Impact of Time to Antibiotics on In-hospital Mortality in Neutropenic Sepsis: A Prospective Multicenter Cohort Study.

Kim, Dongju; Kim, Youn-Jung; Ryoo, Seung Mok; Kim, Won Young; Korean Sepsis Alliance (KSA) Investigators · Crit Care Med · 2026

prospective_cohort · Level II

Where this comes from

Abstract

Neutropenic sepsis is a high-risk form of sepsis associated with rapid deterioration and high mortality. Although early antibiotic administration is recommended, the benefit of shorter time to antibiotics (TTA) remains uncertain. This study aimed to determine the relationship between TTA and in-hospital mortality for patients with neutropenic sepsis and to identify specific phenotypes most vulnerable to TTA delays. Prospective, multicenter observational cohort study. Twenty tertiary or university-affiliated hospitals in South Korea. In this prospective multicenter cohort study, we analyzed 942 patients with sepsis and neutropenia. None. TTA was categorized into three groups: less than 1 hour (early), 1-3 hours (intermediate), and greater than or equal to 3 hours (delayed). We used inverse probability of treatment weighting (IPTW) with logistic regression to evaluate the association between TTA and in-hospital mortality and a causal forest model to identify characteristics associated with the effect of longer TTA. Overall, 211, 531, and 200 patients belonged to the early, intermediate, and delayed groups, respectively. In IPTW-weighted logistic regression, odds ratio (OR) for in-hospital mortality was significantly higher in the delayed (OR: 1.50; 95% CI, 1.22-1.85; p < 0.001) and intermediate groups (OR: 1.26; 95% CI, 1.04-1.53; p = 0.021) than in the early group. The association between longer TTA and mortality was stronger among patients with septic shock and among those with hematologic malignancy in both models. The causal forest further identified elevated lactate as a potential effect modifier, indicating greater estimated benefit from earlier TTA; however, interaction tests in the weighted logistic model were not significant. Delayed antibiotic administration was associated with increased in-hospital mortality among patients with neutropenic sepsis. The impact on in-hospital mortality is heterogeneous, varying by patient characteristics, particularly septic shock and hematologic malignancy.