Back

Burden and Genomic Landscape of Antimicrobial Resistance in Non-Bloodstream Infections Among Patients with Cirrhosis: A Systematic Review and Meta-Analysis

Paintsil, E. K.; Ozdemir, C.; Paul, T.; Egoh, K.; Wanford, J. J.; Shawcross, D.

2025-12-15 gastroenterology
10.64898/2025.12.12.25342176 medRxiv
Show abstract

BackgroundPatients with cirrhosis are highly susceptible to infections due to immune dysfunction and gut barrier impairment. Non-bloodstream infections frequently trigger decompensation and mortality, yet the global distribution of antimicrobial resistance (AMR) in these infections is poorly characterized. MethodsWe performed a systematic review and meta-analysis of studies reporting AMR in non-bloodstream bacterial infections among patients with cirrhosis. PubMed, Embase, and Web of Science were searched up to 15 September 2025. Pooled prevalence estimates of multidrug-resistant (MDR) and key resistant pathogens were calculated using random-effects models. Subgroup analyses were performed by country income, continent, and bacterial species. ResultsThirty-one studies including 3,162 infections were analysed. Spontaneous bacterial peritonitis predominated (79%), followed by colonisation (12%) and urinary tract infections (7%). Gram-negative bacteria accounted for 60% of infections (Escherichia coli 29%, Klebsiella pneumoniae 11%), while Gram-positive pathogens represented 39% (Enterococcus spp. 14%, Staphylococcus aureus 6%). Overall pooled MDR prevalence was 29%, with higher burdens in lower-middle-income countries (MDR 47% vs. 22-41%; ESBL 24% vs. 10%; VRE 21% vs. 3%; CRE 32% vs. 1%). Genotypic data identified 436 resistance genes with marked continental differences. ConclusionCirrhosis-associated non-bloodstream infections are dominated by Gram-negative bacteria and show high MDR, particularly in lower-middle-income countries. These findings highlight the need for integrated phenotypic and genomic surveillance of resistance patterns in these settings to guide empiric therapy.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.