The role of elective lymph node dissection in the management of patients with thick cutaneous melanoma.

A retrospective search of patients seen at the Duke Melanoma Clinic from 1970 to 1986 identified 308 clinically Stage I patients, with 4.0 to 10.0 mm cutaneous melanomas. Five-year and ten-year survival was 56% and 43%, respectively. Elective lymph node dissection (ELND) was done in 116 patients (37.7%); there was no difference in disease-free interval (DFI) or survival between these patients versus patients treated with wide excision only (P = 0.9). Thirty-two patients (27.6%) had pathologically positive nodes on ELND. These patients had a shorter DFI (P = 0.05) and survival (P = 0.03) compared with patients with negative node dissections. When further divided by Breslow's thickness, this difference persisted in patients with 4.0 to 6.0 mm lesions (P = 0.01). However, for thicker lesions (greater than 6.0 mm), there was no difference in survival between the node-negative and node-positive groups (P = 0.9). The mean follow-up was 7.1 years. Elective lymph node dissection was not done in 192 patients; 78 of these recurred first in the regional nodes. These 78 patients were compared with the 32 patients who had pathologically positive nodes by ELND to see if patient survival was improved by early removal of nodal disease. There was no difference in DFI (P = 0.5) or survival (P = 0.3) between these two groups. It is concluded that ELND may provide prognostic information for patients with thick cutaneous melanomas. However, there was no change in DFI or ultimate survival when patients were followed, and nodes removed when clinically positive. The authors do not recommend ELND for patients with thick melanomas because the risk of distant metastases outweighs any benefit of regional node dissection.