Background: Sentinel lymph node biopsy (SLNB) is used to detect breast cancer axillary metastases. Some surgeons send the sentinel lymph node (SLN) for intraoperative frozen section (FS) to minimize delayed axillary dissections. There has been concern that FS may discard nodal tissue and thus underdiagnose small metastases. This study examines whether evaluation of SLN by FS increases the false-negative rate of SLNB.
Methods: A retrospective analysis of SLNB from 659 patients was conducted to determine the frequency of node positivity among SLNB subjected to both FS and permanent section (PS) versus PS alone. Statistical analysis was performed by the chi(2) square test, and a logistic regression model was applied to estimate the effect of final node positivity between the two groups.
Results: FS was performed in 327 patients and PS was performed in all 659 patients. Among patients undergoing both FS and PS (n = 327), the final node positivity rate was 33.0% compared with 19.6% among patients undergoing PS alone (n = 332). After adjustment for patient age, tumor diameter, grade, and hormone receptor status in a multivariate logistic regression model, there remained an increased likelihood of final node positivity for patients undergoing both procedures relative to PS alone (adjusted odds ratio, 2.1; 95% confidence interval, 1.3-3.6; P = .005).
Conclusions: There was a higher rate of SLN positivity in specimens evaluated by both FS and PS. Therefore, evaluating SLN by FS does not underdiagnose small metastases nor produce a higher false-negative rate. Intraoperative FS offers the advantage of less delayed axillary dissections.