Early Ultrasound and Time to Antibiotic Therapy in Emergency Department Patients With Suspected Infection: A Propensity-Weighted Analysis of 15,660 Encounters

Technical Summary: Early Ultrasound and Time to Antibiotic Therapy in ED Patients With Suspected Infection

Problem Statement
Timely antimicrobial therapy is a cornerstone of emergency care for patients with suspected infection, particularly in the context of sepsis management. However, diagnostic uncertainty in the Emergency Department (ED)—driven by nonspecific symptoms and competing diagnoses—often delays therapeutic decisions. While ultrasound is increasingly integrated into emergency medicine for its rapid, radiation-free diagnostic capabilities, there is limited evidence regarding whether early ultrasound utilization translates into more timely delivery of definitive therapies, such as antibiotic administration. This study addresses the gap between diagnostic acceleration and process-of-care outcomes, specifically investigating if early ultrasound is associated with reduced time-to-treatment in hospitalized ED patients with suspected infection.

Methodology
The authors conducted a retrospective observational cohort study using the MIMIC-IV-ED database linked with hospital, microbiology, medication, radiology, and intensive care datasets from a tertiary-care academic medical center.

• Study Population: The analysis included adult ED encounters resulting in hospital admission. "Suspected infection" was defined using an operational definition adapted from the Sepsis-3 framework, requiring both systemic antibiotic administration within 24 hours of ED arrival and microbiological culture sampling within ±24 hours of arrival.
• Exposure: The primary exposure was "early ultrasound," defined as any documented ultrasound examination performed within 2 hours of ED arrival.
• Outcomes: The primary outcome was the administration of systemic antibiotics within 6 hours of ED arrival. Secondary outcomes included antibiotic administration within 3 hours, time to first antibiotic administration, vasopressor use, ICU admission, hospital length of stay, and hospital mortality.
• Statistical Analysis: To mitigate confounding from non-random ultrasound utilization, the study employed overlap-weighted propensity-score analysis. Covariates included demographics, triage acuity, vital signs, shock-triage status, and ED length of stay. Balance was assessed using standardized mean differences (SMDs). Binary outcomes were analyzed using overlap-weighted logistic regression, continuous outcomes via linear regression, and time-to-event data using a Cox proportional hazards model.

Key Contributions and Results
From a screened cohort of 425,087 ED encounters, 15,660 met the suspected-infection criteria and had complete covariate data. Early ultrasound was performed in 5.1% (791) of these encounters.

• Antibiotic Timeliness: Early ultrasound was significantly associated with improved timeliness of antibiotic administration.
  • Odds Ratio: Patients receiving early ultrasound had 1.46 times higher odds of receiving antibiotics within 6 hours (95% CI 1.13–1.87; p = 0.003).
  • Time Reduction: The mean time to first antibiotic administration was reduced by 0.27 hours (approximately 16 minutes) in the early ultrasound group (95% CI −0.44 to −0.11; p = 0.001).
  • Time-to-Event: In the weighted Cox analysis, early ultrasound was associated with earlier antibiotic administration (Hazard Ratio 1.12; 95% CI 1.04–1.20; p = 0.002).
• Clinical Outcomes: Despite the improvement in process metrics, early ultrasound showed no significant association with:
  • Hospital mortality (OR 1.04; p = 0.804).
  • ICU admission (various time windows).
  • Vasopressor use within 6 hours.
• Length of Stay: Patients receiving early ultrasound had a statistically longer hospital length of stay (mean difference +1.19 days; p = 0.002), though the authors note this may reflect residual confounding by diagnostic complexity rather than harm.
• Subgroup and Sensitivity Analyses: The association with antibiotic timeliness was stronger in patients with positive shock-triage status. Sensitivity analyses using 1-hour and 6-hour exposure windows yielded broadly consistent results, though effect sizes varied.

Significance and Claims
The paper concludes that early ultrasound functions primarily as a diagnostic process accelerator within ED infection care pathways. The authors posit that the modest reduction in time to antibiotics (approx. 16 minutes) likely results from ultrasound reducing diagnostic uncertainty, thereby increasing clinician confidence to initiate antimicrobial therapy.

The study explicitly states that these process improvements did not translate into measurable reductions in hospital mortality or ICU utilization in this broad cohort. The authors argue that this does not diminish the clinical value of ultrasound; rather, it suggests that a 16-minute acceleration in treatment, while beneficial for workflow efficiency, may be insufficient to independently alter mortality outcomes in a heterogeneous population of suspected infection cases. The findings support the integration of early ultrasound to improve diagnostic efficiency but call for prospective studies to determine if structured early-ultrasound pathways can improve source identification and patient-centered outcomes in higher-risk populations.