Effect of chronic extreme hyperglycemia on FDG-PET/CT

1698 Objectives Hyperglycemia can reduce 18F-fluorodeoxyglucose (FDG) uptake in tumors considerably, but it is also reported that the adverse effects caused by chronic hyperglycemia in interpreting PET images was minimum. The purpose of this study was to evaluate the diagnostic performance of FDG-PET/CT, which was performed under chronic extreme hyperglycemic status. Methods Among 11,589 FDG-PET/CT scans between April 2009 and March 2014, 18 scans in 18 patients (pts), who had high plasma glucose level (>300 mg/dl) at FDG injection despite at least 4-h fast, were retrospectively analyzed. Chronic hyperglycemia was derived from uncontrolled diabetes (15 pts) and untreated diabetes (3 pts). Based on the final diagnosis obtained by histopathology or clinical follow-up, the diagnostic performance of visual interpretation of PET/CT was evaluated. Results After excluding two patients without sufficient follow-up, 16 pts with final diagnosis were assessed. In 9 pts with at least one malignant lesion, PET/CT was true positive in 6 pts, and false negative (FN) in 3 pts. The FN cases were hepatocellular carcinoma in 2 pts and post-chemotherapeutic state of pancreas cancer in 1 pt. Therefore, FN results may not have been caused by hyperglycemia. There were two metastatic lesions detected by CT portion of PET/CT in these 2 FN cases. In the remaining 7 pts who were confirmed to have no malignancy, PET/CT was true negative in 6 pts and false positive in 1 pt (inflammation). Overall, the patient-based sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy were 67%, 86%, 86%, 67% and 75%, respectively. Conclusions FDG uptake maintained a sufficiently high level for visual interpretations even under chronic extreme hyperglycemia in most cases, and false negative findings may have been caused by characteristics of lesions. FDG-PET/CT scan can be performed in diabetic patients with chronic hyperglycemia.