Long term safety and outcomes after atrial shunting for heart failure with preserved or mildly reduced ejection fraction: 5-year and 3-year follow-up in the REDUCE LAP-HF I and II trials.

Background: There is a little evidence regarding long-term safety and efficacy for atrial shunt devices in heart failure (HF).
Methods: The REDUCE LAP-HF I (n = 44) and II (n = 621) trials (RCT-I and -II) were multicenter, randomized, sham-controlled trials of patients with HF and ejection fraction >40%. Outcome data were analyzed from RCT-I, a mechanistic trial with 5-year follow-up, and RCT-II, a pivotal trial identifying a responder group (n = 313) defined by exercise PVR <1.74 WU and no cardiac rhythm management device with 3-year follow-up.
Results: At 5 years in RCT I, there were no differences in cardiovascular (CV) mortality, HF events, embolic stroke, or new-onset atrial fibrillation between groups. After 3 years in RCT II, there was no difference in the primary outcome (hierarchical composite of CV mortality, stroke, HF events, and KCCQ) between shunt and sham in the overall trial. Compared to sham, those with responder characteristics in RCT-II had a better outcome with shunt (win ratio 1.6 [95% CI 1.2-2.2], P = .006; 44% reduction in HF events [shunt 9 vs. control 16 per 100 patient-years], P = .005; and greater improvement in KCCQ overall summary score [+17.9 ± 20.0 vs. +7.6 ± 20.4], P < .001), while nonresponders had significantly more HF events. Shunt treatment at 3 years was associated with a higher rate of ischemic stroke (3.2% vs. 0%, 95% CI 2%-6.1%, P = .032) and lower incidence of worsening kidney dysfunction (10.7% vs. 19.3%, P = .041).
Conclusions: With up to 5 years of follow up, adverse events were low in patients receiving atrial shunts. In the responder group, atrial shunt treatment was associated with a significantly lower HF event rate and improved KCCQ compared to sham through 3 years of follow-up.
Gov Registration: NCT02600234, NCT03088033.
Competing Interests: Conflict of Interest None.
(Published by Elsevier Inc.)