Results of a prospective randomized clinical trial conducted by the WHO Collaborating Centers for the Evaluation of Methods of Diagnosis and Treatment of Melanoma are reported. Five‐hundred‐fifty‐three Stage I patients whose limbs were affected entered the study; 267 were submitted to wide excision and immediate node dissection and 286 had wide excision and node dissection at the time clinically positive nodes were detected. Survival curves of the two treatment groups could be superimposed. No subsets of patients benefitted from immediate node dissection. The authors conclude that delayed node dissection is as effective as the immediate dissection in Stage I melanoma of the extremities if the patient can be checked every three months. If the quarterly follow‐up is not guaranteed, immediate node dissection is advisable, at least for melanomas thicker than 2 mm.

553 patients with stage I malignant melanoma of the limbs entered a prospective randomized clinical trial carried out by the W.H.O. Collaborating Centres for Evaluation of Methods of Diagnosis and Treatment of Melanoma from September 1967 to January 1974. 286 patients were submitted to wide excision of primary and node dissection at the time as appearance of regional lymph node metastases and 267 to wide excision and immediate node dissection. Survival was identical in the 2 groups. Different subsets of patients were evaluated to assess whether some groups of patients may benefit from immediate node dissection. As regards sex, females and a significantly higher survival rate than males (p < 0.05), but results were not improved by immediate node dissection. Maximum diameter and elevation of primary melanoma were significantly related to survival but also in these cases immediate node dissection did not achieve better results. 63 patients had an excisional biopsy of their melanoma within 4 weeks before final treatment. This procedure did not worsen survival and also in this case immediate node dissection did not improve survival. 273 cases were classified according to histologic type: survival of superficial spreading and nodular melanoma was not different at a statistically significant level after the 2 treatment modalities. 325 cases were considered classifiable according to Clark's levels, out of these 165 were submitted to immediate node dissection. Neither level III nor level IV cases showed higher survival rate after immediate node dissection. Maximum tumor thickness according to Breslow was evaluated in 338 cases: 188 were submitted to wide excision and immediate node dissection. In no clusters of thickness did the enlarged surgical procedure achieve better results. The authors conclude that there is good evidence that in stage I melanoma of the extremities delayed dissection.