Bloodstream infections in allogeneic haematopoietic cell recipients from the Swiss Transplant Cohort Study

Abstract

Empiric antibiotic treatment for bloodstream infection (BSI) after haematopoietic cell transplantation (HCT) in the era of antimicrobial resistance is challenging. We conducted a nested project in the Swiss Transplant Cohort Study 09/2009 to 10/2018 assessing BSI incidence, pathogen distribution and antimicrobial resistance. Cox proportional hazards models were fitted to detect factors associated with BSI and outcome. We included 1 364 adults (61.1% male) receiving 1 432 HCTs complicated by 676 BSIs in 451 (33%) patients. Cumulative incidence of BSI was 20.5% (95%CI 18.4 – 22.7) until day 100 after the first HCT. Gram-positive pathogens, mostly coagulase-negative staphylococci, dominated (454, 58.1%). Production of extended-spectrum beta-lactamases and quinolone-resistance were observed in 41/195 (21%) and 81/195 (41.5%) of Enterobacterales, respectively. Carbapenem-resistance was documented in 10/49 (20.4%) of Pseudomonas aeruginosa isolates. Multivariable cox regression analysis within the first-year post-transplant identified male gender, comorbidities, myeloablative regimen, acute graft-versus-host disease grade III/IV and haematological progression/relapse after HCT as independent risk factors for BSI. One-year mortality was 26.6% with BSI modeled as a time-dependent covariable independent risk factor for mortality. To conclude, BSI remains a frequent complication after HCT, with stable rates of antibiotic resistance in BSI-causing pathogens in Switzerland. BSI was a major contributor to mortality.