Novel Model To Assess Prognosis of Transplant Vasculopathy Following Cardiac Transplantation

Purpose There is growing evidence that non-stenotic epicardial vasculopathy affects survival after heart transplantation. We have developed a model that estimates the risk of death due to vasculopathy by combining the ISHLT working formulation and the consensus statement of the German task force for cardiac allograft vasculopathy (CAV). Methods and Materials Coronary angiograms (n=996, 906 pts) performed at 30 days, 1 and 5 yrs post-transplant were evaluated according to the ISHLT (CAV0-CAV3) and the German task force (epicardial remodeling, peripheral obliteration) working formulations. Cox models were generated to estimate the risk of death, given these predictors. A combination model was designed to estimate vasculopathy-related risk of death at 10 yrs post-transplant. Results CAV0 was present in 85%, 78% and 45% of pts at 30 days, 1 year, and 5 years post-transplant; CAV1 was found in 14%, 17% and 36% of pts; CAV2 and CAV3 in 2%, 5% and 17% of pts; and remodeling was found in 10%, 18% and 45% of pts at 30 days, 1 and 5 yrs, respectively. Relative risks are presented in the table. Ten-year survival using the ISHLT classification showed sensitivity of 70%, specificity of 52% and a false classification rate of 48%. Ten-year survival as classified by epicardial remodeling showed sensitivity of 70%, specificity of 45% and a false classification rate of 43%. The combination of the ISHLT and German classification schemes had sensitivity of 88%, specificity of 29% and false classification rate of 39%. Conclusions A combination model incorporating the ISHLT working formulation and the presence of non-stenotic epicardial vasculopathy shows improved ability to identify pts at risk for death post-transplant and reduces false classification rates. 30 days 1yr 5yrs Parameter HR p HR p HR p CAV1 1.690 p 1.715 p 1.522 p=0.009 CAV2 1.765 p=0.138 2.043 p=0.063 1.928 p=0.029 CAV3 2.964 p 3.160 p 2.099 p=0.001 Remodeling 1.293 p=0.039 1.426 p=0.007 1.383 p=0.024