OBJECTIVE In routine practice, axillary lymph node dissection is performed in early invasive breast cancer patients with positive sentinel node biopsy. However, sentinel node is the only involved axillary node in 40-70% of patients, and determining factors that predict axillary non-sentinel node involvement will therefore prevent unnecessary axillary lymph node dissection and decrease morbidity. METHODS In this study, 119 invasive breast cancer patients with sentinel node metastasis who underwent axillary lymph node dissection between 1998-2009 at our institution were studied. Primary tumor characteristics and features of the metastatic tumors in sentinel nodes, such as microanatomic location, size of metastasis, and the ratio of metastatic tumor area to the total sentinel node area were evaluated. Student's t-test and multivariate logistic regression were used for statistical analysis. RESULTS The mean age of the patients was 50.7 years (28-80). Forty-three patients (36%) had invasive ductal and 25 patients (21%) had invasive lobular carcinoma. Most of the patients had either pT1 (44%) or pT2 (54%) tumors. Fifty-four patients (45%) had no further positive nodes in the axilla. The metastatic deposits in the sentinel node were subcapsular in 16 patients (13%). The percent area of sentinel node occupied by tumor (p < 0.001), number of sentinel nodes (p=0.041), and microanatomic location of metastatic tumor (p=0.002) were significantly associated with non-sentinel node metastasis in univariate analysis. The percent area of sentinel node occupied by tumor (p < 0.001) and number of sentinel nodes (p=0.033) remained significantly associated with non-sentinel node involvement in multivariate analysis. CONCLUSIONS In patients with invasive breast cancer and positive sentinel node, area percent of sentinel node occupied by tumor, and the number of sentinel nodes removed are independently predictive of non-sentinel node involvement.