Two dimensional strain echocardiography for risk stratification in Desmoplakin cardiomyopathy

Background

Desmoplakin gene (DSP) mutation-associated cardiomyopathy carries a high risk of life-threatening ventricular arrhythmias (LTVA) and heart failure (HF) and can present as arrhythmogenic right ventricular (RV) cardiomyopathy (ARVC), left-ventricular (LV) dominant arrhythmogenic cardiomyopathy, or biventricular involvement. A multiparametric DSP-related VA risk score keeps a wide range of incertain risk prediction, highlighting the need for additional risk markers.

Purpose

To evaluate the prognostic value of 2D-strain echocardiography-derived parameters on the risk of LTVA and HF in DSP cardiomyopathy patients.

Methods

This was a retrospective monocentric study that included all DSP pathogenic or likely pathogenic variant carriers with an adequate window at baseline echocardiography for 2D-strain analysis. Offline 2d-strain analysis consisted in LV global longitudinal strain (GLS), RV free wall GLS, and LV and RV mechanical dispersion (MD) measurements. 2D-strain and conventional clinical and imaging parameters association with the risk of presenting LTVA, defined by sudden cardiac death (SCD), aborted SCD, appropriate implantable defirbirllator therapy or sustained ventricular tachycardia, or a HF hospitalisation at follow-up was analyzed.

Results

Of the 124 patients identified with a DSP pathogenic or probably pathogenic variant, 79 had an echocardiography at baseline suitable for offline 2D-strain analysis and were therefore included in the study. 28 (36%) were probands and 51 (64%) relatives, among 47 families. Mean age at inclusion was 41 ± 18 years, with female predominance (54%).

13 patients (16.5%) had an ARVC phenotype, 20 (25 %) a LV-dominant phenotype according to the Padua criteria, 34 (43%) biventricular involvement, and 12 (15%) no cardiomyopathy phenotype.

Over a median follow-up of 55 months [29,5;81] , 9 patients (11%) experienced a LTVA episode, and 6 (8%) HF. LVEF < 50%, LV GLS and LV MD were echocardiographic parameters significantly associated with the occurence of LTVA or HF at follow-up, contrarily to RV function and RV strain parameters, presence of late gadolinium enhancement (LGE) at baseline cardiac magnetic resonance as well as the type of DSP cardiomyopathy phenotype (Table 1). LV GLS > -16.3% and LV MD > 55 ms were the best thresholds to differentiate patients presenting a LTVA event during follow-up, with respectively 89% and 78% sensitivity, 61% and 76.5% specificity, AUC 0.72 [0.56;0.88 CI] and 0.78 [0.63;0.93 CI].

Kaplan-Meier event-free curves for LV LGS and MD thresholds are presented in Figure 1.

Conclusion

Neither the type of cardiomyopathy phenotype, nor RV dysfunction and strain parameters, and LGE presence at diagnosis were associated with the occurrence of LTVA or HF during follow-up in DSP cardiomyopathy patients. On the contrary, LV strain-derived parameters GLS and MD seemed to be markers of interest for better risk stratification on top of LVEF.Table 1For image description, please refer to the figure legend and surrounding text.Figure 1For image description, please refer to the figure legend and surrounding text.