Risk Analytics Clinical Decision Support Decreases Duration of Vasoactive Infusions Following Pediatric Cardiac Surgery: A Multicenter Before and After Clinical Trial

Document Type

Journal Article

Publication Date

4-29-2025

Journal

Critical care medicine

DOI

10.1097/CCM.0000000000006682

Keywords

cardiac intensive care unit; cardiac surgery; clinical decision support system; inadequate delivery of oxygen index; inotropes and vasoactive medications; pediatrics

Abstract

OBJECTIVES: Our primary objective was to quantify the effect of a clinical decision support system (CDSS) informed by inadequate delivery of oxygen index (IDo2), an algorithm that estimates the probability of inadequate oxygen delivery on inotropes and vasoactive infusions (IVAI) postoperatively. DESIGN: Multicenter before-and-after study conducted between June 01, 2020, and December 31, 2022. SETTING: Three pediatric cardiac ICUs (CICUs) at tertiary academic children's hospitals. PATIENTS: Infants older than 36 weeks estimated gestational age and 2 kg following cardiac surgery with cardiopulmonary bypass, treated with IVAI greater than or equal to 6 hours. INTERVENTIONS: Clinicians committed to consider data from a CDSS informed by 6 hours rolling IDo2 average in their plan to wean IVAI postoperatively. Within the system, documentation of decisions incorporating these data prompted patient inclusion in the intervention arm. MEASUREMENTS AND MAIN RESULTS: The primary outcome measure was IVAI duration. Secondary outcomes were CICU length of stay (LOS), rate of cardiac arrest (CA), and rate of IVAI weaning failure. Negative binomial and logistic regression models were used to assess differences between intervention and control groups. Age, sex, Society of Thoracic Surgeons-European association for Cardio-Thoracic Surgery (STAT) category, site, admit time, single-ventricle diagnosis, Vasoactive-Inotropic Score, and trial group were tested as input covariates. Three hundred forty-three and 432 patients were enrolled in the intervention and control groups, respectively. The intervention group included a higher proportion of STAT 4 patients and a lower proportion of STAT 1 patients (81 vs. 62; p < 0.01 and 43 vs. 108; p < 0.01, respectively). A risk-adjusted negative binomial regression model showed 29% shorter vasoactive infusion duration in the intervention group (95% CI, 14-42%; p < 0.01). There were no significant trends observed in CICU LOS, or occurrence rate of CA or weaning failure. CONCLUSIONS: A CDSS informed by IDo2 significantly decreased vasoactive infusions duration. There was no discernible change in ICU LOS, CA rate, or weaning failure rate.

Department

Pediatrics

Link to Full Text

Share

COinS