2022. Modeling Improved Patient Management and Hospital Savings with SeptiCyte LAB in the Diagnosis of Sepsis at ICU Admission

Abstract

BackgroundThe ability to accurately diagnose sepsis at ICU admission is key to effective clinical management, patient safety, and efficient hospital resource utilization. Most tests used for sepsis diagnosis, including pathogen detection and host-based biomarker approaches, are lacking in either sensitivity or specificity, resulting in a clinical assumption of sepsis, patient overtreatment with antimicrobials, and hospital or ICU admission as a precaution. SeptiCyte LAB is the only FDA-cleared test that analyzes the patient’s own host transcriptional response in systemically inflamed patients, to discriminate between infectious and non-infectious underlying causes.MethodsThe SeptiCyte LAB test was validated in a multi-site prospective trial in the ICU (N = 447). Test performance and length of stay in both the ICU and hospital were recorded. Length of stay associated with initial diagnoses was directly recorded, and a model was created in which costs/savings due to the added deployment of the SeptiCyte LAB assay, in comparison to the standard of care, was projected. Assumptions for the timing of antimicrobial de-escalation and hospital stay costs were drawn from published literature. Assumptions for changes in clinical management were based on required minimum negative and positive predictive values.ResultsThe estimated savings due to reduced length of stay per patient, as a consequence of the information generated by SeptiCyte LAB was found to be approximately $1,600 in the ICU, and an estimated $900 on the ward. Matched procalcitonin results from the same trial and exclusively used for de-escalation as specified by the manufacturer resulted in per-patient savings of $80.ConclusionModeling of a rapid and accurate diagnosis sepsis diagnostic (SeptiCyte LAB) at ICU admission projects significant savings and improvements in patient management.Disclosures L. Mchugh, Immunexpress: Employee, Salary.