STUDY OBJECTIVE

Debate exists about the mortality benefit of administering antibiotics within either 1 or 3 hours of sepsis onset. We performed this meta-analysis to analyze the effect of immediate (0 to 1 hour after onset) versus early (1 to 3 hours after onset) antibiotics on mortality in patients with severe sepsis or septic shock.

METHODS

This review was consistent with the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Searched databases included PubMed, EMBASE, Web of Science, and Cochrane Library, as well as gray literature. Included studies were conducted with consecutive adults with severe sepsis or septic shock who received antibiotics within each period and provided mortality data. Data were extracted by 2 independent reviewers and pooled with random effects. Two authors independently assessed quality of evidence across all studies with Cochrane's Grading of Recommendations Assessment, Development and Evaluation methodology and risk of bias within each study, using the Newcastle-Ottawa Scale.

RESULTS

Thirteen studies were included: 5 prospective longitudinal and 8 retrospective cohort ones. Three studies (23%) had a high risk of bias (Newcastle-Ottawa Scale). Overall, quality of evidence across all studies (Grading of Recommendations Assessment, Development and Evaluation) was low. Pooling of data (33,863 subjects) showed no difference in mortality between patients receiving antibiotics in immediate versus early periods (odds ratio 1.09; 95% confidence interval 0.98 to 1.21). Analysis of severe sepsis studies (8,595 subjects) found higher mortality in immediate versus early periods (odds ratio 1.29; 95% confidence interval 1.09 to 1.53).

CONCLUSION

We found no difference in mortality between immediate and early antibiotics across all patients. Although the quality of evidence across studies was low, these findings do not support a mortality benefit for immediate compared with early antibiotics across all patients with sepsis.