Abstract
Background/Aims In 1999, the National Institutes of Health Consensus Conference recommended breast conserving therapy (BCT) as a viable alternative to mastectomy based on several long-term studies and meta-analyses. Since then BCT has been the procedure of choice for most women with early stage breast cancer. However, several recent studies have reported a shift toward increasing use of total mastectomy (TM). We evaluated patterns of mastectomy use and factors associated with mastectomy rates in the Breast Cancer Surgical Outcomes (BRCASO) study.
Methods BRCASO is a collaborative study between University of Vermont, Kaiser Permanente, Colorado; Group Health, Seattle; and Marshfield Clinic, Wisconsin and includes detailed data on 4580 women diagnosed with incident invasive or in situ breast cancer between 2003–2008 obtained from administrative data and medical record abstraction. For this analysis, we included 2384 women with an incident diagnosis of stage 1–3 breast cancer with no history of chest radiation. We excluded cases of inflammatory or multifocal breast cancer.
Results We found rates of TM as the initial surgical procedure varied by study site (p<0.0001), age at diagnosis (p=0.0005), and by use of preoperative MRI (p<0.0001). TM frequency ranged from 10.4% – 21.0% across the four institutions. Increased TM use was associated with increasing pre-operatively estimated tumor size (p<0.0001), ER/PR negative status (p=0.0007), high grade (p<0.0001) and lobular (vs. ductal) carcinoma (p=0.001). We did not find TM increasing over time. When we limited our analysis to the 1712 cases with tumor size <20 mm, we continued to observe variability in initial TM frequency across institutions from 5.2% to 14.2% (p<0.0001).
Discussion Our TM rates were low compared to those reported in previous studies. Variation in TM rates was associated with several patient, tumor, and facility characteristics and may reflect differences in the patient population presenting with breast cancer, or in patients’ treatment preferences. This variation persisted even when limited to women with small tumors. While TM rates were relatively low, the two-fold variation by study site suggests inconsistent surgical practices across facilities despite several studies showing BCT and mastectomy provide equal survival benefit.
