Completion axillary lymph node dissection minimizes the likelihood of false negatives for patients with invasive breast carcinoma and cytokeratin positive only sentinel lymph nodes.

Objective: To document the incidence of metastatic disease in complete axillary lymph node dissections (CALND) of patients with invasive carcinoma after a sentinel lymph node (SLN) biopsy, positive only by immunohistochemical staining for cytokeratin (CK-IHC).
Methods: Sections of all SLNs, negative by routine histology, were immunostained and examined for cytokeratin positive cells. Sections of lymph nodes from CALND specimens were interpreted using routine hematoxylin and eosin (H&E) staining.
Results: A total of 409 patients (29.6%) had metastatic disease in at least one sentinel lymph node on H&E examination. Of 971 H&E negative patients, 78 (8.0%) were positive only by CK-IHC. Sixty-two of the CK-IHC positive only patients underwent CALND. Nine of these 62 patients (14.5%) had metastases identified in the CALND specimen.
Conclusions: Because 14.5% of patients with invasive breast cancer and SLNs positive only by CK-IHC were found to have H&E positive lymph nodes on CALND, we conclude first, that CK-IHC should be used to evaluate SLNs, and second, that CALND should be considered when SLNs are positive by CK-IHC only. This approach will result in an absolute reduction of the false negative rate (absolute false negative rate reduced by 2.6% in our series).