BACKGROUND: Changes in QRS duration (∆QRS) are often used in the clinical setting to evaluate the effect of cardiac resynchronization therapy (CRT), although an association between ∆QRS and outcomes is not firmly established. We aimed to assess the association between mortality and ∆QRS after CRT in patients from the DANISH (Danish Study to Assess the Efficacy of ICDs in Patients with Non-Ischemic Systolic Heart Failure on Mortality) study.

METHODS: We included all patients from DANISH who received a CRT device and had available QRS duration data before and after implantation. Cox proportional hazards models were used to assess associations between ∆QRS (post-CRT QRS minus pre-CRT QRS) and mortality.

RESULTS: Complete data were available in 572 patients. Median baseline QRS duration was 160 ms (IQR [146;180]). Post-CRT QRS was recorded a median of 48 days (IQR [33;86]) after implantation, and the median ∆QRS was -14 ms (IQR [-38;-3]). During a median follow-up of 4.1 years (IQR [2.5;5.8]), 106 patients died. In crude Cox regression, all-cause mortality was reduced by 6% per 10 ms shortening of QRS (HR 0.94; CI: 0.88-1.00, p = 0.04). The effect did not remain significant after multivariable adjustment (HR 1.01, CI: 0.93-1.10, p = 0.77). Further, no association was found between ∆QRS and improvement of New York Heart Association functional class at 6 months (OR 1.03, CI: 0.96-1.10, p = 0.42).

CONCLUSION: In a large cohort of patients with non-ischemic cardiomyopathy, reduction of QRS duration after CRT was not associated with changes in mortality during long-term follow-up.