Is Preoperative Lymphoscintigraphy Needed for Sentinel Node Procedures in Breast Cancer?

Abstract

Sentinel node biopsy techniques have gained rapid acceptance in the medical community as being an accurate tool for the pathologic staging of regional lymph nodes in patients with clinically node-negative breast cancer. Two large multicenter clinical trials, the National Surgical Adjuvant Breast and Bowel Project B32 and the American College of Surgeons Oncology Group Z10 trials, have recently completed accrual and should give us definitive proof that these minimally invasive procedures provide regional disease control and patient survival equivalent to those with standard axillary node dissection, with less morbidity. A great deal of attention is now being paid to determining the most effective methods for performing these procedures and what factors may influence their success. In this issue of Annals of Surgical Oncology, Rousseau et al. describe the factors that may influence the ability to visualize sentinel nodes on preoperative lymphoscintigraphy and the effect that nonvisualization has on the outcome of the sentinel node procedure. Lymphoscintigraphy has been used routinely in sentinel node procedures for melanoma and has been quite useful in this role because of the variability in lymphatic drainage patterns seen in this disease. However, lymphoscintigraphy has been met with mixed enthusiasm for sentinel node identification in breast cancer. From the surgeon s perspective, for lymphoscintigraphy to be a clinically useful tool, it should have a high rate of success in identifying the location of the sentinel node, and there should be some ambiguity as to which nodal basin the sentinel node will be located in before the scan is performed. The track record for lymphoscintigraphy in breast cancer has been for only moderate success in identifying the sentinel node site (only 78.5% successful in this study), whereas the presence of unsuspected nodal drainage sites is relatively uncommon, especially if one considers the internal mammary basin as a known potential drainage site. In this study, the authors made sure to use a technique that would be expected to give a high level of success for visualizing sentinel nodes. All injections were given in the periareolar region, were of small volume (.1 mL), and were of adequate dose (30–40 MBq). Common problems identified in the past that have inhibited the success of lymphoscintigraphy in breast cancer have been the overlap of injection site activity with nodal drainage sites and a lack of sufficient tracer reaching the sentinel nodes. Each of these should have been avoided by the technique used. The 21.5% nonvisualization rate in this study is in fact better than rates in many reports in the literature, but from the standpoint of the surgeon, this rate is still relatively high. This is further borne out by the fact that in 84.6% of patients in whom there was nonvisualization of a sentinel node by scan, a sentinel node was found by the surgeon using the gamma detector at the time of operation. Additional sentinel nodes were also found when blue dye was included, and this increased the sentinel node identification rate to 88.4%. As would be expected, if a sentinel node was identified on preoperative lymphoscintigraphy, the sentinel node identification rate was higher (93.2%), but it was Received March 10, 2005; accepted April 17, 2005; published online j. Address correspondence and reprint requests to: Seth P. Harlow, MD; E-mail: seth.harlow@uvm.edu.