The expanding use of (18)F-fluorodeoxyglucose positron emission tomography ((18)F-FDG PET) has led to the identification of increasing numbers of patients with an incidentaloma in the thyroid gland. We aimed to review the proportion of incidental thyroid cancers found by (18)F-FDG PET or PET/computed tomography imaging. Studies evaluating thyroid carcinomas discovered incidentally in patients or healthy volunteers by (18)F-FDG PET were systematically searched in the PubMed database from 2000 to 2011. The main exclusion criteria were known thyroid disease, lack of assigned diagnoses, investigation of diffuse uptake only, or investigation of patients with head and neck cancer, or cancer in the upper part of the thorax. Twenty-two studies met our criteria comprising a total of 125,754 subjects. Of these, 1994 (1.6%) had unexpected focal hypermetabolic activity, while 999 of 48,644 individuals (2.1%) had an unexpected diffuse hypermetabolic activity in the thyroid gland. A diagnosis was assigned in 1051 of the 1994 patients with a focal uptake, 366 of whom (34.8%) had thyroid malignancy. Likewise, a diagnosis was assigned in 168 of 999 patients with a diffuse uptake, 7 of whom (4.4%) had thyroid malignancy. In the eight studies reporting individual maximum standardized uptake values (SUV(max)), the mean SUV(max) was 4.8 (standard deviation [SD] 3.1) and 6.9 (SD 4.7) in benign and malignant lesions, respectively (p<0.001). Incidentally found thyroid nodules, using (18)F-FDG PET, are at high risk of harboring malignancy if uptake is focal. SUV are significantly higher in malignant than in benign nodules. The pronounced inhomogeneity and other shortcomings of the studies are discussed.