C54-06 Integrating a Novel Cellular Host Response Test for Early Sepsis Detection: A Quality Improvement Initiative in the Emergency Department

Abstract Rationale Sepsis, life-threatening organ dysfunction caused by a dysregulated host response to infection(1), remains a leading cause of morbidity, mortality and healthcare utilization(2). Early recognition and intervention are critical, yet diagnosis often relies on nonspecific findings and inconsistent clinical judgement(3). With the high mortality associated with delayed intervention, there is a tendency to overtreat(4), leading to increased length of stay, hospital mortality, and frequency of Clostridoides difficile infections(5). Cytovale® IntelliSep® test is an FDA cleared cellular host response (CHR) test that delivers results within 8 minutes, stratifying sepsis likelihood into three bands (Band 1-low to Band 3-high). We present a Quality Improvement (QI) Study that focuses on the clinical utility of Band 1 results in ruling out sepsis. Methods Patients presenting to a community hospital over 3 months (06/17/2025 to 09/16/2025) were included. Those suspected of infection - identified using Triage Warning Alerts (infection-associated chief complaint, nurse suspicion, abnormal vitals, EMR alert or provider’s discretion) - were evaluated. Patients with Band 1 results were compared to a baseline group of 195 patients (observed from 04/30/2025 to 05/15/2025) whose clinicians were blinded to the test results. Sepsis was identified using ICD-10 codes. Populations not approved by FDA for the test (age18, hematologic malignancies, on chemotherapy, etc) were excluded from the study. The impact of the test was assessed by comparing ED blood culture and bacteremia rates, antibiotic use and disposition outcomes. Results Of 12444 ER encounters, CHR test was used in 1433. Performance characteristics over the three-month period revealed AUC 0.85, sensitivity 91.5%, Specificity 90.6%, NPV 97.5%, and PPV 52.1%. Blood culture orders for Band 1 patients decreased from 34.5% to 16.5%. Broad-spectrum antibiotic utilization declined from 20% to 15.8%, although overall IV antibiotic utilization was similar (45.5% vs 43.1%). Sepsis prevalence (2.7% vs 2.5%), ED discharge rate (37.3% vs 37.9%), hospital admission (observation 10.0% vs 10.0%; inpatient 48.2% vs 47.4%) and 3-day return for sepsis (none) remained largely unchanged. Conclusion Incorporating CHR testing helped prioritize resources in this community hospital. Despite a steep decrease in antibiotic and blood culture orders, sepsis prevalence and discharge rates remained stable. Results demonstrated excellent negative predictive values for ruling out sepsis and minimizing unnecessary interventions in low likelihood patients, while preserving diagnostic accuracy for the truly ill. Our study supports incorporating CHR testing as another tool into sepsis evaluation and leveraging its rapid turnaround time to guide clinical decision-making in the acute care setting. This abstract is funded by: N/A