Abstract
1394
Objectives The aim of this study was to evaluate the capacity of combined dynamic and static F-18 FDG PET/CT imaging for differentiating benign from malignant solitary pulmonary lesions.
Methods A total of 49 patients had solitary pulmonary nodules or masses with histopathological diagnoses and dynamic and static F-18 FDG PET/CT imaging. For dynamic imaging acquired for 50 minutes immediately after F-18 FDG injection, time-SUV (TS) curves were generated and divided into three phases. The SUVs of the curve peak of the phase I and the slopes of the phase III were calculated and defined as SUVp and SUVslp respectively. For early and delayed static imaging acquired 1 and 3 hours respectively after F-18 FDG injection, the SUV1h, SUV3h and retention index (RI) were calculated.
Results There were 37 malignant and 12 benign pulmonary lesions (BPL). The SUV3h/SUVp, which was a parameter containing both dynamic and static information, had the highest sensitivity, specificity and accuracy (87%, 92% and 88% respectively) for diagnosing malignant pulmonary lesions (MPL). Meanwhile, the SUVslp had higher specificity than RI (75% and 50%, respectively) for diagnosing MPL. In addition, the TS curves of MPL and non-granulomatous BPL could be differentiated by that the former had a rising and the latter had a decreasing pattern in phase III; whereas MPL and granulomatous BPL had similar phase III patterns but significantly different SUVslp values (0.072 ± 0.048 and 0.030 ± 0.017, respectively; p = 0.03).
Conclusions The combined dynamic and static F-18 FDG PET/CT SUV analyses were valuable in diagnosing MPL. TS curves generated by dynamic F-18 FDG PET/CT imaging were helpful for characterizing pulmonary lesions.