Robichaux, Chad

Abstract Background and Aims Women are underdiagnosed and undertreated for cardiovascular disease (CVD). Automatic quantification of breast arterial calcification (BAC) on screening mammography can identify women at risk for CVD. This study aimed to determine whether artificial intelligence–based automatic quantification of BAC from screening mammograms predicts CVD and mortality beyond PREVENT scores in a large, racially diverse, multi-institutional population. Methods This retrospective cohort study included 123 762 women from two healthcare systems who had screening mammograms. Breast arterial calcification was quantified using a transformer-based neural network for segmentation. Breast arterial calcification severity was categorized as zero (0 mm2), mild (>0–10 mm2), moderate (>10–25 mm2), and severe (>25 mm2). Kaplan–Meier analysis, Cox proportional hazards, and Fine–Gray competing event models were used to examine the association between BAC and major adverse cardiovascular events (MACE). Results Breast arterial calcification was detected in 16.1% (internal cohort) and 20.6% (external cohort) of women and provided significant prognostic value incremental to the PREVENT score. In PREVENT adjusted models, a clear dose–response was observed. Compared with zero BAC, mild [internal: hazard ratio (HR) 1.32, 95% confidence interval (CI) 1.10–1.59; external: HR 1.28, 95% CI 1.17–1.39], moderate (internal: HR 1.75, 95% CI 1.23–2.50; external: HR 1.79, 95% CI 1.55–2.06), and severe BAC (internal: HR 3.29, 95% CI 2.15–5.05; external: HR 2.80, 95% CI 2.36–3.32) were all prognostic for any MACE. Each 1 mm2 increase in BAC conferred an additional 2%–3% risk for MACE (P < .001). Conclusions Automatically quantified BAC is an independent predictor of MACE and mortality, adding prognostic value to the PREVENT score. This approach may provide an opportunistic cardiovascular risk assessment during routine mammography screening without additional radiation exposure to guide earlier and more effective preventive care for women.

Abstract Background Clostridioides difficile infection (CDI) incidence estimates vary between geographic regions; few studies have evaluated the impact of CDI test order frequency on estimated CDI incidence. We evaluated this impact in a sample of hospitals at two CDC Emerging Infections Program (EIP) sites. Methods Daily surveillance was conducted for diarrhea among inpatients at 5 acute care hospitals (2,379 beds) in EIP sites in NY (2 hospitals) and GA (3 hospitals) during two 10-workday periods in 2020 and 2021. Diarrhea onset, test orders, and specimen collection status were ascertained. Stools were tested by PCR/NAAT initially or after negative EIA toxin. Differences in diarrhea incidence, testing frequency, and CDI positivity across site, care locations and hospitals were compared using Wilcoxon rank sum test. Correlates of CDI testing and positivity were assessed using modified Poisson regression. Estimates of incidence using EIP methodology at 5 hospitals was compared between sites using Mantel-Hanzel summary rate ratio. Results Surveillance of 38,365 patient-days (PD) identified 860 diarrhea cases from 107 patient-care locations mapped to 26 unique NHSN defined location-types. Incidence of diarrhea was 22.4/1000 PD (medians 25.8 NY, 16.2 GA, P< 0.01); with similar proportions of diarrhea being hospital-onset (66%) and CDI positive (17%) by site. Overall, 35% were tested for CDI (21% NY, 49% GA, P< 0.01). Percent tested varied by NHSN location type (Figure). Regression models identified location-type (oncology, critical care), laxatives use, chemotherapy, and residing in EIP catchment area predictive of testing (Figure). Adjusting for these factors, NY was 49% less likely than GA to test (aRR 0.51, 95% CI 0.40-0.63). Simulation of EIP methods estimated NY had a 38% lower incidence of CDI than GA (summary rate ratio 0.62, 95% CI, 0.54-0.71). Figure Incidence of Diarrheal Episodes (A) and Proportions Tested (B) among Hospitalized Patients, 2021 (solid, ward; open, critical care; grey, oncology), and adjusted relative risk (solid circles) with 95% confidence intervals (whisker) of independent predictors of testing for CDI (C). Conclusion After adjusting for patient characteristics (e.g., location-type, laxative use), the likelihood of testing still differed between NY and GA sites; the magnitude of the differences in testing was similar to that observed in estimated CDI incidence. Testing practices likely influence surveillance data and is a consideration when comparing data across regions. Disclosures Scott Fridkin, MD, Pfizer: Grant/Research Support christopher J. Myers, MS, Infectious diseases, Pfizer: Grant/Research Support Udodirim N. Onwubiko, MBBS MPH, Pfizer: Grant/Research Support William C. Dube, MPH, Pfizer: Grant/Research Support Sahil Khanna, MBBS, MS, Pfizer: Grant/Research Support Joann M. Zamparo, MPH, Pfizer: Employee|Pfizer: Stocks/Bonds Frederick J. Angulo, DVM PhD, Pfizer Vaccines: Employee|Pfizer Vaccines: Stocks/Bonds Ghinwa Dumyati, MD, Pfizer: Grant/Research Support.