CMV and Cardiovascular Disease: Reply

We read with interest the comments by Arasaratnam [1]. Even though he agrees that we provided a convincing argument for the role of cytomegalovirus (CMV) exposure in posttransplant cardiovascular (CV) events [2], he also highlights an inadequate calibration of our model. We acknowledge that the goodness of fit of our model was insufficient, suggesting missing parameters. However, the area under the receiving operating characteristic curve was 0.7 (0.66–0.73), indicating that our model enables good discrimination. In other words, our model is relevant but probably lacks accuracy. Predicting posttransplant cardiovascular risk is challenging. The increased risk of CV disease is primarily due to an