Complete axillary lymph node dissection for stage I-II carcinoma of the breast.

We reviewed the complete axillary dissection specimens of 136 patients with stage I-II breast cancer to clarify the distribution of axillary lymph node metastases in this disease. Our series included 71 patients undergoing axillary dissection as part of a modified radical mastectomy (MRM) and 65 patients undergoing axillary dissection in conjunction with conservative surgery of the breast and definitive postoperative breast radiotherapy (CAD). These two groups of patients were comparable according to age, menopausal status, tumor size, and clinical stage. In all patients the pectoralis minor muscle was excised and all axillary tissue removed. Each specimen contained a median of 23 lymph nodes. The axillary levels (I, II, III) were determined according to the relationship of axillary tissue to the pectoralis minor muscle (lateral, inferior, medial). Thirty-nine percent of the lymph nodes were contained in level I, 41% in level II, and 20% in level III. There were no significant differences noted in the number of lymph nodes or in the distribution of lymph nodes according to axillary level between dissections performed as part of the MRM or those done as a single procedure (CAD). Sixty-five patients (47.8%) had one or more positive lymph nodes in their axillary specimen. The clinical and pathologic stage was determined and compared for all patients. Among patients judged to have a clinically negative axilla, 37.6% had histologically positive lymph nodes (clinical false-negative rate). For patients with a clinically positive axilla, 11.1% had, histologically, no evidence of metastatic disease (clinical false-positive rate). When the distribution of lymph node metastases according to axillary level was studied, it was found that 29.2% of lymph node-positive patients (or 14.0% of all patients) had metastases only to level II and/or III of the axilla, with level I being negative (skip metastases). This incidence of skip metastases was greater among clinically node-negative than among clinically node-positive patients, but was not related to the size or location of the primary tumor in the breast. In addition, it was found that 20.0% of lymph node-positive patients (or 9.6% of all patients) were converted from three or fewer to four or more positive nodes by analysis of lymph nodes contained in levels II and III. This conversion from three or fewer to four or more positive nodes was due primarily to information contained in level II, with level III contributing to a smaller degree.(ABSTRACT TRUNCATED AT 400 WORDS)