Emulating a Target Trial to Evaluate the Impact of Empiric Antimicrobial Strategies on Mortality in Enterococcal Bloodstream Infections

Background: Enterococcal bloodstream infections (BSI) caused by Enterococcus faecalis and Enterococcus faecium pose substantial clinical and public health challenges, especially amid rising vancomycin resistance (VRE). Empiric regimens initiated before pathogen identification may lead to suboptimal coverage. We aimed to estimate the causal effect of two empiric strategies (immediate VRE coverage (daptomycin or linezolid) versus initial vancomycin or ampicillin with the option to escalate) by emulating a target trial using electronic health records. Methods: We identified adults (18+ years) with positive blood cultures for E. faecalis or E. faecium between January 2017 and January 2023. Patients had data compatible with : (1) immediate assignment to VRE coverage on day 0, or (2) initial VSE active therapy on day 0 with potential escalation within three days if clinical deterioration (e.g., ICU admission) or new microbiologic data warranted. We adjusted for baseline confounders (age, sex, Charlson comorbidity index, prior antibiotic use, immunosuppression) and for time varying factors (change in severity) using inverse probability weighting (IPW). We estimated both intention to treat (ITT) and per protocol effects on 30 day all cause mortality; sensitivity analyses assessed 10 and 15 day mortality for early outcome detection. Results: Among 799 patients, 45 (5.6%) received VRE targeted therapy at baseline; the remainder started on VSE active agents. The ITT analysis suggested that universal immediate VRE coverage would reduce 30 day mortality by 1.35 percentage points (95% CI: -7.2% to 7.9%), while the per protocol analysis yielded a risk difference of -3.76 points (95% CI: -14.9% to 20.4%). Sensitivity analyses of early mortality showed similarly null estimated. Conclusions: These results indicate clinical equipoise between immediate VRE coverage and a standard approach with potential escalation, with no clear survival benefit from empirical broad spectrum therapy in this setting. Further studies, prefereably randomized trials or emulation of them with larger datasets, remain necessary to better understand potential benefits in changing empiric treatment protocols for enterococcal BSI.