Prognostic value of myocardial work index dispersion for ventricular arrhythmias in heart failure with reduced ejection fraction

Background

Left ventricular (LV) myocardial work (MW), derived from pressure–strain analysis, integrates myocardial deformation with estimated LV pressure and provides load-adjusted indices of LV mechanics. Dispersion of segmental MW may reflect electromechanical heterogeneity associated with arrhythmogenesis. We aimed to investigate whether MW dispersion, a potential marker of ventricular arrhythmias (VAs) risk, is relevant in patients with heart failure with reduced ejection fraction (HFrEF) considered for implantation of a cardioverter-defibrillator (ICD) or cardiac resynchronization therapy (CRT).

Methods

We prospectively enrolled 96 patients with HFrEF and no history of atrial fibrillation scheduled for elective ICD/CRT implantation. All assessments, including comprehensive echocardiography, were performed the day before the procedure. Myocardial work index dispersion (MWI Dispersion) was assessed using two complementary measures: standard deviation (SD) of peak myocardial work indexes across all 17 LV segments, and Range, defined as the difference between the maximum and minimum segmental MWI values. Ventricular arrhythmias were assessed at 1 year by routine device follow-up with interrogation of the implanted ICD/CRT system.

Results

A total of 86 patients (mean age 65±12 years, 84.9% male, mean LV ejection fraction 30.4±6.6%) were included. The overall incidence of VAs (sVT/nsVT/VF) during 1 year of follow-up was 32.6%. No difference in VAs occurrence was observed between ischemic and non-ischemic HFrEF. The groups did not differ significantly in age, body mass index, or LV dimensions. Patients with VAs showed markedly impaired atrial mechanics (lower left atrial reservoir and contraction strain and higher LA stiffness). Independently, LV MW indices demonstrated greater segmental heterogeneity in the arrhythmia group, reflected by higher MW dispersion metrics (Table 1).

Mechanical dispersion of MW (MWI dispersion by SD or Range) emerged as the strongest individual predictor of VAs. In multivariable analysis, both MWI dispersion and right atrial (RA) area remained independent predictors of arrhythmic events (OR per 10 units 1.011 and 1.276, respectively). Two-variable models combining MWI dispersion (SD or Range) with RA area demonstrated the highest discriminative performance (AUC 0.912 [95% CI 0.849–0.975] and 0.911 [95% CI 0.849–0.973]) and the greatest explained variance (R² 0.583 and 0.592), outperforming more complex models (Table 2). These results indicate that LV mechanical heterogeneity, when considered together with RA remodeling, yields accurate and parsimonious risk stratification for VAs.

Conclusions

Myocardial work dispersion is a powerful marker of arrhythmic risk in HFrEF. When combined with RA area, it provides the most accurate prediction of VAs, highlighting the value of integrating LV mechanical heterogeneity with right-sided remodeling for risk stratification.