Atrioventricular delay programming and the benefit of cardiac resynchronization therapy in MADIT-CRT

Background The optimal atrioventricular pacing delay (AVD) in cardiac resynchronization therapy (CRT) remains to be determined. Objective To determine whether programming CRT devices to short AVD (S-AVD) will improve clinical response secondary to greater reductions in dyssynchrony. Methods The study population comprised 1235 patients with left bundle branch block enrolled in Multicenter Automatic Defibrillator Implantation Trial in Cardiac Resynchronization Therapy (MADIT-CRT). We assessed the relationship between AVD and outcomes. Patients programmed to S-AVD (median value of <120 ms; n = 337) vs long AVD (L-AVD; ≥120 ms; n = 390) were assessed for the end points of heart failure (HF) or death, death alone, and echocardiographic response to the CRT at 1-year follow-up. Outcomes were also compared to the left bundle branch block implantable cardioverter-defibrillator-only group (n = 508). Results Multivariate analysis showed that patients programmed to S-AVD experienced a significant 33% (hazard ratio [HR] 0.67; 95% confidence interval [CI] 0.44-0.85; P =.037) reduction in the risk of HF or death and a 47% (HR 0.53; 95% CI 0.29-0.94; P =.031) reduction in death alone as compared with those programmed to L-AVD. Patients with CRT-programmed S-AVD and L-AVD experienced 63% (HR 0.37; 95% CI 0.26-0.53; P <.001) and 46% (HR 0.54; 95% CI 0.31-0.96; P <.001) reduction, respectively, in the risk of HF or death compared to patients with implantable cardioverter-defibrillator alone. At 1 year of follow-up, S-AVD vs L-AVD was associated with a greater reduction in left ventricular end-systolic volume (34.2% vs 30.8%; P =.002) along with a significantly greater improvement in dyssynchrony (22.3% vs 9.4%; P =.036). Conclusions Our findings indicate that in MADIT-CRT programming, the CRT AVD <120 ms was associated with a greater clinical and echocardiographic response to CRT.