Purpose:

The goal of this review was to synthesize existing evidence regarding outcomes (mortality) for patients who present to the emergency department, are administered antibiotics immediately (within 1 hour) or later (>1 hour), and are diagnosed with sepsis.

Methods:

A search of PubMed, EMBASE, Cochrane Central Register of Controlled Trials, and CINAHL, using the MeSH descriptors “sepsis,” “systemic inflammatory response syndrome,” “mortality,” “emergency,” and “antibiotics,” was performed to identify studies reporting time to antibiotic administration and mortality outcome in patients with sepsis. The included studies (published in English between 1990 and 2016) listed patient mortality based on time to antibiotic administration. Studies were evaluated for methodologic quality, and data were extracted by using a data extraction form tailored to this study. From an initial pool of 582 potentially relevant studies, 11 studies met our inclusion criteria, 10 of which had quantitative data for meta-analysis. Three different models (a random effects model, a bias-adjusted quality-effects [synthetic bias] model, and an inverse variance heterogeneity model) were used to perform the meta-analysis.

Findings:

The pooled results suggest a significant 33% reduction in mortality odds for immediate (within 1 hour) compared with later (>1 hour) antibiotic administration (OR, 0.67 [95% CI, 0.59–0.75]) in patients with sepsis.

Implications:

Immediate antibiotic administration (<1 hour) seemed to reduce patient mortality. There was some minor negative asymmetry suggesting that the evidence may be biased toward the direction of effect. Nevertheless, this study provides strong evidence for early, comprehensive, sepsis management in the emergency department.