Background: Due to the availability of new devices, the use of ventricular assist devices (VADs) in children has been increasing; however, patient selection and optimal timing of device implantation in this population remains uncertain.
Methods: A retrospective review of the United Network for Organ Sharing dataset identified 5,200 listings without mechanical circulatory support (MCS) for isolated pediatric heart transplant, 1995 to 2012. Patients were randomly divided into a derivation and validation cohort. A multivariable logistic regression model predicting the likelihood of death or need for MCS within 60 days was built using the derivation cohort and tested in the validation cohort. A simplified score (PedsMCS score) was developed and evaluated for accuracy.
Results: The predictive model consisted of variables present at listing (age, albumin level, creatinine clearance, serum bilirubin, mechanical ventilation, and inotropic support). It had good predictive ability (C statistic 0.7304) within the validation cohort. The simplified PedsMCS score was also predictive (C statistic 0.7217) and there was a strong correlation between predicted and expected outcomes (r=0.91, p<0.0001). Patients with PedsMCS score 16 or greater had a significantly higher risk of death or MCS within 2 months (36.6%) than those with low scores (<6) (1.5%, p<0.0001). A single point increase in PedsMCS score was associated with a 16.7% increase in the risk of death or MCS with 2 months (p<0.0001).
Conclusions: We have developed and validated a simplified score to predict the need for MCS based on risk factors present at listing. This will provide more accurate prognostication in children awaiting heart transplant, and may improve patient selection.
Copyright © 2014 The Society of Thoracic Surgeons. Published by Elsevier Inc. All rights reserved.