Effect of atrial shunt device on indices of cardiac mechanics in heart failure with preserved and mildly reduced ejection fraction

Background The effects of atrial shunt devices on cardiac mechanics in heart failure with preserved or mildly reduced ejection fraction (HFpEF/HFmrEF) are not well understood. Methods Patients with HF and LVEF ≥40% randomized in REDUCE LAP-HF I underwent both echocardiography and invasive exercise hemodynamics at baseline and during the 1-month follow-up visit. Left ventricle (LV) global longitudinal strain (GLS), RV free wall strain (RVFWS), left atrial reservoir strain (LARS), and right atrial reservoir strain (RARS) were measured on the baseline and 1-month echocardiograms blinded to treatment assignment and invasive hemodynamics. We examined the correlation of indices of cardiac mechanics and resting and exercise invasive hemodynamics at baseline; compared differences of changes in indices of cardiac mecahnics between treatment groups; and evaluated correlations of 1-month changes in indices of cardiac mechanics with changes in invasive hemodynamics. Results A total of 44 patients (mean age 70±9 years, 50% female) were randomized 1:1 to atrial shunt (n=22) or sham control (n=22). Baseline GLS (absolute values) and LARS were inversely correlated with resting PCWP (R=-0.43, P=0.007; and R=-0.41, P=0.01, respectively). One month after randomization, atrial shunt treatment led to significant improvement in GLS (placebo-adjusted +3.7 %-units [95% CI +0.5, +6.9]; P=0.027). LARS showed a trend to improvement (placebo-adjusted +5.6 %-units [95% CI -1.3, +12.6]; (P=0.11). Within-group change in LARS was significant in the atrial shunt group (+5.4 %-unit [95% CI +0.2, +10.6]; p=0.041) but not in the sham group (-1.0 %-units [95% CI -6.7, +4.7]; P=0.72). No between-group differences were observed in delta RVFWS (P=0.70) or delta RARS (P=0.10). Among the rest, legs up, and exercise invasive hemodynamic measures, 1-month change in LARS correlated best with 1-month change in legs up PCWP (R=-0.60, P=0.0006). Conclusions Compared to sham, atrial shunt treatment improved LV and LA mechanics, without differences in RV or RA mechanics after 1 month of follow-up. Improvement in LA strain may reflect a reduction in LA pressure over time and may be a novel indicator of change in LA pressure in response to HFpEF/HFmrEF treatment.