BACKGROUND: Clostridioides difficile infection (CDI) and gastrointestinal acute graft-versus-host disease (GI-GVHD) frequently co-occur after allogeneic hematopoietic cell transplantation (allo-HCT), and a bidirectional relationship has been proposed. The prior CIBMTR analysis associated antecedent GI-GVHD with CDI but modeled GI-GVHD as a fixed exposure and could not test whether CDI predicts subsequent GI-GVHD. METHODS: We analyzed a publicly available CIBMTR dataset of patients undergoing first allo-HCT for AML, ALL, or MDS (2013-2018), each complication was modeled as a time-dependent exposure using cause-specific Cox regression, with death and relapse treated as competing events, adjusting for donor type, disease, conditioning intensity, and (reverse direction) GVHD prophylaxis type. Sensitivity analyses included restriction to adults and exclusion of CDI diagnosed before post-transplant cyclophosphamide PTCy. RESULTS: CDI within 100 days occurred in 777 patients (11%). CDI did not predict subsequent GI-GVHD (HR 1.15; 95% CI 0.90-1.48; p = 0.26). GI-GVHD predicted subsequent CDI (HR 1.54; 95% CI 1.08-2.19; p = 0.016), unchanged after adjustment for prophylaxis type and stronger in adults (HR 1.73; 95% CI 1.18-2.55). An apparent PTCy-CDI association (HR 1.38; 95% CI 1.08-1.75; p = 0.009), was driven by CDI with onset by Day 4, before PTCy, and was no longer significant once excluded (HR 1.11; 95% CI 0.80-1.52). CONCLUSIONS: When event timing and competing risks are respected, the GI-GVHD predicts subsequent CDI whereas the reverse is not observed. GI-GVHD identifies patients at elevated risk who are candidates for heightened surveillance. The disappearance of the PTCy signal once drug timing was respected illustrates how event ordering can misattribute risk.
No takes yet. Share an insight, caveat, or question.
Alzibdeh et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: