Background: This study evaluated the use of axillary surgery (AS), including sentinel lymph node biopsy (SLNB), for patients with ductal carcinoma in situ (DCIS) and the factors associated with its use. To determine whether utilization of SLNB is appropriate, predictors of SLNB performance were compared with factors predictive of tumor upstaging.
Methods: The National Cancer Data Base was utilized to identify patients with American Joint Committee on Cancer (AJCC) clinical stage 0 breast cancer treated from 2004 to 2013. DCIS with microinvasion was excluded. Chi square tests and logistic regression were used to examine patient, tumor, and facility features associated with SLNB and tumor upstaging.
Results: Of the 218,945 total patients, 155,093 (70.8 %) underwent lumpectomy, and 63,852 (29.2 %) underwent mastectomy. SLNB was performed for 19.0 % of lumpectomy patients and 63.5 % of mastectomy patients. Multivariate analysis for 2012-2013 demonstrated that estrogen receptor (ER)-negative and grade 3 tumors were more likely to be treated with SLNB in both groups. Tumor size was significant only for the lumpectomy patients who underwent one operation. Further, 22.8 % of lumpectomy patients and 18.7 % of mastectomy patients who underwent AS were upstaged compared with 1.8 % of lumpectomy and 3.6 % of mastectomy patients who did not undergo AS. Tumor upstaging was predicted by ER-negative status (odds ratio [OR] 2.99; 95 % confidence interval [CI] 2.76-3.24) but not by higher grade or larger tumor size.
Conclusions: Use of SLNB for DCIS is high with mastectomy, and nearly one fifth of the lumpectomy patients underwent SLNB. However, the performance of AS was strongly associated with the likelihood of upstaging in both groups, suggesting that surgical judgment plays an important role in this decision.