Few long-term follow-up studies prove sentinel lymph node biopsy (SLNB) effectively stages breast cancer without the further evaluation of a completion axillary dissection. Our prospective study addressed this issue, enrolling 345 women with clinically node-negative breast cancer who underwent SLNB from October 1997 through December 2000. The median age of the patients in the study was 56.7 years. Average primary tumor size was 1.42 cm. Ninety-three patients had a positive sentinel lymph node (27%); 70 (75.3%) of these patients underwent completion axillary dissection, while 23 patients (24.7%) declined further surgery. Most (91.3%) of the patients who declined further surgery had evidence of micrometastatic disease only. The median follow-up period for all patients was 60 months. No tumor recurrences in the axilla were reported in either sentinel node-negative or -positive patients. The local and systemic recurrence rates were 3.1% and 4% in node-negative patients and 2.2% and 4.3% in node-positive patients. Two patients (0.9%) in the node-negative group and 6 (6.5%) in the node-positive group died of their disease. Estimated 5-year disease-free survival rates were 96% for node-negative patients and 87% for node-positive patients (P = 0.02). The clinical false-negative rate of the SLNB in this study was 0%. This long-term validation trial proves the accuracy of the SLNB and its extremely low false-negative rate. The findings indicate that patients with a positive SLNB have significantly different survival rates than patients with a negative SLNB.