Hidden Toll of CRE Colonization: Tripled Three-Year Mortality Risk and Increased Bloodstream Infection Burden After Allo-HSCT—A Propensity-Adjusted Study

Yuan Zhang,1 Rong Wang,1 Guoqing Lyu,1– 3 Lihua Wang,1 Xiankai Liu,1 Linlin Tian,1 Yan Huang,1– 3 Jiesen Kou,1 Sun Wu,1– 3 Weijie Cao4 1Department of Hematology, the First Affiliated Hospital of Xinxiang Medical University, Weihui, Henan Province, 453100, People’s Republic of China; 2Key Laboratory for Leukemia Molecular Diagnosis and Treatment in Xinxiang City, Weihui, Henan Province, 453100, People’s Republic of China; 3Key Laboratory for Lymphoma Molecular Diagnosis and Treatment in Xinxiang City, Weihui, Henan Province, 453100, People’s Republic of China; 4Department of Hematology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, People’s Republic of ChinaCorrespondence: Sun Wu, Department of Hematology, the First Affiliated Hospital of Xinxiang Medical University, Weihui, Henan Province, 453100, People’s Republic of China, Email [email protected] Weijie Cao, Department of Hematology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, People’s Republic of China, Email [email protected]: Carbapenem-resistant Enterobacteriaceae (CRE) colonization poses a major threat to the success of allogeneic hematopoietic stem cell transplantation (allo-HSCT). However, evidence on its long-term impact and the role of decolonization strategies remains limited.Methods: This study included 240 allo-HSCT recipients prospectively screened for pre-transplant rectal CRE colonization (colonized n=80; non-colonized n=160; 1:2 matching). Inverse probability of treatment weighting (IPTW) was applied to achieve covariate balance (standardized mean difference < 0.10). Key outcomes included CRE bloodstream infection (BSI) incidence, hematopoietic engraftment time, graft-versus-host disease (GVHD) onset, and 3-year overall survival (OS), analyzed using Kaplan-Meier curves and Cox proportional hazards models. Strain and carbapenemase enzyme concordance between colonizing and infecting isolates were evaluated in BSI cases.Results: CRE BSI occurred almost exclusively in colonized patients (26.6%, 21/79), with 81.0% strain concordance (predominantly Escherichia coli and Klebsiella pneumoniae) and 77.8% enzyme profile concordance (mainly metallo-β-lactamase/NDM) between pre-transplant colonizing and post-transplant infecting isolates. An 81.0% concordance rate was observed between CRE strain types isolated from perianal colonization samples and those from bloodstream infections. BSI-attributable 100-day mortality was 34.8% (8/23). Colonized patients exhibited delayed engraftment (neutrophils: 14 vs 13 days, P=0.044; platelets: 15 vs 14 days, P=0.014) and earlier chronic GVHD onset (150 vs 235 days, P=0.004), with comparable GVHD incidence. Both unweighted and IPTW-adjusted 3-year OS were markedly lower in colonized patients (62.5%/58% vs 85.0%/83%; HR 3.49, 95% CI 1.91– 6.38, P< 0.001).Conclusion: Pre-transplant CRE colonization is strongly associated with poorer allo-HSCT outcomes, including increased CRE BSI incidence and 100-day mortality, delayed engraftment, accelerated cGVHD onset, and approximately threefold higher 3-year mortality risk. These findings position CRE colonization as a potentially modifiable driver of both early and long-term morbidity, highlighting the critical need for routine screening and targeted decolonization strategies to improve survival in this high-risk population.Keywords: carbapenem-resistant Enterobacteriaceae, allogeneic hematopoietic stem cell transplantation, graft-versus-host-disease, overall survival: non-relapse mortality