In melanoma patients lymph node metastasis is an important prognostic factor that indicates the need for therapeutic lymph node dissection. Preoperative lymphoscintigraphy mapping associated with radioguided sentinel lymph node biopsy has become a well established procedure for cutaneous melanoma patients without clinically detectable lymph node metastases (stage I, II). This technique is a versatile way of characterizing the lymphatic basin at risk for metastases and identifying involved lymph nodes. The purpose of the present study was to examine the reproducibility of lymphoscintigraphy and sentinel lymph node biopsy in detecting micro metastases in cutaneous melanoma. The study was a single-institution prospective analysis of 74 melanoma patients, with primary tumors having Breslow thickness > 0.7 mm, who underwent lymphoscintigraphies between May 2002 and September 2003. Technetium-99m sulfur colloid was injected intradermally at the primary tumor site and dynamic images were obtained for 40 minutes. Two observers evaluated the images. One to two weeks after the first lymphoscintigraphy, radioguided lymph node biopsy was performed. For the biopsy, technetium-99m sulfer colloid was injected intradermally in the same manner as performed before. Lymph nodes were identified and removed with the aid of a gamma ray detecting probe (GDP), and were submitted to histopathological analysis. The histopathological analysis of the sentinel lymph nodes collected during surgery was performed in a sequential manner. First, frozen sections were analyzed during surgery. The lymph nodes considered negative by frozen section were analyzed by H&E staining. Subsequently, the slides considered negative with H&E were sent for immunohistochemical analysis. Lymphoscintigraphy identified at least one sentinel lymph node in all patients. Sentinel node biopsy detected metastases in 20 patients (27.2%). In all cases the lymph node basins identified during lymphoscintigraphy were found to have at least one sentinel lymph node during surgery. Frozen section analysis of the lymph node was only able to identify the disease in 35% of the patients eventually found to have micrometastases with H&E and immunohistochemistry. Two lymph nodes were negative with H&E and positive with immunohistochemical analysis. In conclusion, lymphoscintigraphy is a simple procedure that is well tolerated by patients. It is useful in locating sentinel lymph nodes in patients with melanoma and is an important tool in the clinical practice of oncology. We recommend performing H&E, and if necessary, immunohistochemical analysis of all sentinel lymph nodes because of the high rate of false negative results with frozen sections alone.