CRT should be recommended for all heart failure patients with mechanical dyssynchrony and a ventricular conduction delay regardless of the QRS morphology

Introduction Two recent meta-analyses displayed the heterogeneous effect of cardiac resynchronization therapy (CRT) in patients with a non-left bundle branch block (non-LBBB) and found an increased burden of ventricular arrhythmias after CRT (1,2). Careful patient selection is therefore paramount. Evidence suggests the assessment of mechanical dyssynchrony may guide decision-making to CRT in this challenging population. Purpose To assess the effect of CRT in patients with and without mechanical dyssynchrony compared to medication only, and in particular in patients with a non-LBBB morphology. Methods We retrospectively screened for symptomatic heart failure patients who presented between 2012 and 2018 at our clinic with a QRS duration ≥130ms and a left ventricular ejection fraction (LVEF) ≤35%. We identified 361 patients who received CRT and 1107 patients who could have been considered for CRT. Patients were excluded if echocardiographic data was unavailable, they received CRT after 2018, were 80 years or older, or had other significant co-morbidities. We enrolled the remaining 220 CRT patients ("CRT") and 124 patients treated with medication only ("Control") for analysis. Patients were categorized for the presence of mechanical dyssynchrony on echocardiography, defined as the presence of either septal flash, apical rocking, or both. Multivariable-adjusted Cox regression analyses, including stratification by the presence of dyssynchrony, was performed to compare the effect of CRT to medication only. Hazard ratios (aHR) were obtained after adjustment for age, sex, atrial rhythm, QRS duration, QRS morphology, ischemic disease, kidney function, medication and presence of a defibrillator. Sensitivity analysis was performed in patients with a non-LBBB morphology. Results The CRT and Control group were comparable except for the QRS duration (163±22ms vs 152±16ms, p<0.001), the proportion of patients on optimal medical therapy (64% vs 47%, p=0.001), patients with a defibrillator (81% vs 19%, p<0.001), and patients with a LBBB morphology (75% vs 47%, p<0.001). During an average follow-up of 6.0(±3.2) years, 162 patients (47%) died. In the overall cohort, CRT did not significantly improve survival (aHR: 0.66, p=0.065). There was, however, a marked survival benefit for CRT over medication only in patients with mechanical dyssynchrony (aHR: 0.37, p=0.001). In patients without dyssynchrony there was a trend towards harm due to CRT (aHR: 2.05, p=0.062). These results were consistent in the sub-cohort of patients with a non-LBBB morphology. Conclusion CRT significantly improves the life expectancy of patients with mechanical dyssynchrony, including patients with a non-LBBB morphology. These results add to the evidence that all heart failure patients with LVEF≤35%, QRS≥130ms and mechanical dyssynchrony should be recommended (Class I) for CRT regardless of QRS morphology, while awaiting the results of ongoing randomized controlled trials.