We prospectively studied the feasibility of sentinel lymph node biopsy (SLNB) after neoadjuvant chemotherapy by comparing the identification rate and the false-negative rate (FNR) with the results obtained from the patients without chemotherapy. From October 2001 to March 2003, a total of 284 consecutive patients who underwent SLNB and axillary lymph node dissection (ALND) at the Center for Breast Cancer, National Cancer Center were enrolled. Of the 284 patients, 54 underwent neoadjuvant chemotherapy prior to operation. The sentinel lymph node (SLN) was mapped by radioactive colloid alone or in combination with blue dye. All SLNs were evaluated by 2 mm serial sections after hematoxylin-eosin staining. The overall SLN identification rate was 91.9% (261/284): 72.2% (39/54) of the patients after chemotherapy and 96.5% (222/230) of the patients without chemotherapy. These results suggest that preoperative chemotherapy significantly affects lymphatic mapping ( p< 0.001). Among the patients with chemotherapy, there were 3 false negatives in 39 successfully mapped tumors, yielding an FNR of 11.1% (3/27), a negative prediction value (NPV) of 80.0% (12/15), and an accuracy of 92.3% (36/39). There were 10 false negatives among 222 successfully detected patients without chemotherapy, yielding an FNR of 9.9% (10/101), an NPV of 92.4% (121/131), and an accuracy of 95.5% (212/222). These results were not statistically different when compared ( p > 0.05). Although the SLN identification rate significantly decreased after neoadjuvant chemotherapy, SLNB could accurately predict axillary status. Thus SLNB can be an alternative to ALND even after neoadjuvant chemotherapy in cases of successful identification of the SLN.