Abstract Background Cytomegalovirus (CMV) causes substantial morbidity and mortality after hematopoietic stem cell transplantation (HCT), and antibiotic exposure after HCT is increasingly recognized as a risk for negative clinical outcomes. Recent studies identified an association between receipt of antibiotics with an anaerobic spectrum of activity and infection with cytomegalovirus (CMV) after HCT in adults. We sought to evaluate the association between anaerobic antibiotic exposure and CMV viremia in a well-characterized cohort of pediatric HCT recipients. Method We conducted a single-center retrospective cohort study of HCT recipients <18 years of age who underwent HCT at Duke University between 2000 and 2016. Routine pre-emptive surveillance testing for CMV was performed; CMV viremia was defined as a positive whole blood hybrid capture CMV DNA assay or quantitative plasma CMV PCR. We used log-logistic regression to evaluate associations between anaerobic antibiotic exposure (vancomycin, metronidazole, meropenem, piperacillin-tazobactam) and CMV viremia-free survival at 100 days after HCT. The first model considered anaerobic antibiotic exposure in the first 14 days after HCT as a binary variable. The second model considered anaerobic antibiotic exposure in the first 100 days as a time-varying variable; once a subject was exposed to a specific antibiotic, they were considered exposed for the remainder of the study period. Both models adjusted for year of HCT, age, race, HCT type and donor cell source, underlying diagnosis, conditioning regimen, CMV serostatus, and treatment for graft-versus-host-disease. Results Among 966 children, median (interquartile range) age was 6.5 (3.1, 11.5) years. Hematologic malignancy was the indication for transplantation in 417 (43%) subjects. Of 770 allogeneic HCT, 566 (74%) were from an umbilical cord blood donor and 204 (26%) were from a bone marrow donor. The majority of HCT were recipient CMV-seronegative and donor CMV-seronegative (R-/D-; 537, 56%), with 24 (2%) R-/D+, 270 (28%) R+/D-, and 135 (14%) R+/D+ transplants. Within the first 100 days after HCT, 692 (72%) children received vancomycin, 363 (38%) received metronidazole, 231 (24%) received meropenem, and 121 (13%) received piperacillin-tazobactam. Exposure to vancomycin (OR=0.86; 95% CI: 0.59, 1.26), metronidazole (OR=0.76; 95% CI: 0.53, 1.10), meropenem (OR=1.19; 95% CI: 0.74, 1.90), or piperacillin-tazobactam (OR=1.03; 95% CI: 0.55, 1.92) in the first 14 days after HCT was not associated with the risk of CMV viremia-free survival at 100 days. Similarly, we found no increased risk of CMV viremia with anaerobic antibiotic exposure in the first 100 days after HCT (vancomycin OR=0.90; 95% CI: 0.60, 1.34; metronidazole OR=0.80; 95% CI: 0.53, 1.21; meropenem OR=0.98; 95% CI: 0.60, 1.62; piperacillin-tazobactam OR=0.69; 95% CI: 0.38, 1.23). Conclusion In this study of pediatric HCT recipients, we did not find evidence of an association between anaerobic antibiotic exposures and CMV viremia risk. However, future pediatric and adult studies are warranted to evaluate this previously reported association in other populations, including studies evaluating for a possible role of the gut microbiome in modifying CMV viremia risk after HCT.
Read full abstract