Risk stratification and prediction of cancer of focal thyroid fluorodeoxyglucose uptake during cancer evaluation

Focal thyroid incidentaloma by F-18 2-deoxy-2-F18-fluoro-d-glucose (FDG) positron emission tomography (PET) has been reported 1-4% of cancer patients and normal healthy population, with a risk of cancer ranging 14-50%. The aim of this study was to investigate the prevalence of thyroid incidentaloma in F-18 FDG PET/CT and risk of cancer, usefulness of visual and SUVmax and SUVmean differentiating malignant nodules and to define the predictable variables. A total 159 patients with focal thyroid FDG incidentaloma during cancer evaluation with non-thyroid cancer were enrolled. After F-18 PET/CT, we analyzed the image visually and obtained semiquantitative indices. The incidence of focal FDG thyroid incidentaloma is 1.36% and cancer risk is 23.3%. The incidence of focal thyroid FDG uptake was significantly higher in women (2.88 vs. 0.31%; chi (2) = 136.4, p < 0.0001). Malignant thyroid incidentalomas show statistically significant higher value of SUVmax (malignant: median 4.53, range 2.1-12.0; benign: median 3.08, range 1.6-35, p = 0.0093). However, SUVmean have no statistical differences (malignant: median 2.17, range 1.77-3.19; benign: median 2.05, range 1.15-5.77, p = 0.0541). In ROC analyses, the optimal visual grades were > grade 3, and the optimal semiquantitative indices were 4.46 for SUVmax, 2.03 for SUVmean. The visual grade was superior to other variables for the differentiation malignant from benign thyroid incidentalomas. The size and visual grade was the potent predictor by logistic regression analysis. Focal thyroid FDG incidentalomas in non-thyroid cancer patients during evaluation have a high risk of malignancy. The size and visual grade are potential predictors for malignant thyroid incidentaloma.