Lymph node dissection for clinically evident lymph node metastases of malignant melanoma.

AIMS: A considerable number of melanoma patients present with clinically evident regional lymph node metastases. Factors influencing prognosis following therapeutic lymph node dissection (TLND) were evaluated.
METHODS: In total 140 patients (68 women, 72 men, median age 53 years) with established regional lymph node metastases, but without clinically detectable distant metastases, received cervical, axillary or ilioinguinal TLND between 1978 and 1997 and were retrospectively reviewed. Uni- and multivariate survival analysis was performed.
RESULTS: Median survival for all 140 patients was 25 months; the observed overall 5 year survival rate was 30%. Age greater than 50 years, primary tumour site on the trunk, more than three lymph node metastases and extracapsular spread were associated with a poor prognosis. In multivariate analysis age (< or =50 years vs >50 years, P=0.02), location of the primary tumour (non-truncal vs truncal, P=0.005), number of lymph nodes involved ( n< or =3 vsn >3, P=0.01) and extracapsular spread (none vs present, P=0.04) proved to be independent prognostic factors.
CONCLUSIONS: TLND is worthwhile and offers a potential chance of cure in about one-third of melanoma patients with established regional lymph node metastases. There are subgroups with a particularly poor prognosis in whom the benefit of radical surgery alone is limited.