Exercise oscillatory breathing in heart failure with reduced ejection fraction: clinical implication

Abstract

Abstract Aim The aim of the study is (i) to evaluate the impact of exercise oscillatory ventilation (EOV) in patients with heart failure (HF) with reduced left ventricular ejection fraction (HFrEF) during cardiopulmonary exercise testing (CPET) compared with no EOV (N-EOV); (ii) to identify the influence of EOV persistence (P-EOV) and EOV disappearance (D-EOV) during CPET on the outcomes of mortality and hospitalization in HFrEF patients; and (iii) to identify further predictors of mortality and hospitalization in patients with P-EOV. Methods and results Three hundred and fifteen stable HFrEF patients underwent CPET and were followed for 35 months. We identified 202 patients N-EOV and 113 patients with EOV. Patients with EOV presented more symptoms [New York Heart Association (NYHA) III: 35% vs. N-EOV 20%, P < 0.05], worse cardiac function (LVEF: 28 ± 6 vs. N-EOV 39 ± 1, P < 0.05), higher minute ventilation/carbon dioxide production (V̇E/V̇CO2 slope: 41 ± 11 vs. N-EOV 37 ± 8, P < 0.05) and a higher rate of deaths (26% vs. N-EOV 6%, P < 0.05) and hospitalization (29% vs. N-EOV 9%, P < 0.05). Patients with P-EOV had more severe HFrEF (NYHA IV: 23% vs. D-EOV: 9%, P < 0.05), had worse cardiac function (LVEF: 24 ± 5 vs. D-EOV: 34 ± 3, P < 0.05) and had lower peak oxygen consumption (V̇O2) (12.0 ± 3.0 vs. D-EOV: 13.3 ± 3.0 mLO2 kg–1.min–1, P < 0.05). Among P-EOV, other independent predictors of mortality were V̇E/V̇CO2 slope ≥36 and V̇O2 peak ≤12 mLO2 kg–1 min–1; a V̇E/V̇CO2 slope≥34 was a significant predictor of hospitalization. Kaplan–Meier survival analysis showed that HFrEF patients with P-EOV had a higher risk of mortality and higher risk of hospitalization (P < 0.05) than patients with D-EOV and N-EOV. Conclusion In HFrEF patients, EOV persistence during exercise had a strong prognostic role. In P-EOV patients, V̇E/V̇CO2 ≥36 and V̇O2 peak ≤12 mLO2 kg–1 min–1 had a further additive negative prognostic role.