Comparison of permanent left ventricular and biventricular pacing in patients with heart failure and chronic atrial fibrillation: a prospective hemodynamic study.

Background: Left ventricular pacing (LVP) and biventricular pacing (BVP) have been proposed as treatments for patients with advanced heart failure complicated by discoordinate contraction due to intraventricular conduction delay. For patients in sinus rhythm, BVP works in part by modulating the electronic atrial-ventricular time delay and thus optimizing contractile synchrony, the contribution of atrial systole, and reducing mitral regurgitation. However, little is known of the mechanisms of BVP in heart failure patients with drug-resistant chronic atrial fibrillation. HYPOTHESIS AND METHODS: LVP differs from BVP because hemodynamic and clinical improvement occurs in association with prolongation rather than shortening of the QRS duration. We sought to determine if LVP or BVP improves mechanical synchronization in the presence of atrial fibrillation. Thirteen patients with chronic atrial fibrillation, severe heart failure and QRS >or=140 ms received (after His bundle ablation) a pacemaker providing both LVP and BVP. The mean age was 62 +/- 6 years and left ventricular ejection fraction was 24 +/- 8%. After a baseline phase of one month with right ventricular pacing, all patients underwent in random order 2 phases of 2 months (LVP and BVP). At the end of each phase, an echocardiogram, a hemodynamic analysis at rest and during a 6-minute walking test and a cardio-pulmonary exercise test were performed.

Results: LVP and BVP provided similar performances at rest (p = ns). The 6-minute walking test revealed similar performances in both pacing modes but patients were significantly more symptomatic at the end of the test with LVP ( p = 0.035). The cardio-pulmonary exercise test showed higher performances with BVP (92 +/- 34 Watts) vs. LVP (77 +/- 23; p = 0.03). LVP was associated with significantly more premature ventricular complexes recorded during the 6 minute walking test (49 +/- 71) than BVP (10 +/- 23; p = 0.04).

Conclusions: In this small series of patients with atrial fibrillation, congestive heart failure and a prolonged QRS duration, LVP and BVP provided similar hemodynamic effects at rest whereas BVP was associated with better hemodynamic effects during exercise and fewer premature ventricular complexes. Although the mechanisms for the observed differences are uncertain, it is possible that there is worsening of right ventricular function due to a rise in left-to-right electromechanical delay during exercise. Increased catecholamines release might contribute to the lower exercise tolerance and greater number of premature ventricular complexes recorded during exercise observed during LVP compared to BVP.

Recommendations: Patients with atrial fibrillation, heart failure and QRS prolongation who are candidates for His-bundle ablation and cardiac resynchronization therapy may respond better to BVP rather than to LVP.