Saar 2016
Saar 2016
DOI 10.1007/s00268-016-3416-2
Abstract
Background Despite significant progress in surgery, controversy persists about timing of appendectomy. Objective
of this prospective observational study was to determine associations between time interval from onset of symptoms
in appendicitis to appendectomy and postoperative complications.
Methods After institutional review board approval, all adult consecutive patients subjected to emergency appen-
dectomy between 1/9/2013 and 1/12/2014 were prospectively enrolled. Data collection included demographics, open
vs. laparoscopic appendectomy, comprehensive complication index (CCI), and 30-day follow-up. To determine time-
dependent associations between delay of surgery and complications all patients were stratified into subgroups based
on 12-h time intervals from onset of abdominal pain to surgery. Primary outcome was complications per CCI in
correlation to delay from symptoms to appendectomy. Secondary outcomes included duration of surgery, hospital
length of stay (HLOS), and incidence of complication within 30-day follow-up.
Results A total of 266 patients with a mean age of 35.4 ± 14.8 years met inclusion criteria. Overall, 83.1 % of
patients were subjected to laparoscopic appendectomy. Delay to surgery in 12-h increments showed stepwise-
adjusted increase in complications per CCI (adj. P = 0.037). Also, delay to appendectomy increased significantly
duration of surgery and HLOS, respectively (adj. P \ 0.001 and adj. P \ 0.001). Overall, 5.7 % of patients
developed a surgical site infection after hospital discharge.
Conclusion Extended time interval from the onset of initial symptoms to appendectomy is associated with increased
complications per CCI, duration of surgery, and HLOS in acute appendicitis. Prompt appendectomy in acute
appendicitis is warranted.
Introduction
& Peep Talving
[email protected]
Appendectomy is the most frequently performed emergent
1
School of Medicine, University of Tartu, Puusepa 8, abdominal operation with a lifetime risk of appendicitis at
51014 Tartu, Estonia 9 % [1, 2]. Despite significant advancements in surgery and
2
Department of Surgery, Tartu University Hospital, antimicrobial therapy, many controversies in the manage-
Puusepa 1a, 50406 Tartu, Estonia ment of acute appendicitis persist.
3
Department of Surgery, School of Medicine, Many studies have observed that the risk of perforation
University of Tartu, Puusepa 8, 51014 Tartu, Estonia is time-dependent and extended time to surgery results in
4
Division of Trauma, Hand and Reconstructive Surgery, poor outcomes [3, 4]. However, other investigations have
Goethe University Hospital, Frankfurt/Main, Germany reported that spontaneous resolution of low-grade appen-
5
Division of Acute Care Surgery, Los Angeles dicitis is common and nonoperative management in these
County ? University of Southern California Medical Center, instances may result in good outcomes [5, 6]. Given the
Los Angeles, CA, USA
123
World J Surg
paucity of data, we set out to determine prospectively the was administered at the discretion of treating physician.
relationship between time interval from onset of symptoms Decision for drainage was made by attending surgeon and
to surgery and incidence of complications while comparing was utilized in general peritonitis, periappendicular
disease severity per histology in all specimens. abscess, or when highly contaminated surgical field was
encountered.
All surgical specimens were subjected to histology.
Methods and patients Based on the histological diagnosis, patients were stratified
into five subgroups using disease severity score (DSS):
Following institutional review board approval, all consec- grade 1 (G1), inflamed; grade 2 (G2), gangrenous; grade 3
utive patients C18 years of age subjected to emergency (G3), perforated with localized free fluid; grade 4 (G4),
appendectomy due to acute appendicitis between 1/9/2013 perforated with regional abscess and grade 5 (G5), perfo-
and 1/12/2014 were prospectively enrolled. Informed rated with diffuse peritonitis [10].
consent was obtained from all patients. Patients subjected Follow-up at 30 days after discharge included compli-
to nonoperative management were excluded. Data collec- cations per CD classification and CCI. All complications
tion included demographics, time of onset of abdominal and readmissions within 30 days after discharge were
pain, antibiotic administration, radiological investigations, documented.
surgical approach, post-operative complications, histology The P values for categorical variables were derived from
reports, hospital length of stay (HLOS), and follow-up the Chi-square or 2-sided Fisher’s exact test and for con-
within 30-day. tinuous variables the Student’s t test or the Mann–Whitney
To determine time-dependent associations between tests were deployed. Logistic regression analysis was used
delay of surgery and complications all patients were to compare the CCI, complications, antimicrobial therapy,
stratified into subgroups of 12-h time increments from duration of surgery, conversion rate, placement of a drain,
onset of abdominal pain to surgery. Primary outcome was HLOS, readmission rate, and the rate of SSI between
complications per comprehensive complication index subgroups. Values are reported as mean ± standard devi-
(CCI) [7] in correlation with time from onset of the ation (SD) for continuous variables and as percentages for
abdominal pain to appendectomy. CCI is a recently pub- categorical variables. P values \ 0.05 were set as statisti-
lished postsurgical morbidity assessment scale based on cal significance level. All analyses were performed using
Clavien-Dindo (CD) classification [8]. While the CD scale the Statistical Package for Social Sciences (SPSS for
includes only the most severe surgical complication, the MacÓ), version 16.0 (SPSS Inc., Chicago, IL).
CCI includes all surgical complications in a scale ranging
from 0 to 100. The CCI score of zero means no compli-
cations and a score of 100 is translated into a complication Results
burden that results in patient’s death. All complications are
weighted for their severity per CCI. Overall, during the 15-month study period, 270 appen-
Complication in this investigation was defined as any dectomies were performed and 266 (98.5 %) of patients
deviation from normal postoperative course with 30-day were included. Four patients declined to participate in the
follow-up included. Secondary outcomes were duration of study.
operation, HLOS, and incidence of complications within Table 1 depicts demographic and clinical characteristics
30-day follow-up after hospital discharge. of the cohort. The sole significant difference in demo-
Definitions of surgical site infections were based on the graphic profile was the mean age of the study groups that
CDC definitions [9]. Wound infection was defined as a was entered into logistic regression model. The mean in-
surgical site infection (SSI) divided into a superficial hospital time to surgery was 6.3 ± 5.7 hours and did not
incisional SSI involving skin and subcutaneous tissue and a differ significantly between the groups.
deep SSI involving deep soft tissues of the incision. Intra- Mean WBC on admission was 13.1 ± 3.9 9 109/L and
abdominal infection was defined as an organ/space SSI mean CRP was 47.8 ± 71.9 mg/L. Overall, 88.6 % of
involving any part of the body deeper than fascial/muscle patients were undergoing ultrasound investigation and
layers. 20.1 % computed tomography scan, respectively, before
All imaging studies and surgical interventions, laparo- the operation.
scopic or open appendectomy, were performed at the dis- A total of 95.1 % of patients received perioperative
cretion of an attending surgeon. Blood samples including intravenous antibiotics at the discretion of the treating
C-reactive protein (CRP) and white blood cell count physician. Overall, 4.9 % (n = 13) of patients did not
(WBC) were obtained in all the patients after admission to receive perioperative antibiotics. Two out of 13 cases
Emergency Department. Perioperative antibiotic therapy (15 %) of patients not receiving perioperative antibiotics
123
World J Surg
Age (years) 35.4 ± 14.8 33.1 ± 14.0 32.4 ± 14.4 36.3 ± 13.8 40.6 ± 14.8 39.0 ± 15.9 0.020
BMI 25.4 ± 4.5 24.4 ± 4.9 25.3 ± 4.6 25.9 ± 4.5 25.0 ± 5.1 25.6 ± 3.9 0.661
Male 51.1 % (136) 67.9 % (19) 49.5 % (50) 55.2 % (32) 42.3 % (11) 45.3 % (24) 0.272
Previous IA surgery 9.8 % (26) 7.1 % (2) 7.9 % (8) 8.6 % (5) 3.8 % (1) 18.9 % (10) 0.152
Perioperative AB 95.1 % (252) 100 % (27) 96.0 % (97) 96.6 % (56) 96.2 % (25) 88.7 % (47) 0.159
In-hospital delay 6.3 ± 5.7 4.6 ± 1.7 6.3 ± 3.3 5.8 ± 3.1 6.4 ± 3.0 8.0 ± 11.1 0.107
Pre-hospital delay 27.1 ± 26.9 4.9 ± 2.4 11.3 ± 4.3 22.7 ± 4.1 36.4 ± 4.3 69.7 ± 31.7 <0.001
Abbreviations: hrs hours, n number of patients, BMI body mass index, IA intra-abdominal, AB antibiotic
developed one superficial SSI and one intra-abdominal subgroup compared to the \12 h subgroup. The follow-up
infection. demonstrated no significant outcome differences between
Surgical interventions are depicted in Table 2. Most the subgroups (Table 5).
patients were subjected to laparoscopic appendectomy at
83.1 %. Duration of operation and intraoperative place-
ment of a drain increased significantly with longer time Discussion
intervals to surgery (adj. P \ 0.001). Three patients
required conversion from laparoscopy to open midline We hypothesized that delayed presentation with an extended
laparotomy due to severe intra-abdominal adhesions. All time interval from onset of abdominal pain to surgery may
the primarily performed midline laparotomies were per- increase post-appendectomy complications. The present
formed due to general peritonitis. study proved our hypothesis and demonstrated that post-
Histology findings based on DSS are shown in Table 3. appendectomy complications are associated with extended
In patients with G4 and G5 acute appendicitis, the overall overall time interval from the very first symptoms to surgery.
time interval from onset of symptoms to operation was Also, all the advanced DSS cases (G4 and G5) were found in
beyond 36 hours (P \ 0.001 for G4 and P = 0.01 for G5). the subgroups of patients with appendicitis subjected to
For G1 cases, time interval was mostly \24 h (P \ 0.001). appendectomy beyond 36 h from the onset of symptoms.
In 1.1 % (n = 3) of patients histology revealed malignancy Multiple previous investigations have evaluated the
(carcinoids) and in 8.3 % non-inflamed appendix was impact of time interval from hospital admission to surgery
encountered. on outcomes in patients with acute appendicitis [11–20].
A detailed description of post-appendectomy compli- Busch et al. observed that an in-hospital delay [12 h
cations and antibiotic therapy are presented in Table 4. The increased considerably the risk of perforation and adverse
prevalence of postoperative complications and need for events [11]. Also, Teixeira and colleagues in an analysis of
antimicrobial therapy correlated with extended time inter- 4529 appendectomies noted that in-hospital delay [6 h
val to surgery. All the patients with DSS [2 received independently increased the risk of surgical site infections
intravenous antibiotics post-appendectomy. [12]. Likewise, in a study by Udgiri et al., an in-hospital
Complications per CD classification did not show sta- delay to surgery beyond 10 hours significantly increased
tistically significant correlation with prolonged time inter- incidence of complications [13]. Nevertheless, some stud-
val to surgery. However, CCI experienced a stepwise- ies have found contrary results. In a study by Shin and co-
adjusted increase in complications with delays from authors, no significant difference in outcomes was noted
symptoms to surgery (adj. P = 0.037) (Figure 1). Overall, between the subgroups waiting for operation less than 8
grade I and grade II type of CD complications predomi- hours versus more than 8 hours [14]. In another large study
nated. A total of 7.1 % (n = 19) of patients developed a by Drake and colleagues, no associations between in-hos-
SSI during hospital stay (20 %) or post-discharge (80 %), pital time prior to surgery and appendiceal perforations
respectively. Intra-abdominal infections, at 80 % (n = 8), were detected [15]. Some other studies have found an in-
were the predominant cause of all readmissions (n = 10) in hospital delay \24 h being a safe practice [16–19].
addition to one Clostridium difficile colitis and one deep Our study documented prospectively the entire time
SSI. No mortalities occurred. segment from the onset of very first symptoms prior to
HLOS was significantly longer with increasing time to hospital admission to the subsequent surgery. Very few
surgery and was about three-fold higher in the [48 h previous studies have evaluated the total time interval from
123
123
Table 2 Duration of operation and surgical interventions of all patients
Time from onset of abdominal pain to surgery Total (n = 266) 0–12 h (n = 28) 13–24 h (n = 101) 25–36 h (n = 58) 37–48 h (n = 26) [48 hrs (n = 53) P adj. P*
Duration of OP (min) 41.2 ± 18.3 33.1 ± 11.9 36.6 ± 16.5 41.3 ± 16.8 46.4 ± 21.0 51.2 ± 20.1 <0.001 <0.001
Operations
Open 13.9 % (37) 17.9 % (5) 9.9 % (10) 15.5 % (9) 11.5 % (3) 18.9 % (10) 0.549 0.297
Laparoscopic 83.1 % (221) 82.1 % (23) 90.1 % (91) 84.5 % (49) 76.9 % (20) 71.7 % (38) 0.056 0.019
Conversion 1.1 % (3) 0 % (0) 0 % (0) 0 % (0) 11.5 % (3) 0 % (0) <0.001 0.273
ML 1.9 % (5) 0 % (0) 0 % (0) 0 % (0) 0 % (0) 9.4 % (5) <0.001 0.989
Drain placement 22.2 % (59) 3.6 % (1) 7.9 % (8) 20.7 % (12) 42.3 % (11) 50.9 % (27) <0.001 <0.001
Abbreviations: hrs hours, n number of patients, OP operation, min minutes, Conversion conversion from laparoscopic to midline laparotomy, ML midline laparotomy
*Adjusted for age
Table 3 Histology findings and distribution by disease severity score (G1–G5) of all patients
Time from onset of abdominal pain to surgery Total (n = 266) 0–12 h (n = 28) 13–24 h (n = 101) 25–36 h (n = 58) 37–48 h (n = 26) [48 h (n = 53) P
Normal appendix 8.3 % (22) 14.3 % (4) 9.9 % (10) 1.7 % (1) 3.8 % (1) 11.3 % (6) 0.179
G1 59.4 % (158) 71.4 % (20) 75.2 % (76) 58.6 % (34) 46.2 % (12) 30.2 % (16) <0.001
G2 13.9 % (37) 10.7 % (3) 11.9 % (12) 25.9 % (15) 7.7 % (2) 9.4 % (5) 0.056
G3 12.8 % (34) 3.6 % (1) 3.0 % (3) 13.8 % (8) 34.6 % (9) 24.5 % (13) <0.001
G4 2.6 % (7) 0 % (0) 0 % (0) 0 % (0) 3.8 % (1) 11.3 % (6) <0.001
G5 1.9 % (5) 0 % (0) 0 % (0) 0 % (0) 3.8 % (1) 7.5 % (4) 0.01
Carcinoid 1.1 % (3) 0 % (0) 0 % (0) 0 % (0) 0 % (0) 5.7 % (3) 0.016
Abbreviations: hrs hours, n number of patients, G1 grade 1 (inflamed), G2 grade 2 (gangrenous), G3 grade 3 (perforated with localized free fluid), G4 grade 4 (perforated with regional abscess),
G5 grade 5 (perforated with diffuse peritonitis)
World J Surg
World J Surg
Adj. P*
onset of symptoms to operation [3, 4, 21, 22]. A large
<0.001
0.002
0.374
0.270
0.254
0.992
0.963
0.037
retrospective study by Ditillo et al. demonstrated that the
–
–
severity of histology and complication rate was time-
dependent and delay of appendectomy was associated with
<0.001
<0.001
0.348
0.322
0.347
0.401
0.401
0.038
poor outcomes. These investigators noted that the risk of
–
–
P
6.09 ± 12.49
83.0 % (44)
28.3 % (15)
Abbreviations: hrs hours, n number of patients, post-OP postoperative, AB antibiotic, hosp hospital, CD Clavien-Dindo, CCI comprehensive complication index
surgery on outcomes [4, 22]. Sadot et al. found that a total
0 % (0)
0 % (0)
time interval beyond 48 hours significantly increased the
risk for perforation [22]. Likewise, Bickell et al. observed
similar results when overall interval from admission to
37–48 h (n = 26)
tive design [4, 21, 22] nor did these studies analyze post-
operative adverse events [3, 22]. With regard to HLOS,
5.35 ± 9.91
65.5 % (38)
17.2 % (10)
12.1 % (7)
13.8 % (8)
0 % (0)
0 % (0)
0 % (0)
0 % (0)
9.9 % (10)
0 % (0)
0 % (0)
0 % (0)
0 % (0)
0 % (0)
14.3 % (4)
3.6 % (1)
4.07 ± 9.09
13.2 % (35)
10.5 % (28)
9.8 % (26)
CD V
CD II
CCI
123
World J Surg
HLOS (h) 61.5 ± 111.6 35.6 ± 23.6 40.5 ± 27.2 60.7 ± 50.9 71.9 ± 56.1 111.0 ± 231.6 0.003 <0.001
Post-discharge SSI 5.7 % (15) 3.6 % (1) 5.0 % (5) 10.3 % (6) 7.7 % (2) 2.0 % (1) 0.401 0.957
Readmission 3.8 % (10) 0 % (0) 4.0 % (4) 5.2 % (3) 3.8 % (1) 4.1 % (2) 0.842 0.551
Abbreviations: hrs hours, n number of patients, HLOS hospital length of stay, SSI surgical site infection
*Adjusted for age
123
World J Surg
4. Bickell NA, Aufses AH Jr, Rojas M, Bodian C (2006) How time 17. Nagpal K, Udgiri N, Sharma N, Curras E, Cosgrove JM, Farkas
affects the risk of rupture in appendicitis. J Am Coll Surg DT (2012) Delaying an appendectomy: is it safe? Am Surg
202(3):401–406 78(8):897–900
5. Andersson R, Hugander A, Thulin A, Nyström PO, Olaison G 18. Abou-Nukta F, Bakhos C, Arroyo K, Koo Y, Martin J, Reinhold
(1994) Indications for operation in suspected appendicitis and R, Ciardiello K. Effects of delaying appendectomy for acute
incidence of perforation. BMJ 308(6921):107–110 appendicitis for 12 to 24 hours. Arch Surg 141(5):504-506
6. Ciani S, Chuaqui B (2000) Histological features of resolving 19. United Kingdom National Surgical Research Collaborative,
acute, non-complicated phlegmonous appendicitis. Pathol Res Bhangu A (2014) Safety of short, in-hospital delays before sur-
Pract 196(2):89–93 gery for acute appendicitis: multicentre cohort study, systematic
7. Slankamenac K, Graf R, Barkun J, Puhan MA, Clavien PA review, and meta-analysis. Ann Surg 259(5):894–903
(2013) The comprehensive complication index: a novel contin- 20. Stahlfeld K, Hower J, Homitsky S, Madden J (2007) Is acute
uous scale to measure surgical morbidity. Ann Surg 258(1):1–7 appendicitis a surgical emergency? Am Sur 73(6):626–629
8. Clavien PA, Sanabria JR, Strasberg SM (1992) Proposed classi- 21. Ditillo MF, Dziura JD, Rabinovici R (1995) Is it safe to delay
fication of complications of surgery with examples of utility in appendectomy in adults with acute appendicitis? Ann Surg
cholecystectomy. Surgery 111(5):518–526 244(5):656–660
9. Centers for Disease Control and Prevention (2015) Injury pre- 22. Sadot E, Wasserberg N, Shapiro R, Keidar A, Oberman B,
vention and control. http://www.cdc.gov/nhsn/PDFs/pscManual/ Sadetzki S (2013) Acute appendicitis in the twenty-first century:
9pscSSIcurrent.pdf, Accessed 7 Nov 2014 should we modify the management protocol? J Gastrointest Surg
10. Garst GC, Moore EE, Banerjee MN, Leopold DK, Burlew CC, 17(8):1462–1470
Bensard DD, Biffl WL, Barnett CC, Johnson JL, Sauaia A (2013) 23. Cheng Y, Zhou S, Zhou R, Lu J, Wu S, Xiong X, Ye H, Lin Y,
Acute appendicitis: a disease severity score for the acute care Wu T, Cheng N (2015) Abdominal drainage to prevent intra-
surgeon. J Trauma Acute Care Surg 74(1):32–36 peritoneal abscess after open appendectomy for complicated
11. Busch M, Gutzwiller FS, Aellig S, Kuettel R, Metzger U, Zingg appendicitis. Cochrane Database Syst Rev 7(2):CD010168.
U (2011) In-hospital delay increases the risk of perforation in doi:10.1002/14651858.CD010168
adults with appendicitis. World J Surg 35(7):1626–1633. doi:10. 24. Marudanayagam R, Williams GT, Rees BI (2006) Review of the
1007/s00268-011-1101-z pathological results of 2660 appendicectomy specimens. J Gas-
12. Teixeira PG, Sivrikoz E, Inaba K, Talving P, Lam L, Demetriades troenterol 41(8):745–749
D (2012) Appendectomy timing: waiting until the next morning 25. Ma KW, Chia NH, Yeung HW, Cheung MT (2010) If not
increases the risk of surgical site infections. Ann Surg appendicitis, then what else can it be? A retrospective review of
256(3):538–543 1492 appendectomies. Hong Kong Med J 16(1):12–17
13. Udgiri N, Curras E, Kella VK, Nagpal K, Cosgrove J (2011) 26. Matthyssens PE, Ziol M, Barrat C, Champault GG (2006) Rou-
Appendicitis, is it an emergency? Am Surg 77(7):898–901 tine surgical pathology in general surgery. Br J Surg
14. Shin CS, Roh YN, Kim JI (2014) Delayed appendectomy versus 93(3):362–368
early appendectomy in the treatment of acute appendicitis: a 27. Salminen P, Paajanen H, Rautio T, Nordström P, Aarnio M,
retrospective study. World J Emerg Surg 9(1):8 Rantanen T, Tuominen R, Hurme S, Virtanen J, Mecklin JP, Sand
15. Drake FT, Mottey NE, Farrokhi ET, Florence MG, Johnson MG, J, Jartti A, Rinta-Kiikka I, Grönroos JM (2015) Antibiotic therapy
Mock C, Steele SR, Thirlby RC, Flum DR (2014) Time to vs appendectomy for treatment of uncomplicated acute appen-
appendectomy and risk of perforation in acute appendicitis. dicitis: the APPAC randomized clinical trial. JAMA
JAMA Surg 149(8):837–844 313:2340–2348
16. Giraudo G, Baracchi F, Pellegrino L, Dal Corso HM, Borghi F 28. Wilms IM, de Hoog DE, de Visser DC, Janzing HM (2011)
(2013) Prompt or delayed appendectomy? Influence of timing of Appendectomy versus antibiotic treatment for acute appendicitis.
surgery for acute appendicitis. Surg Today 43(4):392–396 Cochrane Database Syst Rev 11:CD008359
123