Usefulness of tissue Doppler velocity and strain dyssynchrony for predicting left ventricular reverse remodeling response after cardiac resynchronization therapy

Abstract

The assessment of systolic dyssynchrony by echocardiography is useful in predicting a favorable response to cardiac resynchronization therapy (CRT). Tissue Doppler velocity and tissue Doppler longitudinal strain have been suggested for this purpose. This study compared parameters of systolic dyssynchrony derived from these 2 imaging modalities for their predictive values of CRT response. Two hundred fifty-six patients from 3 different centers who received CRT were followed for 6 +/- 3 months. Parameters of systolic dyssynchrony based on tissue Doppler velocity and strain imaging were assessed for the prediction of left ventricular (LV) reverse remodeling (reduction of LV end-systolic volume > or =15%). These included time to peak systolic velocity (or peak strain) of 12 LV segments to calculate the SD (Ts-SD or Tepsilon-SD), maximal difference in delay (Ts-Diff or Tepsilon-Diff), and opposite wall delay (Ts-OW or Tepsilon-OW). The septal-to-lateral delay (Ts-Sep-Lat or Tepsilon-Sep-Lat) was also measured. LV reverse remodeling, defined as improvement in end-systolic volume > or =15%, was observed in 141 patients (55%). All 4 tissue velocity parameters predicted LV reverse remodeling, and the areas under the receiver-operating characteristic curves were 0.86, 0.85, 0.84, and 0.79 for Ts-SD, Ts-Diff, Ts-OW, and Ts-Sep-Lat, respectively (all p <0.001). The cut-off values derived from receiver-operating characteristic curve analysis were 33 ms for Ts-SD, 100 ms for Ts-Diff, 90 ms for Ts-OW, and 60 ms for Ts-Sep-Lat, and their sensitivities were 93%, 92%, 81%, and 70%, with specificities of 78%, 68%, 80%, and 76%, respectively. In contrast, none of the longitudinal strain parameters predicted LV reverse remodeling. The areas under the receiver-operating characteristic curves ranged from 0.49 to 0.53 (all p = NS). The same conclusions were obtained in subgroup analyses of QRS duration (120 to 150 vs >150 ms) and ischemic or nonischemic cause of heart failure. In conclusion, parameters of tissue Doppler longitudinal velocity, but not longitudinal strain, predicted LV reverse remodeling after CRT.