PULM11: Regional Variation in Severe COVID-19 Therapeutics and Outcomes Within a Single Large Healthcare System: Insights from the ECMOCARD Study

Purpose: Despite extensive descriptions of patients and outcomes associated with critical illness due to COVID-19, there is paucity of data from individual large healthcare systems. Here we describe our experience at Baylor Scott & White Health in Texas. Methods: Patients, treatments, and outcomes were compared among our 4 ECMO centers in Central (Temple) and North Texas (Baylor All Saints (BAS), Baylor University Medical Center (BUMC), The Heart Hospital Baylor-Plano (THHBP)) via the ECMOCARD Registry (Covid-19 Critical Care Consortium) using Bonferroni-corrected multivariable statistics and post-hoc comparisons. Results: Of 228 patients (mean age 57.9 y; 56% men), 26% were treated at BAS, 27% at BUMC, 34% at Temple, and 13% at THBBP. All were admitted to the ICU or high dependency unit with a median length of stay of 17 days. The majority were on inotropes or vasopressors. Overall, 85% received some level of O2 support including 39% non-invasive, 67% high flow nasal cannula, and 34% ECMO, non-exclusive. The 28-day survival was 63% overall with some regional differences (BAS, 47%; BUMC, 72%; Temple, 64%; THHBP, 75%; P=0.02). Patient demographics differed by center with Temple having older patients and BUMC the largest proportion of Black patients (P<0.001). The most prevalent comorbidities were hypertension (59%), obesity (39%), Type 2 diabetes (36%), and chronic cardiac disease (29%). Temple patients had the highest prevalence of chronic neurological disorder, hypertension, and chronic kidney disease. The most frequent level of support at BAS, BUMC, and THHBP was invasive ventilation, whereas Temple patients most frequently were on high flow nasal O2. ECMO support was used most commonly at BUMC and least frequently at Temple. Summary: In this large hospital system there was significant variation among characteristics and treatments for patients with severe-COVID 19 illness despite system-wide protocolized care. Regional impact on patient characteristics and health outcomes warrants further study.