Effect of bloodstream infection on survival in COVID-19 patients admitted to an intensive care unit in Colombia: a matched cohort analysis

Summary: Aim: To determine the impact of bloodstream infection (BSI) and other risk factors for mortality in patients with COVID-19 admitted to the intensive care unit (ICU). Methods: A retrospective cohort was carried out at the Hospital Universitario Nacional (HUN) between March 29 and December 19, 2020. Patients with COVID-19 admitted to the Intensive Care Unit (ICU) were paired 1:4 in two groups, one with BSI and the other without, according to hospital stay and the month of admission. The primary outcome was mortality at 28 days. A Cox proportional hazards model was used to estimate differences in mortality risk. Results: 456 patients were identified and 320 were included in the final cohort, 18% (n = 59) in the BSI group and 82% (n = 261) in the control group. 125 (39%) patients died, 30 (51%) in the BSI group and 95 (36%) in the control group (P = 0.040). BSI was associated with an increased risk of in-hospital mortality at 28 days, [HR] 1.77 (95% CI: 1.03–3.02; P = 0.037). Invasive mechanical ventilation (IMV) and age were associated with increased mortality risk. Some months of the year of the hospital stay were associated with a reduced risk of mortality. There was no difference in mortality between inappropriate and appropriate empirical antimicrobial use. Conclusion: BSI in patients with COVID-19 in ICU increases in-hospital mortality to 28 days. Other risk factors for mortality were IMV and age.