Prospective Validation of a Novel Vasoactive-Ventilation-Renal Score as a Predictor of Outcomes After Pediatric Cardiac Surgery

Abstract

Background: We sought to further validate the novel vasoactive-ventilation-renal (VVR) score in a prospective study of a heterogeneous cohort of children undergoing cardiac surgery that includes patients with single-ventricle anatomy and residual mixing lesions.

Methods: We prospectively performed an observational study of all children less than 18 years of age who underwent surgery for congenital heart disease at our center from November 2013 to June 2014. We calculated VVR score as follows: vasoactive-inotrope score + ventilation index + (change in serum creatinine from baseline × 10). Admission, peak, and 48-hour measurements were recorded. Outcomes of interest were prolonged duration of mechanical ventilation and intensive care unit and hospital stays, represented by the upper 25% for all patients. Areas under the receiver-operating characteristic curves (AUC) were determined for all study timepoints and outcome variables.

Results: Ninety-two patients were analyzed; their median age was 0.65 (range, 3 days to 17.9 years), and 17 (18%) had single-ventricle anatomy. The VVR measurements outperformed vasoactive-inotrope scores in isolation at all timepoints, with higher AUC values for all outcomes. Of the three timepoints assessed, the 48-hour VVR score most consistently predicted poor outcome, especially with regard to prolonged duration of mechanical ventilation (AUC 0.980) and prolonged intensive care unit stay (AUC 0.919).

Conclusions: In a heterogeneous population of children undergoing cardiac surgery, the 48-hour VVR score was a very strong predictor of outcomes, and outperformed the more traditional vasoactive-inotrope score. The VVR score, therefore, represents a novel and potentially powerful means of predicting clinical outcomes relatively early in the hospital course of these patients.