Development of a new score to predict left ventricular reverse remodelling in patients with nonischemic cardiomyopathy

Abstract

Background and purpose Implantable cardioverter-defibrillator is indicated in patients with dilated non-ischemic cardiomyopathy (DCM) and severely depressed left ventricular ejection fraction (LVEF) after a wait-and-see period of 3–9 months under optimised medical therapy. However, in the first 6 months after the disease debut, around 2% of patients might suffer life-threatening arrhythmias. The appearance of left ventricular reverse remodelling (LVRR) in patients with DCM is associated with a lower incidence of ventricular tachyarrhythmias. Therefore, it is relevant an early identification of the patients who will experience LVRR during the follow-up. Our aim was to develop a score to predict the appearance of LVRR in patients with DCM. Methods From 2014 to 2021, 201 patients with DCM and LVEF ≤45% were prospectively evaluated in our tertiary care hospital. All patients underwent a transthoracic echocardiogram and 1.5 Tesla scanner cardiac magnetic resonance (MR) as part of the diagnostic workup. LVRR was defined as an increase in LVEF ≥10 points or absolute LVEF ≥50% associated with a reduction in left ventricular end-diastolic diameter ≥10%. Results The median age of our cohort (n=201) was 61.6 (14.7) years, and 68% were male. Most patients (>90%) were treated with beta-blockers or RASS blockers, and 72% with mineralocorticoid receptor antagonists. During a mean follow-up period of 37.6 (33.9) months 45% of patients had LVRR. Patients with LVRR had a lower cardiovascular mortality (3.33 vs 9.59%; p=0.153), lower mortality due to heart failure (0% vs 8.22%; p=0.023), and a lower incidence of ventricular tachyarrhythmias (1.67% vs 19.18%; p=0.001). Table 1 shows the echocardiographic, MR and clinical characteristics of patients who experienced LVRR. Variables significantly associated with LVRR in the univariable analysis and considered clinically relevant were included in a multivariable logistic regression analysis. The final model included the presence of right ventricular end systolic volume index (RVESVi) >50 ml/m2 (2 points), left bundle brach block (LBBB) echo pattern (1 point), female gender (1 point) and tachycardiomyopathy/idiopathic/alcoholic/chemotherapy induced cardiomyopathy as the potencial cause of DCM (1 point). The score showed a good discrimination, with an area under the ROC curve of 0.82 (95% CI 0.69 to 0.94), 84% sensitivity and 80% specificity. The presence of 3 or more points was associated with a high probability to had LVRR (0 points: 1%; 1 points: 17%; 2 points: 38%; 3 points: 64%; 4 points: 84%%; 5 points: 94% and 6 points: 98%) (Figure 1). Conclusion A new score with four variables (RVESV, LBBB echo pattern, female gender and tachycardiomyopathy/idiopatic/alcoholic/chemotheapy induced cardiomyopathy as potential cause) accurately predicts the probability of LVRR. Considering patients who experience LVRR have less cardiovascular events, this score may be a helpful tool for patients' risk stratification. Funding Acknowledgement Type of funding sources: None.