Performance of Screening Ultrasound as an Adjunct to Screening Mammography in Women Across the Spectrum of Breast Cancer Risk

Importance: Whole-breast ultrasound has been advocated to supplement screening mammography to improve outcomes in women with dense breasts. Objective: To determine the performance of screening mammography plus screening ultrasound compared with screening mammography alone in community practice. Design: Observational cohort study. Setting: The study was IRB approved and HIPAA compliant. Two Breast Cancer Surveillance Consortium (BCSC) registries provided prospectively collected data on screening mammography with vs. without same-day breast ultrasound from 2000-2013. Participants: 6,081 screening mammography plus same day screening ultrasound examinations in 3,385 women were propensity score matched 1:5 to 30,062 screening mammograms without screening ultrasound in 15,176 women from a sample of 113,293 mammograms. Exclusion criteria included personal history of breast cancer and self-reported breast symptoms. Exposure: Screening mammography with versus without screening ultrasound. Main Outcomes and Measures: Cancer detection rate (CDR), and rates of interval cancer, false-positive (FP) biopsy recommendation, short-interval follow-up (SIFU), and positive predictive value of biopsy recommendation (PPV2) were estimated and compared using logbinomial regression. Results: Screening mammography with vs without ultrasound examinations were performed more often in women with dense breasts (74% vs 36% in the overall sample); who were younger than 50 years (50% vs 32%), with a family history of breast cancer (43% vs 15%). While 21% of screening ultrasound examinations were performed in women with high or very high (>2.50%) BCSC 5-year risk scores, 53% had low or average (<1.67%) risk. Comparing mammography plus ultrasound to mammography alone, CDR was similar: 5.4 vs. 5.5 per 1000 examinations (adjusted relative risk [RR]=1.14 95% confidence interval [CI]: 0.76-1.68); as were interval cancer rates: 1.5 vs. 1.9 per 1,000 examinations (RR=0.67, 95%CI: 0.33-1.37); FP biopsy rates were significantly higher: 52.0 vs. 22.2 per 1000 examinations (RR=2.23, 95%CI: 1.93-2.58); as was SIFU: 3.9% vs. 1.1% (RR=3.10, 95%CI: 2.60-3.70); PPV2 was significantly lower: 9.5% vs. 21.4% (RR=0.50 95%CI: 0.35-0.71). Conclusions and Relevance: In a relatively young population of women at low, intermediate, and high breast cancer risk, our results suggest that the benefits of supplemental ultrasound screening may not outweigh associated harms.