Morbidity and recurrence after completion lymph node dissection following sentinel lymph node biopsy in cutaneous malignant melanoma

OBJECTIVE: To assess the nature and rates of complications and recurrences after completion lymph node dissection (CLND) following positive sentinel lymph node biopsy (SLNB) in melanoma patients. SUMMARY BACKGROUND DATA: In contrast to SLNB, CLND is associated with considerable morbidity. CLND delays nodal recurrence, thereby prolonging disease-free survival (DFS), but not overall melanoma-specific survival. Elaborate studies on morbidity and recurrence rates after CLND are scarce. Therefore, many controversies concerning extent and nature of CLND exist. METHODS: We conducted a retrospective study on 100 melanoma patients, on whom we performed CLND between October 1999 and December 2005. The median observation period was 38.8 months. RESULTS: We performed a total of 102 CLNDs, [46.1% axillary (47/102), 42.2% groin (43/102), 11.8% neck (12/102)]. Groin dissection (GD) and axillary dissection (AD) led to comparable morbidity (47.6% and 46.8%), but complications were more severe in GD, mandating additional surgery in 25.6% (11/43), versus 8.5% (4/47) in AD. Of the GD patients, 18.5% (8/43) were readmitted for complications compared with 10.4% (5/47) of AD patients. Only 8.3% (1/12) of ND patients suffered complications, mandating neither readmittance nor further surgery. During the median observation period, 65 (65%) of these patients showed DFS, and 35 (35%) exhibited recurrences after a median DFS of 12.5 months. Of the recurrences, 31.4% were nodal, 42.9% distant, and 25.7% local/in-transit. Of our AD patients, 28.3% suffered recurrences (13/46), as did 33.3% of the GD (14/42) and 66.7% of the ND patients (8/12). CONCLUSIONS: CLND is fraught with considerable morbidity. Local control of the dissected nodal basins was achieved with a modified radical approach in ADs (levels I + II only) and, to a lesser extent, GDs, but not in NDs. Clinical trials are necessary to establish guidelines on the extent of lymphatic dissection