The Impact of Peri-Transplant Antibiotic Exposure on Outcomes in Children Undergoing Allogeneic Hematopoietic Stem Cell Transplantation

Background Survival after allogeneic hematopoietic stem cell transplantation (ASCT) has significantly improved with early recognition of sepsis and prompt empiric antibiotic (AB) therapy during periods of neutropenia. However, empiric AB treatment and prophylaxis practices have increased the duration of AB exposure in ASCT patients. Recent studies in adult ASCT patients have shown that perturbations in intestinal microbiome diversity are associated with the development of acute graft vs. host disease (aGVHD). Furthermore, even brief exposure to broad-spectrum ABs in the early phase of transplant is an independent risk factor for 5-year transplant-related mortality (TRM) in adults. However, the impact of AB exposure in the pediatric ASCT population is not understood. Objective To describe the association between AB exposure during the peri-transplant period (day-28 through day+28) and the development of aGVHD and/or TRM within the first year in children undergoing ASCT. Methods This is a retrospective cohort study of children who underwent ASCT at Boston Children's Hospital between 1/1/2012 and 12/31/2016 (n=296). We collected demographics, comorbidities, transplant information, infections, ABs, and clinical outcome data. The primary exposure is overall AB exposure from day -28 through day +28 measured as cumulative days of therapy (DOT) where each individual AB received within a 24-hour window counts as one DOT. This was further divided into cumulative AB exposure within 2-week periods during this time frame. The primary outcome is the development of aGVHD and/or TRM within the first year post-transplant. Results Of the 296 patients included in the analysis, 20.9% developed aGVHD and 9.46% developed TRM at one year post-transplant (overall mortality = 21.6%). The median cumulative AB DOT during the period from day -28 to day +28 of transplant was 63.3 days [IQR 45.5-85.5]. The top five ABs received, in decreasing order of frequency, were: trimethorprim/sulfamethoxazole, piperacillin/tazobactam, ciprofloxacin, meropenem, and gentamicin. In a multivariate model, patients who received > 60 DOT of ABs during the 57 days peri-transplant had higher odds of aGVHD and/or TRM (OR = 2.20, 95% CI: 1.21 to 4.02; P = 0.01). Exposure to more than 14 DOT of any AB in the first 2-weeks post-ASCT (day 0 to day +14) had significant risk of aGVHD and/or TRM in the first year compared to other 2-weeks periods in the 57-days peri-transplant (OR = 2.28, 95% CI: 1.03 to 5.02; P= 0.042). Conclusion Our preliminary analysis indicates that high AB exposure during the peri-transplant period in children may be associated with higher risk of development of aGVHD and/or TRM in the first year post-transplant. This study prompts future investigation of the mechanistic relationship between the gut microbiome and pediatric ASCT outcomes and underscores the importance of antimicrobial stewardship in this population.