Sentinel lymph node evaluation has enabled identification of patients with cutaneous melanoma who might benefit from elective regional lymph node dissection. Sentinel nodes are currently assessed by histologic and reverse transcription polymerase chain reaction (RT-PCR) evaluation for melanocyte-specific markers. The clinical significance of positive findings by RT-PCR in the absence of histologic evidence of metastasis (HIS(NEG)/PCR(POS)) remains unclear. Examination of 264 lymph nodes from 139 patients revealed histopathologic positivity in 34 patients (24.5%), in which 26 also demonstrated simultaneous RT-PCR positivity (HIS(POS)/PCR(POS)). Of 35 HIS(NEG)/PCR(POS) patients (25.2%), five also had nodal capsular nevi. In total, capsular nevi were detected in 13 patients (9.4%). A total of 70 patients (50.4%) had negative sentinel nodes by both histopathology and RT-PCR (HIS(NEG)/PCR(NEG)). Over a median follow-up of 25 months, local and/or systemic recurrence developed in 31 patients (22.3%). Recurrence rates were similar among patients with histopathologic evidence of sentinel lymph node metastasis, irrespective of RT-PCR status (HIS(POS)/PCR(POS) 62%; HIS(POS)/PCR(NEG) 75%). In contrast, only 10% of HIS(NEG)/PCR(NEG) patients developed recurrence, significantly less than those in either HIS(POS) group (P<0.0001). Recurrence in the HIS(NEG)/PCR(POS)/CN(NEG) group (7.7%) was comparable to that in HIS(NEG)/PCR(NEG) patients and significantly lower than that in either HIS(POS) group (P<0.0001). The only independent prognostic factors identified by multivariate analysis were the Breslow thickness of the primary tumour and histopathologic positivity of sentinel nodes. Our findings support previous observations that histopathologic evidence of metastatic melanoma in sentinel lymph nodes is an independent predictor of disease recurrence. In contrast, detection of tyrosinase mRNA by RT-PCR alone does not appear to increase the likelihood of short-term disease recurrence.