ObjectivePositron emission tomography with [18F]fluorodeoxyglucose (FDG-PET) is of proven value in the detection of metastases in patients with cutaneous melanoma. However, little is known about its value in uveal melanoma (UM). In this study the results of FDG-PET in patients with UM were evaluated.MethodsPatients with UM recorded in the Sydney Melanoma Unit database who had been assessed with FDG-PET were selected. Comparative data (imaging or histopathology) providing information about metastatic disease were obtained within 14 weeks of the FDG-PET study and compared with the FDG-PET result. Sensitivity, specificity, accuracy, and positive and negative predictive values for the detection of liver metastases (LMs) by FDG-PET were calculated.ResultsFDG-PET was performed in 22 patients with UM between April 1993 and March 2003. The presence of at least one focus of metastatic melanoma was confirmed in 14 of 18 patients with positive FDG-PET, and three of four negative FDG-PET studies were confirmed. LMs were demonstrated by FDG-PET in 17 patients. In 15 of these patients this finding was confirmed with anatomical imaging. In two patients LMs indicated by FDG-PET initially appeared to be false positive, but in one of them the diagnosis was confirmed after longer follow-up. Seven of the confirmed lesions were isolated LMs. For LMs FDG-PET showed sensitivity, specificity and accuracy of 100%, 67% and 90% respectively, a positive predictive value of 88% and a negative predictive value of 100%.ConclusionFDG-PET is a valuable investigation for the detection of LMs in UM patients. It appears to be particularly useful in the detection of isolated LMs that are potentially resectable.