Preoperative breast MRI: Surgeons' patient selection patterns and potential bias in outcomes analyses

OBJECTIVE. The purpose of this study is to determine which patient- and tumor-related and clinical variables influence dedicated breast surgeons' and general surgeons' referrals for preoperative breast MRI for patients with newly diagnosed breast cancer. MATERIALS AND METHODS. Surgeons who perform breast surgery responded to a survey from June 16, 2014, through August 11, 2014. Participants self-identified as breast or general surgeons and provided professional practice details. They used Likert scores (range, 1-7 with increasing likelihood to order MRI) to weigh numerous patient- and tumor-related and clinical variables. Mean likelihood scores were calculated and compared using a linear mixed model. A p ≤ 0.05 was considered statistically significant. RESULTS. Two hundred eighty-nine surveys from 154 (53%) breast surgeons and 135 (47%) general surgeons showed an overall likelihood to refer for patients with a BRCA mutation (mean Likert score, 6.17), familial (mean Likert score, 5.33) or personal (mean Likert score, 5.10) breast cancer history, extremely dense breasts (mean Likert score, 5.30), age younger than 40 years (mean Likert score, 5.24), axillary nodal involvement (mean Likert score, 6.22), tumor that is mammographically occult (mean Likert score, 5.62) or fixed to the pectoralis (mean Likert score, 5.49), tumor that is a candidate for neoadjuvant treatment (mean Likert score, 5.38), multifocal or multicentric disease (mean Likert score, 5.22), invasive lobular carcinoma (mean Likert score, 5.20), T3 (mean Likert score, 4.48) or T2 (mean Likert score, 4.41) tumor, triplenegative breast cancer (mean Likert score, 4.66), a patient who is a candidate for mastectomy requesting breast conservation therapy (mean Likert score, 5.27), and radiologists' recommendations (mean Likert score, 5.19). Across all patient ages, breast surgeons referred more often than did general surgeons (mean Likert score, 4.32 vs 3.92; p = 0.03), especially for patients with BRCA mutation (mean Likert score, 6.39 vs 5.93; p = 0.01) and tumors smaller than 1 cm (mean Likert score, 3.84 vs 3.40; p = 0.002). Breast surgeons referred less often than did general surgeons for multifocal or multicentric disease (mean Likert score, 5.02 vs 5.44; p = 0.001). Breast surgeons and general surgeons similarly weighed other variables. CONCLUSION. Preoperative breast MRI referral trended with certain higher risk patient- and tumor-related and clinical variables and were nonuniform between the breast surgeons and general surgeon cohorts. Selection bias could affect outcomes analyses for preoperative breast MRI.