Objectives: The purpose of this study was to evaluate the adverse effects of chronic marked hyperglycemia on clinical diagnostic performance of positron emission tomography (PET) using (18)F-fuorodeoxyglucose (FDG).
Methods: Fifty-seven scans of 54 patients, who received FDG-PET for the diagnosis of various cancer(s), and who showed high plasma glucose level of more than 200 mg/dl at the time of administration of FDG in spite of at least 4-h fasting, were retrospectively analyzed. In the clinical follow-up, this high plasma glucose was confirmed as chronic hyperglycemia derived from uncontrolled diabetes (n = 32) and untreated diabetes (n = 25). Based on the final diagnosis of malignancy obtained by histopathology or clinical follow-up for at least 6 months, the diagnostic performance of visual PET analysis was evaluated.
Results: Excluding nine scans of nine patients without sufficient follow-up, final diagnosis was obtained in 48 scans of 45 patients. In 36 scans of 36 patients, at least one malignant lesion was finally confirmed, and true-positive and false-negative results were obtained in 30 and six cases, respectively. Six cases showed false-negative results due to low FDG-avid pathological characteristics (hepatocellular carcinoma, etc.), chemotherapeutic effect or small tumor size. Overall, the patient-based sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy were 83, 83, 94, 63 and 83%, respectively. In lesion-based diagnosis, 56 of 75 lesions (74%) were depicted by PET, while 19 lesions were negative on PET, also due to low FDG-avid characteristics or small size (less than 15 mm).
Conclusions: At the time of chronic hyperglycemia (not acute hyperglycemia), the adverse effect caused by high plasma glucose level was minimum. The FDG uptake of the tumor maintained a sufficiently high level for visual clinical diagnosis in most cases, except in the cases of low FDG-avid tumors or small lesions (15 mm in size).