Background: Atrial Flow Regulator (AFR) implantation in chronic heart failure has been shown to reduce pulmonary capillary wedge pressure (PCWP) and was associated with submaximal exercise capacity and health-related quality of life in both heart failure with reduced (HFrEF) and preserved ejection fraction (HFpEF). However, the impact of AFR on survival is unknown.

Methods: The prospective, multicenter, open-label, non-randomized AFR-Prelieve Trial (NCT 03030274) assessed the safety and efficacy of the Occlutech Atrial Flow Regulator (Occlutech AFR) device in patients with symptomatic HFrEF (left ventricular ejection fraction (LVEF) ≥15% and <40%) or HFpEF (LVEF ≥40% and <70%) and elevated PCWP (≥15mmHg at rest or ≥25mmHg during exercise). In this interim analysis after the first 60 patients completed 12 months of follow-up, the theoretical impact of AFR implantation was assessed by comparing the observed survival rate with the median predicted probability for one-year survival. The risk of mortality was predicted for each subject from individual baseline data using the Meta-analysis Global Group in Chronic Heart Failure (MAGGIC) prognostic model. The significance of the paired difference between the observed and predicted mortality rate was established by its confidence interval (CI) applying a bootstrapping procedure using 1000 subsamples: if the CI did not contain zero, significance was concluded (p<0.05).

Results: Of the 87 patients with successful device implantation, 46 (53%) had HFrEF and 41 (47%) had HFpEF. At the time of data analysis, 60 patients had a complete 12-month follow-up. The median follow-up duration was 353 days. Compared with baseline (22.5; SD: 5.3), the mean MAGGIC score was significantly reduced by -4.6 (SD: 5.1; p<0.001) at 12 months of follow-up. A total of 6 (6.9%) patients died during follow-up (8.6 deaths per 100 patient-years), all of which had HFrEF. Of these, 4 deaths were rated as heart failure-related deaths (5.7 heart failure-related deaths per 100 patient-years in the overall population; 10.8 heart failure-related deaths per 100 patient-years in the HFrEF subgroup). For the overall study population, the median predicted mortality rate was 12.2 deaths per 100 patient-years. While the observed mortality rate (0 deaths per 100 patient-years) was significantly lower than the median predicted mortality rate (9.3 deaths per 100 patient-years) in patients with HFpEF (-9.3; confidence intervals: -11.1 to -8.4), there was no difference in patients with HFrEF (observed mortality rate of 16.2 deaths per 100 patient-years versus median predicted mortality rate of 16.8 deaths per 100 patient-years) (Figure).

Conclusion: In patients with HFpEF, the mortality rate following AFR implantation was lower than the predicted mortality rate. These findings need to be confirmed by larger randomized, controlled trials.


Figure: Kaplan-Meier survival curves for the observed survival during the study.

The stars (★) indicate the median predicted survival at 12 months using the MAGGIC prognostic model. Colored bands around lines represent estimated 95% confidence intervals for survival curves. Log-rank test for the difference between patients with HFrEF and HFpEF.