Background and Goal of Study:Primary Graft Dysfunction (PGD) often occurs intraoperatively and is the primary cause of early mortality after heart transplantation(HT). PGD is characterized as circulatory failure due to ventricular dysfunction occurring within 24 hours of reperfusion requiring inotropic and mechanical support. The RADIAL score is a simple grading system for predicting severe PGD and is calculated by adding a point for each of the following categories; Right atrial pressure>10mmHg, recipient Age>60y, Diabetes mellitus, Inotrope dependence, donor Age>30y, Length of ischemic time>240. Higher scores are predictive of PGD. However, at our center HT recipients predominately rely on ventricular assist device (VAD) support for exceptionally long periods and the RADIAL score has not been validated in such patients groups. Our goal was to examined the performance of the the RADIAL score for predicting PGD at our single center. 
Methods:After review board approval, we conducted a retrospective analysis of 63 adult HT recipients at our hospital from 2010 to 2016. Collected data included age, sex, reliance and length of VAD support, and the occurrence of PGD. The RADIAL score was calculated for each case. Our primary outcome was the correlation of the RADIAL score with the occurrence of PGD. To assess the predictive value of the RADIAL score a logistic-regression model was constructed from the collected results and the area under a receiveroperating characteristic (ROC) curve was calculated. A value over 0.7 was considered as an acceptable predictive performance model. 
Results: We included 63 HT cases for the final analysis; average right recipient atrial pressure 8±4 (mmHg), recipient mean age 39±13 (y) , diabetes mellitus (n=0), inotrope dependence (n=35), average donor age 42±13 (y), length of ischemic time 217 ± 45 (min). Recipients predominantly required VAD support (90.5%) with a mean support period of 2.4 ± 1.1(y). There were 27 cases of PGD (mild 9, moderate 13, severe 5). The average RADIAL score for PGD and non-PGD cases was, 2.4 ± 1.0 and 1.9 ± 1.0, respectively (p = 0.06) . The area under the ROC curve calculated from our model was 0.61.
Conclusion:The performance to for the RADIAL score to predict PGD was low in our model. Although this was a small single center analysis, the RADIAL score may not be useful in predicting PGD in patients undergoing HT after surviving longer periods of VAD support.

10 / The performance of the RADIAL score to predict primary graft failure in heart transplant recipients surviving prolonged ventricular assist device support.