Risk stratification of surgical site outcomes by body mass index in autologous breast reconstruction: An analysis of a globally federated electronic medical record network in improving patient safety

AIM: With an annual US incidence exceeding 300,000, breast cancer remains an extremely devastating disease1. Furthermore, the average American Body Mass Index (BMI) of 26.5 indicates that a large proportion of the population is at increased risk for adverse outcomes following reconstructive surgery2,3. This study aims to utilize a continuously updated, federated network of 65 million EMRs (TriNetX Inc, Cambridge, MA) for analysis of 90-day post-operative outcomes in autologous breast reconstruction patients by increasing BMI; enabling clinical stratification of risks associated with overweight and obese patients. METHODS: We used TriNetX, a novel research network providing statistics on EMRs from 45 Health Care Organizations globally. The de-identified records of 29,453,000 females, age 18-99, were retrospectively screened. A cohort of 7,136 patients undergoing autologous breast reconstruction via transverse rectus abdominus muscle (TRAM) flap, deep inferior epigastric perforator (DIEP) artery flap, or latissimus flap was categorized by BMI into 5 different subgroups: Normal (n = 3,568), Overweight (n = 1,239), and Class I (n = 1,166), Class II (n = 807), & Class III (n = 356) Obesity. The Normal BMI cohort was then compared to each elevated BMI cohort. BMI strata were then analyzed for risk of surgical site occurrences (SSO) within 90 days of index event using EMR CPT codes. Outcomes were assessed with propensity score matching for confounders including age, race, ethnicity, neoplastic history, chemotherapy exposure, irradiation, & lifestyle hazards including smoking. RESULTS Regarding 90-day SSOs, statistically significant (p