Use of a Predictive Analytics Algorithm to Optimize Weaning of Inotropes Following Pediatric Cardiac Surgery

Overview

The objective of this study is to determine the effectiveness of a real time continuous risk analytics algorithm in the successful de-escalation of vasoactive and inotropic support in pediatric patients following cardiac surgery.

IDO2 utilizes high fidelity continuous and intermittent patient data to feed a Bayesian model which predicts the risk of inadequate tissue oxygen delivery. This index (the IDO2) is FDA 510k cleared for pediatric patients 2 kg up to 12 years of age to continuously report the risk for inadequate tissue oxygen delivery (defined as a mixed venous oxygen saturation less than 40%). The index is continuously displayed in graphical form at the bedside. Clinical decision support systems (CDSS) such as IDO2 may inform inform the clinician when it is appropriate to de-escalate care on a critically ill. Appropriate de-escalation plays a role in the safe, efficient utilization of resources in the CICU, and may reduce duration intervals of care such as duration of support with vasoactive and inotropic drugs, mechanical ventilation and length of stay. In support of this hypothesis, the investigators for this proposed study have recently completed a retrospective, multi-center analysis of 2,556 patient encounters demonstrating that elevated IDO2 during a wean off inotropic agents is associated with weaning failure. When compared to conventional markers of cardiac output, 6-hour average IDO2 was superior to lactate elevation, fall in base deficit, and fall in urine output in discriminating inotrope weaning success from failure. Additionally, for those patients who failed inotrope wean, rescue with re-starting an inotrope was associated with a concomitant fall in IDO2. This analysis supports an underlying hypothesis that IDO2 reflects underlying patient stability, and the rescue of deteriorating physiology leads to an improved physiologic state, and hence a lower IDO2.7 Data suggest that simple CDSS which prompt discussion about management decisions, such as those made during daily ICU rounds, may improve outcomes. These CDSS are often in the form of checklists and apply to usual or standardized practices. These simplified mechanisms may not apply to the more dynamic clinical situations in which specific and intensively monitored patient populations can demonstrate variable response to drugs and recovery. In these circumstances, a CDSS utilizing a 6-hour rolling average value of IDO2, in which the physiologic response to decisions is demonstrated continuously, may inform a more rapid more efficient and safe de-escalation of vasoactive and inotropic drugs when implemented on each of the twice daily clinical work rounds in the CICU.

Conditions

Interventions

Eligibility

Locations (1)

Outcomes

Central Contacts