The sentinel lymph node biopsy (SLNB) procedure is an alternative method for assessing the axillary lymph node (ALN) status in patients with breast cancer. The SLNB carries the risk of a false-negative result, with patients harboring positive ALNs in the face of a negative SLNB examination. In addition, the significance of a SLNB with cells identified only with keratin or with deposits less than 0.2 mm remains unresolved. We analyzed our SLNB data over the past 5 years in order to determine the relationship between SLN tumor burden and ALN tumor burden. Pathology files for the past 5 years at Magee-Womens Hospital were searched for all SLNB cases that had an axillary lymph node dissection (ALND). Each SLNB case was reviewed and tabulated for breast tumor size, SLN tumor size, and largest tumor size in the ALND. Correlation and frequency distribution were performed for the status of all SLNs and ALNDs. Patterns of lymph node metastasis were recorded and the sizes of the SLN metastases were reported according to the recent Philadelphia Consensus Conference on Sentinel Lymph Nodes and the revised American Joint Committee on Cancer (AJCC) staging. SLN metastases were classified as immunohistochemistry (IHC) positive if only single keratin-positive cells or clusters were present and were not observed with standard tissue stains, as submicrometastatic (SMM) if tumors were less than 0.2 mm (excluding IHC positive), as micrometastatic if tumors were larger than 0.2 mm but </=2 mm, or as macrometastatic if tumors were larger than 2 mm. A total of 445 patients had both SLNB and ALND. Fifty percent (224/445) of cases were SLN positive, including 58 SLN positive/ALN positive cases and 166 SLN positive/ALN negative cases. Of the 221 patients in the SLN-negative group, 4 were ALN positive (false-negative SLN). The incidence of SLN metastases increased with tumor stage, with the percentage of SLN positives as follows: T1a, 2.1%; T1b, 10.9%; T1c, 51.7%; and T2, 35.3%. There were 4 of 41 patients (10%) with SLNs that were IHC positive that had macrometastases in a solitary ALN. Three of 22 patients (13.6%) that were SMM positive had ALN macrometastasis in a solitary ALN. Four of 49 patients (8.1%) with micrometastatic SLNs had a solitary positive ALN, 3 of which were macrometastases (6.1%). Overall a total of 10 of 112 patients (9.0%) with traditionally defined SLN micrometastases of 2.0 mm or less had a solitary ALN macrometastasis. The vast majority (90%) of these macrometastases were found with T1c and T2 breast tumors. There was a significant difference in the means of SLN tumor sizes for the SLN-positive/ALND-negative (4.5 mm) versus SLN-positive/ALND-positive (10.1 mm) patients, although the range of SLN tumor sizes within each group were similar. There is an increasing incidence of SLN-positive and ALN-positive cases with increasing T stage. Overall in this series, 9% of patients with SLN metastases </=2 mm had a solitary axillary macrometastasis. Ninety percent of these metastases occurred with T1c/T2 breast tumors, indicating the important codependence of T stage. Overall there is a subset of patients who are IHC positive, SMM positive, or micrometastatic positive with ALNs that are macrometastatic who are at risk of harboring axillary macrometastases. Keratin IHC of breast SLNs is useful for defining these subsets.