| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
EDUCATION EXHIBIT |
1 Department of Radiology, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania
I thank the editor of RadioGraphics for the opportunity to comment on the preceding article by Chong et al (1) in this issue of the journal. The authors present an excellent pictorial essay on the application of integrated 18F-FDG PET for differentiating benign from malignant adrenal lesions in cancer patients. The finding at cross-sectional imaging of an adrenal lesion in a patient with a known malignancy poses a diagnostic dilemma: It represents either a metastatic lesion or a functioning or nonfunctioning adrenal adenoma unrelated to the primary tumor, a so-called incidentaloma. These lesions are found in up to 5% of adrenal nodules at CT performed in patients with a primary malignancy (1). Because adrenal metastases are generally FDG avid, whereas most adrenal adenomas are not, FDG PET has been successfully used in this patient population.
Common causes of false-negative FDG PET for adrenal metastases include necrotic metastases, metastatic lesions from bronchoalveolar cell tumors and carcinoid tumors, and small metastases (ie, <1 cm). The latter should be considered indeterminate for metastatic disease. False-positive results occur when adrenal adenomas concentrate FDG, a finding that presumably reflects enhanced metabolic activity at the time of imaging (ie, a functioning adenoma). Pheochromocytomas are also potentially false positive, with 58% of benign pheochromocytomas and 88% of malignant pheochromocytomas demonstrating elevated FDG uptake (2). Chong at el (1) present a case of adrenal cortical hyperplasia and a case of adrenal endothelial cyst, both of which lesions concentrated FDG.
Using a specific interpretative criterion—namely, the adrenal uptake of FDG relative to the hepatic uptake—Yun et al (3) demonstrated a sensitivity of 100% for FDG PET in distinguishing metastatic from benign adrenal lesions, a specificity of 94%, and an overall accuracy of 96%. An adrenal lesion was considered positive for metastases if the FDG uptake in the lesion was equal to or greater than that in the liver. In their series, none of the metastatic lesions had uptake less than that of the liver (ie, a zero false-negative rate).
Integrated PET-CT has the theoretic advantage of superimposing the FDG uptake (or lack thereof) onto the lesion at CT, resulting in exact localization. The largest series to make use of integrated PET-CT in this clinical scenario was published by Metser et al (4) in the January 2006 issue of the Journal of Nuclear Medicine. In the evaluation of 175 adrenal lesions with PET-CT, the authors reported a sensitivity of 100%, a specificity of 98%, and an accuracy of 99%. They used a maximum SUV of 3.1 as a threshold for malignancy; in other words, an SUV of 3.1 or more was taken to indicate a metastatic lesion. However, as pointed out by Chong et al (1), the use of a particular number for a maximum SUV as a cutoff for malignancy should be a guideline only, since there will always be benign adrenal lesions whose SUV exceeds 3.1, as well as adrenal metastases whose SUV is less than 3.1. Case in point: In Figures 7–9 of the preceding article, the SUVs were all less than 3.1, all three cases were interpreted as negative for adrenal metastases, and all were falsely negative. In two of these cases, the adrenal lesions were smaller than 1 cm, potentially resulting in the so-called partial volume averaging effect, which leads to underestimation of the SUV. At our institution, the interpretative criterion used is that of Yun et al (3), namely, that an adrenal focus is considered positive for metastasis if its activity is equal to or greater than that of the liver. If a lesion is smaller than 1 cm, it is considered nondiagnostic or indeterminate because the degree of uptake can be underestimated.
My greatest departure from the authors concerns their approach to laboratory testing. If a cancer patient with an extraadrenal malignancy is not clinically suspected of having pheochromocytoma or Cushing syndrome, there is no need for the extensive biochemical testing and assays recommended by the authors. Urine catecholamine and cortisol levels, multiple serum cortisol levels, aldosterone and renin sampling, and especially the overnight dexamethasone suppression test—all constitute a typical endocrine work-up and would be considered "overkill" for a cancer patient with an adrenal mass. The expense of such a work-up is prohibitive, especially in the current climate of healthcare cost containment.
 |
References |
|---|
 Top References References |
|---|
2 Department of Radiology and Center for Imaging Science
3 Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea
We thank Dr Intenzo for his commentary on our article.
Most incidentalomas identified in patients without known malignancy are benign adrenocortical tumors that do not secrete excess hormone, but the differential diagnosis includes adrenal adenoma causing Cushing syndrome or primary hyperaldosteronism, pheochromocytoma, adrenocortical carcinoma, and metastatic cancer. Patients should be evaluated for symptoms that suggest pheochromocytoma (episodic headache, palpitations, and sweating) and signs of Cushing syndrome. Plasma potassium and detailed hormonal levels are assessed. The prevalence of metastatic nodules among detected adrenal nodules in cancer patients is high, ranging from 38% to 57% (1–3). Therefore, in these patients, such incidentalomas are relatively less likely to be a functioning adenoma than in patients without extraadrenal malignancy. Thus, detailed endocrine hormonal work-up may not be needed unless the patient has clinically suspected pheochromocytoma or Cushing syndrome. However, in patients who have potentially resectable cancer elsewhere and in whom an adrenal metastasis must be excluded with adrenal needle biopsy, pheochromocytoma should be excluded with measurement of 24-hour urine catecholamine levels prior to biopsy to avoid hypertensive crisis during the procedure.
We agree with Dr Intenzo that the use of a specific SUV as a cutoff for differentiating metastatic from benign adrenal nodules should be considered as only a guideline. SUVs can vary depending on factors such as region of interest shape within which to average, partial volume averaging and spillover effects, attenuation correction, reconstruction method and scanning parameters, counts’ noise bias effect, time of SUV evaluation, competing transport effects, and body size (4). Therefore, SUVs are not a fixed value for differentiating benign from malignant tissue. The cutoff for distinguishing malignant from benign nodules may vary from one PET scanner or institution to another.
18F-FDG PET is inherently limited in the characterization of subcentimeter nodules. This limitation is probably due to the limited resolution (currently an in-plane resolution of 5 mm) and the resulting partial volume averaging effect in the characterization of subcentimeter meta-static adrenal nodules, which usually contain microscopic foci of malignant tissue (5). If 18F-FDG uptake is not increased at PET, integrated PET-CT cannot provide further information. Therefore, as suggested by Dr Intenzo, adrenal nodules less than 10 mm in diameter may be considered nondiagnostic or indeterminate, leading to underestimation of the extent of FDG uptake. Respiratory gating can be accomplished by acquiring continuous CT scans over a single respiratory cycle at each bed position and coregistering the images with the PET images according to the respiratory phase (so-called four-dimensional PET-CT) (6). PET-CT performed with this respiratory gating technique allows more accurate lesion localization and quantification of metabolic activity in a subcentimeter adrenal nodule, thereby making possible more accurate and reproducible SUV calculations.
|
References |
|---|
|
TopReferencesReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOGRAPHICS | RADIOLOGY | RSNA JOURNALS ONLINE |
