From the 2022 HVPA National Conference
George Carey MD (Yale Medical School), Jurgen Holleck MD, Ritujith Jayakrishnan MD, Kirsha Gordon PhD, Alyssa Grimshaw MSLIS, Craig Gunderson MD
Background
In United States (U.S.) hospitals only 0.3% of hospitalized patients have documented infections by methicillin-resistant Staphylococcus aureus (MRSA), yet 12% of patients receive at least one dose of vancomycin and vancomycin accounts for 25% of all antimicrobial use. The 2021 Surviving Sepsis Campaign strongly recommends that patients with sepsis at high risk for MRSA should be treated empirically using antibiotics active against MRSA but does not define high risk or quantify the benefit of empiric antibiotics for MRSA.
Objective
To estimate the benefit of active empiric antibiotics for patients found to have infections with MRSA and to develop a framework to guide empiric antibiotic use for MRSA.
Methods
A systematic literature search was conducted using Embase, Medline, PubMed, Web of Science, Cochrane, Scopus and Google Scholar from earliest entry through April 26, 2022. We included any study of patients hospitalized with culture positive MRSA infections that compared mortality rates depending on whether patients received effective empiric antibiotics or not. The primary outcome was adjusted odds ratio for mortality.
Results
We found 32 studies with documented MRSA infection and mortality, of which 17 reported adjusted analyses. The overall pooled mortality odds ratio was 0.73 (95% CI, 0.59-0.92), favoring effective empiric antibiotics. Eleven of the 17 studies were of patients with bacteremia (aOR 0.68, 95% CI 0.51-0.91), four were of patients with pneumonia (aOR 1.04, 95% CI 0.83-1.29), and two were a mix of infection types (aOR 0.51, 95% CI 0.39-0.66). Heterogeneity was moderate for studies of bacteremia (I2= 65%) and low for studies of pneumonia and mixed infection (I2= 0% for both). A framework was made to estimate the benefit for patients presenting with infection based on the estimated probability of MRSA infection, the average mortality by infection type, and the benefit of active empiric antibiotics based on our meta-analysis. The estimated absolute risk reduction in mortality was 0% for pneumonia, 0.045% for cellulitis and urinary tract infections, and 0.45-0.65% for sepsis.
Conclusions
Based on seventeen studies that reported adjusted odds ratio, the benefit for active empiric antimicrobials for patients found to have MRSA infection was 0.73 overall and 0.68 for bacteremia. We found no benefit for active empiric antimicrobials for MRSA pneumonia. For the three most common infections (pneumonia, cellulitis, and urinary tract infections), the absolute benefit of empiric vancomycin was less than 0.1%. Meaningful benefit of empiric antimicrobials against MRSA is likely to be limited to patients with approximately 30% mortality and 10% prevalence of MRSA, such as patients admitted to intensive care units with possible catheter-related bloodstream infections or endocarditis.
Clinical Implications
Contrary to common practice in the U.S., empiric antibiotics effective against MRSA are unlikely to confer a mortality benefit for most common infections.
Funding: none (PROSPERO: CRD42021289691)
