The usefulness of QRS Index for prediction of echocardiographic response in cardiac resynchronization therapy: a multicenter study

Background: The association between QRS narrowing and response to cardiac resynchronization therapy (CRT) has been investigated by several studies, but their findings remain inconclusive. Aim of our study was to explore the relationship between QRS Index and echocardiographic response to CRT. Methods: This multicenter, retrospective analysis included 326 consecutive patients (mean age was 70.0±10.1 years old; males 76.7%) who underwent CRT-D implantation in primary and secondary prevention between 2018 and 2020. The estimation of QRS shortening after CRT-D implantation was precisely assessed through the QRS Index, calculated as follows: [(QRS duration before implantation - paced QRS duration)/QRS duration before implantation]*100. Results: After a mean follow-up of 12.7±4.5 months, 55.2% (180/326) of the patients showed an echocardiographic response to CRT. The median [25-75th] QRS Index was 3.85% [-14.1% - +13.9%]. The best predictive cut-off value of QRS Index was 1.40% (sensitivity 70.4%, specificity 64.5%, AUC 0.70). In patients with left bundle branch block, the median [25-75th] QRS Index was 9.85% [+3.87% - +16.7%]. In this subgroup, the AUC was 0.737 and the best predictive cut-off of QRS Index was 2.20% (sensitivity 78.3%, specificity 67%). The multivariable model showed that only left ventricular ejection fraction and QRS Index were independently associated with CRT response (respectively OR 0.92, CI 95% 0.86-0.98, P=0.01 and OR 1.057, CI 95% 1.026-1.089, P<0.001). Conclusions: The QRS Index tightly correlated with CRT response. Only LVEF and QRS Index were independently associated with echocardiographic response to CRT.