Bloodstream infection following intestinal colonization with Carbapenem-resistant enterobacterales: A systematic review and meta-analysis of incidence proportion, colonization-to-bsi interval, and risk factors.

BACKGROUND

Intestinal colonization with carbapenem-resistant Enterobacterales (CRE) is a major reservoir for invasive infection, but subsequent bloodstream infection (BSI) risk, observed time to detection, and associated factors remain poorly defined. We quantified the BSI incidence proportion, observed colonization-to-BSI interval, and associated factors.

METHODS

We searched PubMed, Embase, Web of Science, and CNKI from inception to May 1, 2026 for observational studies of microbiologically confirmed intestinal CRE colonization. BSI incidence proportions were logit-transformed and pooled using random-effects models. Reported interval medians were converted to approximate means and SDs and pooled. Adjusted relative association estimates reported in at least three studies were synthesized. Heterogeneity and robustness were assessed using I², prediction intervals, and subgroup and sensitivity analyses.

RESULTS

Nineteen studies including 13,375 colonized patients were analysed. The pooled BSI incidence proportion was 10.9% (95% CI, 7.5%-15.7%; I² = 96.8%; 95% prediction interval, 2.1%-41.5%). The pooled approximate mean observed colonization-to-BSI interval was 20.5 days (95% CI, 9.9-31.0 days; I² = 98.7%). Neutropenia (pooled adjusted estimate = 12.00, 95% CI, 5.73-25.14), multisite colonization (5.92, 95% CI, 2.52-13.90), carbapenem exposure (2.72, 95% CI, 1.28-5.76), and ICU admission (1.95, 95% CI, 1.37-2.76) were associated with BSI.

CONCLUSION

BSI incidence proportions and observed intervals varied substantially across settings. These estimates may support context-specific risk stratification, infection-prevention and stewardship planning, and parameterization of hospital CRE models, but require local calibration and prospective validation integrating host and microbiological data.