RESUMO
PURPOSE: To perform a systematic review and meta-analysis of published data to determine the diagnostic utility of adrenal fluorine 18 fluorodeoxyglucose (FDG) positron emission tomography (PET) for distinguishing benign from malignant adrenal disease. MATERIALS AND METHODS: Data on FDG PET assessment in MEDLINE and other electronic databases (from inception to November 2009) and in subject matter-specific journals were evaluated and compared with histologic diagnoses and/or established clinical and imaging follow-up results. Methodologic quality was assessed by using Quality Assessment of Diagnostic Accuracy Studies criteria. Bivariate random-effects meta-analytical methods were used to estimate summary and subgroup-specific sensitivity, specificity, and receiver operating characteristic curves and to investigate the effects of study design characteristics and imaging procedure elements on diagnostic accuracy. RESULTS: A total of 1391 lesions (824 benign, 567 malignant) in 1217 patients from 21 eligible studies were evaluated. Qualitative (visual) analysis of 841 lesions (in 14 reports) and quantitative analyses based on standardized uptake values (SUVs) for 824 lesions (in 13 reports) and standardized uptake ratios (SURs) for 562 lesions (in eight reports) were performed. Resultant data were highly heterogeneous, with a model-based inconsistency index of 88% (95% confidence interval [CI]: 79%, 98%). Mean sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio values for differentiating between benign and malignant adrenal disease were 0.97 (95% CI: 0.93, 0.98), 0.91 (95% CI: 0.87, 0.94), 11.1 (95% CI: 7.5, 16.3), 0.04 (95% CI: 0.02, 0.08), and 294 (95% CI: 107, 805), respectively, with no significant differences in accuracy among the visual, SUV, and SUR analyses. CONCLUSION: Meta-analysis of combination PET-computed tomography (CT) reports revealed that FDG PET was highly sensitive and specific for differentiating malignant from benign adrenal disease. Diagnostic accuracy was not influenced by the type of imaging device (PET vs PET/CT), but specificity was dependent on the clinical status (cancer vs no cancer).