Effects of Inappropriate Administration of Empirical Antibiotics on Mortality in Adults With Bacteraemia: Systematic Review and Meta-Analysis

Introduction: Bloodstream infections are associated with high mortality rates and contribute substantially to healthcare costs, but a consensus on the prognostic benefits of appropriate empirical antimicrobial therapy (EAT) for bacteraemia is lacking. Methods: We performed a systematic search of the PubMed, Cochrane Library, and Embase databases through July 2021. Studies comparing the mortality rates of patients receiving appropriate and inappropriate EAT were considered eligible. The quality of the included studies was assessed using Joanna Briggs Institute checklists. Results: We ultimately assessed 198 studies of 89,962 total patients. The pooled odds ratio (OR) for the prognostic impacts of inappropriate EAT was 2.06 (P < 0.001), and the funnel plot was symmetrically distributed. Among subgroups without between-study heterogeneity (I² = 0%), those of patients with severe sepsis and septic shock (OR, 2.14), Pitt bacteraemia scores of ≥4 (OR, 1.88), cirrhosis (OR, 2.56), older age (OR, 1.78), and community-onset/acquired Enterobacteriaceae bacteraemia infection (OR, 2.53) indicated a significant effect of inappropriate EAT on mortality. The pooled adjusted OR of 125 studies using multivariable analyses for the effects of inappropriate EAT on mortality was 2.02 (P < 0.001), and the subgroups with low heterogeneity (I² < 25%) exhibiting significant effects of inappropriate EAT were those of patients with vascular catheter infections (adjusted OR, 2.40), pneumonia (adjusted OR, 2.72), or Enterobacteriaceae bacteraemia (adjusted OR, 4.35). Notably, the pooled univariable and multivariable analyses were consistent in revealing the negligible impacts of inappropriate EAT on the subgroups of patients with urinary tract infections and Enterobacter bacteraemia. Conclusion: Although the current evidence is insufficient to demonstrate the benefits of prompt EAT in specific bacteraemic populations, we indicated that inappropriate EAT is associated with unfavorable mortality outcomes overall and in numerous subgroups. Prospective studies designed to test these specific populations are needed to ensure reliable conclusions. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/, identifier: CRD42021270274.