Time to initial glycopeptide therapy and 30-day mortality in methicillin-resistant Staphylococcus aureus bacteremia: a retrospective cohort study

TL;DR

This study found that the time to start glycopeptide therapy for MRSA bacteremia does not significantly affect 30-day mortality, even when adjusting for risk factors.

Contribution

The study introduces a novel approach using landmark analyses to adjust for immortal-time bias in assessing the timing of antibiotic therapy.

Findings

Glycopeptide therapy within 3, 6, 12, 24, 48, or 72 hours was not significantly associated with mortality after adjusting for covariates.

Metastatic solid tumor, initial septic shock, pneumonia, and unknown focus were independent risk factors for death.

A modest delay in AAT may have a slight effect on outcomes in patients without septic shock or other risk factors.

Abstract

The optimal time cut-off for appropriate antibiotic therapy (AAT) in methicillin-resistant Staphylococcus aureus (MRSA) bacteremia remains uncertain. We assessed the effects of time to AAT on 30-day in-hospital mortality in patients with MRSA bacteremia who received glycopeptides as initial therapy, and applied landmark analyses to adjust for immortal-time bias.

We conducted a retrospective cohort study of adults with MRSA bacteremia who received an initial course of glycopeptide therapy at two university-affiliated hospitals between 2018 and 2023. Multivariable logistic regression was used to identify covariates. These covariates were then included in six separate landmark models that assessed the effect of AAT within pre-specified cut-off time of 3, 6, 12, 24, 48, and 72 h after the index blood culture collection.

Among 220 patients, osteoarticular infections were the focus of bacteremia in 25.5%, and septic shock occurred as a complication in 28.2%; 30-day in-hospital mortality was 24.1%. Vancomycin was administered as initial therapy in 63.2% of patients, and the median time to glycopeptide initiation was 36 h (interquartile range, 19–67.5). Metastatic solid tumor, initial septic shock, pneumonia, and unknown focus were independent risk factors for death according to multivariable analysis. After covariate adjustment, glycopeptide therapy administered within any of the six pre-defined cut-off time was not significantly associated with mortality.

No significant time cut-off for glycopeptide initiation was associated with increased mortality. In patients without septic shock or other mortality-related risk factors, a modest delay in AAT may have slight effect on outcomes.

The online version contains supplementary material available at 10.1186/s12879-025-12040-9.

Figures