Catheter ablation for atrial fibrillation in patients with left ventricular systolic dysfunction. A systematic review...

Atrial fibrillation (AF) and heart failure are often coexisting major public health burdens. Although several studies have reported partial restoration of systolic left ventricular (LV) function after catheter ablation for AF, the method is not widely applied in patients with LV dysfunction. We reviewed the results of AF ablation in patients with systolic LV dysfunction.

PubMed was searched for studies published after 2000 reporting original data on AF catheter ablation in adult patients with systolic LV dysfunction. Primary end point was the change of LV ejection fraction (LVEF) after catheter ablation; secondary endpoints were the changes of exercise capacity and quality of life after the procedure. We calculated mean difference (MD) of LVEF and 95% confidence interval (95% CI) using random-effects models. Heterogeneity was investigated by I(2) statistic, publication bias with Egger's test. The impact of covariates on LVEF improvement was evaluated with meta-regression analyses. Nine studies with a total of 354 patients with systolic LV dysfunction were analyzed. Study patients were mainly male with mean age 49 to 62 years, LVEF was moderately impaired and ranged in all but 1 study from 35% to 43%. LVEF improved after ablation with a MD of 11.1% (95% CI: 7.1-15.2, P < .001). Heterogeneity among analyzed studies was significant (I(2) = 92.9, P < .001). No potential publication bias was found. In meta-regression analyses, the proportion of patients with coronary artery disease was inversely related with LVEF improvement (P < .0001) whereas there was no association between the LVEF change and the proportion of patients with nonparoxysmal AF or the proportion of patients without AF recurrences during follow-up.

AF ablation in patients with systolic LV dysfunction results in significant improvement of LV function, but the extent of this improvement is heterogeneous. Patients with coronary artery disease seem to benefit less than patients with other underlying diseases. These results may be explained by patient selection.