Long-term outcomes in patients with non-ischemic dilated cardiomyopathy and improved ejection fraction

Heart failure with improved ejection fraction (HFimpEF) is a distinct heart failure phenotype with favorable long-term outcomes. With the optimal use of guideline-directed medical and device therapy, most patients have a sustained improvement in left ventricular ejection fraction (LVEF). However, a significant proportion of these patients also have an eventual relapse in the LVEF that predicts future cardiac events. Reliable predictors of relapse of LVEF would enable targeted optimization of therapy and improve their long-term outcomes.

It is a single-center retrospective study wherein the long-term follow-up data of patients in the non-ischemic dilated cardiomyopathy (DCM) cohort (LVEF <40% at baseline) who showed improvement in LVEF (>40% with an absolute increase in LVEF of ≥10%) was analyzed. Subgroup analyses were done based on – sex, age group, and outcome. Univariate and multivariate analyses revealed the predictors of relapse in patients with HFimpEF.

Out of the 548 patients in the non-ischemic DCM cohort, 131 (23.9%) had improvement in LVEF after a mean duration of 23.16 ± 26.06 months on guideline-directed medical therapy. In the improved LVEF group, 72 (55%) patients had a sustained recovery of ejection fraction, while 59 (45%) patients had a relapse of LVEF after a mean duration of 69.05 ± 56.04 months. The mean follow-up duration of the study was 117.75 ± 61.62 months. On univariate analysis, diabetes mellitus (p value – 0.04), new-onset renal dysfunction (p value – 0.04), wider QRS width at diagnosis (p value – 0.001) and recovery (p value – 0.013), and a higher left ventricular internal diameter at end-diastole (LVIDd) at recovery (p value <0.001) were predictors of relapse. On multivariable analysis, only diabetes mellitus (p value – 0.044) and a higher LVIDd at recovery (p value – 0.004) were independent predictors of relapse. There were 20 deaths in the study population with a higher proportion in the relapsed EF group (23.7%) compared to the sustained recovery group (8.3%).

Improvement in LVEF is possible in patients with non-ischemic DCM; however, the improvement merely denotes a transient remission in the disease process as the cardiac functions are predisposed to re-worsening. Predictors of relapse in LVEF, like diabetes mellitus and a higher LVIDd at recovery, may be targeted from a therapeutic standpoint to improve the long-term outcomes of this heart failure phenotype.