Systolic function recovery in Heart Failure: Frequency, prognostic impact and predictors

2019 
Abstract Background Systolic function recovery in patients with Heart failure (HF) with reduced ejection fraction (EF) is well recognized but not completely understood. We aimed to characterize HF patients with systolic function recovery, its prognostic impact and predictors. Methods We analysed patients followed in a HF clinic (2006–2015) with 2 echocardiograms performed. Partial recovery: EF recovery without attaining EF ≥ 50%; total recovery: patients reached EF ≥ 50%. Median follow-up from first echocardiogram: 69 months. Multivariate logistic regression models to determine recovery predictors. Results We analysed 304 patients with at least mild left ventricular dysfunction. During a median 34 months between echocardiogram re-evaluation 150 (49.3%) patients showed no EF recovery; 55 (18.1%) had partial recovery and 99 (32.6%) totally recovered. Mean patients age: 66; 71.1% men, high comorbidity burden; ischemic HF: 35.5%. Non-recovered patients were mostly men (80.7% vs 61.8 in partially; 61.6% in fully-recovered) with ischemic HF (46.0% vs 32.5% in partially; 21.2% in fully-recovered). Comorbidity burden, NYHA class and therapy were similar. During follow-up, 156 patients (46.7%) died. Patients with total recovery had a multivariate-adjusted 54% lower risk of dying when compared to non-recovered. Partially-recovered patients showed a non-significant adjusted 8% mortality reduction. Independent predictors of systolic function recovery were female gender(OR: 2.17, 95% CI 1.11–4.35), non-ischemic aetiology (OR: 2.78, 95% CI 1.35–5.56), and end diastolic left ventricular diameter  Conclusions HF-recovered patients were mainly women with non-ischemic HF and smaller left ventricles. These patients had significantly better prognosis than those with persistently reduced EF.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    28
    References
    0
    Citations
    NaN
    KQI
    []