Sentinel-lymph-node-based management or routine axillary clearance? One-year outcomes of sentinel node biopsy versus...

We sought the extent to which arm morbidity could be reduced by using sentinel-lymph-node-based management in women with clinically node-negative early breast cancer. One thousand eighty-eight women were randomly allocated to sentinel-lymph-node biopsy followed by axillary clearance if the sentinel node was positive or not detected (SNBM) or routine axillary clearance (RAC, sentinel-lymph-node biopsy followed immediately by axillary clearance). Sentinel nodes were located using blue dye, alone or with technetium-labeled antimony sulfide colloid. The primary endpoint was increase in arm volume from baseline to the average of measurements at 6 and 12 months. Secondary endpoints were the proportions of women with at least 15% increase in arm volume or early axillary morbidity, and average scores for arm symptoms, dysfunctions, and disabilities assessed at 6 and 12 months by patients with the SNAC Study-Specific Scales and other quality-of-life instruments. Sensitivity, false-negative rates, and negative predictive values for sentinel-lymph-node biopsy were estimated in the RAC group. The average increase in arm volume was 2.8% in the SNBM group and 4.2% in the RAC group (P = 0.002). Patients in the SNBM group gave lower ratings for arm swelling (P < 0.001), symptoms (P < 0.001), and dysfunctions (P = 0.02), but not disabilities (P = 0.5). Sentinel nodes were found in 95% of the SNBM group (29% positive) and 93% of the RAC group (25% positive). SNB had sensitivity 94.5%, false-negative rate 5.5%, and negative predictive value 98%. SNBM was successfully undertaken in a wide range of surgical centers and caused significantly less morbidity than RAC.