Background and method: This paper reviews and discusses the feasibility and accuracy of sentinel lymph node (SLN) biopsy in breast cancer. A standardized method of identifying the SLN and detecting micrometastases is suggested, along with a strategy for the elimination of routine axillary lymph node dissection (ALND). Results: Although the SLN can be identified successfully by experienced practitioners using either the dye-guided or gamma probe-guided method, identification is facilitated when the two techniques are combined. To improve the likelihood of spotting metastases in the SLN, it is desirable to perform step sectioning combined with haematoxylin and eosin staining and immunohistochemistry of permanent and frozen sections. SLN biopsy is as accurate for T-2 tumours as it is for T-1 tumours. However, it is highly unlikely that all false-negative cases can be eliminated, even by detailed histological examination. Nevertheless, patients with T-1 tumours with micrometastases in the SLN have shown no evidence of turnout in the non-sentinel nodes. In other words, ALND can be avoided in these patients, even if histological examination of the SLN fails to detect micrometastasis. Conclusion: In practice, routine ALND can be avoided in patients with T-1 tumours when the identified SLN proves to be histologically negative. However, investigation of long-term regional controls and of survival in a prospective randomized trial is necessary before SLN biopsy can replace routine ALND, particularly for patients with T-2 tumours.