The most powerful predictor for recurrence and survival in patients with primary malignant melanoma is the presence or absence of lymph node metastases. In the present study, 18-fluoro-2-deoxyglucose positron emission tomography (FDG-PET) findings were compared with histopathological results of sentinel lymph node biopsy (SNB). The purpose was to determine the value of FDG-PET in predicting regional lymph node involvement in patients with primary malignant melanoma stage I and II. Forty-eight consecutive patients with primary cutaneous melanoma stage I (Breslow thickness > 1 mm) and II underwent FDG-PET scans, preoperative lymphoscintigraphy, and SNB. The FDG-PET and SNB results were interpreted independently of each other and then compared. Of the 48 patients included in the study, eight (16.7%) had a positive SNB. PET was positive in only one patient with a positive SNB, yielding a sensitivity of 13%. All other positive sentinel nodes could not be detected by metabolic FDG-PET imaging. Our study revealed that FDG-PET is obviously not an adequate screening test for subclinical and sonographically inconspicuous lymph node metastases in patients with malignant melanoma stage I and II. The low sensitivity is probably due to the small size of metastatic deposits in sentinel nodes. Therefore, SNB remains the technique of choice for evaluating the histological status of lymph node basins in patients with early-stage cutaneous melanoma.