Comparison of measures of ventricular delay on cardiac resynchronization therapy response.

Background: Left ventricular (LV) pacing at sites of prolonged LV delay (QLV) or at long interventricular delay (right ventricle [RV]-LV) is strongly associated with cardiac resynchronization therapy (CRT) response. QLV and RV-LV have been independently evaluated, but little is known regarding the interrelationship between these measures or of delay to the RV.

Objective: The purpose of this study was to evaluate the relationship between measures of electrical delay on CRT response in the SMART-AV (SmartDelay Determined AV Optimization: A Comparison to Other AV Delay Methods Used in Cardiac Resynchronization Therapy) trial.

Methods: In 419 patients, QLV and RV-LV were measured. CRT response was defined as a >15% reduction in LV end-systolic volume from implant to 6 months. The correlation between QLV and RV-LV and the clinical variables associated with the difference between QLV and RV-LV (QRV) were determined. Multivariable logistic regression was used to analyze the association between these measures on CRT response. A machine learning algorithm was used to construct a classification tree to predict response to CRT.

Results: The cohort was 66% male (age 66 ± 11 years), 75% had left bundle branch block; and QRS was 150 ± 25 ms. QLV and RV-LV were highly correlated (R = 0.71). A longer QRV was observed among patients with right bundle branch block, ischemic cardiomyopathy, and increased QRS. In a multivariable model including QLV, RV-LV, and other known predictors of CRT response, RV-LV, but not QLV, remained associated with CRT response (odds ratio 1.13; 95% confidence interval 1.02-1.26; P = .017). Combining the 2 measures achieved better prediction of CRT response in the group with intermediate RV-LV.

Conclusion: RV-LV is a better predictor of CRT response than QLV. There is incremental value in using both measurements or QRV in certain subpopulations.