Wang, Chun‐Li

Abstract Aims This study aimed to evaluate the predictive value of a baseline speckle tracking strain rate imaging-derived discoordination index for response to cardiac resynchronization therapy (CRT). Methods and results Ninety-seven patients with QRS ≥120 ms and left ventricular (LV) ejection fraction ≤35% were prospectively followed after CRT in the Mayo CRT Registry. The LV discoordination index (stretch/shortening or thinning/thickening during ejection) was calculated from three types of deformation, radial, circumferential, and longitudinal, using two-dimensional speckle tracking strain rate imaging. The benefit of CRT was evaluated by reverse remodelling (i.e. reduction of LV end-systolic volume ≥15% at 6-month follow-up) and survival. The optimal cut-off value of the baseline discoordination index in discriminating responders from non-responders was determined by receiver operating characteristic curve analysis. Significant differences in baseline indices between responders and non-responders were noted for radial and circumferential discoordination indices. A mid-ventricular radial discoordination index (RDI-M) >38% best predicted responders, especially in patients with ischaemic cardiomyopathy (area under the curve 0.86 for all patients, sensitivity 80%, and specificity 91%). Death occurred in 28 patients over a median follow-up of 3.2 years. When adjusted for confounding variables, lack of significant discoordination (RDI-M <38%) before CRT was associated with a particularly high mortality (hazard ratio 7.05, 95% confidence interval 2.45–26.0). Conclusion LV discoordination assessed by speckle tracking RDI-M imaging was able to predict reverse remodelling at 6 months and survival of patients who received CRT.