Background: Recurrence after treatment of stage I-II melanoma involves regional lymph nodes in about 50%of patients. A reliable method is needed to evaluate lymph node status (metastatic or not) in the case of palpable lymph nodes. Objectives: To evaluate the efficiency of fine-needle aspiration biopsy (FNAB) in examining clinically detected suspicious lymph node in patients followed up after surgical removal of stage I-II melanoma. Patients and methods: One hundred and twenty FNABs were performed in 67 patients with a suspicious node in an open study conducted in a French melanoma regional referral centre, Hpital de l’Htel-Dieu. Cytodiagnosis was classified as positive, negative, inadequate or inconclusive. Sensitivity, specificity, positive and negative predictive values and positive and negative likelihood ratios were calculated after final histopathological evaluation. Results: Fifty-eight of 120 FNABs were positive (48%), 50 of 120 (42%) were negative, four of 120 (3%) were inconclusive and eight of 120 (7%) were inadequate. Among the 108 FNABs in which a definitive diagnosis could be given, sensitivity was 98.2%[95%confidence interval (CI) 90.7-99.9] and specificity was 96.1%(95%CI 86.8-98.9). Conclusions: FNAB under ultrasound guidance is an efficient tool to discriminate better between cases in which surgical treatment of the lymph node basin should be performed and patients who should return for follow-up. Surgical treatment appears to be required in cases of positive FNAB or in inconclusive cases. Background: Recurrence after treatment of stage I-II melanoma involves regional lymph nodes in about 50% of patients. A reliable method is needed to evaluate lymph node status (metastatic or not) in the case of palpable lymph nodes. Objectives: To evaluate the efficiency of fine-needle aspiration biopsy (FNAB) in examining clinically detected suspicious lymph node in patients followed up after surgical removal of stage I-II melanoma. Patients and methods: One hundred and twenty FNABs were performed in 67 patients with a suspicious node in an open study conducted in a French melanoma regional referral center, Hpital de l’Htel-Dieu. Sensitivity, specificity, positive and negative predictive values ​​and positive and negative likelihood ratios were calculated after final histopathological evaluation. Results: Fifty-eight of 120 FNABs were positive (48%), 50 of 120 (42%) were negative, four of 120 (3%) were Among the 108 FNABs in which a definitive diagnosis could be given, sensitivity was 98.2% [95% confidence interval (CI) 90.7-99.9] and specificity was 96.1% (95% CI 86.8-98.9). Conclusions: FNAB under ultrasound guidance is an efficient tool to discriminate better between cases in which surgical treatment of the lymph node basin should be performed and patients who should return for follow-up. Surgical treatment appears to be required in cases of positive FNAB or in inconclusive cases.