Optimization of Antimicrobial Prophylaxis in Patients Undergoing Hematopoietic Stem Cell Transplantation Utilizing an Absolute Neutrophil Count Driven Approach

Background Hematopoietic stem cell transplantation (HSCT) recipients often develop profound neutropenia and are ultimately at an increased risk of developing life-threatening infections peri-engraftment. There is discordance in the current guidelines for antibacterial prophylaxis (AP) in patients undergoing HSCT. While ASTCT, formerly ASBMT, guidelines suggest beginning antibacterial prophylaxis on the day of stem cell infusion, NCCN and ASCO/IDSA guidelines recommend beginning at time of neutropenia. These recommendations for timing of antimicrobials are unsubstantiated by any comparative analyses. Methods A retrospective study was conducted across two cohorts over a two-year period (November 2016-2018). The pre-intervention cohort called for initiation of AP on Day -1 while the post-intervention cohort initiated AP when patients’ absolute neutrophil count (ANC) was ≤500 cells/mm3. The preferred AP agent used was levofloxacin, which was discontinued in both cohorts when patients attained a post-nadir ANC of ≥500 cells/mm3. The primary outcome was frequency of febrile occurrences, defined as a temperature of ≥38°C. Secondary outcomes included days of AP, positive blood cultures, all-cause mortality, length of stay (LOS), acute graft-versus-host disease (GvHD) through Day +100, and Clostridiodies difficile infection rates. Patients were excluded if they received a haploidentical transplant or inappropriately initiated AP per cohort approach. Results A total of 248 patients were included in the final analysis with 130 and 118 patients in the pre- and post-intervention cohorts, respectively. The majority of patients included were male (60%) and autologous transplant recipients (83%), with no clinically significant difference in major patient characteristics between both cohorts. There were lower rates of fever occurrences for autologous HSCT recipients (83 pre- vs 69% post-intervention; p=0.019), but no difference in allogeneic (64 vs 73%; p=0.542) HSCT patients. A significant reduction in the mean AP days per patient was identified (10.3 vs 4.9 days; p Conclusion Delaying AP until severe neutropenia showed a reduction in primary outcome of fever in autologous patients, and no difference in allogeneic patients or other significant clinical outcomes regarding patient safety. This approach is associated with a drastic reduction in antimicrobial exposure.