Abstract 15844: Predicting the Risk of Significant Coronary Artery Disease in Liver Transplant Patients: The CAD-LT Score

Abstract

Background: Patients with significant coronary artery disease (CAD) are more likely to develop post-liver transplant (LT) cardiac events. We developed the CAD-LT screening score and testing algorithm to predict the risk of significant CAD in LT candidates. Methods: Patients who underwent pre-LT evaluation at Indiana University (2010-2017) were studied retrospectively (n=1814). Stress tests (ST) (n=1677) and cardiac catheterization (CATH) reports (n=1300) were reviewed. CATH was performed in patients with predefined CAD risk factors. Significant CAD was defined as disease requiring percutaneous or surgical intervention. Multivariable estimates (Adjusted Odds Ratio i.e. AOR [95%CI]) with assessment of model performance using Receiver Operating Curve analysis were used to compute a point-based risk score and stratify patients. A 10-fold internal cross-validation (CV) model was done. Results: There were 950 LT and 864 no-LT patients. The risk-adjusted predictors of significant CAD were older age (AOR 1.06 [95%CI 1.03-1.09]), male gender (1.69 [1.13-2.50]), diabetes (1.44 [1.01-2.06]), hypertension (1.50 [1.05-2.15]), current smoking history (1.81 [1.16-2.82]), family history of CAD (1.76 [1.24-2.50]), and personal history of CAD (5.41 [3.48-8.43]). The CAD-LT score is shown in Table 1. Figure 1 is an algorithm for its use. The mean CV Area Under the Curve [95% CI] was 0.75 [0.71-0.79]. The algorithm detected 97% of the patients with significant CAD and would decrease the number of ST by 718 (43%; 671 in high-risk group and 47 in low-risk group) and CATH by 409 (30%). Conclusion: The CAD-LT score identifies LT candidates at high risk for significant CAD and guides pre-LT testing.