Introduction: Cardiac resynchronization therapy (CRT) has been shown to reverse left ventricular (LV) remodeling and improve symptoms in heart failure patients with wide QRS complexes; however, its role in patients with mildly prolonged QRS complexes is unclear. This study investigated if CRT benefited patients with mildly prolonged QRS complexes >120 to 150 ms and explored if the severity of systolic asynchrony determined such a response.
Methods and results: Fifty-eight patients (age 66 +/- 11 years, 66% male) who had undergone CRT were studied prospectively. Of these patients, 27 had QRS duration between 120 and 150 ms (group A), and 31 had QRS duration >150 ms (group B). Tissue Doppler echocardiography and clinical assessment were performed at baseline and 3 months after CRT. Both groups had significant reduction of LV volume and increased ejection fraction, +dP/dt, and sphericity index (all P < 0.05). These improvements were greater in group B and were explained by the higher prevalence of systolic intraventricular asynchrony. Significant reverse remodeling (reduction of LV end-systolic volume >15%) was evident in 46% of group A patients and 68% of group B patients. Improvement in clinical endpoints was observed in both groups (all P < 0.01), although the changes in metabolic equivalent and New York Heart Association functional class were greater in group B. In both groups, systolic asynchrony index (TS-SD) was the most important predictor of reverse remodeling (r =-0.78, P < 0.001) and was the only independent predictor in the multivariate model (beta=-1.80, confidence interval =-2.18 to -1.42, P < 0.001); QRS duration was not. A predefined TS-SD value >32.6 ms had a sensitivity of 94% and specificity of 83% to predict reverse remodeling. Improvement of intraventricular asynchrony after CRT was evident only in responders (P = 0.01).
Conclusion: Improvement of LV remodeling and clinical status is evident after CRT in heart failure patients with QRS duration >120 to 150 ms. These responders are closely predicted by the severity of prepacing intraventricular asynchrony but not QRS duration.