Utility of Positron

Emission Tomography
in the Mediastinum:
Moving Beyond Lung
and Esophageal Cancer
Staging
Varun Puri, MD, Bryan F. Meyers, MD, MPH*

KEYWORDS
 PET  Mediastinum  Lymphoma  Thymoma
 Germ cell tumor

Broadly, all imaging can be divided into functional/ used most commonly because of its convenient
molecular and anatomic. Functional imaging may half-life (110 minutes) and easy availability.
be defined as the ‘‘in vivo characterization and Lack of detailed resolution and inability to accu-
measurement of biologic processes at the cellular rately localize anatomy on PET images have been
and molecular level.’’1 Positron emission tomogra- the inherent drawbacks of PET technology. These
phy (PET) is the most widely employed functional weaknesses have led scientists to investigate the
imaging technique, while the modern-day proto- possibility of combining PET with the precise ana-
type of anatomic imaging is the CT scan. tomic imaging offered by CT. Two dichotomous
The last decade has seen a rapid increase in the approaches have been pursued: software fusion
use of PET technology in clinical medicine, and the and hardware fusion. Attempts to align or register
trend is expected to continue (Fig. 1). CT and PET data sets by using fusion software
Much of this has been in the field of oncology, have proven successful imaging the brain. In the
with PET being used to diagnose and stage tumors remainder of the body, however, differences in
and to assess response to therapy, usually in scanner bed profiles, external patient positioning,
conjunction with other imaging modalities. and internal organ movement present challenges
to the pure software approaches.2 Hardware fu-
TECHNICAL ASPECTS sion has largely resolved these issues with the in-
troduction of the combined PET/CT scanner. This
PET is based upon a chemical process termed technology has been commercially available since
‘‘annihilation,’’ which involves the collision of an 2001 and most PET scanners in the United States
electron with a positron that is emitted by a radio- are now combined CT/PET scanners. One must
isotope. The process results in the production of understand, however, that the CT scan performed
two photons that simultaneously move in diametri- as part of the CT/PET is usually a noncontrast
cally opposite directions and are detected by study using a lower radiation dose and provides
a PET scanner that encircles the body. Several less anatomic detail than a contrast-enhanced
positron-emitting isotopes have been developed standard CT scan, especially in the mediastinum.
(Table 1), but 18F fluorodeoxyglucose (FDG) is Other attempts at combining functional and
thoracic.theclinics.com

Division of Cardiothoracic Surgery, Washington University School of Medicine, Barnes- Jewish Hospital,
Queeny Tower, 3108, One Barnes-Jewish Hospital Plaza, St. Louis, MO 63110, USA
* Corresponding author.
E-mail address: [email protected] (B.F. Meyers).

Thorac Surg Clin 19 (2009) 7–15

doi:10.1016/j.thorsurg.2008.09.006
1547-4127/08/$ – see front matter ª 2009 Elsevier Inc. All rights reserved.
8 Puri & Meyers

2.0 Fig.1. Historic and forecast positron emission

volume in millions Total PET procedures tomography (PET) procedure volume for car-
PET procedure

Oncology imaging
1.5 Cardiology imaging diology and oncology from 1999 to 2009.
(Data from Podoloff DA, Advani RJ, Allred
1.0 C, et al. NCCN Task Force report: positron
emission tomography (PET)/computed to-
0.5 mography (CT) scanning in cancer. J Natl
Compr Canc Netw 2007;5(2):S1–22.)
0
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

anatomic imaging are ongoing and simultaneous nonsmall cell lung cancer and esophageal cancer.
PET-MRI scanning was reported recently.3 For lung cancer, PET provides a significant advan-
tage when combined with conventional anatomic
INTERPRETATION OF POSITRON EMISSION imaging, in diagnosing and staging.4 In a meta-
TOMOGRAPHY analysis, PET was 95% sensitive and 77% specific
FDG is the most commonly used tracer, and it de- in providing a diagnosis.5 Similarly, for esophageal
lineates the uptake of glucose in various tissues of cancer, PET has been used in the diagnostic set-
the body. Substitution of fluoride for a hydroxyl ting6 and in monitoring response to treatment.4
group allows cellular uptake but prevents metabo- The discussion about PET in the thorax has been
lism of the tracer; thus tracer accumulation and ac- dominated by lung and esophageal malignancies
tivity are proportional to glycolysis in the anatomic because of the sheer number of papers and vol-
area. Most tumors are obligate consumers of glu- ume of information. For the purpose of this article,
cose and avidly accumulate FDG. Such uptake is the authors will exclude the extensive and volumi-
also increased in various benign pathologies, nous literature surrounding lung and esophageal
such as inflammatory conditions, regenerating tis- cancer.
sue in trauma, infection, persistent muscular activ-
ity, and granulomatous diseases. Familiarity with THE MEDIASTINUM
normal background activity in various tissues is
essential in detecting abnormalities. The mediastinum is populated by structures
PET results are often reported with an SUV derived from all three germ lines with various basal
(standardized uptake value): a semi-quantitative metabolic rates. This unique anatomic and func-
measure of FDG uptake and activity in tissue that tional arrangement makes the mediastinum partic-
frequently has been criticized but also widely ularly suited to accurate evaluation with CT-PET.
accepted because of its relative simplicity. The Myocardial and background mediastinal activity
general formula for calculating SUV is activity per must be understood to accurately interpret PET
unit volume/injected activity/body weight. Maxi- imaging in this area. Mediastinal anatomy can be
mum SUV is a better indicator than average SUV categorized along conventional lines7 (Fig. 2),
and provides more clinical information when the and this helps in classification of mediastinal pa-
there is a large difference between the area of thology (Box 1).
interest and background activity.
The Anterior Mediastinum
POSITRON EMISSION TOMOGRAPHY
FOR THORACIC MALIGNANCIES Information about the use of PET in the anterior
mediastinum is dominated by the application of
In the last decade, PET has been employed widely PET to assessment of the thymus gland. FDG up-
in the thorax in the diagnosis and follow-up of take is commonly seen in the thymus and can be

Table 1
Positron-emitting isotopes in clinical use

Positron Isotope Half-Life (Minutes) Use

F18 109.7 Metabolism, bone imaging
N13 9.96 Perfusion
C11 20.4 Multiple uses
Rb82 1.3 Perfusion
O15 2.07 Perfusion, blood flow, blood volume, metabolism
 Utility of PET in the Mediastinum 9

The radiologic appearance of the so-called

physiologic thymic uptake at PET is characteristic.
It appears as a triangular retrosternal region of in-
creased uptake. Although data are scant, when
the SUVmax is greater than 4 in the thymus, this
should not be considered physiologic or related
to hyperplasia.11 An uncommon scenario occurs
when a patient who has a known thymic abnormal-
ity, often detected on a CT scan, undergoes PET to
differentiate hyperplasia from neoplasia and,
within neoplasia, to distinguish thymoma from
thymic carcinoma. This is a clinical situation that
occurs, but that does not have a large database
of outcomes to offer guidance for decision making.
Liu and colleagues12 reported the preliminary find-
Fig. 2. Three-compartment model of mediastinal anat- ings of a small study with 10 cases of thymoma
omy. (From Raymond DP, Daniel TM. Mediastinal anat- and two cases of thymic hyperplasia. The findings
omy and mediastinoscopy. In: Selke FW, del Nido PJ, suggested diffuse uptake of FDG in thymic hyper-
Swanson SJ, editors. Sabiston and Spencer surgery of plasia, confined focal uptake, in stage 1 and 2 thy-
the chest. 7th edition. Philadelphia: Elsevier Saunders; momas, and multiple discrete foci of FDG uptake
2005. p. 668; with permission.) in stage 3 and 4 advanced invasive thymomas.
Similarly, another small study showed that the pat-
tern of uptake could predict hyperplasia versus
considered normal when there is no clinical, radio- thymoma in patients with myasthenia reliably.13
logic, or hematologic evidence of thymic pathol- Despite these small studies suggesting value in
ogy. It generally has been accepted that diffuse PET, it has not become standard practice to
thymic FDG uptake is seen in more than half of evaluate thymic neoplasia with PET.
prepubertal individuals and rapidly diminishes af- The use of the semi-quantitative SUVmax can
terwards.8 Subsequent studies, however, have help differentiate between thymoma and thymic
shown that physiologic thymic uptake can be carcinoma, but PET has been inconsistent in dis-
seen in patients well beyond puberty. One study tinguishing noninvasive from invasive thymoma
examined thymic uptake of 18F-FDG PET in 94 (Fig. 3).
patients ranging from 18 to 29 years of age and Sasaki and colleagues14 found a mean SUV of
found that 32 of these patients exhibited physio- 7.2  2.9 for patients who had thymic carcinoma.
logic thymic uptake.9 Diffuse FDG uptake in the This value was significantly greater than the values
thymus does not discriminate normal from dis- found for invasive thymoma (3.8  1.3) and nonin-
eased thymus in the absence of other radiological vasive thymoma (3.0  1.0). By using an SUV of
or clinical information. 5.0 as a cut-off, the authors achieved acceptable
A study by Brink and colleagues10 examined sensitivity (84.6%), specificity (92.3%), and accu-
18F-FDG PET findings in four sets of patients racy (88.5%) in differentiating thymic carcinoma
who had known malignancy: (1) children who had from thymoma. Given the relative rarity of thymic
malignancy before chemotherapy, (2) children carcinoma, it seems likely that a similar sensitivity
who had malignancy following chemotherapy, (3) and specificity would be obtained with a purely
adults who had lymphoma before chemotherapy, clinical assessment based on CT scanning. They
and (4) adults who had lymphoma following found no difference in SUV between invasive and
chemotherapy. The authors found increased noninvasive thymomas. An earlier study by Liu
18F-FDG uptake in 73% of the children who had and colleagues,12 however, indicated that PET
malignancy before chemotherapy and in 75% of may be able to distinguish invasive from noninva-
the children who had malignancy following che- sive thymoma based upon pattern of FDG uptake.
motherapy. They also found increased uptake in Kubota and colleagues15 also showed that FDG
5% of the adults who had lymphoma following uptake in invasive thymoma was significantly
chemotherapy, however. None of the adults who higher than in noninvasive thymoma. How that
had lymphoma before chemotherapy exhibited discrimination would be translated into a different
increased thymic uptake. Thus, although thymic treatment approach is not immediately clear.
uptake is common in children, it occasionally In a retrospective review study group of eight
may be seen in adults after chemotherapy, possi- patients, El Bawab and colleagues13 found that
bly related to an immunologic rebound. FDG—PET was a very sensitive diagnostic tool in
10 Puri & Meyers

Box 1 Lymphoma
Mass lesions of the mediastinum Soft tissue tumor
Anterior mediastinum Infectious/inflammatory
Neoplastic Fungal adenopathy
Substernal goiter Sarcoidosis
Thymoma Mycobacterial disease
Thymic hyperplasia Fibrosing mediastinitis
Thymic cyst Mediastinal abscess
Ectopic parathyroid mass Cystic lesions
Lymphoma Bronchogenic
Teratoma Esophageal
Germ cell tumor (GCT) Others
Seminoma Extramedullary hematopoiesis
Nonnseminomatous GCT

Soft tissue tumor detecting residual, recurrent, and metastatic

Infectious/inflammatory disease in patients after thymectomy for thymoma,
Mediastinitis but it was not superior to CT scan.
Vascular
Aneurysm of innominate vein
Lymphoma
Superior vena cava aneurysm Lymphoma represents less than 4% of all neopla-
sia but may account for up to 50% of PET scans
Middle visceral mediastinum
performed at referral centers.1 Lymphoma can in-
Neoplastic volve any organ system and mimic other tumors in
Tracheal tumors its radiologic appearance. The mediastinum may
Esophageal tumors be the primary site of disease, like in Hodgkins’s
lymphoma and primary mediastinal B cell lym-
Lymphoma phoma, or it may be involved as part of systemic
Infectious/inflammatory disease.
Fungal adenopathy Like all malignancies, the diagnosis of
lymphoma requires histologic assessment, and
Sarcoidosis
PET is not likely to be used as an initial diagnostic
Mycobacterial disease modality. Routine staging of a biopsy- established
Fibrosing mediastinitis lymphoma involves a CT scan with contrast, from
Mediastinitis the neck to the pelvis, and a PET scan in addition.
PET virtually has replaced gallium scan in the
Cystic lesions
pretreatment work-up of lymphoma. A health tech-
Bronchogenic nology assessment report from the United King-
Esophageal dom4 summarized that PET was 90% sensitive
and 90% specific when used in the initial staging
Pleuropericardial
of lymphoma. They also noted that the addition
Posterior paravertebral mediastinum of PET changed the staging/management in up
Neoplastic to 20% of patients. One caveat was that staging
Neurogenic tumors PET scan can be omitted as part of staging when
management is to consist of observation alone,
Neurofibroma like in some patients who have follicular lym-
Neurofibrosarcoma phoma. The International Harmonization Project
Ganglioneuroblastoma has provided guidelines for the use of PET in lym-
phoma.16 They recommend that a pretherapy
Parganglioma
baseline PET is not obligatory for assessing
response after treatment of patients who have
some subgroups of lymphoma, because these
 Utility of PET in the Mediastinum 11

Fig. 3. A 33-year-old woman with stage IIa thymoma (case 9). (A) Axial CT shows solid homogenous mass in the
anterior mediastinum. (B) 18F fluorodeoxyglucose (FDG)—positron emission tomography shows intense FDG up-
take in the anterior mediastinum and contiguous pleura suggesting thymoma. (From El-Bawab H, Al-Sugair AA,
Rafay M, et al. Role of flourine-18 fluorodeoxyglucose positron emission tomography in thymic pathology. Euro
J Cardiothorac Surg 2007;31(4):734; with permission.)

lymphomas routinely are FDG-avid. Pretherapy confound the interpretation of PET scans, the Inter-
PET, however, may be valuable, because it can national Harmonization Project recommends that
facilitate the interpretation of post-therapy PET. scans should not be performed sooner than
PET has had the highest impact in management 3 weeks after chemotherapy and should preferably
of lymphoma in the assessment of response to occur 8 to 12 weeks after completion of radiother-
therapy and guiding decisions about further treat- apy. Response assessment at the conclusion of
ment (Fig. 4). Juweid and colleagues17 assessed therapy is recommended; however, no specific
response to therapy in 54 patients who had guidelines about midtherapy assessment are avail-
aggressive non-Hodgkin’s lymphoma using the able. Visual assessment alone appears to be ade-
established criteria in conjunction with PET scan quate for determining whether PET is positive or
results and compared response with progres- negative at the conclusion of therapy, and quantita-
sion-free survival. Using the CT-based criteria tive or semiquantitative approaches (eg, using the
alone, 17 patients experienced a complete re- SUV) do not seem necessary.16
sponse, and seven experienced a less than com- The true incidence of false-positive scans in fol-
plete response. In contrast, when PET results low-up, however, remains unresolved. Zinzani
were incorporated, 35 patients experienced and colleagues19 followed 151 patients who had
a complete response, and no patients experi- mediastinal lymphoma (57 with Hodgkin’s disease
enced less-than-complete response. Therefore, and 94 with aggressive non-Hodgkin’s lymphoma
a negative PET scan even in the presence of a re- after the end of front-line treatment. Patients who
sidual mass was interpreted correctly as a com- had a positive PET scan of the mediastinum under-
plete response. Subsequently, the International went CT scanning and surgical biopsy. In 30 cases,
Harmonization Project has published revised a suspicion of lymphoma relapse was raised based
response criteria for treating lymphoma incorpo- on positive mediastinal PET scanning. Histology
rating PET.18 confirmed this suspicion in 17 out of 30 patients,
Post-therapy inflammatory changes can be seen whereas either benign (nine fibrosis, three sar-
after chemotherapy and radiation. To minimize the coid-like granulomatosis) or unrelated neoplastic
frequency of these changes, which potentially conditions (one thymoma) were demonstrated in
12 Puri & Meyers

Fig. 4. Positron emission tomography (PET)-positive residual mass in a patient with initial diagnosis of stage IIA
nodular sclerosis Hodgkin’s disease. This patient underwent a restaging PET/ CT scan 2 months after treatment
with six cycles of doxorubicin, bleomycin, vinblastine, and dacarbazine. (Top panel) Fused PET/CT images show
increased uptake clearly greater than that of mediastinal blood pool structures within the residual anterior me-
diastinal mass, consistent with persistent disease. (Middle panel) A follow-up PET/CT scan performed 2 months
later shows substantially more intense uptake in the same area, again suggesting persistent disease. (Lower
panel) A third follow-up scan performed 4 months thereafter also demonstrates intense uptake in the same
area but now with additional new sites of disease in the right paratracheal, right hilar, and subcarinal regions,
indicating frank disease progression. The patient then underwent high-dose chemotherapy and autologous
stem cell transplantation. (Courtesy of M. Juweid, MD, Iowa City, IA.)

the remaining 13 patients (43%). SUVmax was sig- mediastinum and frequently present as a very
nificantly higher among the 17 patients who had large mass with local compression. The mediasti-
signs of relapse (median 5.95, range 3.5 to 26.9) num can also be involved by metastatic germ cell
than among the 13 patients who stayed in remis- tumor from the gonads. Chest CT and evaluation
sion (median 2.90, range 1.4 to 3.3). They recom- of serum tumor markers provide the critical
mended routine biopsy before further therapy in workup before tissue diagnosis is obtained. PET
this setting. has found a niche in the evaluation of these meta-
In addition, PET is useful for directing biopsy of bolically active tumors.
the most suspicious areas based on SUV.1 Seminomas are extremely sensitive to chemo-
therapy and radiation and primarily are treated
Germ Cell Tumors
nonsurgically. The dilemma lies in managing post-
Mediastinal germ cell tumors are uncommon tu- chemotherapy residual masses. Traditionally,
mors that occur predominantly within the anterior masses larger than 3 cm on CT have been
 Utility of PET in the Mediastinum 13

considered to potentially harbor residual disease selective venous sampling, and FDG PET and
and thus offered further therapy, usually resection. 11C methionine PET. Beggs and colleagues26 ret-
This approach was re-evaluated with addition of rospectively evaluated 51 patients presenting with
FDG PET studies in patients with metastatic pure hyperparathyroidism, in whom other imaging tech-
seminoma who had radiographically defined post- niques had failed to definitely identify the site of
chemotherapy residual masses.20 The PET studies suspected parathyroid adenoma. All patients had
were correlated with either the histology of the re- undergone 11C methionine PET scanning.
sected lesion or the clinical outcome documented Patients were followed up by surgical histology,
by CT, tumor markers, or physical examination or clinically if surgery was not performed. 11C
during follow-up. All 19 cases with residual lesions methionine PET scanning was found to have a sen-
greater than 3 cm and 35 (95%) of 37 with residual sitivity of 83%, a specificity of 100%, and an accu-
lesions less than or equal to 3 cm were predicted racy of 88% in successfully locating parathyroid
correctly by FDG PET. FDG-PET was superior to adenomas. Most false negatives were caused by
CT scan in detecting residual disease (specificity adenomas in the lower mediastinum that were
100% versus 74%, sensitivity 80% versus 70%, outside the area of scanning.
positive predictive value 100% versus 37%, and
negative predictive value 96% versus 92%,
Middle Mediastinum
respectively).
In another report, FDG PET was shown to be ac- Lymphoma, once again, is a prominent differential
curate in the initial staging of germ cell tumors and diagnosis in middle mediastinal neoplasia. Inflam-
in detecting unsuspected metastatic disease.21 matory involvement of mediastinal lymph nodes
Residual masses in nonseminomatous germ cell (LNs) may be acute, subacute, or chronic. Acute
tumors (NSGCT) may contain necrotic tissue, lymphadenopathy, often related to pneumonia, is
viable NSGCT, or teratoma. Kollmannsberger FDG-avid, but the activity diminishes with follow-
and colleagues22 studied patients who had up studies. Subacute or chronic LN involvement
NSGCT; independent reviewers who were blinded may be fungal or mycobacterial. Mediastinal histo-
to each other’s results evaluated the PET results plasmosis is a frequently seen entity in the mid-
and corresponding CT scan and tumor marker re- western United States and can masquerade as
sults in 85 residual lesions from 45 patients. a malignant mediastinal process in patients who
Resulting sensitivities and specificities for the pre- have known malignancy elsewhere.27 The pres-
diction of residual mass viability were as follows: ence of calcification in LNs on CT and concomitant
PET, 59% sensitivity and 92% specificity; granulomas in the lungs may provide a clue to di-
radiologic monitoring, 55% sensitivity and 86% agnosis, but tissue diagnosis frequently is
specificity; and tumor markers (TUM), 42% sensi- required.
tivity and 100% specificity. The positive and nega- The assessment of mediastinal LNs with PET is
tive predictive values for PET were 91% and 62%, even more problematic in tuberculosis-endemic
respectively. This study shows that a positive PET countries. In a study from Korea involving 674 pa-
image after treatment was a strong predictor for tients who had lung cancer,28 nodes with greater
the presence of viable carcinoma/teratoma. Con- FDG uptake than the mediastinum at PET without
versely, a negative PET after the end of treatment benign pattern calcification or high attenuation
does not allow one to avoid surgery, because 37% (greater than 70 Hounsfield units) on unenhanced
of all negative PET masses either progressed dur- CT scans were regarded as being positive for ma-
ing 6 months of follow-up, or led to a histologic ex- lignancy. The overall sensitivity, specificity, and
amination showing mature teratoma. accuracy of PET/CT for mediastinal nodal staging
Only anecdotal reports show the usefulness of were 61%, 96%, and 86%, respectively. The high
PET in mediastinal germ cell tumors,23,24 but the specificity was achieved at the expense of sensi-
identical biologic nature of disease in the medias- tivity by interpreting calcified nodes or nodes
tinum and elsewhere makes extrapolation from with high attenuation at CT, even with high FDG
previous reports meaningful. uptake at PET, as benign.
One scientist’s false positive is another’s true
PARATHYROID ADENOMA positive. Although sarcoidosis and other granulo-
matous lesions of the mediastinum can create a di-
The incidence of mediastinal parathyroid glands lemma on PET imaging in patients who have
requiring surgery was 3% in a large database.25 known malignancy, scientists have found a unique
Strategies to localize parathyroid tissue before application for PET in assessment of sarcoidosis.
parathyroid surgery encompass CT, MRI, 99Tc FDG uptake in sarcoidosis is nonspecific in both
Sestamibi scan, ultrasound, arteriography, intensity and pattern, and it is not useful in making
14 Puri & Meyers

an initial diagnosis. Diffuse metastatic disease and REFERENCES

lymphoma need to be ruled out. FDG uptake, how-
ever, can decrease when sarcoidosis is treated, 1. Podoloff DA, Advani RJ, Allred C, et al. NCCN task
and PET might be useful in monitoring the effec- force report: positron emission tomography (PET)/
tiveness of therapy.29 computed tomography (CT) scanning in cancer.
Brown adipose tissue can be variably FDG-avid. J Natl Compr Canc Netw 2007;5(2):S1–22.
PET/CT scans were obtained in a study of 845 2. Blodgett TM, Meltzer CC, Townsend DW. PET/CT:
patients who had known cancer.30 Fifteen patients form and function. Radiology 2007;242(2):360–85.
(1.8%) who had focal hypermetabolic mediastinal 3. Judenhofer MS, Wehrl HF, Newport DF, et al.
brown fat were identified. Hypermetabolic medias- Simultaneous PET-MRI: a new approach for functional
tinal brown fat (mean SUVmax 5.7) was more com- and morphological imaging. Nat Med 2008;14(4):
mon in children (four of eight) than in adults 11 of 459–65.
837) and more common in women (9 of 372) than 4. Facey K, Bradbury I, Laking G, et al. Overview of the
in men (2 of 465). Foci of hypermetabolic brown clinical effectiveness of positron emission tomogra-
fat were localized to the paratracheal, paraeso- phy imaging in selected cancers. Health Technol
phageal, prevascular, and pericardial regions; in- Assess 2007;11(44):1–285.
teratrial septum; and azygoesophageal recess. 5. Gould MK, Maclean CC, Kuschner WG, et al. Accu-
Ten patients also had extramediastinal hypermet- racy of positron emission tomography for diagnosis
abolic brown fat in the neck, thorax, and abdomen. of pulmonary nodules and mass lesions: a meta-
The authors concluded that precise localization of analysis. JAMA 2001;285(7):914–24.
brown fat with fusion PET/CT is important in pre- 6. Meyers BF, Downey RJ, Decker PA, et al. The utility
venting misinterpretation as malignancy. Premed- of positron emission tomography in staging of poten-
ication with diazepam and fentanyl has been used tially operable carcinoma of the thoracic esophagus:
to suppress FDG uptake in brown fat.31 results of the American College of Surgeons Oncol-
ogy Group Z0060 trial. J Thorac Cardiovasc Surg
2007;133(3):738–45.
7. Raymond DP, Daniel TM. Mediastinal anatomy and
Posterior Mediastinum
mediastinoscopy. In: Selke FW, delNido PJ,
Cystic lesions arising from the foregut are not FDG- Swanson SJ, editors. Sabiston and Spencer surgery
avid. Neurogenic tumors are the most common of the chest. 7th edition. Philadelphia: Elsevier Saun-
mass lesions unique to the posterior mediastinum. ders; 2005. p. 657–66.
Schwannomas are benign peripheral nerve sheath 8. Patel PM, Alibazoglu H, Ali A, et al. Normal thymic
tumors than can arise from intercostal nerves. Stud- uptake of FDG on PET imaging. Clin Nucl Med
ies have tried to differentiate schwannomas from 1996;21(10):772–5.
sarcoma preoperatively based on PET. Although 9. Nakahara T, Fujii H, Ide M, et al. FDG uptake in the
a cutoff SUVmax of 1.8 has been proposed,32 sub- morphologically normal thymus: comparison of
sequent studies33,34 have shown significantly high- FDG positron emission tomography and CT. Br J
er SUVmax in benign schwannomas, and the Radiol 2001;74(885):821–4.
diagnosis must be considered despite high FDG 10. Brink I, Reinhardt MJ, Hoegerle S, et al. Increased
avidity. Neurofibrosarcomas of the posterior medi- metabolic activity in the thymus studied with FDG
astinum, like sarcomas elsewhere, are expected PET: age dependency and frequency after chemo-
to have significant FDG avidity. therapy. J Nucl Med 2001;42(4):591–5.
11. Fernidand B, Gupta P, Kramer EL. Spectrum of thy-
mic uptake at 18F-FDG PET. Radiographics 2004;
SUMMARY 24(6):1611–6.
12. Liu RS, Yeh SH, Huang MH, et al. Use of fluorine-18
PET and PET-CT are rapidly evolving as modalities fluorodeoxyglucose positron emission tomography
of thoracic imaging. In the mediastinum, PET can in the detection of thymoma: a preliminary report.
provide information to distinguish thymic hyper- Eur J Nucl Med 1995;22(12):1402–7.
plasia from neoplasia, although the use of this im- 13. El-Bawab H, Al-Sugair AA, Rafay M, et al. Role of
aging for this purpose is not accepted uniformly as flourine-18 fluorodeoxyglucose positron emission to-
necessary. PET is the standard of care in staging mography in thymic pathology. Eur J Cardiothorac
and follow-up of mediastinal lymphoma and in fol- Surg 2007;31(4):731–6.
low-up of metastatic seminomas after chemother- 14. Sasaki M, Kuwabara Y, Ichiya Y, et al. Differential
apy. Mycobacterial/fungal infections, sarcoidosis, diagnosis of thymic tumors using a combination of
and brown fat can mimic malignant findings on 11C-methionine PET and FDG PET. J Nucl Med
PET in the mediastinum. 1999;40(10):1595–601.
 Utility of PET in the Mediastinum 15

15. Kubota K, Yamada S, Kondo T, et al. PET imaging of teratoma syndrome or cancer recurrence? J Clin
primary mediastinal tumours. Br J Cancer 1996; Oncol 2007;25(28):4494–6.
73(7):882–6. 25. Nilubol N, Beyer T, Prinz RA, et al. Mediastinal
16. Juweid ME, Stroobnts S, Hoekstra OS, et al. Use of hyperfunctioning parathyroids: incidence, evolving
positron emission tomography for response assess- treatment, and outcome. Am J Surg 2007;194(1):
ment of lymphoma: consensus of the imaging sub- 53–6.
committee of international harmonization project in 26. Beggs AD, Hain SF. Localization of parathyroid
lymphoma. J Clin Oncol 2007;25(5):571–8. adenomas using 11C-methionine positron emission
17. Juweid ME, Wiseman ME, Vose JM, et al. Response tomography. Nucl Med Commun 2005;26(2):
assessment of aggressive non-Hodgkin’s lymphoma 133–6.
by integrated International Workshop Criteria and 27. Mackie GC, Pohlen JM. Mediastinal histoplasmosis:
fluorine-18-fluorodeoxyglucose positron emission to- F-18 FDG PET and CT findings simulating malignant
mography. J Clin Oncol 2005;23(21):4652–61. disease. Clin Nucl Med 2005;30(9):633–5.
18. Cheson BD, Pfistner B, Juweid ME, et al. Revised 28. Kim YK, Lee KS, Kim BT, et al. Mediastinal nodal
response criteria for malignant lymphoma. J Clin staging of nonsmall cell lung cancer using inte-
Oncol 2007;25(5):579–86. grated 18F-FDG PET/CT in a tuberculosis-endemic
19. Zinzani PL, Tani M, Trisolini R, et al. Histological ver- country: diagnostic efficacy in 674 patients. Cancer
ification of positive positron emission tomography 2007;109(6):1068–77.
findings in the follow-up of patients with mediastinal 29. Nishiyama Y, Yamamoto Y, Fukunaga K, et al. Com-
lymphoma. Haematologica 2007;92(6):771–7. parative evaluation of 18F–FDG PET and 67Ga scin-
20. deSantis M, Becherer A, Bokemeyer C, et al. 18fluoro- tigraphy in patients with sarcoidosis. J Nucl Med
deoxy-D-glucose positron emission tomography is 2006;47(10):1571–6.
a reliable predictor for viable tumor in postchemother- 30. Truong MT, Erasmus JJ, Munden RF, et al. Focal
apy seminoma: an update of the prospective multi- FDG uptake in mediastinal brown fat mimicking
centric SEMPET trial. J Clin Oncol 2004;22(6):1034–9. malignancy: a potential pitfall resolved on PET/CT.
21. Hain S, O’Doherty M, Timothy A, et al. Fluorodeoxy- Am J Roentgenol 2004;183(4):1127–32.
glucose PET in the initial staging of germ cell 31. Gelfand MJ, O’hara SM, Curtwright LA, et al. Pre-
tumours. Eur J Nucl Med 2000;27(5):590–4. medication to block [(18)F]FDG uptake in the brown
22. Kollmannsberger C, Oeschle K, Dohmen B, et al. adipose tissue of pediatric and adolescent patients.
Prospective comparison of [18F]fluorodeoxyglucose Pediatr Radiol 2005;35(10):984–90.
positron emission tomography with conventional 32. Watanabe H, Shinozaki T, Aoki J, et al. Glucose met-
assessment by computed tomography scans and abolic analysis of musculoskeletal tumors using fluo-
serum tumor markers for the evaluation of residual rine-18-FDG PET as an aid in preoperative planning.
masses in patients with nonseminomatous germ J Bone Joint Surg Br 2000;82(5):760–7.
cell carcinoma. Cancer 2002;94(9):2353–62. 33. Ahmed AR, Watanabe H, Aoki J, et al. Schwannoma
23. Bachman J, Ernestus K, Werner T, et al. Detection of of the extremities: the role of PET in preoperative
primary choriocarcinoma in the mediastinum by planning. Eur J Nucl Med 2001;28(10):1541–51.
F-18 FDG positron emission tomography. Clin Nucl 34. Beaulieu S, Rubin B, Djang D, et al. Positron emis-
Med 2007;32(8):663–5. sion tomography of schwannomas: emphasizing its
24. Aide N. Enlarging residual mass after treatment of potential in preoperative planning. Am J Roentgenol
a nonseminomatous germ cell tumor: growing 2004;182(4):971–4.