Clinical and histological factors associated with sentinel node identification in breast cancer

Abstract

<h4>Background</h4>Although sentinel lymph node biopsy is likely to be offered as a method of assessing nodal status in primary breast cancer, the inability to identify the sentinel node at the time of surgery will limit the number of patients who may benefit from the procedure. The purpose of the present study was to identify factors that are associated with intraoperative identification of the sentinel node(s).<h4>Methods</h4>Between September 1995 and May 1999, lymphatic mapping using a combination of preoperative lymphoscintigraphy and/or blue dye was performed on 169 consecutive patients with clinically lymph node-negative primary operable breast cancer. Clinical and histological factors were assessed using univariate and multivariate analysis to determine those that were associated with intraoperative identification of the sentinel node.<h4>Results</h4>The sentinel node was identified at the time of surgery in 142 cases (84%). Of the clinical factors assessed, preoperative identification of the sentinel node on lymphoscintigraphy (P < 0.0001), use of blue dye in combination with isotope (P = 0.001), symptomatic palpable tumours (P < 0.05) and the experience of the surgeon (P = 0.03) were significant in identifying the sentinel node at operation. No histological factor was associated with intraoperative identification of the sentinel node. Using multivariate analysis, positive identification of the sentinel node on lymphoscintigram, the experience of the surgeon and the use of both blue dye and isotope for sentinel node mapping were independent factors associated with intraoperative sentinel node identification. The lymphoscintigram result was the strongest independent factor according to its beta value, a measure of the weight of significance.<h4>Conclusion</h4>Patients undergoing sentinel lymph node mapping and biopsy should be warned of the possibility of failure of sentinel node identification at operation. Our results suggest that the best predictor of intraoperative sentinel node identification is the visualization of the sentinel node on preoperative lymphoscintigraphy. The result of the lymhoscintigram may allow for additional preoperative counselling of the patient regarding the success or failure of sentinel node biopsy. Technical factors such as the experience and diligence of the surgeon, as well as the sentinel node mapping technique, are also important in determining the success of the procedure.