Clinical and Microbiological Outcomes of Critically Ill Patients With Monomicrobial Versus Polymicrobial Bacterial Bloodstream Infections: A Retrospective Cohort Study

Le résumé fourni par la source

Background: The impact of polymicrobial bloodstream infections in critically ill patients is uncertain. Methods: This observational cohort study included patients admitted to the ICU for ≥3d. We compared patients with monomicrobial versus polymicrobial (ie, ≥1 positive blood culture with ≥2 isolates) bloodstream infections. Inverse-probability weighting and covariate-adjusted generalized linear regression models were used to calculate adjusted differences in risks and means. The primary outcome was 90-day all-cause mortality from index culture. Secondary outcomes included 30-day all-cause mortality, microbiological failure, ECMO at 30 days, ARDS at 7 days, fever at 7 days, length of ICU and hospital stay. A separate analysis was performed excluding common skin commensals. Results: Between 03/2014 and 03/2024, 3197 patients were included in the overall and 1669 in the common commensal-free cohort. In the overall cohort, 90-day mortality occurred in 749 (28.3%, adjusted 28.3%) of 2648 patients in the monomicrobial group and 164 (29.9%, adjusted 31.1%) of 549 patients in the polymicrobial group (adjusted risk difference, 2.84% [95%CI, -1.19-6.88]). All-cause 30-day mortality, microbiological failure (adjusted, 4.67% vs 6.56%), initiation of ECMO (adjusted, 3.65% vs 2.99%), ARDS (adjusted, 17.4% vs 15.7%), fever (adjusted, 19.1% vs 20.4%), length of ICU stay (adjusted means, 19.6 vs 19.8 days), and length of hospital stay (adjusted means, 42.6 vs 45.2 days) did not differ significantly between the groups. Outcomes did not differ in the common commensal-free cohort. Conclusions: Clinical and microbiological outcomes were comparable between critically ill patients with monomicrobial and polymicrobial bloodstream infections, regardless of whether suspected contaminants were excluded.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.