A Clinical Mathematical Model Estimating Postoperative Urine Output in Children Underwent Cardiopulmonary Bypass for Congenital Heart Surgery

Abstract

Abstract Objectives Our objective was to build a proof of concept of the clinical mathematical model estimating postoperative urine output (UOP) utilizing preoperative, intraoperative, and immediate postoperative variables in children who underwent cardiopulmonary bypass (CPB) for congenital heart surgery. Methods This was a single-center, retrospective cohort study in a university-affiliated children's hospital. Patients younger than 21 years old who underwent CPB for congenital heart surgery and were postoperatively admitted to West Virginia University Children's Hospital's pediatric intensive care unit (PICU) between September 1, 2007 and June 31, 2013 were included in the study. Body surface area, CPB duration, first measured hematocrit, serum pH, central venous pressure, and vasoactive-inotropic score in the PICU were used to build the mathematical model. A randomly selected 50% of the dataset was used to calculate model parameters. A cross-validation was used to assess model performance. Results A total of 256 patients met the inclusion criteria. The model was able to achieve mean absolute error of 1.065 mL/kg/h (95% confidence interval (CI): 1.062–1.067 mL/kg/h), root mean squared error of 1.80 mL/kg/h (95% CI: 1.799–1.804 mL/kg/h), and R2 of 0.648 (95% CI: 0.646–0.650) in estimating UOP in the first 32 hours of postoperative period. Conclusions The mathematical model utilizing preoperative, intraoperative, and immediate postoperative variables may be a potentially useful clinical tool in estimating UOP in the first 32 hours postoperative period.