Background: To evaluate the accuracy of 5node sampling as an alternative to classic axillary dissection in operable breast cancer(stage I and II )     Method and Material: 5 largest nodes of level one were sampled in 26 consecutive patient with breast cancer undergoing modified radical mastectomy and axillary clearance between june 2002 to march 2004.

Result: False negative rate for 5 node sampling was 7.7%, of 14 cases with negative 5 node sampling 2 (7.7%) were found to have disease elsewhere in the axilla.

Conclusion: Sampling of 5 largest nodes accurately (92%) identifies patients with metastatic nodes.