Background: 18-F-fluorodeoxyglucose positron emission tomography (FDG-PET) is valuable in staging of esophageal cancer. However, FDG-PET may falsely upstage patients leading to incorrect exclusion from surgical treatment. This study was performed to determine the false-positive rate and possible causes.
Methods: The rate of false-positive lesions on FDG-PET was documented in 86 out of a group of 98 patients. Lesions were defined as false positive when pathological examination was negative or as absence of tumor activity within 6 months of follow-up. To evaluate the influence of a learning curve on the false-positive rate, the PET scans were revised recently.
Results: False-positive lesions were found in 13 patients (13 of 86; 15%). FDG-PET incorrectly revealed only locoregional node metastases in 5 patients in whom surgery with curative intent was performed. Ten lesions in the other 8 patients were classified as distant organ or as nonregional node metastases (M1a/1b). Finally, 5 patients upstaged to M1a/1b underwent a curative resection. The number of false-positive lesions decreased from 16 to 5 (6%) after revision.
Conclusions: Proper interpretation of FDG-PET in staging esophageal cancer is impeded by false-positive results. Even after completion of the learning curve, positive FDG-PET findings still have to be confirmed by additional investigations.