# P-2129. Prolonged infusion of β-lactams for Enterobacterales bacteremia in high-risk neutropenic patients: Does it improve outcomes?

## TL;DR

This study found no improvement in outcomes for high-risk neutropenic patients with Enterobacterales bacteremia when using prolonged infusion of β-lactams.

## Contribution

The study is the first to investigate the clinical benefits of prolonged infusion of β-lactams in this specific patient group.

## Key findings

- Prolonged infusion did not reduce 30-day mortality in high-risk neutropenic patients with Enterobacterales bacteremia.
- Septic shock was a strong risk factor for 30-day mortality, while 7-day clinical response was protective.
- No significant differences were observed in clinical outcomes between prolonged and standard infusion methods.

## Abstract

No evidences have been reported on the clinical benefits of prolonged infusion (PI) of β-lactams in febrile neutropenic patients (FNP) with Enterobacterales bacteremia (EB).

A prospective observational multicenter study was conducted in 9 referral academic centers in Argentina between February 2019 and December 2024. The first episodes of EB in high-risk FNP who received appropriate empirical treatment (AET) with piperacillin-tazobactam (PT), cefepime (C), or meropenem (M) were included. Clinical, epidemiological, and outcome variables were compared in patients receiving either standard infusion (SI) or PI AET. A propensity score (PS) to balance baseline covariates was used. Adjusted conditional multivariate logistic regression analysis to PS was used to identify independent risk factors for 30-day mortality.

201 patients were included (116 SI and 85 PI). Median age was 49 years (IQR: 35-63), and the most common underlying diseases were acute leukemia (61.69%) and lymphoma (18.91%). The microorganisms most frequently observed in both groups were E. coli (58.71%) and Klebsiella spp. (32.84%). Bacteremia presented with a clinical source in 58.82% of PI cases vs. 55.17% of SI cases, p = 0.60. Hypotension and shock were present in PI vs. SI in 28.4% vs. 25%, p=0.60, and 17.65% vs. 14.66%, p=0.56, respectively. Intensive care unit admission and multiorgan failure occurred in PI vs. SI: 18.82% vs. 12.07%, p = 0.56, and 16.47% vs. 11.21%, p = 0.28, respectively. Seven-day

clinical response between IP and IB was 90.59% vs. 93.10%, p = 0.51. Thirty-day overall mortality and EB-related mortality were 14.12% vs. 10.34% (p = 0.41) and 8.24% vs. 3.45% (p = 0.20) for PI and SI, respectively. Septic shock was the risk factor associated with 30-day mortality (OR: 11.65, 95% CI, 3.66-36.91, p < 0.001), while 7-day clinical response was a protective factor for survival (OR: 0.06, 95% CI, 0.01-0.24, p < 0.001). PI did not correlate with impact on mortality (OR: 1.78, 95% CI: 0.54-5.79, p = 0.34).

In our high-risk FNP cohort with EB, AET with PI did not improve outcomes. However, randomized trials are needed to confirm the study findings.

Alberto Carena, n/a, Roche Diagnostics: Medical affairs lead