Association of COVID-19 coinfection with increased mortality among patients with

Objective: Design: Retrospective cohort study. Setting: Veterans Health Administration. Patients: Hospitalized patients with PA-BSI in pre-COVID-19 (January 2009 to December 2019) and COVID-19 (January 2020 to June 2022) periods. Patients in the COVID-19 period were further stratified by the presence or absence of concomitant COVID-19 infection. Methods: We characterized trends in resistance, treatment, and mortality over the study period. Multivariable logistic regression and modified Poisson analyses were used to determine the association between COVID-19 and mortality among patients with PA-BSI. Additional predictors included demographics, comorbidities, disease severity, antimicrobial susceptibility, and treatment. Results: A total of 6,714 patients with PA-BSI were identified. Throughout the study period, PA resistance rates decreased. Mortality decreased during the pre-COVID-19 period and increased during the COVID-19 period. Mortality was not significantly different between pre-COVID-19 (24.5%, 95% confidence interval [CI] 23.3-28.6) and COVID-19 period/COVID-negative (26.0%, 95% CI 23.5-28.6) patients, but it was significantly higher in COVID-19 period/COVID-positive patients (47.2%, 35.3-59.3). In the modified Poisson analysis, COVID-19 coinfection was associated with higher mortality (relative risk 1.44, 95% CI 1.01-2.06). Higher Charlson Comorbidity Index, higher modified Acute Physiology and Chronic Health Evaluation score, and no targeted PA-BSI treatment within 48 h were also predictors of higher mortality. Conclusions: Higher mortality was observed in patients with COVID-19 coinfection among patients with PA-BSI. Future studies should explore this relationship in other settings and investigate potential SARS-CoV-2 and PA synergy.