An echocardiogram-based 16-segment model for predicting left ventricular ejection fraction improvement.

Abstract

An important goal of cardiac revascularization is to improve the left ventricular ejection fraction, which is an important clinical determinant of the long-term outcome for patients with coronary artery disease. Regional myocardium function improvement may be expected from revascularization when viable myocardium is detected using non-invasive cardiac imaging. However, the quantitative relation between regional myocardial function recovery and global heart function improvement has not been determined and there is no tool to predict the amount of ejection fraction improvement prior to revascularization. A 16 segment biomechanical model of the left ventricle is proposed to establish the relationship between the ejection fraction improvement and the viable segments detected by echocardiography. With the assumption that the viable segments would potentially improve contractility after revascularization, the ejection fraction improvement is estimated for all possible wall motion score improvement in viable segments. The model shows that the ejection fraction improvement is linearly related to the contractility in the normal segments and a weighted sum of the numbers of viable segments that recover to normal or hypokinetic contractility. The predictive value of the model is illustrated for a group of patients reported in the literature. The model predictions of the post-revascularization ejection fraction are very close to the follow-up data with a very strong correlation (R2 = 0.92). By predicting the ejection fraction improvement, the model may provide a tool for evaluating the efficacy of revascularization and for selecting patients who would benefit from revascularization.