Cumulative Effect of Preoperative Risk Factors on Mortality after Pediatric Heart Transplantation.

Year of Publication


Date Published

2018 Apr 20

ISSN Number



<p><strong>BACKGROUND: </strong>Risk assessment in heart transplantation is critical for candidate selection, but current models inadequately assess individual risk of postoperative mortality. We sought to identify risk factors and develop a scoring system to predict mortality following heart transplantation in children.</p>

<p><strong>METHODS: </strong>The records of patients undergoing heart transplantation at our institution from 2010 - 2016 were reviewed. Clinical characteristics were recorded and compared between survivors and non-survivors. Using Cox proportional hazard modeling, a risk factor score was developed using factors associated with postoperative mortality.</p>

<p><strong>RESULTS: </strong>Seventy-four patients underwent heart transplantation at a mean age of 8.8 ± 6.6 years. Congenital heart disease was the most common indication, comprising 48.6% of the cohort. Overall mortality was 18.9%, with 10/14 dying ≤30 days of operation or during initial postoperative admission (early mortality). The following preoperative factors were associated with overall mortality: single ventricle congenital heart disease (HR 3.2, p = 0.042), biVAD (HR 4.8, p = 0.043), history of ≥4 sternotomies (HR 3.9, p = 0.023), panel reactive antibody &gt; 10% (HR 4.4, p = 0.013), any previous surgery at an outside institution (HR 3.2, p = 0.038), and pulmonary vein disease (HR 4.7, p = 0.045). Each risk factor was assigned a point value, based on similar magnitude of the hazard ratios. A score of ≥4 predicted mortality with 57% sensitivity and 90% specificity.</p>

<p><strong>CONCLUSIONS: </strong>In this single-center pediatric cohort, post-heart transplantation mortality could be predicted using patient-specific risk factors. The cumulative effect of these risk factors predicted mortality with high specificity.</p>



Alternate Title

Ann. Thorac. Surg.




Subscription is not available for this page.