Fluoroquinolone Susceptibility and Comparative Outcomes of Gram-negative Bloodstream Infection in Pediatric Patients With Hematologic Malignancy.

BACKGROUND

Bacterial bloodstream infections (BSIs) cause morbidity and mortality in children with chemotherapy-associated neutropenia. While levofloxacin prophylaxis reduces BSI incidence, breakthrough infections occur in ~20%, often with fluoroquinolone non-susceptible (FQN-NS) gram-negative (GN) pathogens. The impact of FQN-NS on outcomes of GN BSI remains unknown. This study compares outcomes in pediatric patients with hematologic malignancies following FQN-NS vs. fluoroquinolone-susceptible (FQN-S) GN BSI.

METHODS

A single-center cohort of pediatric patients with hematologic malignancies and neutropenia-associated GN BSI between 2014 and 2023 was retrospectively collected. The exposure was dichotomized as FQN-NS vs. FQN-S. Primary outcomes were intensive care unit admission and 30-day all-cause mortality, assessed from the day before to 30 days after blood culture collection. Modified Poisson regression with inverse probability of treatment weighting (IPTW) was used to balance covariates including age, calendar year of diagnosis, time from malignancy onset, cumulative neutropenia, granulocyte colony-stimulating factor use, international status, relapse/refractory disease, and type of underlying hematologic malignancy. A subgroup analysis assessed effect modification by resistance to empiric therapy. Only 5 patients contributed multiple events; clustering adjustments were not made.

RESULTS

Among 119 GN BSI events, 82 (68.9%) were FQN-S and 37 (31.1%) FQN-NS. In IPTW-adjusted analyses, FQN-NS BSI was associated with higher 30-day all-cause mortality (aRR: 2.62, 95% CI: 0.39-17.64) and sepsis-related mortality (aRR: 2.60, 95% CI: 0.16-41.65), although confidence intervals were wide and not statistically significant. In a subgroup restricted to patients who received effective empiric therapy, estimates further reverted toward the null.

CONCLUSIONS

Fluoroquinolone non-susceptible GN BSI during neutropenia after chemotherapy was associated with increased 30-day all-cause and sepsis-related mortality in pediatric patients with hematologic malignancies, although not statistically significant. Effect modification by empiric therapy may have contributed to this association, suggesting that FQN-NS itself may not directly drive poor outcomes, but its impact appears mediated through resistance to empiric therapy, which is associated with worse outcomes.