Although an axillary lymph node dissection (ALND) has been considered as an ultimate procedure for axilla in the breast carcinoma, complications after ALND and conceptual revolution for the role of axillary nodes have made it necessary to look for an alternative. Recent studies of sentinel lymphadenectomy (SLND) have shown that SLND accurately predict axillary nodal status. However, for a satisfying outcome in SLND, a learning period would be required, as other surgical procedures do. In this study, the necessity of the learning period for SLND were examined.

From Nov. 1998 to Dec. 2001, 178 patients with invasive breast carcinoma were treated with SLND simultaneously followed by ALND. The period for the first 54 patients, Nov. 1998 to May. 1999, was set as a 'learning period'. Differences of the detection rate and the false negative rate for the 'learning period' and 'after the learning period' were compared. Also changes in the detection rate and the false negative rate with the accumulated experiences for SLND were evaluated.

The sentinel lymph nodes were not identified in 5 patients through the whole period. Three of them occurred in the 'learning period' (5.6%, 3/54) and the rest occurred 'after the learning period' (1.6%, 2/124)(P=0.04). The false negative rate was 16.7% (4/24) in the 'learning period' and 0.0% in 'after the learning period' (P=0.00). The detection rate and the false negative rate improved with the accumulation of experiences for SLND.

It is certain that 'learning period' for SLND is crucial. During this period, the improvement and stabilization of this skill is achieved.