached to the proximal part No oral intake is allowed for 5 days minimum.
of the specimen to enable pull-up of the gastric During that first week, feeding is provided by the
conduit along the anatomical tract of the feeding jejunostomy. After 5 days without any
esophagus. indication of anastomotic dehiscence, sips of water
The esophagus and surrounding lymph nodes are initiated. If there is no evidence of anastomotic
are pulled into the abdomen under laparoscopic leak, oral intake is gradually supplemented to solid
vision. A 7-cm transverse incision is made at the foods under close supervision of a clinical nutri-
level of the left paraumbilical port for extraction tionist. The feeding jejunostomy is left in situ up to
of the specimen and stomach using a wound 6 weeks after discharge from the hospital. Only
protector. after sufficient intake is maintained, the jejunos-
Outside the abdomen, a 5-cm-wide gastric tomy is removed at the outpatient clinic.
tube is constructed with staplers (GIA TM 80, Pain medication through the epidural catheter
3 · 8 mm; Covidien, Dublin, Ireland), and the is required to improve postoperative ventilation
stapled line is oversewn with 3-0 polydioxa- and coughing. Other strategies to prevent postop-
none. Routine extracorporal oversewing was erative pulmonary complications include eleva-
reintroduced as two serious complications tion of the bed by 15–30°, physical respiratory
occurred when the staple line was not oversewn therapy, and early mobilization.
[14]. The specimen consisting of the esophagus
and cardia of the stomach is sent for pathologi-
cal examination. After the gastric tube has been Results of RAMIE
pulled to the neck, a hand-sewn end-to-side
esophagogastrostomy is performed in the neck To overcome the limitations of conventional (tho-
using 3-0 polydioxanone single-layer running raco)scopic surgery, the robot-assisted minimally
sutures. Excess gastric tubing is removed using invasive thoraco-laparoscopic esophagectomy
a GIA stapler. was developed in the UMC Utrecht in 2003.
A feeding jejunostomy (Freka® FCJ-Set, From our first experience, it was concluded that
Fresenius Kabi AG, Bad Homburg vd H., RAMIE is a feasible and safe technique [5, 7]. It
Germany) is placed at the level of the transverse is associated with reduced blood loss, shorter
incision. intensive care unit stay, and a lower percentage of
cardiopulmonary complications compared to lit-
erature reports of open transthoracic esophagec-
Postoperative Care tomy. Mortality, hospital stay, and lymph node
retrieval were comparable. Short-term oncologi-
Clinical Care cal outcomes were equivalent to results from
open transthoracic surgery. Disadvantages of
Postoperatively, patients are transferred to the robot-assisted surgery compared to open surgery
intensive care unit (ICU). After leaving the oper- are a prolonged operative time, high costs associ-
ating room, mechanical ventilation is continued ated with robot acquisition and maintenance, and
briefly usually extubating later that evening. the use of disposable tools.
196 R. van Hillegersberg et al.

Following our initial report of RAMIE in esophagectomy. Results from this randomized
2009, we analyzed the following consecutive controlled trial are to be expected in 2015 [18].
series of 108 patients until 2011. We found a high
percentage (95 %) of radical resections despite
the high rate of T3 tumors (78 %) and only 64 % References
received neoadjuvant therapy. A median of 26
dissected lymph nodes was retrieved. Follow-up 1. Biere SS, van Berge Henegouwen MI, Maas KW, et al.
was at least 25 months with a median follow-up Minimally invasive versus open oesophagectomy for
of 34 months. Median disease-free survival was patients with oesophageal cancer: a multicentre, open-
label, randomised controlled trial. Lancet. 2012;379:
21 months and median overall survival was
1887–92.
29 months, with a 5-year overall survival of 2. Verhage RJ, Hazebroek EJ, Boone J, Van Hillegersberg R.
40 months. The percentage of in-hospital pulmo- Minimally invasive surgery compared to open proce-
nary infections after RAMIE in our series was dures in esophagectomy for cancer: a systematic review
of the literature. Minerva Chir. 2009;64:135–46.
34 % [15].
3. Luketich JD, Pennathur A, Awais O, et al. Outcomes
This percentage is higher than reported in the after minimally invasive esophagectomy: review of
randomized trial comparing minimally invasive over 1000 patients. Ann Surg. 2012;256:95–103.
esophagectomy (MIE) to open transthoracic 4. Ruurda JP, Draaisma WA, van Hillegersberg R, et al.
Robot-assisted endoscopic surgery: a four-year
esophagectomy. Results from this trial showed a
single-center experience. Dig Surg. 2005;22:313–20.
pulmonary complication rate in the MIE group of 5. van Hillegersberg R, Boone J, Draaisma WA, et al.
12 % [1]. However, different definitions of post- First experience with robot-assisted thoracoscopic
operative pneumonia were used. Our definition esophagolymphadenectomy for esophageal cancer.
Surg Endosc. 2006;20:1435–9.
of pneumonia was defined as the decision to treat
6. Kernstine KH. Robotics in thoracic surgery. Am J
suspected pneumonia (MCDC grade II) [16]. The Surg. 2004;188:89S–97.
definition of pneumonia used in the randomized 7. Boone J, Schipper ME, Moojen WA, et al. Robot-
controlled trial was more strict (i.e., infiltrate on assisted thoracoscopic oesophagectomy for cancer. Br
J Surg. 2009;96:878–86.
pulmonary radiography combined with a positive
8. van Hagen P, Hulshof MC, van Lanschot JJ, et al.
sputum culture) leading to a lower percentage of Preoperative chemoradiotherapy for esophageal or
pneumonia. Applying this definition on our junctional cancer. N Engl J Med. 2012;366:2074–84.
cohort yields a pneumonia rate of 18 %, which is 9. Gebski V, Burmeister B, Smithers BM, et al. Survival
benefits from neoadjuvant chemoradiotherapy or che-
comparable to MIE.
motherapy in oesophageal carcinoma: a meta-analysis.
Our results from robot-assisted esophagec- Lancet Oncol. 2007;8:226–34.
tomy are in concordance with a recently pub- 10. Block BM, Liu SS, Rowlingson AJ, et al. Efficacy of
lished systematic review [17]. This systematic postoperative epidural analgesia: a meta-analysis.
JAMA. 2003;290:2455–63.
review included nine articles (130 cases) describ-
11. Michelet P, Roch A, D’Journo XB, et al. Effect of thoracic
ing robot-assisted esophagectomy. It was con- epidural analgesia on gastric blood flow after oesophagec-
cluded that robot-assisted esophagectomy was a tomy. Acta Anaesthesiol Scand. 2007;51:587–94.
feasible and safe technique. In terms of short- 12. Sato N, Koeda K, Ikeda K, et al. Randomized study of
the benefits of preoperative corticosteroid administra-
term oncological outcomes, RAMIE was at least
tion on the postoperative morbidity and cytokine
equivalent to the open transthoracic approach for response in patients undergoing surgery for esopha-
esophageal cancer. The systematic review geal cancer. Ann Surg. 2002;236:184–90.
strongly emphasized the need for well-conducted 13. Verhage RJ, Boone J, Rijkers GT, Cromheecke GJ,
Kroese AC, Weijs TJ, Borel Rinkes IH, van
randomized controlled trials and long-term sur-
Hillegersberg R. Reduced local immune response
vival to prove the superiority of robot-assisted with CPAP during single lung ventilation for oesopha-
minimally invasive thoraco-laparoscopic esopha- gectomy. Br J Anaesth. 2014;112(5):920–8.
gectomy over open transthoracic esophagectomy. 14. Boone J, Rinkes IH, van Hillegersberg R. Gastric con-
duit staple line after esophagectomy: to oversew or
Therefore, we initiated the ROBOT trial
not? J Thorac Cardiovasc Surg. 2006;132:1491–2.
(ClinicalTrial.gov Identifier: NCT01544790) to 15. van der Sluis PCR, JP, Verhage RJJ, van der Horst S,
compare RAMIE with open transthoracic Haverkamp L, Siersema PD, Borel Rinkes IHM, ten
18 Robotic Utilization in Esophageal Cancer Surgery 197

Kate FJW, van Hillegersberg R. Robot-assisted 17. Clark J, Sodergren MH, Purkayastha S, et al. The role
minimally invasive thoraco-laparoscopic esophagec- of robotic assisted laparoscopy for oesophagogastric
tomy with two-field lymphadenectomy for esopha- oncological resection; an appraisal of the literature.
geal cancer: report of 108 consecutive procedures. Dis Esophagus. 2011;24:240–50.
2014 (submitted). 18. van der Sluis PC, Ruurda JP, van der Horst S, et al.
16. Dindo D, Demartines N, Clavien PA. Classification of Robot-assisted minimally invasive thoraco-laparoscopic
surgical complications: a new proposal with evalua- esophagectomy versus open transthoracic esophagec-
tion in a cohort of 6336 patients and results of a tomy for resectable esophageal cancer, a randomized
survey. Ann Surg. 2004;240:205–13. controlled trial (ROBOT trial). Trials. 2012;13:230.
 Minimally Invasive Intragastric
Surgery 19
Didier Mutter and Marius Nedelcu

Introduction experience allowed us to propose it systematically

as an option for the management of gastric
Laparoscopic intragastric surgery (LIGS) repre- lesions.
sents a minimally invasive technique for lesions, The aim of this chapter is to identify the indi-
which mainly exist in the gastric lumen or at the cations and to describe the technical principles of
gastroesophageal junction. In 1995, Ohashi ini- this novel technique used in our current practice.
tially described this technique to resect early gas- The objective is also to expand the surgeon’s
tric cancer, which could not be treated by armamentarium in order to safely address more
endoscopic mucosal resection (EMR) [1]. Since complex intragastric processes while offering the
then, it has evolved with respect to both techno- benefits of minimal access surgery.
logical advances (e.g., the development of cuffed
or single access ports) and tactical innovations by
many teams (Table 19.1). The good results of this Indications
approach associated with our ever-increasing
Indications for laparoscopic intragastric surgery
(LIGS) can be found for all tumors, which may be
Electronic supplementary material Supplementary resected without systematic gastrectomy or lymph
material is available in the online version of this node resection. It includes anecdotal foreign body
chapter at 10.1007/978-3-319-09342-0_19. Videos can removal [8, 9] and pancreatic pseudocyst drainage
also be accessed at http://www.springerimages.com/
videos/978-3-319-09341-3. [10]. Typical indications include the resection of
benign lesions [11–13], lesions with inconclusive
D. Mutter, MD, PhD, FACS (*)
Department of Digestive and Endocrine Surgery, pathological findings after biopsy, when malig-
IRCAD, IHU, University Hospital of Strasbourg, nancy cannot be ruled out [4, 14], and finally early
Nouvel Hôpital Civil – Pôle Hépato-Digestif, malignancy especially at the level of the esophago-
Hôpitaux Universitaires de Strasbourg, gastric region [2]. At present, these tumors can be
1, Place de l’Hôpital, Strasbourg,
Alsace 67091, France detected by different modalities (upper endoscopy,
e-mail: [email protected]; abdominal CT scan). The most frequent tumors are
[email protected] gastrointestinal stromal tumors (GISTs). These are
M. Nedelcu, MD frequently located at the esophagogastric junction,
Department of Digestive and Endocrine Surgery, not easily accessible for resection via an endo-
University Hospital of Strasbourg, scopic retroflexed view. Upper endoscopy repre-
Nouvel Hôpital Civil – Pôle Hépato-Digestif,
Hôpitaux Universitaires de Strasbourg, sents the standard tool for the diagnosis of such
Strasbourg, Alsace, France lesions, but in certain cases its therapeutic purpose

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 199
DOI 10.1007/978-3-319-09342-0_19, © Springer International Publishing Switzerland 2015
200 D. Mutter and M. Nedelcu

Table 19.1 Review of the literature

Year Authors Journal Number of cases Particularities
1995 Ohashi [1] Surgical Endoscopy 8 cases 6 early gastric cancer, 1 submucosal
leiomyoma, 1 giant polyp
2000 Hiki et al. [2] Der Chirurg 13 cases 1 case of conversion due to
intraoperative hemorrhage
2011 Sahm et al. [3] Surg Laparosc Endosc 7 cases 6 gastrointestinal stromal tumors and
Percutan Tech 1 leiomyoma
2004 Uchikoshi et al. [4] Surg Laparosc Endosc 7 cases 4 cases of gastrointestinal stromal
Percutan Tech tumors, 2 leiomyomas, and 1
schwannoma
2012 Hara et al. [5] Surg Laparosc Endosc 10 cases 1 case of conversion due to technical
Percutan Tech difficulties
2011 Shim et al. [6] J Surg Oncol 6 cases 5 leiomyomas, and one case GIST
2011 Na et al. [7] J Gastric Cancer 7 cases 5 gastrointestinal stromal tumors and
2 leiomyomas

has obvious limitations. The main advantage of Surgical Technique

laparoscopic intragastric surgery is yielded by the
direct approach to this region contrary to the retro- The anatomical localization of the tumor is the
flexed approach provided by endoscopy. Its limita- most important factor to determine the ideal
tion is not only due to the visualization of the lesion resection technique. When facing a tumor of the
but also to the main shortcoming of the indirect anterior gastric wall, the tumor is easy to visual-
endoscopic approach which is the inability to offer ize, and a tangential wedge resection through
sufficient strength and precision to control dissec- conventional transperitoneal laparoscopy is the
tion in a safe manner. Laparoscopic intragastric method of choice [3]. Sometimes, the location of
surgery completely overcomes such limitations the tumor may be confirmed by simultaneous
and also offers an appropriate dissection angle endoscopic exploration. If the tumor is located on
using the basic “triangulation” principle of general the posterior gastric wall, proximal to the cardia
laparoscopy. or pylorus, a conventional wedge resection can-
not be performed with appropriate margins. In
this respect, intragastric surgery could well repre-
Preoperative Workup sent a valid option.
Different surgical techniques have been
For preoperative diagnosis and surgical planning, described for laparoscopic intragastric surgery.
preoperative upper endoscopy is a key step to Our standard approach for a laparoscopic intra-
ascertain the precise localization of the tumor. It gastric surgery is represented by a multiple intra-
is needed to define the anatomical landmarks of gastric port approach, as described in Video 19.1.
the lesions regarding the gastric curvatures, dis- A pneumoperitoneum is briefly established using
tance to the cardia, pylorus, and main vessels. an open access at the umbilical level. It allows to
The lesions are also examined by CT scan and explore the abdomen and determine the ideal
endoscopic ultrasound (US) to determine the position of transgastric laparoscopic ports in rela-
depth of invasion of the gastric wall. Endoscopic tion to the anatomy of the stomach. Endoscopy
US is a major tool to identify contraindications aiming to localize the actual position of the tumor
represented by transmural tumors or local lymph may be completed at this moment. The tumor can
node metastases. Finally, such data should be be located by a mark or a suture on the gastric
confirmed intraoperatively by the excellent vision wall. The location of the stomach wall incision is
provided by the laparoscopic exploration as well then identified. Two transparietal stitches are
as by the endogastric approach. placed adjacent to this area and will be used to lift
19 Minimally Invasive Intragastric Surgery 201

Fig. 19.1 Gastric exposure Fig. 19.4 Tumor resection

Fig. 19.5 Final stapling with suture assistance

Fig. 19.2 Intragastric trocar insertion

high rate of insufflation of the stomach, as the

anterior wall is lifted up by traction applied to the
abdominal wall by means of stitches. Usually, no
distal or proximal balloon blockage is required
due to lower esophageal sphincter and pyloric
resting tone. In order to achieve adequate access,
multiple ports should be positioned according to
general laparoscopic principles in order to
achieve maximum triangulation for the dissec-
tion site. The cardioesophageal junction is a dif-
ficult location, which requires optimal
Fig. 19.3 Intragastric identification of the tumor visualization and triangulation of instruments for
safe surgical maneuvers.
up the stomach and fix it to the parietal wall Resection and suture can be performed as a
(Fig. 19.1). A 12 mm cuffed port is inserted into standard procedure, but most of the time, the use
the stomach under laparoscopic guidance of stapling is preferred for many reasons, includ-
(Fig. 19.2). Two additional 5 mm ports are ing speed, safety, and reliability as illustrated by
inserted and positioned to ensure triangulation in this case (Figs. 19.4 and 19.5). It only requires
relation to the tumor’s position. The two cuffed the replacement of the 5 mm port by a 12 mm
5 mm ports are also introduced into the stomach one. In well-selected cases (e.g., pedunculated
under direct control after partial insufflation. The tumors), the advantage of this technique is to
peritoneal cavity is desufflated, and the stomach obtain resection and hemostasis simultaneously,
is explored (Fig. 19.3). There is no need for a using the same instrument. However, achieving
202 D. Mutter and M. Nedelcu

Discussion

Increased screening of the upper gastrointestinal

tract has led to the discovery of a greater number
of intragastric lesions. Despite the frequent
benign nature of such lesions, complete tumor
removal for pathological examination is recom-
mended in order to rule out any underlying
malignancy. Whenever endoscopic resection is
not feasible, the conventional transperitoneal lap-
Fig. 19.6 Gastrotomy closure
aroscopic approach represents the next least
invasive approach.
Ohashi described laparoscopic “intragastric” or
Table 19.2 Key steps of intragastric surgery “intraluminal” surgery in eight patients: six with
1. Gastric exposure early gastric cancer, one with submucosal leiomy-
2. Intragastric trocar insertion oma, and one with a giant gastric polyp [1]. The
3. Intragastric identification of the tumor current literature on intragastric multiport surgery
4. Tumor resection by transgastric stapling focuses on lesions of the gastroesophageal junc-
5. Final stapling with suture assistance tion which have heterogeneous origins (leiomy-
6. Gastrotomy closure oma, GIST, T1a gastric cancer, etc.). In this
technically challenging location, the multiport
approach can be extremely useful, offering the
adequate margins can be difficult, and the risk of advantage of improved triangulation of instru-
tumor rupture might be increased, particularly in ments in order to facilitate the dissection of the
case of gastrointestinal stromal tumors. In such submucosal layer and suturing of the mucosa.
cases, tricks including traction on the parietal Presently, most manuscripts on intragastric resec-
gastric wall assist in achieving a full-thickness tion describe the placement of several ports into
resection of the stomach wall, preserving safety the gastric lumen [6, 17–20]. The particular com-
margins in case of malignant lesions. The port bined laparoscopic and endoscopic approach with
should be positioned so that stapling can be eas- one intragastric port has been described by Lippert
ily accomplished in the narrow space of the insuf- et al. [3]. Gastroscopy helped to intragastrically
flated stomach, and the use of roticulating staplers localize, visualize, and mobilize the tumor with a
is mandatory. Figure 19.6 depicts the closed polypectomy snare. Resection can be performed
gastrotomy. by means of stapling under laparoscopic control.
In some cases, when tumors are located on the In 2 out of 7 patients, an additional 5 mm port was
posterior and mobile part of the greater curva- used to remove the tumor. Resection can be per-
ture, or if they have a long pedicle, they can be formed oncologically; however, full parietal wall
everted through the gastric incision and presented resection might allow safe margins for large-based
to the peritoneal cavity. This approach has first tumors. In such cases, this approach could well
been described by Morinaga et al. [15] for a represent an alternative method of treatment as
tumor located near the gastroesophageal junc- shown by Pfau et al. [13]. They have described the
tion. Ma et al. [16] have reported a series of 56 successful resection of a giant pedunculated mid-
cases of gastric GIST in which 19 patients under- esophageal lipoma using a laparoscopic stapler
went laparoscopic transgastric tumor-everting through one of the two trocars placed transabdom-
resections. They have even extended the indica- inally and intragastrically. The tumor (3.5 cm in
tions of this technique to posterior wall tumors diameter) was retrieved via an Endopouch® speci-
near the greater curvature in 5 cases. The key men retrieval bag and extracted through the intra-
steps of the procedure are described in Table 19.2. gastric laparoscopic port.
19 Minimally Invasive Intragastric Surgery 203

The transgastric route also allows the of this new minimally invasive approach. This
performance of submucosal resections. A poten- confirms that laparoscopic intragastric surgery
tial perforation will be easily controlled using a offers the greatest advantage over a conventional
laparoscopic approach at the end of the proce- resection for lesions at the gastroesophageal
dure. The need to approximate the mucosa to junction, as gastric resections in this area usually
facilitate the healing process remains debatable. necessitate resection of the gastroesophageal
Closure of the mucosal defect might promote junction. The size of the tumor is not a limitation
rapid healing as demonstrated by Yumiba et al. per se. Laparoscopic intragastric surgery can be
and could prevent an esophagogastric junction applied to large tumors and to those located near
stricture [21]. Other authors such as Uchikoshi the cardia and pylorus as well as on the posterior
et al. [4] do not usually close mucosal defects wall of the stomach, where a conservative laparo-
with sutures unless uncontrollable bleeding is scopic wedge resection is frequently not feasible.
encountered. We prefer the use of full-thickness However, LIGS has no application for tumors on
resections made possible by relying heavily on the anterior wall or showing extragastric growth
laparoscopic staplers. as they can easily be resected by wedge resection,
Malignancy is of critical importance when it using the principles of triangular stapling.
comes to this approach. It can sometimes be dif- After resection, specimen removal can be
ficult to preoperatively determine whether tumors achieved through different ways. It has to be
are benign or malignant, even from intraoperative placed into a bag and should be removed through
frozen sections [22]. Major surgical resections the mouth or should be placed into the abdominal
would be excessive for a benign tumor. Llorente cavity in order to be taken out in the same way as
reported a case of gastric leiomyoma subjected to any laparoscopic surgical specimen. Small-sized
laparoscopic gastric resection [23]. Consequently, specimens can even be extracted through a
enucleation or ideally atypical partial resection 12 mm port.
must be considered for these patients. If the final The perioperative management of these
pathology modifies the initial diagnosis and patients is simple and can follow the principles of
reveals a malignant lesion, a second-look opera- early recovery after surgery (ERAS) or the prin-
tion is necessary [24]. ciples used in bariatric surgery. No drain is
An experience of 27 cases with 3 surgical inserted into the abdominal cavity, and the
approaches (open laparotomy, laparoscopic par- nasogastric tube is removed at the end of the sur-
tial gastrectomy, and laparoscopic intragastric gical procedure. Patients are administered proton
surgery) was reported by Hara et al. [5]. In this pump inhibitors (PPIs) intravenously or orally for
retrospective review, all gastric submucosal 7 days. Liquid intake is allowed the day after sur-
tumors were adjacent to the esophagogastric gery, and patients have their first meal on postop-
junction. Globally, intragastric surgery was pre- erative day 2. Patients receive a single shot of
ferred for lesions with intragastric growth, a stan- perioperative antibiotic prophylaxis. Total hospi-
dard laparoscopic approach was used in cases tal stay lasts between 2 and 5 days depending on
with transgastric or exogastric growth, and the the type of resection as well as on potential
open approach was selected for bulky lesions. resection-related hazards.
Their completion rates were 50 % in the laparo-
scopic group and 90 % in the transgastric group, Conclusions
respectively. Additionally, the overall rate of car- Laparoscopic intragastric surgery offers and
dia preservation was 80 % in the laparoscopic can enhance the typical benefits of laparoscopic
group, 100 % in the transgastric group, and 29 % surgery such as reduced pain, faster recovery,
in the open group, respectively. Although selec- and shorter length of hospital stay. The intra-
tion of the surgical approach based on the sur- gastric approach has reached a wide acceptance
geon’s choice represents the major bias of this from advanced laparoscopic teams and should
manuscript, it demonstrates a significant benefit be systematically proposed as an alternative to
204 D. Mutter and M. Nedelcu

other new minimally invasive approaches, extraluminal approach. Hepatogastroenterology. 1998;

including single-port surgery and natural ori- 45:2215–8.
11. Geis WP, Baxt R, Kim HC. Benign gastric tumors.
fice transluminal endoscopic surgery (NOTES) Minimally invasive approach. Surg Endosc. 1996;10:
[25]. Based on our experience, intragastric sur- 407–10.
gery is ideally suited to address lesions located 12. Watson DI, Game PA, Devitt PG. Laparoscopic resec-
at the posterior or superior aspect of the stom- tion of benign tumors of the posterior gastric wall.
Surg Endosc. 1996;10:540–1.
ach. Such lesions are frequently difficult or 13. Weigel TL, Schwartz DC, Gould JC, Pfau PR.
impossible to access via an endoscopic Transgastric laparoscopic resection of a giant esopha-
approach and would therefore require resection geal lipoma. Surg Laparosc Endosc Percutan Tech.
of the esophagogastric junction if addressed via 2005;15:160–2.
14. Li VK, Hung WK, Chung CK, et al. Laparoscopic
a conventional laparoscopic or open approach. intragastric approach for stromal tumours located at the
An adequate preoperative workup to precisely posterior gastric wall. Asian J Surg. 2008;31:6–10.
determine the optimal indication and strategy, 15. Morinaga N, Sano A, Katayama K. Laparoscopic
including adequate position of intragastric transgastric tumor-everting resection of the gastric sub-
mucosal tumor located near the esophagogastric junc-
ports, is crucial. tion. Surg Laparosc Endosc Percutan Tech. 2004;14:
344–8.
16. Ma JJ, Hu WG, Zang L, et al. Laparoscopic gastric
resection approaches for gastrointestinal stromal
References tumors of stomach. Surg Laparosc Endosc Percutan
Tech. 2011;21(2):101–5.
1. Ohashi S. Laparoscopic intraluminal (intragastric) 17. Ghushe ND, Dulai PS, Trus TL. Laparoendoscopic
surgery for early gastric cancer. A new concept in transgastric resection of a submucosal mass at the
laparoscopic surgery. Surg Endosc. 1995;9:169–71. gastroesophageal junction. J Gastrointest Surg. 2012;
2. Hiki Y, Sakuramoto S, Katada N, et al. Combined 16:2321.
laparoscopic- endoscopic procedure in stomach carci- 18. Heniford BT, Arca MJ, Walsh RM. The mini-
noma. Chirurg. 2000;71:1193–201. laparoscopic intragastric resection of a gastroesopha-
3. Sahm M, Pross M, Lippert H. Intraluminal resection of geal stromal tumor: a novel approach. Surg Laparosc
gastric tumors using intragastric trocar technique. Surg Endosc Percutan Tech. 2000;10:82–5.
Laparosc Endosc Percutan Tech. 2011;21:169–72. 19. Choi YB, Oh ST. Laparoscopy in the management of
4. Uchikoshi F, Ito T, Nishida T, Kitagawa T, Endo S, gastric submucosal tumors. Surg Endosc. 2000;14:
Matsuda H. Laparoscopic intragastric resection of gastric 741–5.
stromal tumor located at the esophago-cardiac junction. 20. Ridwelski K, Pross M, Schubert S, et al. Combined
Surg Laparosc Endosc Percutan Tech. 2004;14:1–4. endoscopic intragastral resection of a posterior stro-
5. Hara J, Nakajima K, Takahashi T, et al. Laparoscopic mal gastric tumor using an original technique. Surg
intragastric surgery revisited: its role for submucosal Endosc. 2002;16:537.
tumors adjacent to the esophagogastric junction. Surg 21. Yumiba T, Ito T, Ikushima H, et al. Effect of mucosal
Laparosc Endosc Percutan Tech. 2012;22:251–4. suture on the healing of mucosal defect in laparoscopic
6. Shim JH, Lee HH, Yoo HM. Intragastric approach for intragastric surgery. Gastric Cancer. 2003;6:96–9.
submucosal tumors located near the Zline: a hybrid 22. Schiu M, Farr G, Papachristou DN, Hajdu S.
laparoscopic and endoscopic technique. J Surg Oncol. Myosarcoma of the stomach: natural history, prog-
2011;104:312–5. nostic factors and management. Cancer. 1982;49:
7. Na JU, Lee SI, Noh SM. The single incision laparo- 177–87.
scopic intragastric wedge resection of gastric submu- 23. Llorente J. Laparoscopic gastric resection for gastric
cosal tumor. J Gastric Cancer. 2011;11:225–9. leiomyoma. Surg Endosc. 1994;8:887–9.
8. El-Hayek K, Timratana P, Brethauer SA, Chand B. 24. Taniguchi E, Kamiike W, Yamanishi H, Ito T, Nezu R,
Complete endoscopic/transgastric retrieval of eroded Nishida T, Momiyama T, Ohashi S, Okada T, Matsuda
gastric band: description of a novel technique and review H. Laparoscopic intragastric surgery for gastric leio-
of the literature. Surg Endosc. 2013;27(8):2974–9. myoma. Surg Endosc. 1997;11(3):287–9.
9. Son T, Inaba K, Woo Y, Pak KH, Hyung WJ, Noh 25. Asakuma M, Perretta S, Allemann P, Cahill R, Con SA,
SH. New surgical approach for gastric bezoar: “hybrid Solano C, Pasupathy S, Mutter D, Dallemagne B,
access surgery” combined intragastric and single port Marescaux J. Challenges and lessons learned from
surgery. J Gastric Cancer. 2011;11:230–3. NOTES cholecystectomy initial experience: a stepwise
10. Holeczy P, Danis J. Laparoscopic transgastric pancre- approach from the laboratory to clinical application.
atic pseudocystogastrostomy — first experience with J Hepatobiliary Pancreat Surg. 2009;16(3):249–54.
 Laparoscopic Partial Gastrectomy
20
Georgios Rossidis

Introduction between microscopically negative and the

microscopically positive resections in tumors with
Laparoscopic partial gastrectomy, also called macroscopically negative margins [4]. A partial
wedge gastrectomy, refers to resection of part of gastrectomy without gastrojejunostomy confers
the stomach without the subsequent need for a gas- the same progression-free survival as a more for-
trojejunostomy, in a laparoscopic fashion. It is an mal gastrectomy, with the added benefit of far
approach that has gained popularity for resection lower postoperative morbidity and mortality.
of lesions that are either benign (ulcers, polyps, Initially, the laparoscopic approach was reserved
cysts, leiomyomas, heterotopic pancreas) or for smaller tumors, in order to minimize the risk of
malignant (gastrointestinal stromal tumors). The tumor spillage and peritoneal seeding. The GIST
surgical approach to all submucosal stromal Consensus Conference recommended that laparo-
tumors is similar, and we will attempt to describe scopic resection for gastric GISTs should be lim-
the laparoscopic approach to the resection of these ited to tumors smaller than 2 cm [5]. More recent
lesions and the challenges associated with the size studies challenge this concept and show that lapa-
and anatomic location. The chapter will describe roscopic approaches can provide comparable
the surgical management of GIST tumors, the oncologic outcomes and better postoperative
most common submucosal lesion of the stomach. recovery, regardless of tumor size or location,
Since the first description of a laparoscopic when compared to open resections, even at sizes
resection of a submucosal tumor by Lukaszczyk bigger than 5 cm [6]. The size of the tumor is not
and Preletz in 1992 [1], advances in laparoscopic the only challenge to a successful laparoscopic
techniques, instruments, and stapling devices have resection. The location of the tumor can also be a
made the laparoscopic approach to a wedge resec- formidable challenge to a laparoscopic approach
tion safe, with excellent outcomes [2, 3], and is especially for lesions located on the lesser curve
now the accepted treatment. The extremely rare and very close to the gastroesophageal junction
lymph node metastasis and the need for only a (GEJ) or in the prepyloric antrum. Many authors
grossly negative margin make the laparoscopic have described these challenges and approaches to
approach even more attractive. A study by resections of these challenging tumors. Privette
DeMatteo et al. showed no survival advantage et al. proposed a classification scheme (Fig. 20.1),
dividing the tumors into three types based on the
anatomic location and offered distinct approaches
G. Rossidis, MD
for each type [7]. Song et al. published successful
Department of Surgery, University of Florida,
100286, Gainesville, FL 32610, USA outcomes of ten patients with lesions very near the
e-mail: [email protected] GEJ proving that a tailored laparoscopic approach

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 205
DOI 10.1007/978-3-319-09342-0_20, © Springer International Publishing Switzerland 2015
206 G. Rossidis

for lesions near the GEJ can be feasible and safe lesion requiring gastric resection. The symptoms
[8]. Japanese authors have published a tailored are usually related to the size and location of the
approach to resection of GISTs based on the tumor. Larger lesions usually present as a palpa-
lesion’s size, anatomic location, and growth ble tumor with symptoms of pressure and abdom-
(Fig. 20.2) [9]. The different surgical approaches inal pain. Smaller lesions may present with acute
will be described in detail. upper GI blood loss or anemia and fatigue.
Dysphagia may be the main symptom in lesions
occurring at the gastroesophageal junction or at
Symptomatology and Diagnosis the pylorus. Many GISTs are asymptomatic and
may be diagnosed during upper endoscopy for the
While GISTs are rare with an annual incidence in workup of other conditions. Computed tomogra-
the United States of 1,000–2,500 cases per year, phy and upper endoscopy are diagnostic for
they are still the most common non-adenomatous GISTs. The classic findings are a submucosal
mass with smooth borders or a rounded appear-
Type I ance or an exophytic lesion. On endoscopy GISTs
Fundus and are firm, smooth, distinct, rounded, or lobulated
greater curvature
submucosal lesions. The above findings are so
characteristic that they exclude the need for a nee-
dle biopsy. Percutaneous biopsy is contraindi-
cated also because of the risk of tumor spillage. In
Type III
Lesser curvature the case of large lesions in need for neoadjuvant
and near GE junction therapy or with associated liver lesions suggestive
of metastatic disease, endoscopic ultrasound
(EUS)-guided needle biopsy may be warranted.

Preoperative Planning
Type II The patient’s overall health status and medical
Prepyloric and antrum
conditions should be assessed, and cardiopulmo-
Fig. 20.1 Privette’s anatomic classification of gastric
nary comorbidities should be evaluated as for any
lesions and the distinct surgical approach to them other major abdominal procedure.

GIST

Tumor size: <2 cm 2−5 cm >5 cm

Follw-up
(every 6 months)
Near EGJ Anterior gastric wall Near pylorus
Tumor location:
Posterior gastric wall Greater and lesser curvature

Tumor growth: Endoluminal Other type All type All type

Approach: Gastrotomy Exogastric Exogastric Laparoscopy –assisted

(SILAS) (manual resection) or open

Fig. 20.2 Therapeutic strategy for suspected gastric GISTs (From Sasaki et al. [9], with permission)
20 Laparoscopic Partial Gastrectomy 207

Fig. 20.3 Patient positioning Anesthesia

and position of primary and
assistant surgeon

Monitor Monitor

Primary surgeon Assistant surgeon

Previous abdominal procedures and operations and the thighs and legs are strapped so as to
should be noted, as intra-abdominal adhesions support the patient during steep reverse
may make a laparoscopic approach far more Trendelenburg position (Fig. 20.4). A Foley cath-
challenging. eter is inserted for precise urine output measure-
All the pertinent imaging and workup must be ments, and an orogastric tube is inserted to
reviewed, and after a detailed discussion of all decompress the stomach.
benefits, risks, and alternatives, an informed con- The different approaches shall be described
sent should be obtained. based on the anatomic location of the lesion.

Surgical Technique Fundus and Greater Curve

The patient is placed in a supine position with The trocar placement for lesions of the greater
both arms extended. The primary surgeon is posi- and lesser curve of the stomach is shown in
tioned on the right side of the patient and the Fig. 20.5. Access to the peritoneal cavity is
assistant surgery on the left side of the patient. obtained via the left subcostal incision with the
Monitors are placed over the patient’s shoulders use of an optical port under direct vision, and
bilaterally (Fig. 20.3). As with all foregut proce- 15 mmHg of carbon dioxide is required to achieve
dures, a footboard is placed at the patient’s feet, pneumoperitoneum. The other ports are then
208 G. Rossidis

Fig. 20.4 A footboard and 2

straps support the patient
during steep reverse
Trendelenburg

Intraoperative esophagogastroscopy is employed

to identify the tumor and its exact location (espe-
cially for endophytic lesions) and also to ensure
adequate margins. The short gastric vessels are
divided with the use of a bipolar energy device
with the surgeon retracting the stomach medially
and the assistant retracting the omentum and the
gastrosplenic ligament laterally. For an anterior
wall lesion, the next step is to elevate the anterior
Liver retractor wall with atraumatic graspers, and an endoscopic
GIA stapler is passed under the tumor incorporat-
5 mm 5 mm ing an adequate margin of normal gastric tissue
port port to ensure negative margins (Fig. 20.6). The lesion
is placed in a laparoscopic extraction bag. A non-
12 mm 5 mm
port port
touch lesion lifting method is described by
Kiyozaki et al., where traction sutures are placed
on the gastric wall over normal stomach 2 cm
away from the lesion and are pulled out through
Fig. 20.5 Trocar placement for extragastric resection of the abdominal wall. Thus the tumor is lifted, and
lesions. The surgeon stands on the right side of the table the GIA stapler is passed under the tumor to
and the assistant on the left
excise it [10]. This method allows the excision of
the tumor with decreased risk of tumor spillage
placed. The camera is inserted in the 5 mm port and also allows the resection the posterior gastric
in the left upper quadrant. The peritoneal cavity wall lesions. With adequate mobilization of the
is inspected, and the patient is placed in steep stomach, rotation of the stomach allows a poste-
reverse Trendelenburg position to expose the rior gastric wall lesion to face anteriorly and
stomach and the hiatus. A liver retractor is intro- therefore to be removed as an anterior wall lesion.
duced in to the peritoneal cavity to elevate the left Larger lesions that are endophytic, requiring
lobe of the liver to provide better visualization. more extensive resection, require the placement
20 Laparoscopic Partial Gastrectomy 209

Lesion

Endo GIA stapler

Fig. 20.7 A lesion located in close proximity to the pylo-

Fig. 20.6 Excision of greater curve lesion using an endo- rus. T tumor, P pylorus
scopic GIA Stapler. Normal gastric tissue is incorporated
to ensure negative margins

of a bougie (40 Fr) to ensure luminal patency of

the stomach post resection of the lesion.

Antrum/Prepyloric Region

While tumors in the distal stomach or prepyloric

region can be excised with the method described
above, tumors adjacent to the pylorus are more
challenging, due to the difficulty in achieving
negative margins without compromising the
patency of the pylorus (Fig. 20.7). Posterior
lesions limited to the mucosa or submucosa can
be excised via an anterior gastrotomy. The access
to the peritoneal cavity and port placement is as Fig. 20.8 Resection of prepyloric lesion. An anterior
described above. Upper endoscopy can localize gastrotomy is created. The lesion can be seen in the gas-
tric lumen
the lesion and assist with the location of the
anterior gastrotomy. A horizontal anterior gas-
trotomy is performed with the use of electrocau- sutures are placed on the anterior gastric wall,
tery or ultrasonic shears and must be made no and a GIA stapler is passed below them to staple
closer than 3–4 cm from the pylorus (Fig. 20.8). the anterior defect (Fig. 20.10). The already
Traction sutures are placed proximal and distal to placed endoscope is then utilized to assess the
the mass, and the lesion is pulled out through the prepyloric area for bleeding, ensure the luminal
gastrotomy into the peritoneal cavity. The lesion patency, and rule out a staple line leak. Larger
is then removed with an endoscopic GIA stapler lesions or lesions that involve the pylorus may
(Fig. 20.9). The horizontal incision is closed in a not be amenable to wedge gastrectomy, and a
vertical fashion in order to not compromise the resection with reconstruction may be required to
luminal diameter of the distal stomach. Traction avoid stenosis of the gastric outlet.
210 G. Rossidis

a b

Fig. 20.9 (a, b) The lesion is pulled out through the gastrotomy into the peritoneal cavity and is then divided with an
endoscopic GIA stapler

divided with bipolar energy device, and branches

of the left gastric artery and coronary vein are
divided as well. Stay sutures placed proximal and
distal to the lesion can lift the lesion, and an
endoscopic GIA stapler is passed below the mass
and the lesion is stapled off. The lesion is
extracted in an extraction bag. Lesions near the
GE junction can be managed in fashion similar to
lesions near the pylorus, through a gastrotomy
and subsequent resection [11]. Placement of a 40
Fr bougie during resection and closure will pre-
vent narrowing of the gastric inlet. For lesions
that are situated at the GE junction, laparoscopic
resection becomes very challenging. Lesions that
do not invade deep to the submucosa can be
excised via enucleation through a combination of
an endoscopic and transgastric laparoscopic
approach. The lesion is first identified via upper
endoscopy, and it is raised via endoscopic injec-
Fig. 20.10 The anterior gastrotomy is approximated with tion of dilute epinephrine. The stomach is then
full-thickness sutures and is stapled off with the use of an distended, and laparoscopic ports are inserted
endoscopic GIA staple into the gastric lumen. A balloon tipped trocar is
used to attach the stomach to the abdominal wall
Lesser Curve/Gastroesophageal and allow the placement of the other trocars.
Junction With an angled laparoscope and electrocautery,
the lesion is enucleated (Fig. 20.11) and retrieved
Lesser curvature lesions can be excised with the via the mouth with the use of the endoscope. The
same method as described above for greater mucosal defect is reapproximated with absorb-
curve lesions. The hepatogastric ligament is able suture. The endoscope then confirms the
20 Laparoscopic Partial Gastrectomy 211

negative margins on all of their patients, and a

long-term disease-free survival of 92 % [2].
Matsuhashi et al. published their experience and
Electrocautory a literature review confirming the safety and
reproducibility of laparoscopic wedge gastrec-
Camera tomy with sound oncologic outcomes [12]. A
Laparascopic grasper
recent meta-analysis by Ohtani et al. including
644 patients showed that laparoscopic surgery for
gastric GIST was associated with a reduction in
intraoperative blood loss, shorter period to flatus,
earlier resumption of oral intake, and shorter
duration of hospital stay over the short term and
with a significantly lower rate of overall recur-
rence, metastatic recurrence, and local recurrence
in the long term compared to open surgery [13].
Fig. 20.11 Lesions located at the GE junction can be
excised via enucleation through transgastric ports and can Conclusion
be retrieved transorally with the use of an endoscope The recent technological strides in both lapa-
roscopy and endoscopy allow surgeons to pro-
patency of the gastroesophageal junction, hemo- vide a minimally invasive approach for
stasis and rules out a leak at the area of palliation and curative resection of gastric
resection. lesions. Advanced preoperative or intraopera-
tive endoscopy allows the localization and
characterization of gastric lesions at different
Postoperative Care locations. With the use of endoscopic, laparo-
scopic, and intragastric approaches, we can
Patients recover on the surgical ward unless provide a minimally invasive approach for
comorbidities dictate a stay in an intermediate care almost all gastric lesions amenable to a wedge
or intensive care unit for better monitoring. The resection. In combination with molecular tar-
Foley catheter is removed, and the patient takes geted adjuvant therapy in the form of Imatinib,
sips of water on postoperative day 1. For more laparoscopic partial gastrectomy for GISTs
complex resections at difficult anatomic locations can provide excellent outcomes and long-term
or in the presence of larger lesions where extensive survival.
resection is required, an upper GI study to rule out
a leak and to confirm luminal patency is per-
formed. The patient is discharged on postoperative
day 3 on full liquid diet, and the diet is advanced to
References
a pure diet, to a soft diet, and ultimately to a regu-
lar diet at 2 weeks increments. 1. Lukaszczyk JJ, Preletz RJ. Laparoscopic resection of
benign stromal tumor of the stomach. J Laparoendosc
Surg. 1992;2:331–4.
2. Novitsky YW, Kercher KW, Sing RF, Heniford
Outcomes BT. Long-term outcomes of laparoscopic resection of
gastric gastrointestinal stromal tumors. Ann Surg.
There is evidence to support that laparoscopic 2006;243:738–45; discussion 745–7.
wedge gastrectomy can be performed safely and 3. Matthews BD, Walsh RM, Kercher KW, et al.
Laparoscopic vs open resection of gastric stromal
is reproducible. Novitsky et al. have published
tumors. Surg Endosc. 2002;16:803–7.
their outcomes showing no major postoperative 4. DeMatteo RP, Lewis JJ, Leung D, Mudan SS, et al.
complications or mortality, short hospitalization, Two hundred gastrointestinal stromal tumors:
212 G. Rossidis

recurrence patterns and prognostic factors for survival. 9. Sasaki A, Koeda K, Obuchi T, Nakajima J, et al. Tailored
Ann Surg. 2000;231:51–8. laparoscopic resection for suspected gastric gastrointes-
5. Demetri GD, van Oosterom A, van Glabbeke M, et al. tinal stromal tumors. Surgery. 2010;147(4):516–20.
Consensus meeting for the management of gastroin- 10. Kiyozaki H, Saito M, Chiba H, Takata O, et al.
testinal stromal tumors. Report of the GIST Consensus Laparoscopic wedge resection of the stomach for gas-
Conference of 20-21 March 2004, under the auspices trointestinal stromal tumor (GIST): non-touch lesion
of ESMO. Ann Oncol. 2005;16:566–78. lifting method. Gastric Cancer. 2014;17:337–40.
6. Lee HH, Hur H, Jung H, Park CH, et al. Laparoscopic 11. Sakamoto Y, Sakaguchi Y, Akimoto H, Chinen Y,
wedge resection for gastric submucosal tumors: a et al. Safe laparoscopic resection of a gastric gastroin-
size-location matched case–control study. J Am Coll testinal stromal tumor close to the esophagogastric
Surg. 2011;212(2):195–9. junction. Surg Today. 2012;42(7):708–11.
7. Privette A, McCahill L, Borrazzo E, et al. Laparoscopic 12. Matsuhashi N, Osada S, Yamaguchi K, Okumura N,
approaches to resection of suspected gastric gastroin- et al. Long-term outcomes of treatment of gastric gas-
testinal stromal tumors based on tumor location. Surg trointestinal stromal tumor by laparoscopic surgery.
Endosc. 2008;22:487–94. Hepatogastroenterology. 2013 [Epub ahead of print].
8. Song KY, Kim SN, Park CH. Tailored-approach of 13. Ohtani H, Maeda K, Noda E, Nagahara H, et al. Meta-
laparoscopic wedge resection for treatment of submu- analysis of laparoscopic and open surgery for gastric
cosal tumor near the esophagogastric junction. Surg gastrointestinal stromal tumor. Anticancer Res. 2013;
Endosc. 2007;21:2272–6. 33(11):5031–41.
 Principles and Practice
of Laparoscopic Gastrectomy 21
with Gastroduodenostomy
(Billroth I)

Sang-Hoon Ahn and Hyung-Ho Kim

Introduction of food, and avoidance of gastrojejunostomy-related

postgastrectomy syndrome (e.g., the afferent loop
Since Kitano first performed laparoscopic distal syndrome). According to the Korean Laparoscopic
gastrectomy (LDG) for early gastric cancer in Gastrointestinal Surgery Study Group (KLASS)
1991, it has become a popular procedure for gas- survey, 63.4 % of all distal gastrectomies performed
tric cancer resection. However, LDG is a com- in 2009 were Billroth I reconstructions [3].
plex, technically demanding procedure, and the Using LDG, Billroth I gastroduodenostomy
learning curve for its use on early gastric cancer can be performed extracorporeally or intracor-
is thought to require experiences with more than poreally. Recently, LDG with intracorporeal
40–50 cases [1, 2]. In addition to an experience, a anastomosis, such as intracorporeal Billroth I
surgeon would require an excellent surgical team, anastomosis (delta-shaped anastomosis) and
proper equipment, and a good facility to obtain a intracorporeal uncut Roux-en-Y gastrojejunos-
consistently good result from LDG. tomy, has become a popular procedure [4, 5].
Because it restores normal bowel continuity, We call this procedure a totally laparoscopic
Billroth I gastroduodenostomy is the most physio- distal gastrectomy (TLDG). In this chapter, the
logic type of gastric resection. It is one of the most current techniques of LDG with Billroth I anas-
common types of reconstruction after distal gastrec- tomosis, including the details and advantages
tomy. The advantages of Billroth I over Billroth II and disadvantages of each technique, are
or Roux-en-Y gastrojejunostomy are the short sur- discussed.
gical time, preservation of the physiologic passage

Indications
Electronic supplementary material Supplementary
material is available in the online version of this LDG is usually indicated for gastric cancer,
chapter at 10.1007/978-3-319-09342-0_21. Videos can which requires lymphadenectomy, and for peptic
also be accessed at http://www.springerimages.com/ ulcer disease and submucosal gastric tumors,
videos/978-3-319-09341-3.
which do not require lymphadenectomy. Billroth
S.-H. Ahn, MD • H.-H. Kim, MD, PhD (*) I anastomosis is usually performed for gastric
Department of Surgery, Seoul National University cancer with negative proximal margins in the dis-
Bundang Hospital,
tal third of the stomach, indicating complete
300 Gumi-dong, Bundang-gu, Seongnam, Gyeonggi
463-708, Republic of Korea tumor excision. It is also indicated for type I gas-
e-mail: [email protected]; [email protected] tric ulcer.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 213
DOI 10.1007/978-3-319-09342-0_21, © Springer International Publishing Switzerland 2015
214 S.-H. Ahn and H.-H. Kim

Contraindication – An HD camera is mandatory in laparoscopic

gastrectomy, and a flexible HD scope is pre-
Contraindications include severe cardiopulmo- ferred because it can visualize the entire
nary disease, hemodynamically unstable patients, intra-abdominal space, especially during
and detection of advanced gastric cancer during suprapancreatic lymph node dissection.
the preoperative workup. Relative contraindica- • 5–12 mm trocars
tions include duodenal ulcer, duodenal ulcer scar, • Video system
and tumor invasion into the pylorus. • Energy device (e.g., harmonic scalpel and
LigaSure)
• Hemoclips
Preoperative Preparation • Linear (45 and 60 mm) or circular staplers (29
or 31 mm)
In general, preoperative nasogastric tube inser- • The skin is prepared in a routine manner.
tion and preoperative bowel preparation are not
mandatory.
Incision and Exposure

Anesthesia and Antibiotic Coverage To introduce the first trocar, Hasson’s open
technique, which involves direct open visual-
General anesthesia with endotracheal intubation ization of the tissues, is the safest. With a No.
and muscle relaxants are usually used. Spinal or 11 blade, the infraumbilical incision, including
epidural anesthesia can be used; however, supple- the half below the umbilicus, is created. The
mentation with intravenous sedative may be indi- subcutaneous fat tissue is then dissected with a
cated to prevent nausea during bowel manipulation. mosquito clamp. The rectus fascia is clamped
A first-generation cephalosporin is administered with mosquito clamps and lifted. The fascia is
for 24 h as prophylactic antibiotic coverage. divided using an electrocautery without expos-
ing the rectus muscle because the incision is
close to the umbilicus. Finally, a Kelly clamp is
Position gently introduced along the anterior peritoneum
until some resistance is felt, but the peritoneum
The patient is placed in the supine position with the is easily opened. An 11- or 12-mm trocar can be
right arm at a right angle and the left arm placed inserted after confirming that the peritoneal
alongside the body. The patient is then moved into cavity is opened. At this point, there should be
a reverse Trendelenburg position with a 10–30° tilt. no resistance while inserting the trocar.
In some cases, a lithotomy position with reverse The peritoneal cavity is insufflated with carbon
Trendelenburg is preferable, especially for single- dioxide at a pressure of 10–13 mmHg. Tilting the
incision laparoscopic distal gastrectomy. operation table to the right or left side 10–20° may
Generally, the operator and scopist sit on the help exposure if necessary. A laparoscope is
patient’s right side, and the first assistant sits on inserted through the port, and four more 5–12-mm
the patient’s left side. Sitting during the operation trocars are placed on the upper abdominal wall
is recommended to reduce surgeon fatigue and to under direct visualization. All trocars are inserted
allow for more stable movement of equipment a fist’s distance apart to avoid interference, and
with reduced tremor. they are positioned on the lateral side of the rectus
muscle so laparoscopic instruments cannot to be
manipulated vertically or in mirror image. The left
Operative Equipment hand 5-mm trocar is inserted on the lateral portion
of the rectus muscle. It is positioned between
• 10-mm, 30° or 45° rigid scope or flexible the rectus muscle’s lateral border and the anterior
high-definition (HD) scope axillary line according to the preference of the
21 Principles and Practice of Laparoscopic Gastrectomy with Gastroduodenostomy (Billroth I) 215

of intracorporeal Billroth I anastomosis, which is

a functional end-to-end gastroduodenostomy
technique using linear staplers [6, 7, 9, 10]. Here,
we describe the delta-shaped Billroth I anasto-
mosis because it is the most popular and is easier
X technically (Table 21.1):
5 mm X
1. Liver retraction
X 5 mm A. Because the left lobe of the liver overlies
12 mm X
most of the lesser curvature of the stomach
12 mm
and the lesser omentum, liver retraction is
absolutely necessary for a TLDG and
X intracorporeal anastomosis.
12 mm B. Begin by penetrating a 2-0 straight Prolene
needle into the abdomen just below the
xiphoid in the midline (on the left side of
the falciform ligament). Then, insert the
needle from the peritoneal cavity to the
Fig. 21.1 Ports placement. Three 12-mm trocars are used outside of the body at the right upper epi-
for ports gastrium (on the right side of the falciform
ligament). Clip the middle portion of the
suture twice with a mid pars condensa. A
operator. When an intracorporeal Billroth I anasto- gauze is placed between the liver and the
mosis is performed, a 12-mm trocar is needed in suture to protect the liver. Both ends of the
the assistant’s left hand to introduce a linear stapler suture are pulled and are grasped snugly
for the duodenum transection and anastomosis over the skin of the anterior abdominal
(Fig. 21.1). wall with a mosquito clamp. This results
in a V-shaped sling that retracts the liver
cranially and anteriorly (Fig. 21.2) [20].
Detailed Procedure: Intracorporeal 2. Duodenum transection (Fig. 21.3)
Delta-Shaped Anastomosis (Video A. After mobilization of the gastroduode-
21.1) num, the duodenum is transected just
below the pylorus using a linear stapler
Due to advances in technology and surgical tech- (blue or purple cartridge). A sufficient
niques, the use of extracorporeal anastomosis is length of duodenum is required. To make
gradually shifting toward intracorporeal anasto- the duodenal stump, a Kocher maneuver is
mosis. Several techniques for intracorporeal sometimes required prior to the anastomo-
Billroth I anastomosis using a linear stapler, cir- sis to minimize tension on the anastomo-
cular stapler, or hand-sewing technique have sis. Clearing adhesions on the transverse
been reported in the literature. Among them, a colon, hepatoduodenal ligament, gallblad-
linear stapler has several advantages over the der, and pancreas head is recommended
other methods. It requires only a 12-mm trocar, B. The stapler is introduced through the left
which is easy to handle, and has three staple lower 12-mm port with the stapler directed
lines, which is thought to be more secure than the posteroanteriorly instead of the usual cra-
two staple lines created by a circular stapler. niocaudal direction. (Its direction is
There are several linear stapler techniques: the rotated 90° compared to the usual posi-
delta-shaped anastomosis [6], the triangulating- tion. This can facilitate a favorable blood
stapling technique [7], the bookbinding tech- supply to the anastomosis and provides a
nique [8], and the linear gastroduodenostomy [9]. wider space for the manipulation of the
The delta-shaped anastomosis is a representative anvil side of a 45-mm linear stapler.
216 S.-H. Ahn and H.-H. Kim

Table 21.1 Brief summary of published reports on intracorporeal Billroth I anastomosis

The details of
Billroth I Number of
Author Year Staplers anastomosis cases Complications Conclusions
Kanaya et al. 2002 Linear Delta shaped 9 No complications Safe and feasible
[6] as a result of the
anastomosis
Kim et al. 2008 Linear Delta shaped 25 12 % (n = 3) Safe and feasible
[11] 1 anastomotic
leakage, 1
anastomotic
stenosis, and 1
delayed gastric
emptying
Tanimura 2008 Linear Triangulating 81 1 anastomotic Safe and feasible
et al. [12] stapling leakage
Song et al. 2008 Linear Delta shaped 20 1 intra-abdominal Shorter bowel
[13] bleeding recovery than
extracorporeal
anastomosis
Kinoshita 2011 Linear Delta shaped 42 14.3 % (n = 6) Faster recovery than
et al. [14] No leakage extracorporeal
Kim et al. 2011 Linear Delta shaped 339/239 3.9 % (n = 9) Better early surgical
[15, 16] 2 anastomotic outcomes than
leakage, 1 extracorporeal
anastomotic anastomosis,
bleeding especially in obese
3.5 % (n = 12) patients
1 anastomotic
leakage, 1
anastomotic
bleeding
Kanaya et al. 2011 Linear Delta shaped 100 1 minor Mean follow-up
[17] anastomotic 54.9 months,
leakage satisfactory
outcomes
Lee et al. [18] 2011 Linear Delta shaped 26 3.9 % (n = 1) Feasible and safe
1 anastomotic
bleeding
Omori et al. 2012 Circular Delta shaped 20 No postoperative Safe and feasible
[19] (single-incision complications
laparoscopic distal
gastrectomy)
Ikeda et al. 2013 Linear Bookbinding 10 No complications Feasible and safe
[8] technique
Omori et al. 2013 Linear Triangulating 45 No anastomotic Safe and feasible
[10] stapling (single- complications
incision
laparoscopic distal
gastrectomy)
21 Principles and Practice of Laparoscopic Gastrectomy with Gastroduodenostomy (Billroth I) 217

accurate proximal resection line because

the lesion cannot be palpated or visual-
ized. In this case, the location of lesion can
be confirmed by intraoperative endoscopy
or comparing the location between the
endoscopic clips and the laparoscopic
clips using intraoperative X-ray [21, 22].
C. The proximal stomach is transected from
the greater curvature by linear stapler
(blue or gold or purple cartridge). Two lin-
ear staplers are enough to transect the
stomach in most cases.
4. Specimen delivery and check of the resection
Fig. 21.2 Combined retraction of the falciform ligament
and the left lateral lobe of the liver margins
A. The specimen is removed through the
extension of the umbilical port after plac-
ing it in a plastic bag. The plastic bag
allows the prevention of wound infection
and potential implantation of tumor cells.
In most cases, a 3–4-cm-long skin incision
with a wound protector is sufficient to
deliver the specimen.
5. Stomach opening (Fig. 21.4)
A. A small opening on the greater curvature
side of the remnant stomach is made using
a laparoscopic electrocautery or harmonic
scalpel. We recommend a harmonic scalpel
to open the stomach because one full bite of
Fig. 21.3 Duodenum transection in a vertical direction
a harmonic scalpel is appropriately the
through the 12 mm port of assistant length of the opening, does not cause any
bleeding, and leaves a “dog ear” that can be
subsequently used for pulling the stomach.
C. Check the color of the duodenal stump. If B. After the formation of the stomach open-
the blood supply to the duodenal stump is ing, aspirate the intraluminal contents of
poor, immediately convert to a Billroth II the stomach using a suction device to pre-
or Roux-en-Y gastrojejunostomy after vent a spillage of the bowel contents.
using an additional linear stapler to tran- Sometimes, irrigation of the remnant
sect the duodenal portion receiving the stomach is recommended to reduce the
poor blood supply. potentially existing cancer free cells.
3. Tumor localization and stomach transection 6. Duodenum opening (Fig. 21.4)
A. After complete D1+ or D2 lymphadenec- A. The small opening on the posterior side of
tomy is performed, the stomach is tran- the duodenal stump is made using a lapa-
sected. If the tumor is located below the roscopic electrocautery. We recommend
angle, the proximal stomach is transected hook or endo-shear with electrocautery to
immediately above the angle without make a sharp incision on the duodenal
checking the location of tumor. edge, which is important to avoid creating
B. If the tumor is located above the angle of a large opening. This is an important tip
the stomach, it is difficult to make an for creating the delta-shaped anastomosis.
218 S.-H. Ahn and H.-H. Kim

Fig. 21.4 (a) Formation of the stomach opening, (b) Formation of the duodenal opening

C. The jaw of the linear stapler is then closed

to avoid slipping of the stomach. The lin-
ear stapler is moved close to the duodenal
hole with the stapler closed.
(i) Place the end of the anvil of the stapler
into the opening of the duodenum.
(ii) Both the distal part and the staple side
of the duodenum are grasped by the
operator, and then the duodenum is
pulled to the anvil. The anvil should not
be pushed or thrust to the duodenum.
D. Before firing, the operator rotates the staple
line of the duodenal stump to the right side,
Fig. 21.5 Side-to-side gastroduodenostomy between the
posterior wall of the stomach and the posterosuperior wall
and the assistant rotates the staple line of
of the duodenum the stomach to the left side to form a side-
to-side gastroduodenostomy between the
7. Linear stapler insertion (Fig. 21.5) posterior wall of the remnant stomach and
A. A 45-mm linear stapler (blue or purple the posterosuperior wall of the duodenum.
cartridge) is inserted through the left lower E. The linear stapler is fired by the assistant
12-mm trocar. after waiting for 15 s.
B. After open the linear stapler, the stapler side 8. Common entry hole closure (Fig. 21.6)
is inserted into the opening in the stomach A. After firing the stapler, a common entry
in a manner similar to pulling up socks. hole is made. Check the staple line for
(i) Place the end of the staple side into anastomotic bleeding through this hole.
the opening of the stomach: B. The operator retracts both ends of the
(ii) Pull the dog ear of the stomach hole previous stapling. The common entry
with the grasper. hole is then closed with one or two con-
(iii) After full insertion of stapler, the sta- secutive firings of 60-mm linear sta-
ple line on the stomach is rotated to plers. There is another method. After
the left side by the operator’s two transient approximation of the entry
graspers, and then the assistant grasps hole using 3 stay sutures (on both ends
and pulls the midportion of the staple of the previous stapling and the midpor-
line to maintain the position of inser- tion), the operator retracts two sutures,
tion and rotation. and the assistant retracts one suture.
21 Principles and Practice of Laparoscopic Gastrectomy with Gastroduodenostomy (Billroth I) 219

Fig. 21.6 (a) Retraction of both ends of the previous stapling by the operator’s grasper. (b) Closure of the common
entry hole using a 60 mm linear stapler

This retraction of the stay sutures allows 2 cm distal to the pylorus, and a 2-0
horizontal alignment of the common straight Prolene needle is inserted through
entry hole. the purse-string clamp. A nylon tape is
C. We recommend placement of reinforce- tied just proximal to the purse-string
ment sutures on the greater curvature side clamp to prevent spillage from the stom-
where there is maximum anastomotic ach, and the duodenum is divided.
tension. 3. Anvil placement into the duodenal stump
A. After the division of the duodenum, the
proximal gastroduodenum is placed into
Brief Description: Extracorporeal the abdominal cavity. This procedure pro-
Billroth I Reconstruction vides a large working space without inter-
ference from the gastroduodenum. The
Extracorporeal Billroth I anastomosis has several anvil of a circular stapler is inserted into
advantages over intracorporeal anastomosis. It the duodenal stump, and a purse-string
allows the proximal stomach to be accurately suture is tied over the anvil. Endoloop
transected because the lesion can be palpated or reinforcement is sometimes useful to
visualized through the gastrotomy, and it requires secure the purse-string suture.
fewer staples than an intracorporeal anastomosis. 4. Tumor localization and stomach transection
It is disadvantageous because it has to be per- A. The proper line for the proximal resection
formed in a narrow space, is technically difficult is confirmed by palpation or direct visual-
in obese patients, and sometimes causes severe ization of the endoscopic intragastric clip,
postoperative pain at the mini-laparotomy site. which was placed preoperatively.
We briefly discuss the extracorporeal end-to-side B. The stomach is transected from the greater
posterior wall anastomosis [1]: curvature to the midpoint of the section
1. Mini-laparotomy line using a Kelly clamp and an Allen
A. A 4–5-cm upper transverse incision is clamp. The remaining proximal stomach
made at the right epigastrium. A plastic (the lesser curvature side) is divided using
wound retractor is recommended to pre- a linear stapler.
vent wound infection. 5. Extracorporeal end to posterior wall of the
2. Duodenum transection stomach Billroth I
A. After complete retrieval of the duodenum A. After the resected stomach and lymph
from the abdominal cavity, a purse-string nodes are removed, the body of a circular
clamp is applied to the duodenum 1 or stapler (29 or 31 mm) is inserted into the
220 S.-H. Ahn and H.-H. Kim

remnant stomach through the opening, Maehara Y. Progression from laparoscopic-assisted to

totally laparoscopic distal gastrectomy: comparison
which was previously closed with an Allen
of circular stapler (i-DST) and linear stapler (BBT)
clamp. for intracorporeal anastomosis. Surg Endosc.
B. The central rod is advanced to penetrate 2013;27:325–32. doi:10.1007/s00464-012-2433-y.
the posterior-greater curvature side wall of 9. Song HM, Lee SL, Hur H, Cho YK, Han SU. Linear-
shaped gastroduodenostomy in totally laparoscopic
the stomach and then connected to the
distal gastrectomy. J Gastric Cancer. 2010;10:69–74.
anvil previously placed in the duodenum. doi:10.5230/jgc.2010.10.2.69.
At least 3 cm in length is needed from the 10. Omori T, Masuzawa T, Akamatsu H, Nishida T. A
proposed closure line of the opening. simple and safe method for Billroth I reconstruction
in single-incision laparoscopic gastrectomy using a
C. After the circular stapler is closed and fired,
novel intracorporeal triangular anastomotic tech-
the anastomotic staple line through the open- nique. J Gastrointest Surg. 2013;18:1–4. doi:10.1007/
ing of the stomach is checked for bleeding. s11605-013-2419-7.
6. Closure of the stomach opening 11. Kim J-J, Song KY, Chin HM, Kim W, Jeon HM, Park CH,
Park SM. Totally laparoscopic gastrectomy with various
A. After the opening of the remnant stomach
types of intracorporeal anastomosis using laparoscopic
is roughly closed with three Allis clamps, linear staplers: preliminary experience. Surg Endosc.
it is completely closed using additional 2008;22:436–42. doi:10.1007/s00464-007-9446-y.
linear staplers. 12. Tanimura S, Higashino M, Fukunaga Y, Takemura M,
Nishikawa T, Tanaka Y, Fujiwara Y, Osugi H. Intracorporeal
Billroth 1 reconstruction by triangulating stapling tech-
nique after laparoscopic distal gastrectomy for gastric can-
References cer. Surg Laparosc Endosc Percutan Tech. 2008;18:54–8.
doi:10.1097/SLE.0b013e3181568e63.
1. Lee SI, Choi Y-S, Park DJ, Kim HH, Yang H-K, Kim 13. Song KY, Park CH, Kang HC, Kim J-J, Park SM, Jun KH,
MC. Comparative study of laparoscopy-assisted distal gas- Chin HM, Hur H. Is totally laparoscopic gastrectomy less
trectomy and open distal gastrectomy. J Am Coll Surg. invasive than laparoscopy-assisted gastrectomy?: prospec-
2006;202:874–80. doi:10.1016/j.jamcollsurg.2006.02.028. tive, multicenter study. J Gastrointest Surg. 2008;12:1015–
2. Kim MC, Jung G-J, Kim HH. Learning curve of 21. doi:10.1007/s11605-008-0484-0.
laparoscopy-assisted distal gastrectomy with systemic 14. Kinoshita T, Shibasaki H, Oshiro T, Ooshiro M,
lymphadenectomy for early gastric cancer. World J Okazumi S, Katoh R. Comparison of laparoscopy-
Gastroenterol. 2005;11:7508–11. assisted and total laparoscopic Billroth-I gastrectomy
3. Jeong O, Park Y-K. Clinicopathological features and for gastric cancer: a report of short-term outcomes.
surgical treatment of gastric cancer in South Korea: Surg Endosc. 2011;25:1395–401. doi:10.1007/
the results of 2009 nationwide survey on surgically s00464-010-1402-6.
treated gastric cancer patients. J Gastric Cancer. 15. Kim MG, Kawada H, Kim BS, Kim TH, Kim KC,
2011;11:69–77. doi:10.5230/jgc.2011.11.2.69. Yook JH, Kim BS. A totally laparoscopic distal gas-
4. Ahn S-H, Son S-Y, Lee C-M, Jung H, Park DJ, Kim trectomy with gastroduodenostomy (TLDG) for
HH. Intracorporeal uncut Roux-en-Y gastrojejunos- improvement of the early surgical outcomes in high
tomy reconstruction in pure single-incision laparo- BMI patients. Surg Endosc. 2011;25:1076–82.
scopic distal gastrectomy for early gastric cancer: doi:10.1007/s00464-010-1319-0.
unaided stapling closure. J Am Coll Surg. 2014;218:e17– 16. Kim MG, Kim KC, Kim BS, Kim TH, Kim HS, Yook JH,
21. doi:10.1016/j.jamcollsurg.2013.09.009. Kim BS. A totally laparoscopic distal gastrectomy can be
5. Hosogi H, Kanaya S. Intracorporeal anastomosis in an effective way of performing laparoscopic gastrectomy
laparoscopic gastric cancer surgery. J Gastric Cancer. in obese patients (body mass index ≥ 30). World J Surg.
2012;12:133–9. doi:10.5230/jgc.2012.12.3.133. 2011;35:1327–32. doi:10.1007/s00268-011-1034-6.
6. Kanaya S, Gomi T, Momoi H, Tamaki N, Isobe H, 17. Kanaya S, Kawamura Y, Kawada H, Iwasaki H, Gomi
Katayama T, Wada Y, Ohtoshi M. Delta-shaped anas- T, Satoh S, Uyama I. The delta-shaped anastomosis in
tomosis in totally laparoscopic Billroth I gastrectomy: laparoscopic distal gastrectomy: analysis of the initial
new technique of intraabdominal gastroduodenos- 100 consecutive procedures of intracorporeal gastro-
tomy. J Am Coll Surg. 2002;195:284–7. duodenostomy. Gastric Cancer. 2011;14:365–71.
7. Tanimura S, Higashino M, Fukunaga Y, Kishida S, doi:10.1007/s10120-011-0054-0.
Nishikawa M, Ogata A, Osugi H. Laparoscopic distal 18. Lee HW, Kim H-I, An JY, Cheong J-H, Lee KY,
gastrectomy with regional lymph node dissection for Hyung WJ, Noh SH. Intracorporeal anastomosis
gastric cancer. Surg Endosc. 2005;19:1177–81. using linear stapler in laparoscopic distal gastrec-
doi:10.1007/s00464-004-8936-4. tomy: comparison between gastroduodenostomy and
8. Ikeda T, Kawano H, Hisamatsu Y, Ando K, Saeki H, gastrojejunostomy. J Gastric Cancer. 2011;11:212–8.
Oki E, Ohga T, Kakeji Y, Tsujitani S, Kohnoe S, doi:10.5230/jgc.2011.11.4.212.
21 Principles and Practice of Laparoscopic Gastrectomy with Gastroduodenostomy (Billroth I) 221

19. Omori T, Tanaka K, Tori M, Ueshima S, Akamatsu H, 21. Kim H-I, Hyung WJ, Lee CR, Lim JS, An JY,
Nishida T. Intracorporeal circular-stapled Billroth I Cheong J-H, Choi SH, Noh SH. Intraoperative por-
anastomosis in single-incision laparoscopic distal table abdominal radiograph for tumor localization: a
gastrectomy. Surg Endosc. 2012;26:1490–4. simple and accurate method for laparoscopic
doi:10.1007/s00464-011-2034-1. gastrectomy. Surg Endosc. 2011;25:958–63.
20. Shabbir A, Lee JH, Lee M-S, Park do J, Kim HH. doi:10.1007/s00464-010-1288-3.
Combined suture retraction of the falciform ligament 22. Park DJ, Lee HJ, Kim SG, Jung HC, Song IS, Lee
and the left lobe of the liver during laparoscopic total KU, Choe KJ, Yang HK. Intraoperative gastroscopy
gastrectomy. Surg Endosc. 2010;24:3237–40. for gastric surgery. Surg Endosc. 2005;19:1358–61.
doi:10.1007/s00464-010-1118-7. doi:10.1007/s00464-004-2217-0.
 Laparoscopic Subtotal Gastrectomy
with Gastrojejunostomy and D2 22
Lymphadenectomy

Joshua Ellenhorn

Clinical Studies slowly. Kitano et al. reported the first

laparoscopic-assisted gastrectomy with lymph-
Laparoscopic surgical procedures have been suc- adenectomy over two decades ago [11]. However,
cess fully adopted for abdominal surgery because despite encouraging results, laparoscopic resec-
of their favorable effects on pain, postoperative tion for gastric cancer is only now gaining accep-
recovery, pulmonary function, and incision- tance in North America. Concerns about technical
related complications [1, 2]. Laparoscopic resec- difficulty, completeness of resection and ade-
tion for colorectal cancer has become a standard quacy of lymphadenectomy have limited enthusi-
of care based on the positive results of several ran- asm for laparoscopic gastrectomy.
domized trials [3–5]. Despite early concerns The first and only prospective randomized
about the adequacy of resection for oncologic trial comparing laparoscopic to open gastrectomy
indications, clinical studies have demonstrated in a Western country was published in 2005 [12].
that laparoscopic resections for abdominal malig- In the small trial, Huscher et al. randomized 59
nancy can be performed with equivalent extent of patients with gastric cancer to laparoscopic or
resection compared to open resection. Clinical open gastrectomy. The laparoscopic approach
studies in pancreatic [6], cervical [7], endometrial was associated with a decreased estimated blood
[8], colorectal [2], prostate [9], and renal carci- loss, earlier oral intake, and a shorter hospital
noma [10] have demonstrated that the laparo- stay. There was no difference in lymph node
scopic approach yields similar margins and nodal count suggesting that the laparoscopic approach
clearance to open surgery. Survival rates are also did not compromise the adequacy of resection.
similar to that seen in traditional open surgery. There was no difference in 5-year disease-free or
For gastric adenocarcinoma, minimally inva- overall survival [12].
sive techniques have been adopted relatively Several prospective randomized trials compar-
ing laparoscopic to open resection have been con-
Electronic supplementary material Supplementary
ducted in Asia. The largest of these is the Korean
material is available in the online version of this Laparoscopic Gastrointestinal Surgery Study
chapter at 10.1007/978-3-319-09342-0_22. Videos can (KLASS) Group trial with over 1,400 patients
also be accessed at http://www.springerimages.com/ recruited [13]. The main endpoint of the trial is
videos/978-3-319-09341-3.
survival which has not yet been published. Interim
J. Ellenhorn, MD analysis revealed no difference in morbidity or
Department of Surgery, Cedars-Sinai Medical Center,
mortality between the groups [14]. In another
8635 West Third Street, Suite 880 West,
Los Angeles, CA 90048, USA Korean study, 82 patients were randomized to
e-mail: [email protected] open gastrectomy and 82 to laparoscopic resection

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 223
DOI 10.1007/978-3-319-09342-0_22, © Springer International Publishing Switzerland 2015
224 J. Ellenhorn

Table 22.1 Summary of randomized trials

Adequacy of
Author Year Lap Open resection Results for lap group Survival
Kitano et al. 2002 14a 14 Identical Less EBL and pain, na
[11] earlier recovery of
bowel function
Hayashi et al. 2005 14a 14 Equally radical Shorter epidural use na
[16]
Lee et al. [17] 2005 24a 23 No significant Fewer pulmonary No difference at
difference complications 14 months
Huscher et al. 2005 30 29 No significant No difference No difference at
[12] difference 5 years
Kim et al. [15] 2008 82a 82 na Less EBL and pain na
medicine, shorter
hospital stay, improved
QOL
Kim et al. [14] 2010 179a 161 na No difference in na
morbidity or mortality
Cai et al. [18] 2011 61a 62 No difference Less pulmonary No difference at
infection 2 years
a
Laparoscopic assisted with an open component
EBL estimated blood loss, Lap laparoscopic, na not available

[15]. Laparoscopic resection was associated with longer operative times, less blood loss, earlier
longer operative times but lower blood loss, a return of bowel activity, and shorter hospital.
shorter hospital stay, and an improvement in qual- They also found that laparoscopic resection
ity of life [15]. Seven randomized trials compar- yielded slightly fewer lymph nodes [20]. Two
ing laparoscopic to open resection have completed meta-analyses focusing on patients with locally
accrual (Table 22.1). The results of the trials sug- advanced gastric cancer had similar conclusions
gest that laparoscopic resection takes longer and except that the lymph node yields of the laparo-
has a similar morbidity and mortality to open scopic groups in these meta-analyses were simi-
resection but is associated with a faster return of lar to the open groups [21, 22].
bowel function, less blood loss, and reduced post- In most of the Asian studies, the laparoscopic
operative pain. groups included a large proportion of patients
Several meta-analyses of randomized and undergoing open intestinal transection and gas-
nonrandomized trials have been published. A trojejunal or gastroduodenal anastomoses. In
recent meta-analysis of laparoscopic versus open Western series, the procedures almost always
distal gastrectomy concluded that laparoscopic include intracorporeal anastomoses and are
gastrectomy was associated with lower blood therefore considered laparoscopic as opposed to
loss, faster return of bowel function, and a shorter laparoscopic-assisted resections. Compared to
hospital stay but a slight reduction in lymph node laparoscopic-assisted gastrectomy, totally laparo-
yield [19]. Although there was no difference in scopic gastrectomy is associated with less blood
the proportion of patients with 15 or more lymph loss, shorter time to first flatus, and shorter post-
nodes in their specimen, the laparoscopic group operative hospital stay [23]. There is no signifi-
had a median of 3.9 fewer lymph nodes than the cant difference in operative time, mean number
open group. The implications of this small differ- of lymph nodes retrieved, and postoperative com-
ence in lymph node yield is unclear [19]. A meta- plications [23]. An evaluation of the largely non-
analysis of laparoscopic versus open gastrectomy randomized Western data is therefore warranted
for early stage gastric cancer concluded that the (Table 22.2). Western series support the conclu-
laparoscopic approach was associated with sions that, compared to open distal gastrectomy,
 22

Table 22.2 Summary of Western distal gastrectomy studies

Open
Open Laparoscopic lymph node Open Laparoscopic Open Laparoscopic Open
Laparoscopic group lymph node harvest – Laparoscopic LOS 30-day 30-day 30-day 30-day
Author Year Country group (n) (n) harvest – mean mean length of stay (days) morbidity morbidity mortality mortality
Reyes et al. 2001 USA 18 18 8 11 6.3 8.6 na na
[24]
Huscher et al. 2005 Italy 30 29 30 33 10.3 14.5 23.3 27.6 3.3 6.7
[12]
Dulucq et al. 2005 France 14 15 17 15 16 25 12.5 17.5 0 0
[28]
Pugliese 2007 Italy 43 32 36 9.8 16.7 10 14 2 3
et al. [29]
Strong et al. 2009 USA 30 30 18 21 5 7 26 63 0 0
[25]
Guzman 2009 USA 30 48 24 26 7 10 30 46 0 0
et al. [26]
Scatizzi et al. 2011 Italy 30 30 31 37 7 9 7 26 0 0
[27]
Chouillard 2010 France 51 79 19 22 8 11.5 12 16 0 2.5
et al. [30]
Laparoscopic Subtotal Gastrectomy with Gastrojejunostomy and D2 Lymphadenectomy
225
226 J. Ellenhorn

laparoscopic distal gastrectomy takes longer with lymphadenectomy can be accomplished

[24–27] but yields a similar lymph node count following neoadjuvant chemotherapy but may be
[12, 26, 28–30] with less blood loss [12, 24, 26, technically demanding in some patients who
29], shorter hospital stay [12, 26–30], and lower have had prior radiation therapy. As with any
postoperative morbidity [25–28]. Short-term [25, laparoscopic procedure, it is incumbent on the
27, 29, 30] and 5-year [12] survivals of laparo- operating surgeon to be cognizant of their own
scopic distal gastrectomy are similar to those fol- surgical limitations and have a low threshold to
lowing open gastrectomy. convert any case to open if it cannot be performed
thoroughly and safely laparoscopically. Poor car-
diopulmonary reserve is a relative contraindica-
Patient Selection tion to laparoscopic gastrectomy because of the
decrease in venous return and increase in pulmo-
The decision to perform a laparoscopic versus an nary resistance associated with prolonged
open gastrectomy depends on several factors. pneumoperitoneum.
The most important consideration is the skill and Hand-assisted laparoscopic gastrectomy has
experience of the operating surgeon. Laparoscopic been advocated by some surgeons in an attempt to
gastrectomy is an operation requiring advanced overcome the technical challenges associated with
laparoscopic skills to perform an adequate a totally laparoscopic approach [32]. Although a
lymphadenectomy and an intestinal anastomosis. hand assist approach may enable a surgeon early
The procedure also requires an operating room in their learning curve to complete a laparoscopic-
team equipped with appropriate laparoscopic assisted distal gastrectomy, it will inevitably
atraumatic graspers, an energy device, liver require an incision significantly larger than that
retractor, wound protector, and laparoscopic required for specimen extraction with a totally
reticulating staplers. The procedure is particu- laparoscopic approach. Routine reliance on the
larly demanding in obese patients. Challenges in use of a hand port might also limit the surgeon’s
laparoscopic resection in obese patients include own technical development and proficiency.
decreased surgical visibility, dissection hindered
by adipose tissue, and difficulty with anastomo-
ses. Higher BMI is an independent risk factor for Patient Positioning and Room Setup
pancreatic fistula following laparoscopic distal
gastrectomy [31]. The patient is placed in the supine position with
Prior upper abdominal surgery can make lapa- both arms tucked. General endotracheal anesthe-
roscopic gastrectomy technically difficult. sia is administered. A nasogastric tube is placed
Laparoscopic or open cholecystectomy often to decompress the stomach, and a Foley catheter
results in adhesion of the first portion of the duo- is inserted into the bladder. Intravenous antibiot-
denum to the gallbladder bed. Careful dissection ics are administered within 1 h of the incision and
can usually allow full mobilization of the duode- redosed as necessary. Deep venous thrombosis
num off the liver bed. An adequate lymphadenec- prophylaxis is accomplished with sequential
tomy can be performed following an open or pneumatic compression stockings on the lower
laparoscopic cholecystectomy as long as the extremities. In addition, all patients are adminis-
patient did not undergo an open common bile tered 5,000 units of subcutaneous unfractionated
duct exploration. Prior right hemicolectomy or heparin in the immediate preoperative period.
splenectomy can complicate laparoscopic distal Eggcrate foam is secured to the operating
gastrectomy. Prior gastric resection or transverse room with wide tape, and the patient is placed on
colectomy is a relative contraindication to laparo- the eggcrate without an intervening bedsheet.
scopic gastrectomy because the plane of dissec- This secures the patient to the surgical bed and
tion for lymphadenectomy will have been altered prevents shifting during maximum reverse
or obliterated. In general, adequate resection Trendelenburg position [33]. The video monitors
22 Laparoscopic Subtotal Gastrectomy with Gastrojejunostomy and D2 Lymphadenectomy 227

Table 22.3 Sequence of laparoscopic subtotal gastrec-

tomy with gastrojejunostomy and D2 lymphadenectomy
1 Omentectomy
2 Transection of right gastroepiploic vessels
3 Transection of postpyloric duodenum
4 Division of lesser omentum
5 Dissection of capsule over superior boarder of
pancreatic neck
6 Dissection of hepatic artery and portal vein nodes
7 Division of origin of the left gastric artery
5 mm 8 Dissection of proximal splenic artery nodes
5 mm 9 Stripping of lesser curvature of proximal stomach
5 mm
5 mm
10 Transection of stomach
12 mm 11 Stapled side-to-side gastrojejunostomy
Extraction site

liver or peritoneal metastatic disease. If metastatic

disease is identified, the operation can be aborted
or converted to a gastrojejunal bypass, if clinically
indicated. If distant metastatic disease is not iden-
tified, an extreme right lateral subcostal port is
Fig. 22.1 Laparoscopic port site placement. A 5 mm then placed. A snake retractor placed through the
supra-umbilical camera port is flanked by the right and
left upper quadrant 5 mm dissection ports. A 12 mm left- right lateral port is triangulated closed and used to
sided stapling port is later enlarged for specimen extrac- elevate the left lobe of the liver. The snake retrac-
tion. A 5 mm right lateral subcostal port site is used for tor is fixed in place with an adjustable robot arm-
placement of a liver retractor type retractor system. A right upper quadrant
5 mm dissecting port is placed. A 5 mm 30°
are positioned near the shoulders on each side of angled camera is used for the entire operation.
the operating table. The procedure is performed With some equipment, the 5 mm camera does not
with a surgeon and an assistant. The surgeon allow enough light to accomplish advanced lapa-
begins on the right side of the table with the assis- roscopy. If the operation cannot be performed
tant on the left side of the table. The scrub nurse safely with a 5 mm camera, the supra-umbilical
is positioned on the right side of the table. 5 mm port can be exchanged for a 10–12 mm port
for a 30° angled 10 mm camera.
The omentum is reflected into the upper abdo-
Operative Procedure (Video 22.1) men and dissected off of the transverse colon
using a 5 mm energy device (Table 22.3).
The procedure is performed with four 5 mm ports Dissection is conducted from the right side of the
and a single 12 mm port for laparoscopic stapling table, beginning at the midline and extending up
(Fig. 22.1). The peritoneal cavity is entered and to the lowest short gastric vessels. For this dissec-
insufflated with a Veress needle after a stab wound tion, the camera can be moved to the left 12 mm
is made in the left subcostal space at Palmer’s port to allow the surgeon to work with both hands
point. A 5 mm camera trocar is placed in the through the right upper quadrant and supra-
supra-umbilical region. A left upper quadrant umbilical port. The omental dissection is then
5 mm dissection port is placed. This port should carried up to the greater curvature. The camera is
be placed laterally enough to allow for placement moved from the left to the supra-umbilical port,
of a 12 mm stapling port midway between the left and the right side of the omentum is dissected
upper quadrant and supra-umbilical ports. The from the transverse colon. This is best performed
abdomen is thoroughly explored for evidence of from the left side of the patient. After the
228 J. Ellenhorn

a b

c d

Fig. 22.2 Lymph node dissection. (a) The lymph node- reflected on to the specimen. (e) Lymph nodes are dis-
bearing tissues are lifted off the hepatic artery and (b) sected off the lesser curvature of the stomach. Hepatic
lymph nodes along the portal vein and reflected to the left. artery (HA), portal vein (PV), left gastric artery (LGA),
(c) The origin of the left gastric artery is skeletonized, and left gastric vein (SV), lymph node (LN)
(d) the lymph nodes along the proximal splenic artery are

complete separation of the omentum from the stapler using 3.5 mm (blue) staples or Tri-
transverse colon, the base of the right gastroepi- StapleTM 2.0, 2.5, and 3.0 mm (tan) staples. Staple
ploic vessels are dissected at the level of the infe- line buttressing material is not used for any of the
rior border of the pancreas. The right stapling in the procedure. None of the staple lines
gastroepiploic vessels are transected using an are imbricated with sutures, and sutures are not
energy device. Attention is then turned to the generally used to take tension off the staple lines.
supra-duodenal region, and the lesser omentum is The lymph node dissection is accomplished by
opened. The first portion of the duodenum is sur- clearing the fat over the portal hepatitis and proxi-
rounded and transected using a 60 mm endoscopic mal hepatic artery (Fig. 22.2). The fat is reflected
22 Laparoscopic Subtotal Gastrectomy with Gastrojejunostomy and D2 Lymphadenectomy 229

to the left. Dissection is carried along the common the abdomen, the ligament of Treitz is identified.
hepatic artery up to the porta hepatis reflecting the The proximal jejunum is carefully followed and
nodal tissues to the left. The portal dissection is reflected over the transverse colon. An area in the
carried up reflecting the nodal tissues from the left proximal jejunum, which approximates the stom-
side of the portal vein. With the nodal packet ach without tension, is chosen for the anastomo-
reflected to the left, dissection is then carried onto sis. The anastomosis is best accomplished from
the proximal proper hepatic artery. The left gastric the right side of the patient. The jejunum is laid
vein is transected at the upper border of the pan- next to the stomach such that the proximal end of
creas. The base of the left gastric artery is dis- the jejunum is to the right and the distal end is to
sected and controlled with Hem-o-lok clips and the left (Fig. 22.3). This is done so that the sta-
transected. The nodal tissue along the proximal pling defect following formation of the anastomo-
splenic artery is dissected, reflecting the nodes off sis is on the afferent limb of the small bowel. Any
of the body of the pancreas, exposing the splenic difficulty with closure of this defect will not affect
artery. This nodal packet is then reflected to the to the efferent limb of the gastrojejunostomy. A
left. The entire nodal packet is reflected off of the long 3-0 Vicryl traction suture is used to approxi-
retroperitoneum. mate the small bowel to the proximal gastric
A gastroscopy is performed using an upper GI pouch just superior to the gastric staple line. This
endoscope. The exact location of the tumor is traction suture is brought through the 12 mm port
noted by endoscopy, while the corresponding (Fig. 22.3). An enterotomy is made with the
serosal area is identified by laparoscopy. This is energy device or hook electrocautery, in the small
best accomplished by pressing on the stomach in bowel and in the stomach just beyond the traction
the region of the tumor with a laparoscopic dis- suture. A side-to-side anastomosis is performed
sector. A proximal gastric transection region is using a 60 mm endoscopic stapler using 3.5 mm
chosen and can be marked with sutures or clips. (blue) staples or Tri-StapleTM 2.0, 2.5, and 3.0 mm
The lesser omentum is transected close to the (tan) staples. It is important to place the stapler
liver. The nodal tissues along the lesser curvature into position in the jejunum and stomach and then
of the stomach are then dissected. This is per- rotated it in a counterclockwise way, so that the
formed by stripping the lymph node-bearing fat anastomosis will be on the anterior wall of the
from the lesser curvature from proximal to distal. stomach and not cross the gastric transection sta-
Following the node stripping, the lesser curvature ple line. The stapling enterotomy defect is closed
often appears somewhat dark or ecchymotic in using two layers of running 3-0 VicrylTM suture.
color, even though its blood supply remains Lapra-TysTM can be used to secure the sutures.
robust. The stomach is then transected with an A feeding jejunostomy tube is generally not
endoscopic stapler, taking an appropriate margin necessary. The nasogastric tube is left in place in
proximal to the tumor. The stomach is transected the gastric pouch and removed on the first post-
using sequential firings of a 60 mm endoscopic operative morning. The patient is advanced from
stapler using 4.1 mm (green) staples or Tri- a clear liquid diet on the first and second
StapleTM 3.0, 3.5, and 4.0 mm (purple) staples. postoperative days to a regular diet by postopera-
The specimen is then grasped with a laparoscopic tive days number three and four. Patients are dis-
instrument. charged when they are able to tolerate a regular
The 12 mm stapling port is enlarged, and a diet.
wound protector is placed. The specimen is with-
drawn through the wound protector and immedi-
ately opened by the pathologist to assess margins. Reconstruction
The wound protector is loosened and turned
around a 12 mm port. A moist laparotomy pad can Reconstruction following distal gastrectomy can
be wrapped around the wound protector and take several forms [34, 35]. Billroth I gastroduode-
secured with a Kocher clamp. After insufflating nostomy is not commonly performed in the United
230 J. Ellenhorn

a b

c d

Fig. 22.3 Gastrojejunal anastomosis. (a)The proximal through enterotomies in the jejunum and stomach. (c, d)
jejunum is approximated to the proximal gastric pouch The stapling defect is closed with two layers of running
with a traction suture, and (b) a 60 mm stapler is inserted suture

States and can be difficult to construct in patients larger gastric remnant. Patients were allocated
for whom over 50 % of the stomach has been into two groups based on the size of their rem-
resected. Roux-en-Y gastrojejunostomy is com- nant gastric pouch, one having at least 70 % of
monly performed but requires two enteric anasto- their stomach resected (small pouch) and the
moses. Billroth II (BII) loop gastrojejunostomy other having less than 70 % of their stomach
requires only one anastomosis and is straightfor- resected (large pouch). Thirty patients consented
ward and easily performed using laparoscopic to participate and completed the EORTC QOL-
techniques. Because BII reconstruction involves STO22 instrument. In general, patients expressed
fewer anastomoses, there is a reduced likelihood few symptoms. There was no significant differ-
of anastomotic leak and internal hernia formation ence between the large and small remnant pouch
than are known to associate with a Roux-en-Y cohorts in overall symptoms or specific symp-
reconstruction. Unfortunately, concern for bile toms (Table 22.4).
reflux gastritis severely limits its use by most gas-
trointestinal surgeons. Instead, Roux-en-Y recon-
struction is favored after gastrectomy and has been Laparoscopic Versus Robotic
recommended in the United States [36–39]. In my Gastrectomy
practice, all patients undergoing laparoscopic dis-
tal gastrectomy are reconstructed with a BII gas- Robotic surgery systems have been introduced as a
trojejunostomy because it is straightforward to solution to minimize the shortcomings of laparos-
perform laparoscopically [26, 40]. copy. Robotics provides definite technical advan-
We compare quality of life of patients at least tages over conventional laparoscopy [41], but its
6 months after laparoscopic partial gastrectomy role for gastric cancer is still unclear [41–44]. Since
with BII reconstruction with a small versus robotic gastrectomy was first reported [41, 45], its
22 Laparoscopic Subtotal Gastrectomy with Gastrojejunostomy and D2 Lymphadenectomy 231

Table 22.4 Quality of life measurement 2. Schwenk W, Haase O, Neudecker J, Muller JM. Short
term benefits for laparoscopic colorectal resection.
Large Small
Cochrane Database Syst Rev. 2005;(3):CD003145.
Entire pouch (16 pouch (14
PubMed PMID: 16034888.
cohort patients) patients)
3. Jayne DG, Thorpe HC, Copeland J, Quirke P, Brown JM,
mean mean mean
Guillou PJ. Five-year follow-up of the Medical Research
scorea score score
Council CLASICC trial of laparoscopically assisted ver-
Symptoms (±STD) (±STD) (±STD) p-value
sus open surgery for colorectal cancer. Br J Surg.
Overall 1.47 1.42 1.53 0.54 2010;97(11):1638–45. PubMed PMID: 20629110.
(±0.44) (±0.44) (±0.46) 4. Lacy AM, Garcia-Valdecasas JC, Delgado S, Castells A,
a
Score scale: 1, not at all; 2, a little; 3, quite a bit; 4, very Taura P, Pique JM, et al. Laparoscopy-assisted colec-
much tomy versus open colectomy for treatment of non-
metastatic colon cancer: a randomised trial. Lancet.
2002;359(9325):2224–9. PubMed PMID: 12103285.
use has been expanding, primarily in high-volume 5. Janson M, Bjorholt I, Carlsson P, Haglind E,
centers. There have been very few studies and no Henriksson M, Lindholm E, et al. Randomized clini-
cal trial of the costs of open and laparoscopic surgery
randomized studies comparing laparoscopic gas-
for colonic cancer. Br J Surg. 2004;91(4):409–17.
trectomy to robotic gastrectomy. A recent meta- PubMed PMID: 15048739.
analysis of studies compared robotic to open 6. Kendrick ML. Laparoscopic and robotic resection for
gastrectomy for gastric cancer [46]. Only six arti- pancreatic cancer. Cancer J. 2012;18(6):571–6.
PubMed PMID: 23187844.
cles compared robotic gastrectomy to laparoscopic
7. Kucukmetin A, Biliatis I, Naik R, Bryant A.
gastrectomy [47–52]. The operative time for Laparoscopically assisted radical vaginal hysterectomy
robotic gastrectomy was significantly longer than versus radical abdominal hysterectomy for the treatment
that for laparoscopic gastrectomy. There was no of early cervical cancer. Cochrane Database Syst Rev.
2013;10:CD006651. PubMed PMID: 24085528.
difference in the number of lymph nodes retrieved.
8. He H, Zeng D, Ou H, Tang Y, Li J, Zhong
The amount of blood loss was significantly less for H. Laparoscopic treatment of endometrial cancer:
robotic gastrectomy than for laparoscopic gastrec- systematic review. J Minim Invasive Gynecol.
tomy, and the length of hospital stay was signifi- 2013;20(4):413–23. PubMed PMID: 23506718.
9. Sandhu GS, Nepple KG, Tanagho YS, Andriole GL.
cantly longer with laparoscopic gastrectomy. Five
Laparoscopic prostatectomy for prostate cancer: con-
of the six series were not truly laparoscopic or tinued role in urology. Surg Oncol Clin N Am.
robotic, utilizing an open incision for the gastric 2013;22(1):125–41, vii. PubMed PMID: 23158089.
transaction and reconstruction. In addition, the 10. Ni S, Tao W, Chen Q, Liu L, Jiang H, Hu H, et al.
Laparoscopic versus open nephroureterectomy for the
robotic procedures were likely performed when the
treatment of upper urinary tract urothelial carcinoma:
surgeons were more accomplished minimally inva- a systematic review and cumulative analysis of com-
sive surgeons. These confounding variables might parative studies. Eur Urol. 2012;61(6):1142–53.
limit the validity of the conclusions favoring the PubMed PMID: 22349569.
11. Kitano S, Iso Y, Moriyama M, Sugimachi K.
robotic approach. While there may be real benefits
Laparoscopy-assisted Billroth I gastrectomy. Surg
to the use of robotics for gastrectomy and lymph- Laparosc Endosc. 1994;4(2):146–8. PubMed PMID:
adenectomy, downsides include significant costs 8180768.
and issues of availability of the technology. It 12. Huscher CG, Mingoli A, Sgarzini G, Sansonetti A, Di
Paola M, Recher A, et al. Laparoscopic versus open
remains to be seen whether the use of the surgical
subtotal gastrectomy for distal gastric cancer: five-
robot can facilitate an oncologically sound mini- year results of a randomized prospective trial. Ann
mally invasive gastrectomy by a surgeon who lacks Surg. 2005;241(2):232–7. PubMed PMID: 15650632.
the skills to perform a straight laparoscopic Pubmed Central PMCID: 1356907.
13. Kim HH, Han SU, Kim MC, Hyung WJ, Kim W,
approach.
Lee HJ, et al. Prospective randomized controlled
trial (phase III) to comparing laparoscopic distal
gastrectomy with open distal gastrectomy for gas-
References tric adenocarcinoma (KLASS 01). J Korean Surg
Soc. 2013;84(2):123–30. PubMed PMID: 23396494.
Pubmed Central PMCID: 3566471.
1. Abraham NS, Young JM, Solomon MJ. Meta-analysis
14. Kim HH, Hyung WJ, Cho GS, Kim MC, Han SU, Kim
of short-term outcomes after laparoscopic resection
W, et al. Morbidity and mortality of laparoscopic gas-
for colorectal cancer. Br J Surg. 2004;91(9):1111–24.
trectomy versus open gastrectomy for gastric cancer:
PubMed PMID: 15449261.
232 J. Ellenhorn

an interim report – a phase III multicenter, prospective, 26. Guzman EA, Pigazzi A, Lee B, Soriano PA, Nelson
randomized Trial (KLASS Trial). Ann Surg. RA, Benjamin Paz I, et al. Totally laparoscopic gastric
2010;251(3):417–20. PubMed PMID: 20160637. resection with extended lymphadenectomy for gastric
15. Kim YW, Baik YH, Yun YH, Nam BH, Kim DH, adenocarcinoma. Ann Surg Oncol. 2009;16(8):2218–
Choi IJ, et al. Improved quality of life outcomes after 23. PubMed PMID: 19444523.
laparoscopy-assisted distal gastrectomy for early gas- 27. Scatizzi M, Kroning KC, Lenzi E, Moraldi L, Cantafio
tric cancer: results of a prospective randomized clini- S, Feroci F. Laparoscopic versus open distal gastrec-
cal trial. Ann Surg. 2008;248(5):721–7. PubMed tomy for locally advanced gastric cancer: a case-
PMID: 18948798. control study. Updates Surg. 2011;63(1):17–23.
16. Hayashi H, Ochiai T, Shimada H, Gunji Y. Prospective PubMed PMID: 21286896.
randomized study of open versus laparoscopy-assisted 28. Dulucq JL, Wintringer P, Stabilini C, Solinas L,
distal gastrectomy with extraperigastric lymph node Perissat J, Mahajna A. Laparoscopic and open gastric
dissection for early gastric cancer. Surg Endosc. resections for malignant lesions: a prospective com-
2005;19(9):1172–6. PubMed PMID: 16132323. parative study. Surg Endosc. 2005;19(7):933–8.
17. Lee JH, Han HS, Lee JH. A prospective randomized PubMed PMID: 15920691.
study comparing open vs laparoscopy-assisted distal 29. Pugliese R, Maggioni D, Sansonna F, Scandroglio I,
gastrectomy in early gastric cancer: early results. Surg Ferrari GC, Di Lernia S, et al. Total and subtotal lapa-
Endosc. 2005;19(2):168–73. PubMed PMID: 15580441. roscopic gastrectomy for adenocarcinoma. Surg
18. Cai J, Wei D, Gao CF, Zhang CS, Zhang H, Zhao T. A Endosc. 2007;21(1):21–7. PubMed PMID: 17031743.
prospective randomized study comparing open versus 30. Chouillard E, Gumbs AA, Meyer F, Torcivia A,
laparoscopy-assisted D2 radical gastrectomy in Helmy N, Toubal M, et al. Laparoscopic versus open
advanced gastric cancer. Dig Surg. 2011;28(5– gastrectomy for adenocarcinoma: a prospective com-
6):331–7. PubMed PMID: 21934308. parative analysis. Minerva Chir. 2010;65(3):243–50.
19. Vinuela EF, Gonen M, Brennan MF, Coit DG, Strong PubMed PMID: 20668413.
VE. Laparoscopic versus open distal gastrectomy for 31. Jiang X, Hiki N, Nunobe S, Kumagai K, Nohara K,
gastric cancer: a meta-analysis of randomized controlled Sano T, et al. Postoperative pancreatic fistula and the
trials and high-quality nonrandomized studies. Ann risk factors of laparoscopy-assisted distal gastrectomy
Surg. 2012;255(3):446–56. PubMed PMID: 22330034. for early gastric cancer. Ann Surg Oncol. 2012;
20. Peng JS, Song H, Yang ZL, Xiang J, Diao DC, Liu ZH. 19(1):115–21. PubMed PMID: 21739317.
Meta-analysis of laparoscopy-assisted distal gastrec- 32. Tanimura S, Higashino M, Fukunaga Y, Osugi H.
tomy and conventional open distal gastrectomy for Hand-assisted laparoscopic distal gastrectomy with
early gastric cancer. Chin J Cancer. 2010;29(4):349– regional lymph node dissection for gastric cancer.
54. PubMed PMID: 20346206. Surg Laparosc Endosc Percutan Tech. 2001;11(3):
21. Martinez-Ramos D, Miralles-Tena JM, Cuesta MA, 155–60. PubMed PMID: 11444743.
Escrig-Sos J, Van der Peet D, Hoashi JS, et al. 33. Klauschie J, Wechter ME, Jacob K, Zanagnolo V,
Laparoscopy versus open surgery for advanced and Montero R, Magrina J, et al. Use of anti-skid material
resectable gastric cancer: a meta-analysis. Rev Esp and patient-positioning to prevent patient shifting dur-
Enferm Dig. 2011;103(3):133–41. PubMed PMID: ing robotic-assisted gynecologic procedures. J Minim
21434716. Invasive Gynecol. 2010;17(4):504–7. PubMed PMID:
22. Qiu J, Pankaj P, Jiang H, Zeng Y, Wu H. Laparoscopy 20471916.
versus open distal gastrectomy for advanced gastric 34. Hoya Y, Mitsumori N, Yanaga K. The advantages and
cancer: a systematic review and meta-analysis. Surg disadvantages of a Roux-en-Y reconstruction after a
Laparosc Endosc Percutan Tech. 2013;23(1):1–7. distal gastrectomy for gastric cancer. Surg Today.
PubMed PMID: 23386142. 2009;39(8):647–51. PubMed PMID: 19639429. Epub
23. Gao J, Li P, Li QG, Chen J, Wang DR, Tang D. 2009/07/30. eng.
Comparison between totally laparoscopic and laparo- 35. Nomura S, Kaminishi M. Surgical treatment of early
scopically assisted distal gastrectomy for gastric cancer gastric cancer. Dig Surg. 2007;24(2):96–100. PubMed
with a short follow-up: a meta-analysis. J Laparoendosc PMID: 17460412. Epub 2007/04/27. eng.
Adv Surg Tech A. 2013;23(8):693–7. PubMed PMID: 36. Burden WR, Hodges RP, Hsu M, O’Leary JP. Alkaline
23678885. reflux gastritis. Surg Clin North Am. 1991;71(1):33–
24. Reyes CD, Weber KJ, Gagner M, Divino CM. 44. PubMed PMID: 1989108. Epub 1991/02/01. eng.
Laparoscopic vs open gastrectomy. A retrospective 37. Sugiyama Y, Sohma H, Ozawa M, Hada R, Mikami Y,
review. Surg Endosc. 2001;15(9):928–31. PubMed Konn M, et al. Regurgitant bile acids and mucosal
PMID: 11605108. injury of the gastric remnant after partial gastrectomy.
25. Strong VE, Devaud N, Allen PJ, Gonen M, Brennan Am J Surg. 1987;153(4):399–403. PubMed PMID:
MF, Coit D. Laparoscopic versus open subtotal gastrec- 3565686. Epub 1987/04/01. eng.
tomy for adenocarcinoma: a case-control study. Ann 38. Osugi H, Fukuhara K, Takada N, Takemura M, Kinoshita
Surg Oncol. 2009;16(6):1507–13. PubMed PMID: H. Reconstructive procedure after distal gastrectomy to
19347407. prevent remnant gastritis. Hepatogastroenterology.
22 Laparoscopic Subtotal Gastrectomy with Gastrojejunostomy and D2 Lymphadenectomy 233

2004;51(58):1215–8. PubMed PMID: 15239282. Epub gastrectomy: a meta-analysis. J Gastric Cancer.

2004/07/09. eng. 2013;13(3):136–48. PubMed PMID: 24156033.
39. Csendes A, Burgos AM, Smok G, Burdiles P, Pubmed Central PMCID: 3804672.
Braghetto I, Diaz JC. Latest results (12-21 years) of a 47. Kim MC, Heo GU, Jung GJ. Robotic gastrectomy for
prospective randomized study comparing Billroth II gastric cancer: surgical techniques and clinical merits.
and Roux-en-Y anastomosis after a partial gastrec- Surg Endosc. 2010;24(3):610–5. PubMed PMID:
tomy plus vagotomy in patients with duodenal ulcers. 19688399.
Ann Surg. 2009;249(2):189–94. PubMed PMID: 48. Pugliese R, Maggioni D, Sansonna F, Costanzi A,
19212169. Epub 2009/02/13. eng. Ferrari GC, Di Lernia S, et al. Subtotal gastrectomy
40. Pigazzi A, Ellenhorn JD, Ballantyne GH, Paz IB. with D2 dissection by minimally invasive surgery for
Robotic-assisted laparoscopic low anterior resection distal adenocarcinoma of the stomach: results and
with total mesorectal excision for rectal cancer. Surg 5-year survival. Surg Endosc. 2010;24(10):2594–602.
Endosc. 2006;20(10):1521–5. PubMed PMID: PubMed PMID: 20414682.
16897284. Epub 2006/08/10. eng. 49. Yoon HM, Kim YW, Lee JH, Ryu KW, Eom BW,
41. Giulianotti PC, Coratti A, Angelini M, Sbrana F, Park JY, et al. Robot-assisted total gastrectomy is
Cecconi S, Balestracci T, et al. Robotics in general comparable with laparoscopically assisted total gas-
surgery: personal experience in a large community trectomy for early gastric cancer. Surg Endosc.
hospital. Arch Surg. 2003;138(7):777–84. PubMed 2012;26(5):1377–81. PubMed PMID: 22083338.
PMID: 12860761. 50. Woo Y, Hyung WJ, Pak KH, Inaba K, Obama K, Choi
42. Hyung WJ. Robotic surgery in gastrointestinal sur- SH, et al. Robotic gastrectomy as an oncologically
gery. Korean J Gastroenterol. 2007;50(4):256–9. sound alternative to laparoscopic resections for the
43. Gutt CN, Oniu T, Mehrabi A, Kashfi A, Schemmer P, treatment of early-stage gastric cancers. Arch Surg.
Buchler MW. Robot-assisted abdominal surgery. Br J 2011;146(9):1086–92. PubMed PMID: 21576595.
Surg. 2004;91(11):1390–7. PubMed PMID: 15386325. 51. Eom BW, Yoon HM, Ryu KW, Lee JH, Cho SJ, Lee
44. Baek SJ, Lee DW, Park SS, Kim SH. Current status of JY, et al. Comparison of surgical performance and
robot-assisted gastric surgery. World J Gastrointest short-term clinical outcomes between laparoscopic
Oncol. 2011;3(10):137–43. PubMed PMID: and robotic surgery in distal gastric cancer. Eur J Surg
22046490. Pubmed Central PMCID: 3205112. Oncol. 2012;38(1):57–63. PubMed PMID: 21945625.
45. Hashizume M, Sugimachi K. Robot-assisted gastric 52. Huang KH, Lan YT, Fang WL, Chen JH, Lo SS,
surgery. Surg Clin North Am. 2003;83(6):1429–44. Hsieh MC, et al. Initial experience of robotic gastrec-
PubMed PMID: 14712877. tomy and comparison with open and laparoscopic
46. Marano A, Choi YY, Hyung WJ, Kim YM, Kim J, gastrectomy for gastric cancer. J Gastrointest Surg.
Noh SH. Robotic versus laparoscopic versus open 2012;16(7):1303–10. PubMed PMID: 22450954.
 Laparoscopic Proximal
Gastrectomy with Double 23
Tract Anastomosis

Yukinori Kurokawa, Noriko Wada, Shuji Takiguchi,

and Yuichiro Doki

Introduction Operative Indication

Proximal gastrectomy is one of the modified Laparoscopic proximal gastrectomy is per-

surgical approaches for early gastric cancer formed for clinical T1 tumors (within the sub-
located in the upper stomach without lymph node mucosal layer) located in the upper stomach
metastasis. It allows for storage, digestion, and without lymph node metastasis in the preopera-
absorption of food and prevents agastric anemia. tive assessment. Two clips to mark the distal
Important things in proximal gastrectomy are the side of the lesion and negative biopsy are rec-
curability and postoperative quality of life. As for ommended preoperatively to determine the tran-
the curability, the range of dissected lymph nodes section line of the stomach. The extent of lymph
should follow the Japanese Gastric Cancer node dissection should be D1+ (station Nos. 1,
Treatment Guidelines (ver. 3) in principle [1], 2, 3a, 4sa, 4sb, 7, 8a, 9, 11p) according to the
and the distance of tumor from the gastric stump Japanese Gastric Cancer Treatment Guidelines
is also important. As for quality of life, it is nec- (ver. 3) [1].
essary to consider the reconstruction method that
reduces reflux esophagitis. Recently, the develop-
ment of instruments and techniques has enabled Surgical Procedures
the performance of laparoscopic or laparoscopy-
assisted proximal gastrectomy. The important Trocar Insertion
points of the surgical technique of laparoscopic
proximal gastrectomy with double tract anasto- The patient is placed in the supine Trendelenburg
mosis are described in this section. position with legs apart. Five trocars are used as
shown in Fig. 23.1. The surgeon stands on the
patient’s right, the assistant stands on the patient’s
Electronic supplementary material Supplementary
material is available in the online version of this left, and the camera operator stands between the
chapter at 10.1007/978-3-319-09342-0_23. Videos can patient’s legs. The trocar for camera is inserted
also be accessed at http://www.springerimages.com/ through the umbilical region. Under a 10 mmHg
videos/978-3-319-09341-3. CO2 pneumoperitoneum, the other four trocars
Y. Kurokawa, MD (*) • N. Wada, MD are inserted. After an inspection of the abdominal
S. Takiguchi, MD • Y. Doki, MD cavity, the round ligament of liver is hung with
Department of Gastroenterological Surgery,
Osaka University Graduate School of Medicine,
string. A retractor is inserted through a pinhole
2-2-E2 Yamadaoka, Suita, Osaka 565-0871, Japan incision at the epigastrium to retract the left lobe
e-mail: [email protected] of the liver.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 235
DOI 10.1007/978-3-319-09342-0_23, © Springer International Publishing Switzerland 2015
236 Y. Kurokawa et al.

5 mm 5 mm

12 mm
12 mm
12 mm

Fig. 23.2 An avascular area of the omentum approxi-

mately 3 cm apart from the gastroepiploic arcade is dis-
sected toward the inferior pole of the spleen

Fig. 23.1 Trocar placement and a mini-laparotomy site

for laparoscopic proximal gastrectomy. Five trocars are
placed in the abdominal wall. The trocar site on the left
rectus muscle is extended to 5 cm when the stomach is
pulled out of the abdominal cavity

Dissection of the Left Gastrocolic

Ligament (Video 23.1)

The assistant holds the left side of the gastroepi-

ploic arcade in the right hand and holds the gas-
trocolic ligament and stretches it to caudal in the
Fig. 23.3 The gastrocolic ligament is dissected toward
left hand. The surgeon holds the right side of the the descending portion of duodenum
gastroepiploic arcade in the left hand and dissects
an avascular area of the omentum approximately
3 cm apart from the gastroepiploic arcade toward the LCS in the right hand (Fig. 23.3). The right
the inferior pole of the spleen using a laparo- gastroepiploic vessels are carefully preserved. If
scopic coagulating shears (LCS) in the right hand there are abnormally enlarged lymph nodes along
(Fig. 23.2). In order to mobilize the stomach the right gastroepiploic vessels or infrapyloric
freely, the physiological adhesions between the lymph nodes, the operative method needs to be
pancreas and the posterior wall of the stomach changed to a total gastrectomy, considering the
should be separated properly. possibility of lymph node metastasis.
To dissect the right gastrocolic ligament, the
standing position of the surgeon and the assistant
is changed. The assistant holds the gastroepiploic Lymph Node Dissection Along the Left
arcade in both hands. The surgeon holds the gas- Gastroepiploic Artery and the Short
trocolic ligament and stretches the gastrocolic Gastric Artery (Video 23.2)
ligament in a caudal direction in the left hand.
The surgeon dissects the gastrocolic ligament The assistant grips and lifts the posterior wall of
toward the descending portion of duodenum with the upper stomach in the right hand and draws
23 Laparoscopic Proximal Gastrectomy with Double Tract Anastomosis 237

Fig. 23.4 The fat tissue around the inferior pole of the Fig. 23.6 The base of the esophageal cardiac branch of
spleen is drawn to identify the left gastroepiploic vessels the left inferior phrenic artery is dissected

gastrosplenic ligaments around the superior pole

of the spleen are short, complete dissection
should be performed during resection of the
esophagus.

Resection of the Abdominal

Esophagus (Video 23.3)

The assistant grips and stretches the lesser

curvature in both hands. If there are enlarged
suprapyloric lymph nodes, the operative method
needs to be changed to the total gastrectomy.
Fig. 23.5 The gastrosplenic ligaments is divided upward The surgeon dissects an avascular area of the
toward the superior pole of the spleen lesser curvature along the hepatic branch of the
vagus nerve toward the abdominal esophagus.
the fat tissue near the inferior pole of the spleen The operating surgeon dissects the esophageal
to the left and expands the field of view in the diaphragmatic ligament and exposes the right
left hand. The surgeon identifies the left gastro- crus and the front wall of the abdominal esopha-
epiploic artery rising from the tail of the pan- gus. The fat tissue surrounding the abdominal
creas, peels off the surrounding fat tissue, and esophagus is peeled off, and the anterior and
dissects the artery distal to the omental branch posterior branches of the vagus nerves are tran-
using the LCS after clipping the artery sected using the LCS. The base of the esopha-
(Fig. 23.4). geal cardiac branch of the left inferior phrenic
The surgeon divides the gastrosplenic liga- artery is dissected using the LCS after clipping
ments upward toward the superior pole of the the vessel (Fig. 23.6). After a detachable vessel
spleen with the LCS (Fig. 23.5). It is of particular forceps are applied to the abdominal esophagus,
concern to avoid injury to the spleen or the short the abdominal esophagus is divided above the
gastric vessels by strong traction when the esophagogastric junction using the LCS
assistant expands the field of view. The short gas- (Fig. 23.7). The residual gastrosplenic ligaments
tric vessels are dissected using the LCS with or are dissected downward using the LCS with or
without clipping these vessels. Because the without clipping of vessels.
238 Y. Kurokawa et al.

Fig. 23.7 The abdominal esophagus is divided above the Fig. 23.9 The lymph nodes along the common hepatic
esophagogastric junction after a detachable vessel forceps artery are dissected just above the nerve plexus along the
are applied artery

Lymph Node Dissection Along

the Common Hepatic Artery
(Video 23.5)

The assistant grips and lifts the fat tissue on the

upper border of the pancreas including the left
gastric artery and stretches the gastropancreatic
ligament in the right hand. The assistant gently
pushes down the pancreatic body in the left
hand holding gauze to show the superior border
of the pancreas. The surgeon dissects the perito-
neum along the upper border of the pancreas
from the right to the left side of the abdomen. If
the left gastric vein appears, it is clipped and
Fig. 23.8 The fat tissues including the lesser curvature
lymph nodes are dissected upward to clear the stomach at dissected.
the line of transection The surgeon holds and lifts the lymph
nodes along the common hepatic artery in the
left hand and dissects it just above the nerve
Dissection of the Lesser Curvature plexus along the common hepatic artery using
(Video 23.4) the LCS (Fig. 23.9 ). The surgeon identifies
and exposes the base of the left gastric artery
The lesser curvature is completely dissected and transects it using the LCS after double
between the terminal branches of the left gastric clipping the vessel (Fig. 23.10 ). The surgeon
artery and right gastric artery. The fat tissues dissects the lymph nodes along the common
including the lesser curvature lymph nodes are hepatic artery from the front of the common
dissected upward carefully using the LCS to clear hepatic artery in a cephalad direction
the stomach at the line of transection of the lesser including the lymph nodes along the celiac
curvature of the stomach (Fig. 23.8). artery.
23 Laparoscopic Proximal Gastrectomy with Double Tract Anastomosis 239

Fig. 23.10 The base of the left gastric artery is exposed

and transected after double clipping the vessel

Lymph Node Dissection Along

the Splenic Artery (Video 23.6)

First, the surgeon dissects the retroperitoneum

from the crus to the pancreas body. The operator
holds and lifts the lymph nodes along the splenic
artery in the left hand and dissects the nodes just
above the nerve plexus along the splenic artery
toward the pancreatic tail using the LCS. The sur-
geon identifies the base of the posterior artery and
dissects it using the LCS after clipping the vessel.
The extent of dissection along the splenic artery is
from its origin to halfway of the splenic artery.
Fig. 23.11 Schema of double tract anastomosis
Then, the whole stomach can be mobilized freely.

Reconstruction (Video 23.7)

Resection of the Stomach
There are mainly three reconstruction methods,
A skin incision (5 cm) is made at the trocar site esophagogastric anastomosis, jejunal interposi-
on the left rectus muscle of abdomen, and the tion [2], and double tract anastomosis [3]. The
stomach is pulled out of the abdominal cavity optimal method has not been established. We
through the mini-laparotomy site. Confirming the prefer a double tract anastomosis method due to
clips which marked the lesion by palpation, the the reasons as follows: the anastomosis can be
stomach is cut at the distal side of the clips with performed even if the remnant stomach is small;
the autosuture linear stapler. Then, the stomach is part of food passes the duodenum; and the rem-
removed. In case of intracorporeal resection, the nant stomach can be checked postoperatively
marking clips should be confirmed by intraopera- using esophagogastroduodenoscopy (Fig. 23.11).
tive esophagogastroduodenoscopy before cutting For the double tract anastomosis, esophagoje-
the esophagus. junostomy is performed with the purse-string
240 Y. Kurokawa et al.

Fig. 23.12 The esophageal stump is sewn over with Fig. 23.14 Esophagojejunostomy is performed using a
interrupted sutures laparoscopically to fix the anvil of a circular stapler introduced into the abdominal cavity
circular stapler through the mini-laparotomy site

Fig. 23.13 The anvil of a circular stapler is inserted into Fig. 23.15 A side-to-side gastrojejunostomy is per-
the esophageal stump formed using a linear stapler of which the forks are
inserted into the holes in the anterior wall of the greater
curvature of the remnant stomach and the jejunal limb
suture method as previously reported [4, 5]. The
esophageal stump is sewn over with interrupted
sutures laparoscopically or by using a device Gastrojejunostomy is performed at a site
called the Endostitch (Fig. 23.12), and the anvil of 25 cm below the esophagojejunostomy. Small
a circular stapler is inserted into the esophageal enterotomies are made in the anterior wall of the
stump (Fig. 23.13). The purse-string suture is tied greater curvature of the remnant stomach and the
and reinforced with a monofilament pretied loop. jejunal limb. A linear stapler is introduced into
The jejunum is transected at a point about 20 cm the abdominal cavity, the forks are inserted into
from the ligament of Treitz. A circular stapler is the holes, and a side-to-side gastrojejunostomy is
inserted into the distal side of jejunum and is performed (Fig. 23.15). The entry hole is closed
introduced into the abdominal cavity through the by laparoscopic hand-sewn technique or standard
mini-laparotomy site, and esophagojejunostomy hand-sewn through the mini-laparotomy site.
is performed (Fig. 23.14). Anastomotic leaks are Finally, jejunojejunostomy is performed at a site
evaluated using air insufflation. of 20 cm below the gastrojejunostomy.
23 Laparoscopic Proximal Gastrectomy with Double Tract Anastomosis 241

Drain Insertion References

The abdominal cavity is washed with saline. After 1. Japanese Gastric Cancer Association. Japanese clas-
sification of gastric carcinoma – 3rd English edition.
confirmation of no bleeding under a pneumoperi- Gastric Cancer. 2011;14:101–12.
toneum, a drain is inserted near the upper part of 2. Katai H, Sano T, Fukagawa T, Shinohara H, Sasako
the pancreas through the right subcostal trocar. M. Prospective study of proximal gastrectomy for
early gastric cancer in the upper third of the stomach.
Br J Surg. 2003;90:850–3.
3. Ahn SH, Jung DH, Son SY, Lee CM, Park DJ, Kim
Short-Term Outcomes in Our HH. Laparoscopic double-tract proximal gastrectomy
Institute for proximal early gastric cancer. Gastric Cancer.
2014;17:562–70.
4. Takiguchi S, Sekimoto M, Fujiwara Y, Miyata H,
Between November 2011 and November 2013, Yasuda T, Doki Y, Yano M, Monden M. A simple
we have performed laparoscopic proximal gas- technique for performing laparoscopic purse-string
trectomy with double tract anastomosis for suturing during circular stapling anastomosis. Surg
13 patients with clinical T1 gastric cancer at the Today. 2005;35:896–9.
5. Wada N, Kurokawa Y, Takiguchi S, Takahashi T,
Osaka University Hospital. The mean operation Yamasaki M, Miyata H, Nakajima K, Mori M, Doki
time was 274 min, and the mean blood loss was Y. Feasibility of laparoscopy-assisted total gastrec-
127 mL. According to the Clavien-Dindo classi- tomy in patients with clinical stage I gastric cancer.
fication, there were one grade II pancreatic fistula Gastric Cancer. 2014;17:137–40.
and one grade III anastomotic leakage complica-
tions. There were no treatment-related deaths or
grade IV complications.
 Laparoscopy-Assisted Total
Gastrectomy 24
Nobuhiko Tanigawa, Sang-Woong Lee,
and George Bouras

Patient Positioning way of the ports. Two seats are placed on either
side of the patient for the assistants. Two video
The patient is positioned supine on a Maquet monitors are positioned on either side of the
operating table (Maquet, Germany), with the patient’s head facing inward toward the primary
right arm adducted and the left arm abducted to operator (Figs. 24.1 and 24.2) [1].
90°. Pneumatic compressors are attached, and
legs are bandaged to the lower limb supports,
which are abducted and hyperextended to make Port Placement
space for the primary surgeon who stands in
between the patient’s legs. The patient is tilted Entry into the abdomen is gained through a 2 cm
head-up in a reverse-Trendelenburg position. vertical skin incision that is made just above the
Cardiac monitor electrodes are placed away from umbilicus and a 12 mm Ethicon Excel blunt port
the ventral abdomen so that they don’t get in the is inserted. The supraumbilical optical port serves
as a reference point for insertion of all other
ports, two on either side of patient’s abdomen.
Electronic supplementary material Supplementary Both operative ports are placed 2 cm above the
material is available in the online version of this umbilical port and at a handbreadths distance lat-
chapter at 10.1007/978-3-319-09342-0_24. Videos can erally (12 mm right-hand and 5 mm left-hand
also be accessed at http://www.springerimages.com/
videos/978-3-319-09341-3.
port). Two further ports (both 5 mm) for the
assistants are placed further superiorly near the
N. Tanigawa, MD, FACS (*)
costal margins, between the mid-clavicular and
Department of Surgery,
Tanigawa Memorial Hospital, anterior axillary lines on the patient’s right and
16-59, Kasuga 1-Chome, Ibaraki, more laterally on the mid-axillary line on the
Osaka 567-0031, Japan patient’s left ensuring that the ports don’t clash
e-mail: [email protected]
with the operative ports [2–4]. The umbilical
S.-W. Lee, MD, PhD wound is extended 3 cm superiorly along the
Department of General and Gastroenterological
midline for proximal procedures such as total and
Surgery, Osaka Medical College,
Takatsuki, Osaka, Japan proximal gastrectomy that require esophagojeju-
nal anastomosis and insertion of the circular sta-
G. Bouras, BMBS, BMedSci, FRCS
Department of Surgery and Cancer, pler through a wound protector in the umbilical
Imperial College London, London, UK wound (Fig. 24.3) [5].

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 243
DOI 10.1007/978-3-319-09342-0_24, © Springer International Publishing Switzerland 2015
244 N. Tanigawa et al.

Fig. 24.1 Patient

positioning

Fig. 24.2 Operating room

setup (From Tanigawa [1]
with permission)
24 Laparoscopy-Assisted Total Gastrectomy 245

Fig. 24.3 Port positioning (From Tanigawa [1]) with permission)

Ports are first inserted into the right side of middle portion of the Penrose drain passed from
the abdomen (5 mm upper lateral port then beneath mounted on a curved Karl-Storz grasper
5 mm lower operative port) followed by the left facing upward. Occasionally (especially with
side (5 mm upper lateral port then 12 mm lower large left lobe of liver), the Penrose drain cannot
operative port). The long needle is inserted be passed easily, and the black tie needs to be
through the skin at the marked site and perito- mounted on the tip of an Endo-Mini retractor to
neal entry in the correct position and direction bring it through.
is confirmed. The needle is then withdrawn, and Once the middle portion is delivered over the
the skin is incised to accommodate the trocar. liver, the white tie is placed to the right and the
Artery forceps are used to separate the muscle dye-stained tie to the left of the patient. Three
before inserting the trocar in the desirable small punctures are made in the epigastric skin
direction while avoiding visceral injury. with a number 11 blade for retrieval of each of
the sutures. The Endo Close is inserted into the
middle hole first and confirmed to enter the abdo-
Liver Retraction men just to the left of the falciform ligament. The
black tie is grasped and brought out to the skin
A 6 mm Penrose drain is prepared outside the where it is clipped secure with a mosquito. The
body with colored sutures placed through it to white tie is brought through the right side skin
provide ties for liver retractions. An incision is hole and similarly the dye-stained tie through the
made on the superior leaf of the left triangular left. The ties are then pulled laterally to lift the
ligament above the left lobe of the liver with the left lobe of the liver. Fine adjustments are made
Opti 2 while the assistants hold down the liver to to achieve adequate exposure for the procedure.
apply tension. A space is created behind between Ties are usually tightened in the order of white
the liver and the diaphragm with atraumatic tie, dye-stained tie, and finally the black tie to
graspers for passage of the middle part of the complete the liver retraction (Figs. 24.4, 24.5,
Penrose drain. The left lobe is then lifted to iden- 24.6, 24.7, 24.8, and 24.9) [1]. There are other
tify the corresponding space from below and the alternative techniques, but this is our preference.
246 N. Tanigawa et al.

Dissection left grasps the gastrocolic ligament at the same

level on the side of the transverse colon. The two
Division of the Gastrocolic Ligament assistants apply tension on the fatty tissue in
to the Left between to allow for the operator standing in
between the patient’s legs to divide the gastro-
The assistant on the patient’s right grasps the colic ligament to enter the lesser sac (bursa
anterior surface of the gastric body near the omentalis). While maintaining a distance of
greater curve, and the assistant on the patient’s 3–4 cm away from the gastroepiploic arcade, the
division of the gastrocolic ligament is continued
proximally toward the patient’s left. Tissue divi-
sion is usually performed using a combination of
the LigaSure and monopolar diathermy.
The assistant on the right retracts the greater
curve in a 10 o’clock position on the video moni-
tor. The assistant on the left retracts the greater
omentum in a 4 o’clock position and adjusts in
real time so that the appropriate amount of ten-
Fig. 24.4 Technique utilized for liver retraction sion is applied on the tissues that are being

Fig. 24.5 Technique utilized for liver retraction (From Tanigawa [1] with permission)

Fig. 24.6 Technique utilized for liver retraction (From Tanigawa [1] with permission)
24 Laparoscopy-Assisted Total Gastrectomy 247

Fig. 24.7 Technique utilized for liver retraction (From Tanigawa [1] with permission)

Fig. 24.8 Technique utilized for liver retraction (From Tanigawa [1] with permission)

Fig. 24.9 Technique utilized for liver retraction (From Tanigawa [1] with permission)
248 N. Tanigawa et al.

Fig. 24.10 Division of left gastroepiploic vessels (From Tanigawa [1] with permission)

divided. Care must be taken not to injure the 4sb lymph nodes dissected on the side of the
transverse colon; the assistant on the left can specimen (Fig. 24.10), [1]. The space beneath the
grasp the tissues close to the colon while protect- gastrosplenic ligament is then entered and an
ing the bowel wall with the grasper as the avascular plane on the greater curve identified.
dissection is progressed toward the spleen. Tissues are divided to reach the gastric wall at
this point to complete the mobilization from this
approach. According to the 13th edition of cancer
Division of the Left Gastroepiploic treatment guidelines, 4sb lymph nodes are con-
Vessels and Dissection of Number sidered as 3rd-tier lymph nodes for gastric antrum
4sb Lymph Nodes lesions, and therefore resection is not mandatory
in all cases.
As the left gastroepiploic pedicle is approached,
the assistant on the right grasps the posterior wall
of the stomach near the greater curve and lifts in Division of the Gastrosplenic
a 10 o’clock position to apply tension to the ped- Ligament and Dissection of Number
icle and gastrosplenic ligament. This helps clar- 4sa Lymph Nodes
ify the anatomy and facilitates dissection around
the root of the left gastroepiploic pedicle. Care Beyond the left gastroepiploic vessels, the
must be taken not to apply too much tension to gastrocolic ligament fuses with the gastrosplenic
avoid traction injury to the spleen. Adhesions to ligament. The gastrosplenic ligament together
the posterior gastric wall or the gastrocolic liga- with short gastric vessels is divided proximally
ment are often present and need to be divided close to the spleen until the fundus of the stomach
sharply with the diathermy or bipolar scissors to is disconnected from the spleen (Fig. 24.11).
improve the exposure. The position of the root of Occasionally, large short gastric arteries are
the left gastroepiploic pedicle is determined from encountered that need to be clipped and coagu-
the inferomedial side (inside the lesser sac) from lated with the LigaSure. Adhesions between the
where the pedicle can be seen arising vertically stomach, spleen, and omentum need to be care-
from the retroperitoneum and running into the fully divided with monopolar diathermy or the
lifted greater curve of the stomach. The perito- LigaSure. Any parenchymal bleeding from the
neal layer is incised, and the vessels are skeleton- spleen caused by traction needs to be avoided as
ized using the Marylands or dissecting forceps. this can be difficult to control. Assistants must take
The vessels are then clipped at their roots and care not to retract on the stomach omentum too
divided by coagulation above the clips with the strongly as there may be adhesions to the spleen.
24 Laparoscopy-Assisted Total Gastrectomy 249

Fig. 24.11 Division of short gastric vessels (From Tanigawa [1] with permission)

Posterior Mobilization mesocolon. The right side assistant adjusts the

of the Proximal Stomach ventral retraction to apply the right amount of
tension on the fat to facilitate division. As the
Adhesions between the retroperitoneum and the transverse mesocolon becomes tented up, it must
posterior wall of proximal stomach are divided to be separated from the omentum with a combina-
free the stomach and expose the left gastric pedi- tion of blunt dissection with graspers and sharp
cle from the left side. Gastrodiaphragmatic dissection with monopolar diathermy to allow it
attachments including those to the left crus of to drop down. As this is continued, the inner wall
diaphragm are divided. The posterior gastric of the duodenal C-loop on 2nd part is reached
artery and the left inferior lateral diaphragmatic with the right gastroepiploic pedicle still attached.
artery need to be identified and clipped prior to The membrane between the duodenum and the
division with the LigaSure [6]. pancreas is divided back toward the pylorus
defining the lateral limit of number 6 lymph node
dissection [1–3, 5–10].
Dissection of 4d Lymph Nodes

Division of the gastrocolic ligament in the same Dissection of Number 6 Lymph Nodes
plane parallel to the gastroepiploic arcade is con-
tinued distally toward the patient’s right. The The position of the pylorus is determined anteri-
assistant on the right grasps the greater curve of orly, and the dissection is commenced over the
the stomach and lifts in a 12 o’clock direction to anterior surface of the pancreas to identify the
create space behind the stomach and apply ten- right gastroepiploic vessels and dissect the sur-
sion on the undivided distal part of the gastro- rounding lymph nodes. The right side assistant
colic ligament. Adhesions to the anterior surface lifts the right gastroepiploic pedicle with perigas-
of the pancreas are divided to allow further lifting tric fat near the pylorus ventrally to apply tension
of the stomach and dissection toward the duode- on the vessels. The root of the right gastroepi-
num. As the gastrocolic ligament is divided, the ploic artery is usually found 2 cm below the pylo-
plane between the transverse mesocolon and the rus in this configuration. The level of the head of
greater omentum is identified, and any adhesions the pancreas and the inferior border of the body
across this space are divided. The peritoneal are used for orientation during this dissection. As
reflection at the right lower limit of the lesser sac the plane anterior to the transverse mesocolon is
is then reached, and the fat is separated beyond followed superiorly toward the pancreas, the
this laterally in the plane anterior to the transverse accessory right colic vein and gastrocolic trunk
250 N. Tanigawa et al.

Fig. 24.12 Division of right gastroepiploic vessels (From Tanigawa [1] with permission)

can be seen within the mesenteric fat. Going over duodenum can be separated toward the pylorus.
the anterior surface of the pancreas, the anterior Further posterior mobilization along the gastro-
superior pancreaticoduodenal vein comes into duodenal artery leads to the identification of the
view and can be followed to its drainage into the branching of the hepatic arteries defining the
right gastroepiploic vein. Once skeletonized, the upper limit of the dissection from this approach.
right gastroepiploic vein is clipped above the Several arterial branches to the lesser curve of the
anterior superior pancreaticoduodenal tributary, duodenum can also be opportunistically sealed
and the LigaSure V or Harmonic can be used to and divided from this direction [1–3, 5–10].
divide above the clip. The anterior surface of the
pancreas can then be exposed further cranially in
search of the artery. Care must be taken not to Duodenal Transection
damage the pancreas as its head is tented up by
the retraction on the artery. Energy devices in A gauze is packed behind the pylorus to help with
particular can cause thermal injury here and dur- mobilization of the lesser curve. The hepatoduode-
ing dissection around the common hepatic and nal ligament is divided above the duodenal cap by
splenic arteries. Any damage can lead to postop- incising the peritoneal layer with monopolar dia-
erative inflammation. At our unit, we apply ice- thermy. In thin patients, the gauze may be visible
cooled wet swabs on the surface of the pancreas through the transparent membrane, and the incision
during dissection in an attempt to minimize can be made in an avascular area. A wide window is
injury. As the fat is divided and the duodenal created by dividing the vascular tissues along the
neck becomes mobilized laterally, small vessels lesser curve using the bipolar diathermy or coagula-
and fibers running between the pancreas and duo- tion devices. Once the duodenal neck is completely
denum require pre-coagulation with the LigaSure mobilized, the linear stapler is inserted for duodenal
V or Marylands before division. Once exposed, transection distal to the pylorus (Fig. 24.13).
the gastroduodenal artery on the pancreas can be
followed to the origin of the right gastroepiploic
artery. Approached from both sides, the artery Reinforcement of the Stapled
can be dissected, clipped at its root, and divided Duodenal Stump
above the clip (Fig. 24.12). The infrapyloric
artery running to the pylorus can often be identi- The staple line is buried with interrupted sero-
fied here and can either be clipped or coagulated. muscular 3/0 Vicryl. Once the corners are buried,
Once more space is created posterior to the duo- 1–2 more sutures are usually enough to bury the
denum, the remaining tissues connected to the middle portion of the staple line [3].
24 Laparoscopy-Assisted Total Gastrectomy 251

Fig. 24.13 Duodenal transection (From Tanigawa [1] with permission)

Division of the Right Gastric Artery Division of the Lesser Omentum

and Dissection of Number 5 and the Peritoneum Overlying
Lymph Nodes the Right Crus

The operator together with foot pedals shifts to The hepatogastric ligament forming the
the right side of the patient. The assistant stands membranous proximal part of the lesser omen-
in between the patient’s legs and holds the cam- tum is divided proximally toward the abdominal
era with the left hand and an atraumatic grasper esophagus. The LigaSure and Harmonic are used
in the right hand. The peritoneal layer of the hep- to divide fatty layers, while membranes are
atoduodenal ligament is already divided and the divided by monopolar diathermy. In 15–20 % of
common hepatic artery exposed from the dissec- cases, an accessory left hepatic artery runs from
tion from below. The dissection can be continued the lesser curve of stomach to the liver [11]. This
toward the porta hepatis to expose the border of artery needs to be dissected and its size and con-
the hepatic artery proper with a combination of tribution to hepatic circulation estimated. In most
coagulation and sharp dissection. Large nerve cases, this artery can be divided. Once the upper
fibers are present in this area and must be distin- limit near the hiatus is reached, the peritoneal
guished from lymphatics and small vessels. membrane overlying the right crus of the dia-
Inferiorly, the plane between the common hepatic phragm is divided to define the superior and right
artery and number 8a lymph nodes is separated lateral limit of dissection of number 9 lymph
cranially. The left side assistant retracts the right nodes in front of the celiac trunk.
gastric pedicle in a 2–3 o’clock direction to facil-
itate dissection around its root. The retraction
can, however, tent up the hepatic artery itself, so Dissection Above the Pancreas
care must be taken not to divide this. The right (Lymph Nodes 8a, 7, 11p, 11d, and 9)
gastric artery can also arise from the left hepatic
artery. Anatomical variations are frequent here, The operator moves to the patient’s right. The
and care must be taken not to cause any hepatic dissection is continued exposing the common
ischemia. Laterally, lymph nodes 12a and 5 are hepatic artery toward the left gastric artery. The
separated from the hepatic arteries, and the root operator gently grasps the lymph nodes with a
of the right gastric artery is clipped before divi- grasper or dissecting forceps in the left hand and
sion (Fig. 24.14). The right gastric vein is also continues separating the plane between the artery
divided when encountered during this dissection and lymph nodes using the monopolar diathermy
[1–3, 5–10]. in soft coagulation mode. This allows for
252 N. Tanigawa et al.

Fig. 24.14 Division of right gastric vessels (From Tanigawa [1] with permission)

Fig. 24.15 Dissection of nodal tissue along the common hepatic artery (From Tanigawa [1] with permission)

hemostatic division of lymphatic tissues with a sharp division with the cutting mode diathermy.
low risk of major vascular injury. The anterior Repetition of these gestures exposes the entire
surface of pancreas is pushed downward by the surface of the common hepatic artery to complete
assistant to facilitate deeper dissection. Vessels the inferior mobilization of number 8a lymph
running from the pancreas to the lymph nodes nodes (Fig. 24.15). The LigaSure can also be
must be pre-coagulated with the soft coagulation used for simultaneous coagulation and tissue
diathermy or bipolar forceps (Marylands) before division. The left gastric vein may be encountered
24 Laparoscopy-Assisted Total Gastrectomy 253

Fig. 24.16 Left gastric vein coursing anterior to the common hepatic artery (From Tanigawa [1] with permission)

Fig. 24.17 Dissection of level 11p nodes along the splenic artery (From Tanigawa [1] with permission)

toward the pedicle. In cases where this vessel splenic artery is followed until the posterior
drains directly into the portal vein or into the gastric artery is reached. The splenic vein running
junction between the portal vein and splenic vein, posterior to the artery can occasionally be
it runs posterior to the common hepatic artery, visualized during this part of the dissection. The
whereas in other cases where the vein drains into loose space next to the left gastric pedicle can be
the splenic vein, it loops anterior to the common opened and the lymphatic tissues divided laterally
hepatic artery (Fig. 24.16). Rarely, the vein runs to dissect number 11p lymph nodes off the splenic
posterior to the splenic artery draining into the artery (Fig. 24.17). Then, attachments to the left
splenic vein. Once identified, the left gastric vein crus are divided to define the left lateral limit of
is clipped and divided. the celiac lymph node dissection. Once tissues are
The left gastric pedicle is grasped near the gas- separated from the crura on both sides, the poste-
tric wall and retracted ventrally with the opera- rior limit of number 9 dissection is defined. The
tor’s left-hand grasper. The dissection on the root of the left gastric artery can then be dissected
common hepatic artery is continued in front of the by dividing the fibrous nerve bundles around the
left gastric artery by dividing fibers running vessel. The artery is then clipped and divided with
between the pancreas and the vessels. The dissect- number 7 lymph nodes on the side of the stomach
ing forceps are used to create gaps in the lym- (Fig. 24.18). A cut finger end of a surgical glove
phatic tissue then the LigaSure V or Harmonic is (Sensi-touch 8.0) is inserted into the abdomen
used to coagulate and cut simultaneously. The through the 12 mm port, and the dissected lymph
254 N. Tanigawa et al.

Fig. 24.18 Division of left gastric artery (From Tanigawa [1] with permission)

Fig. 24.19 Circumferential mobilization of gastroesophageal junction (From Tanigawa [1] with permission)

node is placed inside for retrieval from the right crus, the posterior vagus is divided leading
abdominal cavity A [1–3, 5–10]. to further lengthening of the abdominal part of the
esophagus (Fig. 24.19). The Endo-Mini retractor
is then passed behind and above the angle of His
Hiatal Mobilization of the Abdominal to confirm circumferential mobilization of the
Esophagus gastroesophageal junction [5, 6, 8, 12].

Bearing in mind that the proximal transection will

be on the esophagus, the anterior surface of the Insertion of Anvil and Esophageal
esophagus is approached from the right and freed Transection
from the diaphragm taking lymph nodes around
the hiatus on the side of the stomach. The anterior The nasogastric tube is withdrawn in preparation
vagus nerve running longitudinally on the anterior for the insertion of the anvil of the circular stapler
surface of the esophagus is divided distal to the prior to esophageal transection and subsequent
origin of the hepatic branch preserving the nerve esophagojejunal anastomosis.
supply to the liver. As the dissection is continued The authors have employed the hemi-double
posteriorly separating the esophagus from the stapling technique at esophagojejunostomy,
24 Laparoscopy-Assisted Total Gastrectomy 255

Fig. 24.20 Inserting the anvil through gastrotomy (From Tanigawa [1] with permission)

because it is simple without need of suturing holder in the operator’s right hand, and the anvil
techniques and the most familiar to every surgeon is inserted into the abdominal esophagus
who has some experience of open gastrectomy. (Fig. 24.20). A rotating movement often facili-
Insertion of the anvil is accomplished by two tates insertion. The anvil is pushed in proximally
ways, one through gastrotomy which is made in so that there is adequate distal esophagus beyond
the anterior wall of the fundus and another the anvil rod for transection with the linear sta-
through the mouth by the use of OrVil Tilt-top. pler. A grasper is used to palpate and confirm that
the anvil rod is above the esophageal transection
Insertion of Anvil Through line. The Vicryl tie should still be within the tran-
Gastrostomy (Video 24.1) section line at this stage. The esophagus is then
Using monopolar diathermy, a full-thickness gas- transected with the Echelon 60–3.5 blue inserted
trotomy is made on the lesser curve slightly anteri- through the left lower 12 mm port (Fig. 24.21).
orly near the gastroesophageal junction to enter the Slight angulation of the staple line is appropriate
lumen. The hole is extended longitudinally to about when transection is complete (Fig. 24.22).
3 cm enough for insertion of the 25 mm anvil. If the Once divided, the purple color of the Vicryl tie
tumor is located on the proximal lesser curve, the can be seen on the staple line. This is pulled so
entry hole is made anteriorly to avoid the tumor. A that the center rod abuts the staple line. While
straight atraumatic grasper is inserted into the esoph- pulling on the Vicryl, the stapled line is incised
agus to confirm the direction of anvil insertion. with monopolar diathermy onto the center rod
The anvil of a PCEAA 25 mm stapler is pre- until the metal rod becomes visible and can be
pared outside the body by attaching a 2/0 Vicryl grasped and pulled out of the esophagus
tie onto the center rod. As the Vicryl tie needs to (Fig. 24.23). Once the anvil is in position, the
be removed once the anvil is in position, it is tied Vicryl is cut and removed so that it does not get
loosely so that the tie can be easily cut laparo- in the way when combining the anvil with the cir-
scopically. The length of the tie is left at 5 cm so cular stapler.
that it can be identified once inside the body. The
pneumoperitoneum is temporarily stopped, the Anvil Insertion Through Mouth
umbilical port is removed, and the anvil is placed By the use of OrVil Tilt-top, the anvil is placed
into the abdomen. The umbilical port is then rein- into the abdominal esophagus through the mouth
serted and pneumoperitoneum reestablished. The by pulling down the tube, which is connected
center rod of the anvil is grasped with the anvil with the center rod (Fig. 24.24).
256 N. Tanigawa et al.

Fig. 24.21 Esophageal transection (From Tanigawa [1] with permission)

Fig. 24.22 Vicryl suture tied to the anvil identified post-esophageal transection (From Tanigawa [1] with permission)

Fig. 24.23 Vicryl suture utilized to retrieve the anvil (From Tanigawa [1] with permission)
24 Laparoscopy-Assisted Total Gastrectomy 257

the mini-laparotomy after which the marked jeju-

num is delivered outside [9, 13].

Division of the Jejunum

From the point on the jejunum marked with dye,

the jejunal mesentery is divided for 10 cm dis-
tally using the LigaSure to coagulate vessels. This
bowel becomes ischemic and is the sacrificed por-
tion of jejunum. Another way of creating a gap
in the jejunal mesentery without sacrificing the
bowel is to cut into the mesentery toward its root.
Once ready, the jejunum can be transected distal
to the sacrifice jejunum using the Echelon 60–3.5
blue.

Insertion of the Anvil into

Fig. 24.24 Insertion of anvil through mouth (OrVil TM
) the Jejunum at 20 cm
(From Tanigawa [1] with permission) from the Ligament of Treitz

To perform the jejunojejunostomy using the

Marking of the Jejunum PCEEA 21 mm circular stapler, a purse-string
applicator is placed on the healthy bowel just
The two assistants lift and spread the transverse proximal to the sacrifice jejunum on the oral side.
mesocolon so that the operator can identify the The redundant bowel is removed with the mono-
ligament of Treitz. Using a 10 cm measuring polar diathermy for external use, and a 2/0
tape, a distance of 20 cm is measured from the Prolene with a straight needle on either end is
ligament of Treitz, and the bowel wall is marked used as the purse-string suture. Once in position,
with a dye-stained Endo-peanut. A metal clip is the needles are cut off, and the purse-string appli-
placed immediately distally to avoid confusion. cator is withdrawn. The anvil is then placed into
The left side assistant grasps this part of the the bowel lumen, and the purse-string is tight-
bowel while insufflation is stopped in preparation ened around the center rod of the anvil.
for the open part of the procedure.

Reinforcement of the Staple Line

Extension of the Umbilical Wound on the Divided Jejunum and Creation
(Mini-laparotomy) and Specimen of the Enterotomy for Insertion
Retrieval (Video 24.2) of the Circular Stapler

The umbilical port wound is extended to 4 cm The staple line on the divided distal jejunum of
superiorly using the diathermy onto the shaft of the alimentary limb is reinforced with interrupted
the port while it is still inserted. Once the wound seromuscular 3/0 Vicryl stitches. The last suture
is adequately large, the Lap Protector is inserted on the mesenteric edge of the stump is left long
into the wound. Two towels are placed between at 2 cm so that it can be grasped to maneuver
the wound protector and the skin to prevent the bowel intracorporeally during subsequent
contamination. The specimen is removed through esophagojejunostomy.
258 N. Tanigawa et al.

After measuring 20 cm from the reinforced

jejunal stump, a 25 mm enterotomy is made on
the anti-mesenteric side of the jejunum. This is
used to insert the circular stapler distally for the
jejunojejunostomy and proximally for the esoph-
agojejunostomy. Furthermore, this hole serves as
the site where the subsequent gastrojejunostomy
is created as part of double-tract reconstruction
following proximal gastrectomy.

Extracorporeal Jejunojejunostomy
(Using Circular Stapler)

The PCEEA 21 mm circular stapler is inserted

into the enterotomy created. The shaft of the sta-
pler is fed distally to a point 20 cm from the
enterotomy. The center pin of the stapler is
brought out to pierce the bowel wall here, and the
stapler is connected to its corresponding anvil
already secured in the biliopancreatic jejunum
(on the side of the ligament of Treitz). The stapler
is fired to create a stapled side-to-end jejunojeju-
Fig. 24.25 Rubber band applied to the shaft of the stapler
nostomy. Reinforcement of the staple line is not to stabilize for esophagojejunostomy (From Tanigawa [1]
performed; however, any bleeding points are con- with permission)
trolled with full-thickness hemostatic sutures
across the staple line. The gap in the mesentery is heavy Vicryl tie so that the jejunum does not slip
closed with continuous 3/0 Vicryl to avoid inter- on the shaft during the anastomosis (Fig. 24.25).
nal herniation between the cut edges of the small The whole apparatus including the jejunum con-
bowel mesentery. The completed jejunojejunos- nected to the stapler is inserted through the Lap
tomy is then pushed into the abdomen [3, 14]. Protector taking care not to cause any visceral
damage inside the abdomen. The glove on the
shaft is brought over the Lap Protector and fixed
Laparoscopic Esophagojejunostomy with three arterial clips to seal the gap for pneu-
(Using Circular Stapler) (Video 24.3) moperitoneum. Insufflation is commenced
through the left lower port through which the
The end (2 cm) of the middle finger of a number camera is also inserted. Using graspers, the cen-
8 surgical glove is cut off, and the glove is fed ter rods on the anvil and circular stapler are com-
over the shaft of the 25 mm PCEEA circular sta- bined until securely clicked into position. The
pler through the finger hole. The head of the cir- stapler is slowly closed ensuring that no tissues
cular stapler is then fed into the enterotomy on are caught in the anastomosis. The left-hand
the jejunum created earlier and passed proxi- assistant helps by pulling on the Vicryl on the
mally toward the stump. The center pin is brought jejunal stump for retraction. Once closed and
out to pierce the bowel wall on the anti-mesenteric operators are satisfied, the circular stapler is fired.
side near the stump. A window is made in the The rubber band is cut off and the stapler gently
mesentery 2 cm away from the end to pass a cut removed with some rotation so that it smoothly
rubber band through to the other side. The rubber slips out from inside the bowel lumen (Fig. 24.26).
band is pulled over the shaft and secured with a Pneumoperitoneum is stopped again, and the
24 Laparoscopy-Assisted Total Gastrectomy 259

Fig. 24.26 Completed esophagojejunostomy (From Tanigawa [1] with permission)

glove is detached from the wound protector to

completely free the stapler.
Another method for creation of the esophago-
jejunostomy can be performed through division
of the jejunum and insertion of the stapler through
the end of the distal limb and advancing the head
of the circular stapler distally toward healthy
jejunum. By the use of a rubber band, the jejunal
wall is fixed on the shaft so that the jejunum does
not slip on the shaft during the anastomosis.
After pneumoperitoneum is established again,
the shaft is introduced into the peritoneal cavity,
and the center pin is brought out to pierce the
bowel wall on the anti-mesenteric side. Using
graspers, the center rods on the anvil and circular
stapler are combined until securely clicked into
position. The stapler is slowly closed ensuring
that no tissues are caught in the anastomosis.
After the circular stapler is fired, the rubber band
is cut off and the stapler gently removed with
some rotation so that it smoothly slips out from
inside the bowel lumen.

Closure of the Jejunal Enterotomy

(Used for Insertion of Circular
Staplers) (Video 24.4)

The hole used for insertion of the circular stapler

distally and proximally is closed using a two-layer
Albert-Lembert technique (Fig. 24.27 depicts the Fig. 24.27 Sketch depicting completed anastomoses
completed anastomoses). (From Tanigawa [1] with permission)
260 N. Tanigawa et al.

Drain Insertion and Closure laparoscopic linear staplers: preliminary experience.

Surg Endosc. 2008;22:436–42.
7. Japanese Gastric Cancer Association. Japanese
A 10 mm flat-type Blake drain is inserted through classification of gastric carcinoma. 14th ed. Kanehara
the right lower port, passed between the perito- Co, Ltd; 2010.
neum and abdominal wall and then into the space 8. Ikeda O, Sakaguchi Y, Aoki Y, Harimoto N, Taomoto J,
Masuda T, Ohga T, Adachi E, Toh Y, Okamura T,
just distal to and behind the esophagojejunal
Baba H. Advantages of totally laparoscopic distal
anastomosis. The umbilical wound needs formal gastrectomy over laparoscopically assisted distal
mass closure with 1 Vicryl. Skin closure is per- gastrectomy for gastric cancer. Surg Endosc. 2009;23:
formed with interrupted 3/0 Vicryl subdermal 2374–9.
9. Nomura E, Isozaki H, Fujii K, Toyoda M, Niki M,
stitches with buried knots. Steri-Strip tapes are
Sako S, Mabuchi H, Nishiguchi K, Tanigawa N.
used for accurate apposition of skin edges. Postoperative evaluation of function-preserving gastrec-
tomy for early gastric cancer. Hepatogastroenterology.
2003;50:2246–50.
10. Bouras G, Lee SW, Nomura E, Tokuhara T, Tsunemi S,
References Tanigawa N. Comparative analysis of stration-specific
lymph node yield in laparoscopic and open distal gas-
1. Tanigawa N. A manual for laparoscopic gastric resec- trectomy for early gastric cancer. Surg Laparosc
tion. 1st ed. Osaka: Nagai Co, Ltd; 2009. Endosc Percutan Tech. 2011;21(6):424–8.
2. Shinohara H, Sonoda T, Niki M, Nomura E, 11. Lee SW, Shinohara H, Matsuki M, Okuda J, Nomura E,
Nishiguchi K, Tanigawa N. Laparoscopically-assisted Mabuchi H, Nishiguchi K, Takaori K, Narabayashi I,
pylorus-preserving gastrectomy with preservation of Tangiawa N. Preoperative simulation of vascular anat-
the vagus nerve. Eur J Surg. 2003;169:55–8. omy by three –dimensional computed tomography
3. Takaori K, Nomura E, Mabuchi H, Lee S, Agui T, imaging in laparoscopic gastric cancer surgery. J Am
Miyamoto Y, Iwamoto M, Watanabe H, Tanigawa N. A Coll Surg. 2003;197(6):927–36.
secure technique of intracorporeal Roux-Y reconstruc- 12. Tanigawa N, Nomura E, Lee SW, Kaminishi M,
tion after laparoscopic distal gastrectomy. Am J Surg. Sugiyama M, Aikou T, Kitajima M. Current state of
2005;189:178–83. gastric stump carcinoma in Japan: based on the results
4. Lee SW, Bouras G, Nomura E, Yoshinaka R, Tokuhara of a nationwide survey. World J Surg. 2010;34:1540–7.
T, Nitta T, Tsunemi S, Tanigawa N. Intracorporeal 13. Nomura E, Lee SW, Bouras G, Tokuhara T, Hayashi M,
stapled anastomosis following laparoscopic segmen- Hiramatsu M, Okuda J, Tanigawa N. Functional out-
tal gastrectomy for gastric cancer: technical report comes according to the size of the gastric remnant and
and surgical outcomes. Surg Endosc. 2010;24: type of reconstruction following laparoscopic distal
1774–80. gastrectomy for gastric cancer. Gastric Cancer. 2011;
5. Lee SW, Nomura E, Bouras G, Tokuhara T, Tsunemi S, 14(3):279–84.
Tanigawa N. Long-term oncological outcomes from 14. Lee SW, Nomura E, Tokuhara T, Kawai M, Matsuhashi N,
laparoscopic gastrectomy for gastric cancer: a single- Yokoyama K, Fujioka H, Hiramatsu M, Okuda J,
center experience of 601 consecutive resections. J Am Uchiyama K. Lapascopic technique and initial experi-
Coll Surg. 2010;211:33–40. ence with knotless, unidirectional barbed suture closure
6. Kim JJ, Song KY, Chin HM, Kim W, Jeon HM, Park for staple-conserving, delta-shaped gastroduodenos-
CH, Park SM. Totally laparoscopic gastrectomy with tomy after distal gastrectomy. J Am Coll Surg. 2011;
various types of intracorporeal anastomosis using 213(6):e39–45.
 Robotic Utilization in Gastric
Cancer Surgery 25
Kaitlyn J. Kelly and Vivian E. Strong

Introduction laparoscopically. Controversy exists over the

ability to perform an adequate lymphadenectomy
Utilization of minimally invasive techniques for laparoscopically in cases of locally advanced
gastric cancer surgery has increased in recent years. disease and over the safety of a laparoscopic esoph-
Laparoscopic distal gastrectomy for early-stage, agojejunal anastomosis in total gastrectomy.
distal gastric cancers is well established and is rou- The robotic surgery platform offers several
tinely practiced in the East where gastric cancer technical advantages over laparoscopy. The cam-
screening is routine. More than five randomized, era provides a three-dimensional, magnified,
prospective trials have confirmed improvements in high-definition view that is stable and is con-
short-term outcomes compared to open distal gas- trolled by the primary surgeon. The articulated
trectomy for patients with early-stage disease [1– robotic instruments provide seven degrees of
6]. Laparoscopic resection of locally advanced and freedom and facilitate performance of difficult
proximal gastric cancers, however, is not as well dissection and suturing. These advantages have
studied or widely performed. The two-dimensional led surgeons to investigate the use of the robotic
view provided by the conventional laparoscope and platform for gastrectomy. Robot-assisted gastrec-
limited range of motion of the instruments makes tomy (RG) for gastric adenocarcinoma was first
these complex resections challenging to perform reported in 2003 [7, 8] and was first reported in
the United States in 2007 [9]. Since that time,
Electronic supplementary material Supplementary multiple retrospective series of RG for gastric
material is available in the online version of this adenocarcinoma have been published, almost all
chapter at 10.1007/978-3-319-09342-0_25. Videos can
from the East [10–16]. The conclusions that can
also be accessed at http://www.springerimages.com/
videos/978-3-319-09341-3. be drawn from these retrospective studies are
limited due to great variability in inclusion crite-
K.J. Kelly, MD
Department of Surgery, ria, surgeon experience, type of reconstruction
University of California, San Diego, performed, and the outcomes evaluated.
3855 Health Sciences Drive, This chapter will describe the technical
La Jolla, CA 92093, USA
aspects of RG for gastric cancer and discuss con-
e-mail: [email protected]
siderations regarding the learning curve and
V.E. Strong, MD (*)
patient selection. Additionally, the chapter will
Department of Surgery,
Memorial Sloan-Kettering Cancer Center, summarize current literature on RG for gastric
New York, NY, USA cancer with a focus on outcomes and costs.

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 261
DOI 10.1007/978-3-319-09342-0_25, © Springer International Publishing Switzerland 2015
262 K.J. Kelly and V.E. Strong

Technical Aspects of Robotic

Gastrectomy (Video 25.1)

Patient Positioning and Port

Placement

RAG is performed with the patient in the supine

position on a split-leg table. The patients’ arms
are tucked bilaterally with adequate padding of
elbows and hands to avoid pressure points. The
patient is secured to the table at the shoulders
using foam blocks and heavy-duty adhesive tape
applied circumferentially around the blocks and
10−12 mm 8 mm
the table. Fixation is also applied at the hips with 5 mm 8 mm
a safety belt and circumferentially at the knees. 10 mm
Footboards may also be applied at the feet as fur-
ther means to avoid sliding during reverse
Trendelenburg positioning. Once patient posi-
tioning is completed, it is important to place the
patient in steep reverse Trendelenburg as a test to Fig. 25.1 Recommended port placement for robotic-
assure stability. assisted gastrectomy
Port placement for RAG follows the same
principles as for any robot-assisted procedure extra-gastric disease. If a distal subtotal gastrec-
which include placement of the camera port at a tomy is to be performed and the lesion is not
distance of 15–20 cm from the target anatomy, appreciable on the extraluminal surface, an endo-
placement of robot ports at least 8-cm apart from scope is passed, and the lesion is localized. A silk
each other, and an assistant port at least 5 cm stitch is placed laparoscopically to mark the level
from adjacent robotic ports. While multiple vari- of transection of the stomach that will likely
ations of port placement have been described, the achieve a negative proximal margin. Once this is
placement illustrated in Fig. 25.1 is recom- complete, the patient is placed in steep reverse
mended. Pneumoperitoneum is established with Trendelenburg, and the robot is docked from
a Veress needle just off of the left costal margin. directly over the patient’s head. Arms 1 and 3 are
A 12-mm trocar is then placed in the midline attached to the left-sided ports, and arm 2 is
above or below the umbilicus depending on the attached to the right-sided port within the large
patient’s body habitus but with a goal of port 12-mm port. A fenestrated bipolar grasper is
placement 15–20 cm from the target anatomy. In placed in arm 2, and a harmonic scalpel or mono-
the majority of cases, the infraumbilical position polar scissor is placed in arm 1. A grasping for-
is best. Two additional 8-mm da Vinci ports are cep, preferably a Cardiere, is placed in arm 3.
then placed on the left side, at least 8 cm from
each other and slightly off-set from the plane of
the camera port. An additional 12-mm port is Procedural Steps
placed in the right mid-clavicular line, and an
8-mm robotic port is placed within it. A 5-mm The procedure commences by flipping the greater
assistant port is placed further laterally on the omentum cephalad and locating the transverse
right side, approximately at the anterior colon. The omentum is carefully taken off of the
axillary line. colon proceeding in the direction of the splenic
At this point, the abdomen is explored for flexure. With careful dissection, the omentum is
adhesions and for any evidence of peritoneal or separated from the transverse mesocolon, and the
25 Robotic Utilization in Gastric Cancer Surgery 263

Fig. 25.2 Confluence of right gastroepiploic and right Fig. 25.3 Division of proximal duodenum just distal to
colic veins at anterior border of pancreas pylorus

lesser sac is entered. Visualization of the poste- the suprapyloric region. The gastrohepatic omen-
rior wall of the stomach confirms entry into the tum is incised with hook monopolar cautery or a
lesser sac. The posterior wall of the stomach is harmonic scalpel in arm 1. The right gastric
then grasped by the bedside assistant and is artery is identified and is ligated at its take-off
retracted anteriorly and to the patient’s right side. from the proper hepatic artery with the harmonic
The omentectomy is carried up toward the spleen scalpel. The lymphatic tissue along the hepatic
and is stopped at the edge of the stomach just proper and common hepatic artery is swept medi-
prior to reaching the short gastric vessels in a dis- ally toward the specimen, and a window is cre-
tal subtotal gastrectomy. For a total gastrectomy, ated at the level of the pylorus. The posterior
the omentectomy is carried up to the esophageal aspect of the pylorus and proximal duodenum is
hiatus, and the short gastric vessels are divided. elevated off of the retroperitoneum with a combi-
Once this is complete, the posterior wall of the nation of blunt dissection and use of the harmonic
stomach is grasped with the 3rd arm of the robot scalpel. A blue load of the stapler with bioab-
and is retracted toward the patient’s left shoulder. sorbable reinforcement is then introduced, and
The omentectomy then proceeds toward the the proximal duodenum is stapled and divided
hepatic flexure of the colon and is completed. just distal to the pylorus (Fig. 25.3).
The omentum can be placed in the left upper Once this is complete, the distal stomach can
quadrant on the anterior wall of the stomach at be retracted toward the patient’s left shoulder uti-
this point. The posterior attachments between the lizing robot arm 3. The lymph node dissection
stomach and pancreas are then divided sharply or that was started previously is then continued
with the harmonic scalpel in the direction of the along the common hepatic artery toward the
pylorus. The right gastroepiploic vessels are celiac axis and proximal splenic artery. The left
identified and dissected circumferentially at the gastric artery is identified at the celiac axis and is
level of the anterior border of the pancreas divided at its base with a vascular load of the sta-
(Fig. 25.2). The vessels are divided at their origin pler. The gastrohepatic omentum is further
with a vascular load of a stapler or with clips. If incised up to the level of the esophageal hiatus
the stapler is to be used, arm 2 of the robot with the harmonic scalpel. For distal subtotal
together with its associated 8-mm port is removed gastrectomy, the level 1 and 2 lymph nodes are
from the larger 12-mm port, and the stapler is peeled down off of the proximal stomach down to
passed by the bedside assistant. the level where the stomach will be divided. For
The pylorus is then identified by the vein of a total gastrectomy, the distal esophagus is
Mayo/white line, and attention is turned toward divided with stapler (blue load).
264 K.J. Kelly and V.E. Strong

a b

Fig. 25.4 (a) Creation of stapled side-to-side gastrojejunostomy. (b) Closure of gastroenterotomy

At this point, the specimen is placed in a

specimen retrieval bag and is removed via the
12-mm port site in the right upper quadrant. The
12-mm port is then replaced, and the 8-mm
robotic port attached to arm 2 is placed within it.
Attention is then turned to the reconstruction.
For a distal subtotal gastrectomy in which no
more than half of the stomach was removed, we
prefer an antecolic, Billroth II reconstruction is
preferred. If greater than half of the stomach is
removed or if a total gastrectomy is performed, we
prefer a Roux-Y reconstruction is preferred. The
colon is elevated cephalad, and the ligament of
Treitz is identified. A mobile piece of jejunum
approximately 30-cm downstream is selected and
is used for the reconstruction. For a Billroth II or
Roux-Y reconstruction to a gastric remnant, a
Fig. 25.5 Schematic diagram of stapled esophagojeju-
side-to-side stapled gastrojejunostomy is created
nostomy following total gastrectomy
with a 60-mm laparoscopic stapler. The remaining
enterotomy is sutured closed with a running 3.0
silk stitch with needle drivers in robot arms 1 and Harmonic scalpel. Once the anastomosis is cre-
2 (Fig. 25.4). For an esophagojejunostomy, an ated, the open end of the Roux limb is closed with
end-to-side anastomosis is created with a circular a linear stapler (Fig. 25.5). All staplers are inserted
stapler. To facilitate this, the Orvil of the stapler is via the right upper quadrant 12-mm incision by the
passed transorally on a nasogastric tube which is bedside assistant. For Roux reconstruction, a side-
then pulled through the distal esophagus. The tub- to-side stapled jejunojejunostomy is created
ing is then gently detached from the Orvil and is approximately 70-cm downstream from the proxi-
removed through the 12-mm right upper quadrant mal anastomosis, and the remaining enterotomy is
port. The stapler itself is inserted into the Roux sutured closed. Mesenteric defects are also sutured
limb after removing the staple line with the closed in a running fashion with 3.0 Vicryl.
25 Robotic Utilization in Gastric Cancer Surgery 265

The Learning Curve intestinal-type histology. As a surgeon’s experi-

ence with the procedure increases, the incorpo-
It is hypothesized that the learning curve for RG ration of patients with more advanced disease,
is less than that for LG due to the ergonomic and neoadjuvant treatment, proximal tumors, and
technical advantages provided by the robotic higher BMI is reasonable. It may, in fact, be in
platform. There is evidence suggesting that this is these settings where the robotic platform is most
the case for surgeons already experienced with advantageous over laparoscopy. It is in these sce-
advanced laparoscopy, but the recommended narios where LG with D2 lymphadenectomy is
number of procedures required for learning var- most challenging.
ies. Some authors have suggested 20 cases for We have noted an association between diffuse-
learning RG by advanced laparoscopic surgeons type histology and microscopic proximal margin
[11, 17]. More recently, Kim et al. performed a positivity (R1) in LG and therefore recommend
comprehensive, multidimensional analysis of the selecting patients with intestinal-type histology
learning curve for laparoscopic versus robotic in one’s initial experience with minimally inva-
distal gastrectomy [14]. With their more rigorous sive gastrectomy (unpublished data). LG and RG
statistical analysis of stability in operating time are still feasible in patients with diffuse-type his-
and “surgical success,” they found that 95 cases tology, but a frozen section of the proximal mar-
were required for learning RG, and 270 were gin should be sent intraoperatively in all cases,
required for LG. This was a retrospective analy- and the surgeon should be comfortable with lapa-
sis, however, and the surgeon had completed 177 roscopic or robotic total gastrectomy or with con-
LGs prior to the first robotic case. The number of version to an open procedure if necessary.
robotic cases required might be greater for sur-
geons without prior laparoscopic experience.
There have been no studies to date prospec- Perioperative Outcomes
tively evaluating learning curves from initial sur-
geon experience in RG versus LG. It has been Multiple retrospective series have been published
suggested that experienced open surgeons can evaluating perioperative outcomes following
transition directly to the robotic platform without RG. Three of the largest studies, inclusive of
an intermediate laparoscopic step [18], but for- patients reported in prior smaller studies, are
mal simulation training with the robotic platform summarized in Table 25.1 [10, 11, 15]. It is
with both dry and wet labs is a must. Furthermore, important to note that these studies contained a
one should at least be familiar with laparoscopic predominance of patients with T1–2, N0 disease.
exposure of relevant anatomy and with laparo- An association between decreased blood loss and
scopic tissue handling while still having haptic RG has been shown, but no other measurable
feedback, which is lost with the robotic short-term outcomes have reliably been shown to
platform. be different between RG and LG. The study by
Kim and colleagues was the largest and included
5,839 patients who underwent open (n = 4,542),
Patient Selection laparoscopic (n = 861), or robotic (n = 436) gas-
trectomy for gastric adenocarcinoma [15].
For RG to be performed successfully, good Patients in the open group had more advanced
patient selection is critical. This is especially disease and proximal tumors than those in the LG
true in a surgeon’s initial experience. Ideal can- and RG groups. The authors found no differences
didates for RG are patients with early-stage dis- in overall perioperative morbidity or mortality
ease who have not received neoadjuvant therapy among the groups, although interestingly, the
and those with normal BMI, distal tumors, and types of complications did vary. The open
266 K.J. Kelly and V.E. Strong

Table 25.1 Published studies comparing robotic and laparoscopic gastrectomy for cancer
Positive Length
Op-time Open margin Lymph of stay EBL Morbidity Mortality
Study N (min)a conversion (R1/2) nodes (N) (days) (mL) (%) (%)
Kang et al. (2012) [11]
RG 80 202 ± 52b NR NR NR 10 ± 12 93 ± 85 14 0
LG 282 173 ± 51 NR NR NR 8±4 173 ± 51 10 0
Kim et al. (2012) [15]
RG 436 226 ± 54 NR 1 40 ± 15 8 ± 14 85 ± 160 10 0.5
LG 861 176 ± 63 NR 2 38 ± 14 8±9 112 ± 229 9 0.3
Hyun et al. (2013) [10]
RG 38 234 ± 48 0 0 33 ± 14 10 ± 6 131 ± 10 47.3 0
LG 83 220 ± 61 0 0 33 ± 13 12 ± 10 130 ± 18 38.5 0
RG robotic gastrectomy, LG laparoscopic gastrectomy, NR not reported
a
Data expressed as means ± standard deviation
b
Bolded variables were statistically significantly different

approach was associated with a greater incidence studies was from Italy, and the remaining eight
of postoperative bowel obstruction, ileus, and were from China, Korea, or Japan. In the meta-
abscess formation. The minimally invasive analysis RG was again associated with decreased
approaches were associated with a greater inci- blood loss and increased operative time com-
dence of anastomotic leak. In this study, compli- pared to LG. An association was also observed
cations were tracked out to 30 days postoperatively between RG and a shorter distal margin. There
or to >30 days within the same hospitalization. were no differences in any other short-term out-
Late complications were not reported, and con- comes evaluated, including number of lymph
version rates were not commented on. nodes retrieved, proximal resection margin, rate
One nonrandomized, prospective study of 150 of conversion to open surgery, overall morbidity,
patients undergoing RG (N = 30) or LG (N = 12) has anastomotic leakage or stenosis, intestinal
been reported [17]. In this study, operative time obstruction, time to first flatus, length of hospital
was significantly longer with RG. There were no stay, or perioperative mortality [21].
significant differences in margin status, number of
lymph nodes retrieved, blood loss, length of stay,
perioperative morbidity, or mortality. There were Long-Term Outcomes
no conversions to open surgery in either group.
This study evaluated CRP and IL-6 levels as mark- Very limited data on long-term, oncologic outcomes
ers of surgical stress and found them to be signifi- of RG are available. Pugliese and colleagues
cantly lower with LG. Cost was significantly more reported 18 cases of RG including both early and
with RG (approximately $4400 more per case). advanced disease. At a median follow-up of
The applicability of these studies to Western 28 months, the 3-year overall survival was 78 %.
patients is limited given the high case volume of Four patients (22 %) had recurrence within the fol-
these surgeons, the very low overall morbidity low-up period [22]. Decreased blood loss with RG
reported, the exclusion of patients who received over LG likely reflects the enhanced ability to per-
neoadjuvant therapy, and the predominance of form a delicate lymphadenectomy near the celiac
patients with early-stage disease. axis and major gastric vessels. Whether this advan-
Finally, several meta-analyses of studies com- tage will translate into improvements in recurrence-
paring LG and RG have recently been reported free or disease-specific survival is not known. It was
[19–21]. The most comprehensive of these 15 years before the survival benefit of D2 lymphad-
included nine nonrandomized studies that enectomy in Western gastric cancer patients became
compared the two procedures. One of these apparent [23]. It may therefore be some time before
25 Robotic Utilization in Gastric Cancer Surgery 267

a measureable difference in survival from a more 3. Kim HH, Hyung WJ, Cho GS, et al. Morbidity and
mortality of laparoscopic gastrectomy versus open
precise lymphadenectomy emerges.
gastrectomy for gastric cancer: an interim report – a
phase III multicenter, prospective, randomized Trial
(KLASS Trial). Ann Surg. 2010;251:417–20.
Cost 4. Kim YW, Baik YH, Yun YH, et al. Improved quality
of life outcomes after laparoscopy-assisted distal gas-
trectomy for early gastric cancer: results of a prospec-
The cost of the robotic surgery platform is limit- tive randomized clinical trial. Ann Surg. 2008;248:
ing in the current economy. In Korea, patients 721–7.
pay out of pocket for the extra costs of robotic- 5. Lee JH, Han HS, Lee JH. A prospective randomized
study comparing open vs laparoscopy-assisted distal
assisted procedures. In the United States, hospi-
gastrectomy in early gastric cancer: early results. Surg
tals charge significantly more for robotic-assisted Endosc. 2005;19:168–73.
procedures than for open or laparoscopic surger- 6. Kitano S, Shiraishi N, Fujii K, Yasuda K, Inomata M,
ies to off-set the costs of the robots, instruments, Adachi Y. A randomized controlled trial comparing
open vs laparoscopy-assisted distal gastrectomy for
and support. While the technical advantages of
the treatment of early gastric cancer: an interim report.
the robot definitely allow for better dissection Surgery. 2002;131:S306–11.
and lymphadenectomy in some procedures, par- 7. Hashizume M, Sugimachi K. Robot-assisted gastric
ticularly gastrectomy, prostatectomy, and proc- surgery. Surg Clin North Am. 2003;83:1429–44.
8. Giulianotti PC, Coratti A, Angelini M, et al. Robotics
tectomy, it is unknown whether the increased cost
in general surgery: personal experience in a large
will continue to be justified in the absence of community hospital. Arch Surg. 2003;138:777–84.
measurable clinical benefits over laparoscopy. 9. Anderson C, Ellenhorn J, Hellan M, Pigazzi A. Pilot
series of robot-assisted laparoscopic subtotal gastrec-
tomy with extended lymphadenectomy for gastric
cancer. Surg Endosc. 2007;21:1662–6.
Summary 10. Hyun MH, Lee CH, Kwon YJ, et al. Robot versus lap-
aroscopic gastrectomy for cancer by an experienced
Utilization of the robot in gastrectomy for cancer surgeon: comparisons of surgery, complications, and
surgical stress. Ann Surg Oncol. 2013;20:1258–65.
allows for a more precise dissection and D2
11. Kang BH, Xuan Y, Hur H, Ahn CW, Cho YK, Han SU.
lymphadenectomy than what can be achieved Comparison of surgical outcomes between robotic
with standard laparoscopy. This advantage comes and laparoscopic gastrectomy for gastric cancer: the
with significantly increased cost, however, and it learning curve of robotic surgery. J Gastric Cancer.
2012;12:156–63.
is unclear whether it will translate into clinical
12. Woo Y, Hyung WJ, Pak KH, et al. Robotic gastrec-
benefits for patients. Further controlled, prospec- tomy as an oncologically sound alternative to laparo-
tive studies inclusive of patients with advanced scopic resections for the treatment of early-stage
disease, neoadjuvant treatment, and higher BMI gastric cancers. Arch Surg. 2011;146:1086–92.
13. Song J, Oh SJ, Kang WH, Hyung WJ, Choi SH, Noh SH.
are needed to clarify the role of the robot in gas-
Robot-assisted gastrectomy with lymph node dissection
tric cancer surgery. It may be in these settings, for gastric cancer: lessons learned from an initial 100 con-
where laparoscopy is particularly challenging, secutive procedures. Ann Surg. 2009;249:927–32.
where the robot may be most advantageous. 14. Kim HI, Park MS, Song KJ, Woo Y, Hyung WJ. Rapid
and safe learning of robotic gastrectomy for gastric
cancer: multidimensional analysis in a comparison
with laparoscopic gastrectomy. Eur J Surg Oncol.
2013. Epub ahead of print.
References 15. Kim KM, An JY, Kim HI, Cheong JH, Hyung WJ, Noh
SH. Major early complications following open, lapa-
1. Hayashi H, Ochiai T, Shimada H, Gunji Y. Prospective ran- roscopic and robotic gastrectomy. Br J Surg. 2012;
domized study of open versus laparoscopy-assisted distal 99:1681–7.
gastrectomy with extraperigastric lymph node dissection 16. Kim MC, Heo GU, Jung GJ. Robotic gastrectomy for
for early gastric cancer. Surg Endosc. 2005;19:1172–6. gastric cancer: surgical techniques and clinical merits.
2. Huscher CG, Mingoli A, Sgarzini G, et al. Surg Endosc. 2010;24:610–5.
Laparoscopic versus open subtotal gastrectomy for 17. Park JY, Jo MJ, Nam BH, et al. Surgical stress after
distal gastric cancer: five-year results of a randomized robot-assisted distal gastrectomy and its economic
prospective trial. Ann Surg. 2005;241:232–7. implications. Br J Surg. 2012;99:1554–61.
268 K.J. Kelly and V.E. Strong

18. Coratti A, Annecchiarico M, Di Marino M, Gentile E, laparoscopic gastrectomy for gastric cancer: a
Coratti F, Giulianotti PC. Robot-assisted gastrectomy meta-analysis of 2495 patients. J Laparoendosc Adv
for gastric cancer: current status and technical consid- Surg Tech A. 2013;23(12):965–76.
erations. World J Surg. 2013;37(12):2771–81. 22. Pugliese R, Maggioni D, Sansonna F, et al. Subtotal
19. Liao GX, Xie GZ, Li R, et al. Meta-analysis of out- gastrectomy with D2 dissection by minimally invasive
comes compared between robotic and laparoscopic surgery for distal adenocarcinoma of the stomach:
gastrectomy for gastric cancer. Asian Pac J Cancer results and 5-year survival. Surg Endosc. 2010;24:
Prev. 2013;14:4871–5. 2594–602.
20. Xiong B, Ma L, Zhang C. Robotic versus laparo- 23. Songun I, Putter H, Kranenbarg EM, Sasako M, van
scopic gastrectomy for gastric cancer: a meta-analysis de Velde CJ. Surgical treatment of gastric cancer:
of short outcomes. Surg Oncol. 2012;21:274–80. 15-year follow-up results of the randomised nation-
21. Xiong J, Nunes QM, Tan C, et al. Comparison of wide Dutch D1D2 trial. Lancet Oncol. 2010;11:
short-term clinical outcomes between robotic and 439–49.
 Index

A Azygous vein
Adenocarcinoma division of, 141–142
of esophagus thoracoscopic division of, 150
acute and chronic inflammation, 6–7
gastroesophageal reflux disease, 5–6
genetic factors, 4–5 B
Helicobacter pylori infection, 6 Balloon-mounted esophagoscope, 178–180
incidence, 4 Barrett’s esophagus (BE), 5–6, 11, 17, 21–22
obesity, 6 Barrx™ Ablation System, 14, 15
gastric cancer, 61–65 Billroth I anastomosis
low- and high-risk factors, 12 extracorporeal, 219–220
in United States, 137 intracorporeal
Adjuvant chemotherapy, gastric cancer, 108–109 common entry hole closure, 218–219
Adjuvant therapy duodenum opening, 217–218
with overall survival, 75 duodenum transection, 215, 217
postgastrectomy survival, 85 linear stapler insertion, 218
Advanced gastric cancer liver retraction, 215
definition, 63 published reports, 215, 216
macroscopic appearance, 63, 65 specimen delivery, 217
spreading and prognosis, 67–68 stomach opening, 217, 218
staging of, 65–67 stomach transection, 217
AF. See Atrial fibrillation (AF) Billroth II (BII) loop gastrojejunostomy, 230
Aging British Stomach Cancer Group, 109
esophageal cancer, 54
gastric cancer, 61
Alcohol consumption, 3, 4, 6 C
Alcohol dehydrogenase mutation, 3 CALGB 80101 study, 109, 111
American College of Gastroenterology, 11–12 Cap resection technique, EMR, 16
American Society for Gastrointestinal Endoscopy Cardiac arrhythmias, 43
(ASGE), 21 Cell-sheet technology, 22
American Society of Anesthesiologists, 36 Cervical anastomosis
Amiodarone, 53 creation of, 127
Anastomotic leakage, 43, 161 laparoscopic/thoracoscopic transhiatal
Arrhythmia esophagectomy
cardiac, 43 abdominal dissection, 152–154
esophageal cancer surgery, 53 advantages, 155
perioperative complications, 191 anesthetic/induction phase, 147–148
risks of, 53 cervical dissection, 151–152
ARTIST trial study, 110, 111 disadvantages, 155
ASCC1 signaling, esophageal cancer, 4 patient positioning, 148
ASGE. See American Society for Gastrointestinal postoperative care, 155–156
Endoscopy (ASGE) preoperative preparation, 147
Atrial fibrillation (AF) reconstruction, 154–155
cardiac arrhythmias, 43 thoracoscopic dissection, 148–151
postoperative risk, 53 respiratory complications, 42

S.N. Hochwald, M. Kukar (eds.), Minimally Invasive Foregut Surgery for Malignancy: Principles and Practice, 269
DOI 10.1007/978-3-319-09342-0, © Springer International Publishing Switzerland 2015
270 Index

Chemotherapy outcomes, 135–136

gastric cancer postoperative care, 135
adjuvant chemotherapy, 108–109 techniques, 128
perioperative neoadjuvant chemotherapy, thoracoscopic phase, 130–134
106–108 EGC. See Early gastric cancer (EGC)
perioperative, 75–76 ELND. See Extended LN dissection (ELND)
with radiation, 85 EMR. See Endoscopic mucosal resection (EMR)
salvage, 115 Endoscope
single-agent, 112 enteral access, 185
Chylothorax, 43 for ESD, 97
Cigarette smoking, 2 gastroesophageal junction, 209–210
Circumferential resection margin (CRM), 38–39 Endoscopic ablation therapy
Clinical pathways, for esophagectomy, 26–29 cryotherapy, 14–15
College of American Pathologists (CAP), 38 purpose of, 14
Colonic interposition, 157–162 radiofrequency ablation, 14, 15
Computed tomography (CT) Endoscopic mucosal resection (EMR)
and external US, 37 cap resection technique, 16, 93
and PET-CT, 137 vs. ESD, 20
transmural tumors, 200 esophageal malignancy, 15
CRM. See Circumferential resection margin (CRM) gastric cancer, 76, 93
Cryotherapy, 14–15 ligate-and-cut technique, 16
CTHRC1 signaling, esophageal cancer, 4–5 retrospective study, 17
for staging purposes, 13, 14
techniques, 16–17
D types of, 16
Definitive chemoradiotherapy (dCRT), 37–38 Endoscopic resection (ER)
D2 lymphadenectomy complication associated with, 15
advantages, 231 early gastric cancer
clinical studies, 223–225 with gastric ulcer, 98–102
laparoscopic subtotal gastrectomy with (see well-differentiated, 64
Laparoscopic subtotal gastrectomy) EMR, 16–17
Double tract anastomosis, 235–241. See also Proximal ESD, 17–21
gastrectomy gastric cancer
Dysplasia indication, 93–94
gastric cancer, 62–63 therapeutic efficacy, 94
HGD, 11 goals of, 15
esophageal cancer, 12–13 indication for, 93–94
low- and high-risk factors, 12 oncology nurse coordinator, 27
low-grade vs. high-grade, 62–63 risk factors, 17
SCC, 13
therapeutic efficacy of, 94
E Endoscopic submucosal dissection (ESD)
Early gastric cancer (EGC) complications, 20
definition, 63 early gastric cancer, 64
endoscopic resection, 64 vs. EMR, 20
with gastric ulcer, 98–102 esophageal squamous cell carcinoma, 19
identification and diagnosis, 93 gastric cancer, 76
laparoscopic distal gastrectomy, 213–220 distal attachment, 97–98
Paris classification, 63–64 endoscope, 97
proximal gastrectomy, 235–241 high-frequency generator, 98
spreading and prognosis, 67 high-frequency knives, 95
undifferentiated intramucosal, 102 indication, 93–94
ECHELON FLEX™ Powered ENDOPATH® Stapler, IT-knife type, 95–97
122–123 needle type knives, 95
ECM. See Extracellular matrix (ECM) non-IT-knife type, 97
EEA anastomosis, MIE procedure, 98–100
anesthetic management, 128 scissors forceps type, 97
endoscopic evaluation, 128 severe fibrosis with ulcer, 100–102
laparoscopic phase, 128–130 therapeutic efficacy, 94
McKeown esophagectomy, 128 undifferentiated adenocarcinoma, 102–103
Index 271

injection solutions for, 17 risk factors, 1

needle knives for, 17, 20 SEER analysis, 40
schematic representation, 18 squamous cell carcinoma, 1
Endoscopic ultrasound (EUS) alcohol consumption, 3
esophageal cancer, 36 alcohol dehydrogenase mutation, 3
esophageal-preserving therapy, 13–14 caustic injury, 4
EUS-guided fine-needle aspiration, 137, 177 poverty, 4
transmural tumors, 200 raw fruit vs. vegetable consumption, 4
Epidermal growth factor (EGF), 5 tobacco use, 2–3
Epstein Barr virus (EBV) infection, 62 surgical therapy for
ESD. See Endoscopic submucosal dissection (ESD) dCRT, 37–38
Esophageal benign tumors lymphadenectomy, 39–42
laparoscopic enucleation morbidity, 42–43
anti-reflux mechanism, 180 mortality/quality control, 43–45
surgical indications, 177–178 multidisciplinary approach, 35
surgical procedures, 179–180 operative resection, 35
thoracoscopic enucleation patient selection, 36
balloon-mounted esophagoscope, 178–179 performance indicators, 45
esophageal myotomy, 178 resection margins, 38–39
indications, 177–178 tumor selection, 36–37
procedures, 178–179 T1a/T1b cancer, 12–13
Esophageal cancer in United States, 137
adenocarcinoma Esophageal-preserving therapy
gastroesophageal reflux disease, 5–6 advances in, 22
gender influence, 5 approaches, 11
genetic factors, 4–5 clinical staging, 13–14
Helicobacter pylori infection, 6 HGD, 12–13
incidence, 4 long-term outcome, 21–22
obesity, 6 management after, 21
evaluation of surgery, 51 patient selection, 12–13
HGD, 12–13 T1a/T1b esophageal cancer, 13
at high-volume institutions, 26 transnasal endoscopy, 21
incidence, 1 Esophageal resection
laparoscopic transhiatal clinical pathways associated with, 26–29
esophagectomy, 119 long-term survival after, 44
abdominal dissection, 120–122 minimally invasive techniques for, 123
cervical component, 122–123 volume-outcome relationship, 25–26
complications, 123–124 Esophageal squamous cell cancer
contraindications, 120 endoscopic submucosal dissection, 19
Kocher maneuver, 121 risk factors for, 12–13
outcomes, 123–124 Esophageal stent
patient positioning, 120 advantages/disadvantages, 56, 57
port placement for, 121 obstructive symptoms, 28
short-term results for, 124 placement (see also Minimally invasive feeding tube)
medical evaluation, 51–55 complications, 184
age-related comorbidities, 54 evaluation, 183–184
cardiac, 53 outcomes, 185
hepatic disease, 53–54 steps for, 184–185
obesity, 54–55 Esophagectomy, 11, 13. See also Specific types
pulmonary complications, 53 clinical pathways for, 26–29
venous thromboembolism, 55 esophageal cancer (see Esophageal cancer)
neoadjuvant chemoradiotherapy, 27 hospital volume and, 25
noninvasive staging modalities, 137 laparoscopic/thoracoscopic
nutritional assessment, 55–57 abdominal dissection, 138–140
operative resection, 35 operative description, 138
optimization, 55–57 thoracic anastomosis, 142–144
prior surgical history, 55 thoracic dissection, 140–142
resection minimally invasive vs. open, 30–31
long-term survival after, 44 respiratory failure, 42–43
volume-outcome relationship, 25–26 Esophagotomy, 142, 154
272 Index

European Organization for Research and molecular pathology, 69–70

Treatment of Cancer (EORTC), 108–109, 111 multimodality therapy
EUS. See Endoscopic ultrasound (EUS) adjuvant chemotherapy, 108–109
Extended LN dissection (ELND), 80, 87 incidence, 105
External US, esophageal cancer, 37 perioperative neoadjuvant chemotherapy,
Extracellular matrix (ECM), 5, 14, 22 106–108
postoperative chemoradiation, 111
radiation role, 109–110
G noncurative operative procedure, 73–74
Gastrectomy. See also Specific types operative therapy
gastric cancer historic development, 74
failure pattern, 85 intra- and postoperative considerations, 82–85
survival outcomes, 86 lymph node dissection, 80–82
HDGC, 69 multidisciplinary strategy, 75–76
open standards, 76, 77 preoperative intents, 73–75
palliative-intent, 85 reconstruction technical aspects, 82
prophylactic, 87 technical aspects of resection, 76–80
reconstruction options, 83 perioperative chemotherapy trials, 106
robotic surgery systems, 230–231 REAL-2 study, 113
Gastric Adenocarcinoma and Proximal Polyposis of the risk factors, 61–63
Stomach (GAPPS), 61, 68, 69, 87 robot-assisted gastrectomy (see Robot-assisted
Gastric cancer (GC) gastrectomy (RG))
adenocarcinoma, 61–65 surgical objectives, 73
advanced (see Advanced gastric cancer) TAX325 trial study, 112–113
age-standardized, 61 TNM classification, 67
bevacizumab in, 113 WHO classifications, 62, 65
CALGB 80101 study, 109 Gastric conduit
classification, 65, 66 anatomy, 130
dysplasia, 62–63 aspects, 142
EGC (see Early gastric cancer (EGC)) creation, 139–140, 154
endoscopic submucosal dissection distal portion, 132
distal attachment, 97–98 gastrotomy, 143
endoscope, 97 Gastrocolic omentum, 121, 129, 139
high-frequency generator, 98 Gastroduodenostomy. See Billroth I anastomosis
high-frequency knives, 95 Gastroesophageal anastomosis, 133–134
indication, 93–94 Gastroesophageal reflux disease (GERD), 5–6, 21, 55
IT-knife type, 95–97 Gastrointestinal stromal tumors (GISTs)
needle type knives, 95 diagnosis, 206
non-IT-knife type, 97 esophageal, 178
procedure, 98–100 GIST Consensus Conference, 205
scissors forceps type, 97 laparoscopic partial gastrectomy (see Laparoscopic
severe fibrosis with ulcer, 100–102 partial gastrectomy)
therapeutic efficacy, 94 special considerations, 87
undifferentiated adenocarcinoma, 102–103 symptoms, 206
epidemiology, 61 therapeutic strategy, 206
etiology of, 61–63 Gastrojejunostomy, 240
gastrectomy BII loop gastrojejunostomy, 230
failure pattern, 85 laparoscopic subtotal gastrectomy
prophylactic, 87 clinical studies, 223–225
survival outcomes, 86 meta-analyses, 224
Helicobacter pylori infection, 61–62 patient position, 226–227
hereditary syndromes patient selection, 226
GAPPS, 69 procedure, 227–230
HDGC, 68–69 randomized trials, 224
histological subtypes of, 66 reconstruction, 229–231
Kaplan-Meier curve, 107 robotic gastrectomy, 230–232
macroscopic appearances, 63 room setup, 226–227
metastatic (see Metastatic gastric cancer) Western distal gastrectomy studies, 224–225
microscopic appearances, 63–65 Roux-en-Y gastrojejunostomy, 213
minimally invasive intragastric surgery, 199–204 Gastroscopy, 202, 229
Index 273

Gastrostomy indications, 213–214

advantages/disadvantages, 56 intracorporeal Billroth I anastomosis
anterior, 209–210 common entry hole closure, 218–219
anvil insertion through, 255 duodenum opening, 217–218
closure, 202 duodenum transection, 215, 217
gastric conduit, 143 linear stapler insertion, 218
laparoscopic complications, 57, 187 liver retraction, 215
use of, 56 published reports, 215, 216
GC. See Gastric cancer (GC) specimen delivery, 217
Gender influence, esophagus, 5 stomach opening, 217, 218
Genome-wide association studies, 68–70 stomach transection, 217
GISTs. See Gastrointestinal stromal tumors (GISTs) preoperative preparation, 214
Granulocyte colony-stimulating factor (GCSF), 113 Laparoscopic enucleation, esophageal benign tumors
anti-reflux mechanism, 180
indications, 177–178
H procedures, 179–180
HDGC. See Hereditary diffuse gastric cancer (HDGC) Laparoscopic intragastric surgery (LIGS)
Helicobacter pylori infection advantages, 200
esophageal adenocarcinoma, 6 indications, 199–200
gastric cancer, 61–62 preoperative diagnosis, 200
Hepatic disease, 53–54 procedure, 200–202
Hereditary diffuse gastric cancer (HDGC), 61, 68–69 resection, 202–203
Hereditary syndromes, of gastric cancer surgical planning, 200
GAPPS, 69 Laparoscopic jejunostomy tube placement
HDGC, 68–69 complications, 187
High-frequency generator (HFG), for ESD, 98 extracorporeal technique, 186
High-grade dysplasia (HGD) illustration, 186
esophageal cancer, 12–13 outcomes, 187
low- and high-risk factors, 12 placement steps, 185, 186
prospective study, 14 port sites, 185, 186
and T1a cancer, 11–12, 20, 21 Laparoscopic partial gastrectomy
Human epidermal growth factor receptor 2 (HER2) anatomic classification, gastric lesions, 205, 206
inhibitor, 113–114 antrum/prepyloric region, 209–210
fundus and greater curve, 207–209
lesser curve/gastroesophageal junction, 210–211
I outcomes, 211
Interleukin-18, 5 postoperative care, 211
Intraepithelial neoplasia (IEN), 62–63 preoperative planning, 206–207
IT-knife type, ESD, 95–97 therapeutic strategy, 206
Laparoscopic subtotal gastrectomy
clinical studies, 223–225
J meta-analyses, 224
Jejunostomy patient position, 226–227
advantages/disadvantages, 56–57 procedure, 227–230
feeding jejunostomy tube placement, 130 randomized trials, 224
outcomes for laparoscopic, 187 reconstruction, 229–231
percutaneous endoscopic jejunostomy, 185 robotic gastrectomy, 230–232
room setup, 226–227
Western distal gastrectomy studies, 224–225
K Laparoscopic/thoracoscopic esophagectomy
Kaplan-Meier curve, 107 abdominal dissection, 138–140
Knives for ESD, 95, 97 colonic interposition
Kocher maneuver, 121, 130, 139, 215 abdominal phase, 158–160
morbidity/mortality, 162
patient selection/evaluation, 158
L postoperative care, 160–161
Laparoscopic distal gastrectomy (LDG) thoracic phase, 160
contraindication, 214 operative description, 138
extracorporeal Billroth I reconstruction, 219–220 thoracic anastomosis, 142–144
incision and exposure, 214–215 thoracic dissection, 140–142
274 Index

Laparoscopic transhiatal esophagectomy two-field, 165–166

cervical anastomosis Lymph node dissection (LND), 228, 267
abdominal dissection, 152–154 gastric artery, 236–237
advantages, 155 gastric cancer, 80–82
anesthetic/induction phase, 147–148 gastroepiploic artery, 236–237
cervical dissection, 151–152 hepatic artery, 238–239
disadvantages, 155 resection extent/fields, 40
patient positioning, 148 splenic artery, 239
postoperative care, 155–156
preoperative preparation, 147
reconstruction, 154–155 M
thoracoscopic dissection, 148–151 Malnutrition, factors associated with, 56
esophageal cancer, 119 Medical Research Council (MRC), 107
abdominal dissection, 120–122 Metastatic gastric cancer
cervical component, 122–123 first-line treatment of, 112–113
complications, 123–124 management, 111–113
contraindications, 120 salvage therapy, 114–115
Kocher maneuver, 121 targeted therapies, 113–114
outcomes, 123–124 MicroRNA, gastric cancer, 69
patient positioning, 120 Minimally invasive esophagectomy (MIE)
port placement for, 121 anesthetic management, 128
short-term results for, 124 approaches, 120
Laparoscopy in Barrett’s disease, 120
plastic, 123 description, 137–138
vs. robotic gastrectomy, 230–231 endoscopic evaluation, 128
staging esophageal cancer, 119–121
in gastric cancer, 76 laparoscopic phase
intraoperative, 37 feeding jejunostomy tube placement, 130
Laparoscopy-assisted total gastrectomy gastric mobilization, 129–130
abdominal esophagus, 254 gastric tube creation, 130
dissection patient positioning, 128–129
abdominal esophagus, 254 port placement, 128–129
anvil insertion, 254–257 pyloroplasty, 130
drain insertion, 260 McKeown esophagectomy, 128
duodenal transection, 250, 251 obesity, 55
extracorporeal jejunojejunostomy, 258 vs. open esophagectomy, 30–31
gastrocolic ligament, 246, 248 operative resection, 35
gastrosplenic ligament, 248–249 outcomes, 135–136
jejunum, 257–258 postoperative care, 135
laparoscopic esophagojejunostomy, 258–259 prior surgical history, 55
left gastroepiploic vessels, 248 pulmonary complications, 53
lesser omentum, 251 thoracoscopic phase
lymph nodes, 248–254 dissection, 131–133
pancreas, 251–254 drain placement and closure, 134
proximal stomach, 249 gastroesophageal anastomosis, 133–134
right gastric artery, 251, 252 patient positioning, 131
stapled duodenal stump, 250 port placement, 131, 132
liver retraction, 245–247 preparation for, 130–131
patient position, 243, 244 Minimally invasive feeding tube
port placement, 243, 245 evaluation, 183–184
Laurén’s classification, gastric cancer, 65 extracorporeal technique, 186
Leapfrog Group, 26 laparoscopic jejunostomy tube
Ligate-and-cut technique, EMR, 16 complications, 187
Lymphadenectomy extracorporeal technique, 186
dissection, 139 illustration, 186
D2 lymphadenectomy (see D2 lymphadenectomy) outcomes, 187
esophageal cancer, 39–42 placement steps, 185, 186
intraoperative images, 81 port sites, 185, 186
survival benefit, 41 outcomes, 184, 185
three-field (see Three-field lymphadenectomy) percutaneous endoscopic jejunostomy, 185
Index 275

steps for, 184–185 Pulmonary complications, esophageal

Minimally invasive surgery (MIS), 73, 76 cancer, 53
MSR1 signaling, 4 Pyloromyotomy, 77, 123, 130, 170
Pyloroplasty, 77, 130, 131, 170

N
Nakamura’s classification, gastric cancer, 65 R
Needle knives, for ESD, 17, 20, 95 Radiation, gastric cancer, 109–110
Neoadjuvant chemoradiotherapy Radiofrequency ablation (RFA), 14–15, 17, 22
dCRT and, 37–38 Reactive oxygen species, 3, 6
esophageal cancer, 27 REAL-2 study, gastric cancer, 113
Non-IT-knife type, ESD, 97 Recurrent laryngeal nerve injury, 43
Nutritional assessment, esophageal cancer, 28, 55–57 Respiratory failure, esophagectomy, 42–43
Rethoracoscopy, 173
RFA. See Radiofrequency ablation (RFA)
O Robot-assisted gastrectomy (RG)
Obesity, esophageal cancer, 6, 54–55 advantages, 261
Open esophagectomy cost, 267
MIE vs., 30–31, 124 vs. laparoscopy, 230–231
postoperative morbidity, 162 learning curve, 265
Operative therapy, of gastric cancer oncologic outcomes, 266–267
general objectives, 73 patient position, 262
historic development, 74 patient selection, 265
intra- and postoperative considerations, 82–85 perioperative outcomes, 265–266
lymph node dissection, 80–82 port placement, 262
multidisciplinary strategy, 75–76 procedure, 262–264
preoperative intents, 73–75 Robot-assisted thoracoscopic esophagectomy
reconstruction technical aspects, 82 (RAMIE)
technical aspects of resection, 76–80 disadvantages, 195
indications, 189–190
laparoscopic dissection, 194–195
P perioperative management
Penrose drain (PD) complications, 191
diaphragm level, 193 fluid, 191
distal esophagus, 159, 160 intraoperative, 190
GE junction, 138–141 one-lung ventilation, 190–191
liver retraction, 245 postoperative care, 191
Percutaneous endoscopic jejunostomy (PEJ), 185 preoperative, 190
Percutaneous radiologic gastrostomy (PRG), 56 postoperative care, 195
Perioperative neoadjuvant chemotherapy, 106–108 results, 195–196
Positron emission tomography (PET) thoracoscopic dissection
esophageal cancer, 36, 37 patient position, 192–193
gastric cancer, 76 procedure, 193–194
PRG. See Percutaneous radiologic gastrostomy (PRG) robotic instruments, 191
Prone position. See Thoracolaparoscopic esophagectomy Roux-en-Y gastrojejunostomy, 213, 230
Proximal gastrectomy, 77 Royal College of Pathologists (RCP), 38
dissection RTOG 0114 randomized phase II study,
abdominal esophagus, 237–238 109, 111
hepatic artery, 238–239
left gastrocolic ligament, 236
left gastroepiploic artery, 236–237 S
lesser curvature, 238 Salvage therapy
short gastric artery, 236–237 chemotherapy, 115
splenic artery, 239 metastatic gastric cancer, 114–115
stomach, 239 Scissors forceps type, ESD, 97
drain insertion, 241 Sentinel node (SN) sampling, 37
indication, 235 Signet ring cell carcinoma, 102
outcomes, 241 Smoking
reconstruction, 239–230 esophageal cancer, 2
trocar insertion, 235–236 gastric cancer, 62
276 Index

Squamous cell carcinoma Thoracoscopic enucleation

of esophagus, 1 approach, 178–179
alcohol consumption, 3 esophageal benign tumors
alcohol dehydrogenase mutation, 3 balloon-mounted esophagoscope,
caustic injury, 4 178–179
pathogenesis, 2 esophageal myotomy, 178
poverty, 4 procedures, 178–179
raw fruit vs. vegetable consumption, 4 surgical indications, 177–178
tobacco use, 2–3 for esophageal leiomyoma, 179
in United States, 137 Thoracoscopic transhiatal esophagectomy
Surgical therapy abdominal dissection, 152–154
for esophageal cancer advantages, 155
dCRT, 37–38 anesthetic/induction phase, 147–148
lymphadenectomy, 39–42 cervical dissection, 151–152
morbidity, 42–43 disadvantages, 155
mortality/quality control, 43–45 patient positioning, 148
multidisciplinary approach, 35 postoperative care, 155–156
operative resection, 35 preoperative preparation, 147
patient selection, 36 reconstruction, 154–155
performance indicators, 45 thoracoscopic dissection, 148–151
resection margins, 38–39 Three-field lymphadenectomy
tumor selection, 36–37 abdominal/laparoscopic phase, 169–170
gastric cancer cervical phase, 170–171
historic development, 74 gastric tube formation, 170
intra- and postoperative pyloromyotomy, 170
considerations, 82–85 thoracoscopic phase, 167–169
lymph node dissection, 80–82 TNM classification of gastric cancer, 67
multidisciplinary strategy, 75–76 Tobacco use, of esophagus, 2
preoperative intents, 73–75 Transnasal endoscopy, 21
reconstruction technical aspects, 82 Two-field lymphadenectomy, 165–166
technical aspects of resection, 76–80

U
T Upper endoscopy
Targeted therapies, metastatic gastric cancer, antrum/prepyloric region, 209
113–114 conscious sedation, 21
T1a/T1b esophageal cancer, 11–13 initial/repeat, 51
TAX325 trial study, gastric cancer, 112–113 intraoperative, 158
Thoracic anastomosis, laparoscopic/thoracoscopic laparoscopic intragastric surgery, 199–200
esophagectomy, 142–144 preoperative workup, 200
Thoracic epidural analgesia (TEA), 190
Thoracolaparoscopic esophagectomy
advantages, 166–167 V
postoperative care, 171–172 Venous thromboembolism (VTE), 55
results, 172–173 Volume-outcome relationship, esophageal
rethoracoscopy, 173 resection, 25–26
three-field lymphadenectomy
abdominal/laparoscopic phase,
169–170 W
cervical phase, 170–171 Wedge gastrectomy. See Laparoscopic partial
gastric tube formation, 170 gastrectomy
pyloromyotomy, 170 Will Rogers phenomenon, 41
thoracoscopic phase, 167–169 Worldwide Esophageal Cancer Collaboration
two-hole esophagogastrectomy, 171 (WECC), 41