Unfiltered sulfur colloid and sentinel node biopsy for breast cancer: Technical and kinetic considerations

Background: There are few clinical data on technical limitations and radiocolloid kinetics related to sentinel lymph node (SLN) biopsy for breast cancer. Methods: In 70 clinical node-negative patients, unfiltered ^99mTc sulfur-colloid was injected peritumorally and cutaneous hot spots were mapped with a gamma probe. SLN biopsy was performed followed by axillary lymph node dissection. Missed radioactive nodes (nodes not under hot spots) were removed from axillary lymph node dissection specimens and submitted separately. Results: At least one hot spot was mapped in 69 patients (98%) and SLNs were retrieved in 62 (89%). No radiolabeled nodes were found in five (7%) and only nodes not under hot spots were retrieved in three patients (4%). Residual nodes not under hot spots were retrieved in 17 patients (24%) in whom at least one SLN specimen had been found. Diffuse radioactivity around the radiocolloid injection site impeded identification of all radiolabeled nodes during SLN biopsy, and was responsible for one of two false negatives (20 node-positive patients; false-negative rate 10%). Hot spot radioactivity, number of radiolabeled nodes, and nodal radioactivity did not change with time interval from radiocolloid injection to surgery (0.75-6.25 hours). Conclusions: Although SLN localization rate is high, intraparenchymal injection may predispose to failure of radiocolloid migration, failure to identify SLNs because of high radiation background, and false-negative outcomes. Alternative routes of radiocolloid administration should be explored.