Laboratory‐Detected Clostridioides difficile Stool‐Test Positivity and Mortality in Sepsis: A Retrospective Cohort Study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

ABSTRACT Background Sepsis disrupts intestinal microbiota, and broad‐spectrum antibiotics may worsen dysbiosis. Clostridioides difficile ( C. difficile ) infection (CDI) is a prevalent nosocomial enteric pathogen among critically ill patients, yet its prognostic implications in sepsis remain inadequately studied. Methods We retrospectively analyzed 13,484 adult sepsis hospitalizations from MIMIC‐IV. The exposure was any laboratory‐detected C. difficile‐positive stool test during the index hospitalization. Cox regression, detailed early‐antibiotic adjustment, stabilized inverse probability weighting, timing‐ and assay‐based sensitivity analyses, and a post hoc nested‐model comparison assessed association and incremental prediction. Archived positive‐subgroup machine‐learning models were retained as exploratory risk‐stratification analyses. Results Detailed early‐antibiotic adjustment substantially attenuated the association: the 30‐day hazard ratio (HR) was 0.98 (95% CI, 0.87–1.11; p = 0.767) and the 90‐day HR was 1.08 (95% CI, 0.99–1.18; p = 0.103). Early, landmark, PCR‐only, and toxin‐positive analyses were also nonsignificant, whereas the stabilized‐weighted 90‐day estimate remained modest (HR, 1.13; 95% CI, 1.03–1.24; p = 0.010), indicating inconsistency across analyses. Predictive models showed strong performance (C‐index: 0.81 for in‐hospital mortality; 0.68 for 30‐day mortality). Conclusions While laboratory‐detected C. difficile stool‐test positivity is associated with worse unadjusted outcomes in sepsis, this relationship is substantially attenuated and inconsistent after rigorous adjustment. Accordingly, laboratory positivity may serve as a marker of a complex clinical phenotype. Evidence for its independent prognostic utility remains limited, and its incremental predictive value beyond conventional severity measures is minimal.