Diagnostic performance of unenhanced computed tomography and 18 F- fluorodeoxyglucose positron emission tomography in indeterminate adrenal tumours

Transcription

1 Received: 8 June 2017 Revised: 3 August 2017 Accepted: 11 August 2017 DOI: /cen ORIGINAL ARTICLE Diagnostic performance of unenhanced computed tomography and 18 F- fluorodeoxyglucose positron emission tomography in indeterminate adrenal tumours Danae A Delivanis 1 Irina Bancos 1 Thomas D Atwell 2 Grant D Schmit 2 Patrick W Eiken 2 Neena Natt 1 Dana Erickson 1 Spyridoula Maraka 1,3 William F Young Jr 1 Mark A Nathan 2 1 Division of Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic, Rochester, MN, USA 2 Department of Radiology, Mayo Clinic, Rochester, MN, USA 3 Department of Endocrinology and Metabolism, Center for Osteoporosis and Metabolic Bone Diseases, University of Arkansas for Medical Sciences and the Central Arkansas Veterans Health Care System, Little Rock, AR, USA Correspondence Mark A Nathan, Department of Radiology, Mayo Clinic, Rochester, MN, USA. Nathan.Mark@mayo.edu Summary Objective: Evidence on the diagnostic performance of adrenal imaging is limited. We aimed to assess the diagnostic performance of unenhanced computed tomography (CT) and 18 F- fluorodeoxyglucose ( 18 FDG) positron emission tomography (PET)/CT imaging in a high- risk population for adrenal malignancy using an optimal reference standard. Design: Retrospective cohort study. Methods: Imaging studies of patients with adrenal nodules who underwent adrenal biopsy and/or adrenalectomy between 1994 and 2014 were reviewed and compared to the reference standard of histology. Eighty % of patients presented with known or suspected extra- adrenal malignancy. Results: Unenhanced abdominal CT was performed in 353 patients with adrenal lesions; median size was 3 (0.7-15) cm and median radiodensity was 33 ( 21-78) Hounsfield units (HU). Radiodensity of >10 HU diagnosed malignancy with a sensitivity of 100%, specificity of 33%, positive predictive value (PPV) of 72% and negative predictive value (NPV) of 100%. 18 FDG- PET/CT was performed in 89 patients; median tumour size was 2.1 ( ) cm. Maximum standardized uptake (SUV max) was higher in malignant lesions when compared to benign lesions (median=10 [ ] vs 3.7 [ ], respectively, P<.0001). Similarly, median SUV max lesion to SUV max liver ratio (ALR) in malignant lesions was higher than in benign lesions (median=3 [ ] vs 1.2 [ ], respectively, P<.0001). 18 FDG- PET/CT ALR >1.8 diagnosed malignancy with a sensitivity of 87%, specificity of 84%, PPV of 85% and NPV of 86%. Conclusion: Noncontract CT radiodensity of 10 HU excludes malignancy even in a high- risk population. For indeterminate adrenal lesions, given a superior specificity, 18 FDG- PET/CT could be considered as a second stage imaging study. KEYWORDS 18 F-FDG-PET/CT, adrenal nodule, computed tomography, metastatic cancer, MRI John Wiley & Sons Ltd wileyonlinelibrary.com/journal/cen Clinical Endocrinology. 2018;88:30 36.

2 DELIVANIS et al INTRODUCTION 2 SUBJECTS AND METHODS Adrenal masses are identified in 5% of patients undergoing computed tomography (CT) of the abdomen. 1,2 Adrenal cortical carcinoma (ACC) has been reported in 4.7% 3 of patients with adrenal tumours while other malignant adrenal masses, mainly represented by metastases, are rare in patients without history of extra- adrenal malignancy. 2 However, in patients with newly diagnosed extra- adrenal malignancy or radiographic surveillance for a previously treated malignancy, the likelihood of adrenal malignancy can be as high as 50-75%. 4,5 The adrenal gland is a common site of metastatic disease, deriving mainly from melanoma and pulmonary, gastrointestinal, renal and breast carcinomas. 6-9 Despite the widespread use of imaging tests, a recent systematic review and meta- analysis revealed that current evidence on diagnostic performance of imaging tests in patients with adrenal tumours is scarce, mainly due to small sample size and the lack of an optimal reference standard. 8 Best available evidence on diagnostic imaging studies such as noncontrast abdominal CT suggests that in patients with an adrenal mass with a radiodensity of <10 HU on unenhanced CT, adrenal malignancy can be excluded 6,8 as sensitivity was reported to be %. However, specificity is poor (38-91%) mainly due to a high number of lipid poor benign adrenal adenomas. 10,11 Adrenal 18 F- fluorodeoxyglucose (FDG)- positron emission tomography (PET)/CT imaging has been proposed by the recent guidelines 12 as one of the diagnostic imaging modalities in the evaluation of adrenal masses, especially when suspicion for malignancy is high. However, current published literature on 18 FDG- PET/CT imaging in patients with adrenal incidentalomas, as well as in patients with adrenal masses discovered in the context of cancer surveillance, is composed mainly of studies with a small sample size, suboptimal reference standard, a heterogeneous patient population and high risk of bias. 11,12 Thus, there is an urgent need for well- designed studies to establish the true performance of adrenal imaging and more specifically of 18 FDG- PET/ CT in the evaluation of adrenal nodules. In patients with extra- adrenal malignancy, distinguishing benign from metastatic adrenal tumours is critical in establishing the stage of the disease, predicting prognosis, determining the optimal management plan and finally improving survival benefit as demonstrated in isolated adrenal metastatic disease. 13 To address this knowledge gap, we analysed the results of unenhanced abdominal CT and 18 FDG- PET/CT imaging studies performed in all patients who underwent adrenal biopsy or adrenalectomy at our institution over a 20- year period. The aims of this study were to: determine the diagnostic performance of 18 FDG-PET/CT imaging in diagnosing a malignant adrenal mass identify the optimal FDG uptake in the adrenal nodule on 18 FDG- PET/CT to detect adrenal malignancy. compare the diagnostic performance of abdominal unenhanced CT and 18 FDG-PET/CT employed in a high-risk population for adrenal malignancy using a stringent reference standard. We reviewed a retrospective cohort of 418 patients undergoing 419 adrenal biopsies at Mayo Clinic Rochester, Minnesota, between 1994 and We have previously published our experience on the diagnostic performance of adrenal biopsy in this cohort. 6 The study protocol was approved by the Mayo Clinic Institutional Review Board. Imaging modalities (CT scan and 18 FDG- PET/CT) were reviewed and imaging characteristics recorded by radiologists with expertise in adrenal imaging. Radiodensity on CT was measured for patients with available unenhanced CT imaging. Large adrenal tumours are known to be heterogeneous due to necrosis and haemorrhage. Determining the HU value based on a single image, therefore, may not capture the true mean HU value of the tumour. To ascertain that the HU threshold established in this study still applies to smaller adrenal tumours, we independently reviewed a cohort of patients (n=293) with adrenal tumours measuring 5 cm or less. Medical records were reviewed for clinical information pertinent to the presentation and management of the adrenal mass. All pathology reports were reviewed for information regarding the histologic diagnosis. In addition, all follow- up clinical, imaging and surgical information was collected for each patient FDG- PET/CT The 18 FDG- PET/CT acquisition followed a standard Mayo Clinic clinical protocol. Specifically, weight and blood glucose levels were recorded for all patients. 18 FDG- PET/CT studies (Discovery RX, 690, or 710; GE Healthcare) were performed following injection with 15 mci of 18 F- FDG and imaging began 60 minutes after injection. When possible, patients were imaged with arms up, and body coverage was dictated by patient malignancy history, using a matrix and a rate of 3 minutes per bed position. PET images were reconstructed with a 3- dimensional ordered- subsets expectation maximization algorithm (28 subsets, 2 iterations). Low- dose helical CT images were obtained for attenuation correction and anatomic localization (detector row configuration, mm; pitch, 1.75; gantry rotation time, 0.5 seconds; slice thickness, 3.75 mm; 140 kvp; and a range of mas using automatic current modulation). In addition, a qualitative 18 FDG- PET/CT visual score assessment of the adrenal tumours was performed by our radiologists with expertise on adrenal 18 FDG- PET/CT imaging. The scoring system used was as follows: 1=definitely benign, 2=probably benign, 3=equivocal, 4=probably malignant and 5=definitely malignant adrenal tumour. Equivocal cases were excluded, to allow for appropriate calculations. For 18 FDG- PET, the maximum standardized uptake value (SUV max) was measured in the adrenal mass (SUV max adrenal) and liver (SUV max liver). The SUV max adrenal to SUV max liver ratio (ALR) was calculated.

3 32 DELIVANIS et al. 3.1 Abdominal CT imaging CT imaging was performed using multiple different models of General Electric (GE Medical Systems, Milwaukee, WI, USA) and Siemens (Siemens, Forchheim, Germany) scanners over the 20- year study period. Imaging was generally obtained at 2-3 mm collimation, tube voltage of 120 kv and tube current between 100 and 300 ma. Images were reconstructed at 3- mm intervals, with a 3-5 mm slice thickness. Noncontrast CT images were reviewed on an institutional PACS (picture archiving and communication system). The attenuation in HU of the adrenal lesions was measured using a circular or ovoid ROI (region of interest) set as large as possible, but entirely contained in the adrenal lesion to avoid partial volume averaging with the surrounding tissue. A threshold of 10 HU on noncontrast imaging was set for diagnosing an adrenal adenoma. Contrast- enhanced CT images were not evaluated, and contrast washout calculations of the adrenal lesions were not performed. 3.2 Reference standard The reference standard consisted of adrenal biopsy histology in all patients. In addition to adrenal biopsy histology, a final surgical histopathological result was used in 40 patients who underwent adrenalectomy following the adrenal biopsy. The following were excluded from the analysis: patients in whom the adrenal mass could not be classified as malignant or benign (pheochromocytomas), patients with nonneoplastic adrenal disease (granulomas, haematomas), patients with clearly benign diagnostic imaging characteristics (such as in myelolipoma) (Appendix Table 1), and patients in whom adrenal biopsy failed to provide histology (nondiagnostic adrenal biopsy) and adrenalectomy was not performed (Figure 1). 3.3 Statistics Descriptive statistics were used to determine mean and standard deviation (SD) or median and ranges depending on data distribution, while categorical data are shown as a number (%). Associations between variables were assessed using the Student t test and ANOVA for continuous variables and the chi- square test for categorical variables. P values <.05 were considered significant. Data were analysed using the receiver operating characteristic (ROC) analysis for the attenuation values on the unenhanced CT, and ALR SUV max ratio was generated from the data using JMP software, version 10 (SAS, Carey, NC, USA). For each discriminator, the ROC curve is a plot of the truepositive fraction (TPF=sensitivity) against the false- positive fraction (FPF=1 specificity). A univariate logistic regression model was used with adrenal SUVmax, ALR SUVmax, CT attenuation value and tumour diameter to examine the independence of their relationship with the presence of malignant lesions. 4 RESULTS 4.1 Patients Four hundred and nineteen adrenal biopsies were performed in 418 patients between 1994 and After applying exclusion criteria as described in Figure 1, we analysed 379 patients with adrenal masses evaluated at our institution (Appendix Table 2). Median age was 68 years (range, ) and 237 (62%) were men. Median adrenal mass size for the entire cohort was 3 cm (range, ). The indication for adrenal biopsy was suspected adrenal metastasis from a known or suspected extra- adrenal primary source in the majority of the cases, 302/379 (80%), reflecting a high risk for adrenal malignancy in our cohort of patients. 4.2 Histology The majority of patients were diagnosed with adrenal metastases (218/379, 58%), most originating from a primary lung malignancy of nonsmall (n=115) and small cell origins (n=7), followed by renal cell carcinoma (n=27), and gastrointestinal carcinoma (n=21). Fifteen patients with diagnosis of lymphoma (4%) were also identified. Histology consistent with benign adrenal tissue was obtained in 136 of 379 (36%) adrenal biopsies. Our cohort also included 8 patients with ACC (Table 1). 4.3 Performance of unenhanced CT imaging FIGURE 1 Representation of exclusion criteria Unenhanced abdominal CT was performed in 353 of the patients with adrenal masses (225 malignant, 128 benign adrenal lesions). Median size of adrenal lesions was 3.0 cm (range, ), and median precontrast administration radiodensity was 33 HU (range, 21-78). The diagnostic performance of unenhanced CT to diagnose malignancy when a radiodensity of >10 HU was used as a cut- off threshold resulted in a sensitivity of 100%, specificity of 33%, positive predictive value (PPV)

4 DELIVANIS et al. 33 TABLE 1 Histologic diagnosis (fine needle aspiration or surgical pathology) in 379 adrenal tumours Diagnosis based on adrenal biopsy histology n (%) Benign adrenal cortical tissue/adrenal adenoma 136 (36) Adrenocortical carcinoma 8 (2) Neuroblastoma 1 (<1) Ganglioneuroma 1 (<1) Metastasis (n=218) Lung 122 (32) Kidney 27 (7) Gastrointestinal 21 (6) Melanoma 8 (2) Genitourinary 7 (2) Sarcoma 7 (2) Breast 4 (1) Thyroid 3 (<1) Prostate 2 (<1) Head and neck 1 (<1) Unknown primary 16 (4) Lymphoma 15 (4) Total 379 of 72% and negative predictive value (NPV) of 100% (Table 2). The diagnostic performance of unenhanced CT to diagnose malignancy in patients (n=293) with tumours measuring 5 cm or less was identical; sensitivity of 100%, specificity of 33%, PPV of 67% and NPV of 100%. 4.4 Performance of 18 FDG- PET/CT imaging 18 FDG- PET/CT scan was performed in 89 patients with adrenal masses (44 metastases, 3 lymphomas and 42 adrenal adenomas). The median size of the adrenal masses was 2.1 cm (range, ), median adrenal nodule SUV max was 5.9 (range, ) and median ALR was 2.0 (range, ). The SUV max was higher in malignant lesions when compared to benign lesions (median =10 [ ] vs 3.7 [ ], respectively, P<.0001). Similarly, the ALR in malignant lesions was higher; median =3 (range, ) than in benign lesions, median= 1.2 (range, ), P<.0001 (Table 3). When the ALR cut- off of >1.8 was applied based on optimal performance of ROC analysis, 18 FDG- PET/CT scanning diagnosed malignancy with a sensitivity of 87%, specificity of 84%, PPV of 85% and NPV of 86%. Based on ROC analysis (0.85), a SUV max cut- off value of 4.5 diagnosed adrenal malignancy with a sensitivity of 87%, specificity of 69%, PPV of 76% and NPV of 83%. Qualitative assessment of 18 FDG- PET/CT (scores 1 and 2 considered benign and scores of 4 and 5 considered malignant) by our experienced radiologists resulted in a sensitivity of 91%, specificity 57%, PPV of 74% and NPV of 84%. Nine per cent (8/89) of the cases were considered nondiagnostic and received a score of 3. To evaluate the diagnostic performance of 18 FDG- PET/CT scanning in patients with indeterminate adrenal nodules, a subgroup analysis was performed of 71 patients with adrenal nodules that had a radiodensity >10 HU on unenhanced CT. The median SUV max of these adrenal nodules was 7.4 (range, ), and median ALR was 2.2 (range, ). When an ALR cut- off of>1.8 was used, there were 6 false- negative cases (metastatic lung cancer n=4, sarcoma n=1 and renal cell carcinoma n=1)(figure 2) and 6 false- positive cases (adrenal adenoma n=6)(figure 3). A univariate logistic regression analysis revealed that an ALR (P<.0001), SUV max of adrenal nodule (P<.0001), HU attenuation (P<.0001) and adrenal size (P=.03) were all independent prognostic factors for malignancy. Finally, we analysed a cohort of 81 patients that had undergone both unenhanced CT imaging and 18 FDG- PET/CT scan. We have not observed any significant differences in the performance of any of the imaging studies in this patient population (Table 4). 5 DISCUSSION The optimal diagnostic imaging approach of adrenal nodules in a patient population with a history of extra- adrenal malignancy has not TABLE 2 Diagnostic performance of imaging studies to diagnose adrenal malignancy and diagnostic performance of imaging studies to diagnose adrenal malignancy only in patients with history of known or suspected extra- adrenal malignancy Diagnostic performance of imaging studies to diagnose adrenal malignancy N Sensitivity Specificity PPV NPV Unenhanced CT HU > (malignant 224; benign 129) 100% 33% 72% 100% 18 FDG- PET/CT ALR SUV max > (malignant 47; benign 42) 87% 84% 85% 86% 18 FDG- PET/CT Adrenal SUV max > (malignant 47; benign 42) 87% 69% 76% 83% Diagnostic performance of imaging studies to diagnose adrenal malignancy only in patients with history of known or suspected extra- adrenal malignancy Unenhanced CT HU > (malignant 191; benign 93) 100% 29% 75% 100% 18 FDG- PET/CT ALR SUV max > (malignant 44; benign 36) 86% 86% 88% 84% 18 FDG- PET/CT Adrenal SUV max > (malignant 44; benign 36) 86% 75% 81% 84% PPV, positive predictive value; NPV, negative predictive value; CT, computed tomography; HU, hounsfield units; 18 FDG- PET, 18 F- fluorodeoxyglucose positron emission tomography; ALR, adrenal liver ratio and SUV, standardized uptake value.

5 34 DELIVANIS et al. TABLE 3 Imaging findings in benign and malignant adrenal masses All Benign Malignant P value (benign vs malignant) n Median age, range 68 (0.5-91) 67 (0.5-88) 69 (27-91).42 Male sex, % 237,62% 84, 35% 153, 65%.81 Unenhanced CT radiodensity available, n Radiodensity, HU (range) 33 ( 21-78) 20 ( 21-63) 36 (11-78) <.0001 Size based on CT, cm (range) 3.1 (0.6-21) 2.4 ( ) 3.5 (0.9-15) < FDG- PET/CT obtained, n FDG- PET/CT median SUV max (range) 5.9 ( ) 3.7 ( ) 10 ( ) < FDG- PET/CT median ALR SUV max (range) 2.0 ( ) 1.2 ( ) 3 ( ) <.0001 CT, computed tomography; HU, hounsfield units; 18 FDG- PET, 18 F- fluorodeoxyglucose positron emission tomography; SUV, standardized uptake value and ALR, adrenal liver ratio. FIGURE 2 A 63- year- old man with history of tobacco use presented with weight loss. A, CT shows a 2 cm nodule of the lateral limb of the left adrenal gland (arrow). B, 18 FDG- PET/CT imaging SUV max was 4 and ALR was low at 1.3. Adrenal biopsy revealed metastatic adenocarcinoma of the lung [Colour figure can be viewed at wileyonlinelibrary.com] FIGURE 3 A 56- year- old woman with history of papillary thyroid cancer. A, 18 FDG- PET/CT showed intense FDG uptake (SUV max: 15.3 and ALR: 4.9) in the left adrenal gland. B, Subsequent CT of the abdomen revealed a 2.5 cm mass in the left adrenal gland with an unenhanced radiodensity of 35 HU. Adrenal biopsy revealed oncocytic cells suggestive of oncocytoma or adrenocortical tumour. Final histopathology report after adrenalectomy was consistent with adrenal adenoma [Colour figure can be viewed at wileyonlinelibrary.com] TABLE 4 Comparison of performance of imaging studies in patients that underwent both unenhanced CT and 18 FDG- PET/CT Unenhanced CT HU >10 18 FDG- PET/CT ALR SUV max > FDG- PET/CT Adrenal SUV max >4.5 N Sensitivity Specificity PPV NPV % 27% 62% 100% 81 86% 86% 88% 84% 81 89% 76% 81% 85% been well established. Due to frequent imaging surveillance as part of their oncological follow- up, the prevalence of adrenal masses is high. 14 Optimal diagnostic evaluation of adrenal tumours in this patient population is pivotal not only for assurance of accurate staging and avoidance of futile surgical procedures, but also for avoidance of missed opportunities for curative surgical or ablative treatment. 15,16 In our study of patients at high risk for adrenal malignancy, we demonstrated that when a threshold of >10 HU on noncontrast CT is used, malignancy can be diagnosed with 100% sensitivity and NPV, but a suboptimal specificity and PPV of 33% and 72%, respectively.

6 DELIVANIS et al. 35 In a recent systematic review and meta- analysis on the diagnostic accuracy of imaging characteristics summarizing 5 studies (168 patients with history of extra- adrenal malignancy), CT noncontrast tumour radiodensity cut- off of >10 HU was reported to have a slightly lower sensitivity of 93% and a higher specificity of 71% when compared to our results. 8 Reported HU can be influenced by adrenal mass size, heterogeneity of the lesion and the type of region of interest (ROI) being used. 17 Several cases of malignant adrenal masses have been described in the literature to have radiodensity 10 HU, including metastatic clear renal cell carcinoma and hepatocellular carcinoma. 18 In our cohort of malignant adrenal masses, the lowest measured radiodensity was 11 HU. 18 FDG- PET/CT is a commonly employed imaging study for surveillance for metastatic disease. We found that qualitative assessment of PET positivity by an experienced radiologist had almost similar diagnostic performance to the ALR based on ROC curves analysis a finding that highlights that radiologist experience plays a clinically significant role. 19 We demonstrated that overall the ALR performs better than SUV max of the lesion by itself which is also in accordance with other studies. 8,20 On 18 FDG- PET/CT imaging, malignant lesions had a higher ALR compared to benign lesions, and an ALR ratio cut- off of >1.8 diagnosed malignancy with a sensitivity of 87% and specificity of 84%. Comparison of our results to other studies is challenging due to differences in reported SUV max and ALR cut- offs, heterogeneity of targeted population (incidentalomas versus oncologic populations), differences in reference standards used and differences in imaging modalities such as PET vs PET imaging with CT image fusion. Dinnes et al identified only 3 studies with optimal reference standards examining the diagnostic performance of 18 FDG- PET/ CT in patients with extra- adrenal malignancy. In a pooled analysis of those 3 studies, which included only 156 patients, the sensitivity of 18 FDG- PET/CT was similar to our findings (82% for ALR and 84% for SUVmax); however, specificity was higher (96 and 90%, respectively). In patients with adrenal incidentalomas, 18 FDG- PET/CT performed better (sensitivity 100%, specificity 96%); however, these conclusions were based only on 2 studies of 64 patients, and confidence intervals of estimates were wide. In the current study, false- negative results on 18 FDG- PET/CT were identified in 4 patients with metastatic lung carcinoma, 1 patient with adrenal sarcoma and 1 patient with metastatic renal cell carcinoma. Other false- negative cases have been reported. 21,22 It is known that 18 FDG- PET/CT is less sensitive and less specific for characterizing small adrenal lesions particularly those less than 1 cm. However, in our study, the median size of false- negative adrenal lesions was 1.8 cm (range, ), implying that even larger adrenal masses can be misclassified. We identified 6 cases of false- positive adrenal lesions, median size 2.3 cm (range, ), and the majority of these were located in the left adrenal gland. Similar results have been demonstrated by others. 21,23,24 Although several studies 23,25 have recommended 18 FDG- PET/CT scan as the modality of choice for the characterization of adrenal lesions, particularly in patients with a history of extraadrenal malignancy, our results using an optimal reference standard indicate that both the sensitivity and the specificity of 18 FDG- PET/CT are not perfect; hence, careful clinical judgement is warranted so that cases of adrenal malignancy are not missed. 5.1 Strengths and limitations This study, performed in an institution with expertise in the evaluation of adrenal masses, analysed one of the largest cohorts of patients using stringent inclusion criteria and an optimal reference standard. An adrenal histopathological result was available in all cases, and the adrenal nodules represented a diagnostic dilemma in the patient population studied given that the majority of patients had a known or suspected extra- adrenal malignancy. Imaging studies were reviewed independently by experienced adrenal radiologists. Adrenal pathology that could result in a false- positive (infection, pheochromocytomas) or false- negative diagnosis (haematomas, necrosis) on 18 FDG- PET/CT 24 was purposefully excluded from the analysis (Appendix Table 1). This approach allowed for a distinct differentiation between adrenal adenomas and other malignant processes. The retrospective design of our study and the fact that our institution is a large referral centre could limit the generalizability of our results. Adrenal biopsy was used as the reference standard and this undoubtedly has its own limitations as the outcome depends on biopsy technique, heterogeneity of the adrenal mass and pathologist s experience. 6 We have tried to minimize potential errors in pathologic diagnosis by also including information on adrenalectomy histology when available, as well as clinical and imaging follow- up information. The retrospective nature of this study over an extended time interval resulted in a variation of imaging algorithms and technique between some patients. In addition, patients underwent a variety of different combinations of imaging modalities, thus making it difficult to make firm conclusions on less commonly employed studies such as MRI due to small sample size, and thus MRI data were not included in this study. Finally, given the large size of some of the tumours and inherent pathologic and imaging heterogeneity of such masses, specific imaging characteristics were difficult to define as a single measure. We attempted to mitigate such variances by including a cohort of patients with smaller neoplasms to validate the more generalized results and the resultant findings were similar. 6 CONCLUSION In conclusion, we investigated the diagnostic accuracy and utility of unenhanced CT and 18 FDG- PET/CT in patients with a high suspicion of adrenal malignancy. We found that all malignant tumours demonstrated unenhanced CT attenuation >10 HU (100% sensitivity), while 18 FDG- PET/CT had a superior specificity (84% for >1.8 ALR SUV max ratio) for diagnosis of adrenal malignancy. Therefore, for indeterminate adrenal lesions, 18 FDG- PET/CT could be considered as a second line imaging study in patients at high risk for adrenal malignancy.

7 36 DELIVANIS et al. CONFLICT OF INTEREST Nothing to declare. ORCID Danae A Delivanis REFERENCES 1. Bovio S, Cataldi A, Reimondo G, et al. Prevalence of adrenal incidentaloma in a contemporary computerized tomography series. J Endocrinol Invest. 2006;29: Song JH, Chaudhry FS, Mayo-Smith WW. The incidental adrenal mass on CT: prevalence of adrenal disease in 1,049 consecutive adrenal masses in patients with no known malignancy. AJR Am J Roentgenol. 2008;190: Young WF Jr. Management approaches to adrenal incidentalomas. A view from Rochester, Minnesota. Endocrinol Metab Clin North Am. 2000;29: Lenert JT, Barnett CC Jr, Kudelka AP, et al. Evaluation and surgical resection of adrenal masses in patients with a history of extra- adrenal malignancy. Surgery. 2001;130: Frilling A, Tecklenborg K, Weber F, et al. Importance of adrenal incidentaloma in patients with a history of malignancy. Surgery. 2004;136: Delivanis DA, Erickson D, Atwell TD, et al. Procedural and clinical outcomes of percutaneous adrenal biopsy in a high risk population for adrenal malignancy. Clin Endocrinol (Oxf) 2016;85: Lam KY, Lo CY. Metastatic tumours of the adrenal glands: a 30- year experience in a teaching hospital. Clin Endocrinol (Oxf). 2002;56: Dinnes J, Bancos I, Ferrante di Ruffano L, et al. Management of endocrine disease: imaging for the diagnosis of malignancy in incidentally discovered adrenal masses: a systematic review and meta- analysis. Eur J Endocrinol 2016;175: Bancos I, Tamhane S, Shah M, et al. Diagnosis of endocrine disease: the diagnostic performance of adrenal biopsy: a systematic review and meta- analysis. Eur J Endocrinol 2016;175: Hamrahian AH, Ioachimescu AG, Remer EM, et al. Clinical utility of noncontrast computed tomography attenuation value (hounsfield units) to differentiate adrenal adenomas/hyperplasias from nonadenomas: cleveland clinic experience. J Clin Endocrinol Metab. 2005;90: Caoili EM, Korobkin M, Francis IR, Cohan RH, Dunnick NR. Delayed enhanced CT of lipid- poor adrenal adenomas. AJR Am J Roentgenol. 2000;175: Fassnacht M, Arlt W, Bancos I, et al. Management of adrenal incidentalomas: european society of endocrinology clinical practice guideline in collaboration with the European network for the study of adrenal tumors. Eur J Endocrinol. 2016;175:G1 G Kim SH, Brennan MF, Russo P, Burt ME, Coit DG. The role of surgery in the treatment of clinically isolated adrenal metastasis. Cancer. 1998;82: Mitchell IC, Nwariaku FE. Adrenal masses in the cancer patient: surveillance or excision. Oncologist. 2007;12: Porte H, Siat J, Guibert B, et al. Resection of adrenal metastases from non- small cell lung cancer: a multicenter study. Ann Thorac Surg. 2001;71: Raz DJ, Lanuti M, Gaissert HC, Wright CD, Mathisen DJ, Wain JC. Outcomes of patients with isolated adrenal metastasis from nonsmall cell lung carcinoma. Ann Thorac Surg 2011;92: discussion Park SY, Park BK, Park JJ, Kim CK. CT sensitivities for large (>/=3 cm) adrenal adenoma and cortical carcinoma. Abdom Imaging. 2015;40: Porte HL, Ernst OJ, Delebecq T, Metois D, Lemaitre LG, Wurtz AJ. Is computed tomography guided biopsy still necessary for the diagnosis of adrenal masses in patients with resectable non- small- cell lung cancer?. Eur J Cardiothorac Surg. 1999;15: Boland GW, Blake MA, Holalkere NS, Hahn PF. PET/CT for the characterization of adrenal masses in patients with cancer: qualitative versus quantitative accuracy in 150 consecutive patients. AJR Am J Roentgenol. 2009;192: Watanabe H, Kanematsu M, Goshima S, et al. Adrenal- to- liver SUV ratio is the best parameter for differentiation of adrenal metastases from adenomas using 18F- FDG PET/CT. Ann Nucl Med. 2013;27: Jana S, Zhang T, Milstein DM, Isasi CR, Blaufox MD. FDG- PET and CT characterization of adrenal lesions in cancer patients. Eur J Nucl Med Mol Imaging. 2006;33: Metser U, Miller E, Lerman H, Lievshitz G, Avital S, Even-Sapir E. 18F- FDG PET/CT in the evaluation of adrenal masses. J Nucl Med. 2006;47: Yun M, Kim W, Alnafisi N, Lacorte L, Jang S, Alavi A. 18F- FDG PET in characterizing adrenal lesions detected on CT or MRI. J Nucl Med. 2001;42: Chong S, Lee KS, Kim HY, et al. Integrated PET- CT for the characterization of adrenal gland lesions in cancer patients: diagnostic efficacy and interpretation pitfalls. Radiographics. 2006;26: discussion Schwartz LH, Panicek DM, Koutcher JA, et al. Adrenal masses in patients with malignancy: prospective comparison of echoplanar, fast spin- echo, and chemical shift MR imaging. Radiology. 1995;197: SUPPORTING INFORMATION Additional Supporting Information may be found online in the supporting information tab for this article. How to cite this article: Delivanis DA, Bancos I, Atwell TD, et al. Diagnostic performance of unenhanced computed tomography and 18 F- fluorodeoxyglucose positron emission tomography in indeterminate adrenal tumours. Clin Endocrinol (Oxf). 2018;88: