Background: Our study aims were to investigate breast cancer patients with micrometastases or isolated tumor cells (ITCs) in sentinel lymph nodes (SLNs) to determine the rate of non-SLN metastasis and axillary recurrences, and to compare actual non-SLN metastasis rates with those predicted by the Memorial Sloan-Kettering Cancer Center (MSKCC) nomogram. Methods: We identified 116 stage I to III breast cancer patients who underwent sentinel lymph node biopsy and had micrometastases or ITCs (<2-mm deposits). Patients underwent completion axillary lymph node dissection (ALND) (group 1) or had no further axillary surgery (group 2). P < 0.05 was considered statistically significant. Results: Of 116 patients with micrometastases or ITCs in SLNs, 55 (47%) underwent completion ALND (group 1), and 61 (53%) had no further axillary surgery (group 2). The rate of non-SLN metastases in group 1 patients was 9 (16%) of 55, which was significantly less than that predicted by the MSKCC nomogram (median 30%, P < 0.001). Patient age, race, tumor histology, tumor grade, estrogen receptor/Her-2neu status, and lymphovascular invasion did not differ significantly between group 1 patients with positive non-SLNs and those with negative non-SLNs (P > 0.05 for each), but patients with positive non-SLNs had larger tumors (P < 0.001). No patient in group 1 experienced an axillary recurrence, while only one patient (1.6%) in group 2 experienced axillary recurrence. Conclusions: The actual rate of positive non-SLNs for breast cancer patients with SLN micrometastases or ITCs who underwent completion ALND was significantly less than that predicted by the MSKCC nomogram. The rate of axillary recurrence is negligible, regardless of the extent of axillary staging.