Cost Analysis of Vasopressin Use in Septic Shock and Predictive Performance of GOSSIS in ICU Mortality

Background: Vasopressin is a commonly used adjunctive therapy in the management of septic shock. While vasopressin may provide clinical benefits, at times its acquisition cost has risen to being substantially more expensive than other second-line vasopressors such as epinephrine, phenylephrine, or dopamine. Additionally, mortality prediction tools such as the Global Open-Source Severity of Illness Score (GOSSIS) have emerged as a valuable resource to guide resource allocation in critical care. This study aimed to evaluate the healthcare costs associated with vasopressin use and assess the predictive performance of GOSSIS in patients with septic shock. Methods: We conducted a multi-center retrospective cohort study of adult ICU patients with septic shock treated throughout the UPMC Health System from January 2021 to January 2022. Patients who received norepinephrine as a first-line vasopressor followed by either vasopressin or an alternative vasopressor were included. Generalized linear models with a log link and gamma distribution were used to compare adjusted total healthcare costs. Recycled predictions were used to estimate marginal cost differences. GOSSIS performance was evaluated by comparing predicted versus observed ICU mortality across risk groups. Results: Among 3,898 eligible patients, 1,060 received vasopressin and 2,838 received alternative second-line vasopressors. Vasopressin use was associated with significantly higher adjusted hospital costs (cost ratio = 1.14, 95% CI: 1.08, 1.20). The estimated marginal cost difference was $5,887 per patient. Drug costs accounted for a greater proportion of total hospitalization costs among vasopressin users (20% vs. 12%). GOSSIS demonstrated strong calibration, with predicted ICU mortality closely aligning with observed rates across all severity strata. Conclusions: Vasopressin use in septic shock was associated with a 14% increase in hospitalization costs compared to alternative second-line vasopressors. The GOSSIS demonstrated strong predictive validity for ICU mortality, supporting its potential use in guiding resource allocation decisions in critically ill populations.