Sentinel lymph node biopsy (SLNB) has progressively replaced complete axillary lymph node dissection in the evaluation of breast cancer patients with clinically node-negative disease. Our study investigates the rate of and risk factors involved in sentinel node identification failure.Materials and methods
We collected data on SLNBs performed during 2002–2010, focusing on tumor, patient, and breast characteristics, radioactivity parameters, and operators' experience. Data were analyzed by R (v2.14.2), considering significance at P values lower than 0.05.Results
Among 1050 women who underwent an SLNB, the rate of identification failure was 2% (23/1050), which, on bivariate analysis, was seen to be significantly influenced (P<0.05) by the preoperative and intraoperative low radiotracer uptake (axilla/lesion radiotracer uptake ratio<1%), low level of experience of the specialist in nuclear medicine, luminal A subtype, and radiotracer uptake localization in internal mammary lymph nodes. On multivariate analysis, significant risk factors for sentinel node identification failure were found to be: axilla/lesion radiotracer uptake ratio less than 1%, radiotracer uptake localization in internal mammary lymph nodes, and luminal A subtype. Considering only the preoperative variables in our multivariate analysis, axilla/lesion radiotracer uptake ratio less than 1%, negative lymph node scintiscan, and radiotracer uptake localization in internal mammary lymph nodes had an area under the curve (receiver operating characteristic curve) of 96% (95% confidence interval 92–100%). Further, we built a nomogram based on these simple parameters for counseling the patient about the probability of not finding the sentinel lymph node during the surgical procedure.Conclusion
The relatively low prevalence of SLNB failure (2%) is indicative of the accuracy of the procedure when performed by experienced surgeons. The sentinel node identification failure in our population seemed to be related to biological tumor factors (luminal A subtype) and probably to physiological or pathological variations in the lymphatic drainage (axilla/lesion radiotracer uptake ratio<1% and radiotracer uptake localization in internal mammary lymph nodes).