AIMS: In patients with recently diagnosed non-ischaemic LV systolic dysfunction, left ventricular reverse remodelling (LVRR) and favourable prognosis has been documented in studies with short-term follow-up. The aim of our study was to assess the long-term clinical course and stability of LVRR in these patients. METHODS AND RESULTS: We prospectively studied 133 patients (37 women; 55 [interquartile range 46, 61] years) with recently diagnosed unexplained LV systolic dysfunction, with heart failure symptoms lasting <6 months and LV ejection fraction <40% persisting after at least 1 week of therapy. All patients underwent endomyocardial biopsy (EMB) at the time of diagnosis and serial echocardiographic and clinical follow-up over 5 years. LVRR was defined as the combined presence of (1) LVEF ≥ 50% or increase in LVEF ≥ 10% points and (2) decrease in LV end-diastolic diameter index (LVEDDi) ≥ 10% or (3) LVEDDi ≤ 33 mm/m2. LVRR was observed in 46% patients at 1 year, in 60% at 2 years and 50% at 5 years. Additionally, 2% of patients underwent heart transplantation and 12% experienced heart failure hospitalization. During 5-year follow-up, 23 (17%) of the study cohort died. In multivariate analysis, independent predictors of mortality were baseline right atrial size (OR 1.097, CI 1.007-1.196), logBNP level (OR 2.02, CI 1.14-3.56), and PR interval (OR 1.02, CI 1.006-1.035) (P < 0.05 for all). The number of macrophages on EMB was associated with overall survival in univariate analysis only. LVRR at 1 year of follow-up was associated with a lower rate of mortality and heart failure hospitalization (P = 0.025). In multivariate analysis, independent predictors of LVRR were left ventricular end-diastolic volume index (OR 0.97, CI 0.946-0.988), LVEF (OR 0.89, CI 0.83-0.96), and diastolic blood pressure (OR 1.04, CI 1.01-1.08) (P < 0.05 for all). CONCLUSIONS: LVRR occurs in over half of patients with recent onset unexplained LV systolic dysfunction during first 2 years of optimally guided heart failure therapy and then remains relatively stable during 5-year follow-up. Normalization of adverse LV remodelling corresponds to a low rate of mortality and heart failure hospitalizations during long-term follow-up.
