Echocardiographic post-systolic shortening patterns correlate with ventricular arrhythmias and heart failure in desmoplakin associated cardiomyopathy: a multimodality imaging study
A Salavati, G De Zan, M Perotto, S Frittella, S A Muller, A S Amin, A C Houweling, B K Velthuis, J P Van Tintelen, J Van Setten, P Van Der Harst, M Guglielmo, K Taha, A S J M Te RieleAbstract
Introduction
Desmoplakin (DSP) associated cardiomyopathy (CMP) is characterized by left ventricular (LV) systolic dysfunction with a high risk of sustained ventricular arrhythmias (VA). While strain echocardiography (TTE) has identified specific LV patterns in cardiac amyloidosis and phospholamban p.Arg14del associated CMP, data in DSP CMP are lacking. More insight into segmental LV involvement may help identify DSP CMP early in its disease course and distinguish it from other CMPs.
Purpose
To determine segmental LV abnormalities on imaging in DSP CMP and to correlate them with clinical endpoints.
Methods
Segmental LV strain analysis was performed by an experienced cardiologist on the first available TTE of retrospectively included DSP (likely)pathogenic (LP/P) variant carriers from two tertiary cardiac centers. Data on clinical follow-up, Holter monitoring, ECG and cardiac magnetic resonance (CMR) with late gadolinium enhancement (LGE) within a 1-year period from the TTE was collected. Dilated cardiomyopathy (DCM) was diagnosed per 2023 ESC guidelines and arrhythmogenic right ventricular cardiomyopathy per 2010 Task Force Criteria. Non-sustained ventricular tachycardia (nsVT) was defined as ventricular tachycardia at ≥100 bpm for a duration lasting 3-30 seconds. Circumferential ring-like LGE was defined as ≥3 contiguous LV segments with subepicardial or mid-wall LGE in a single short-axis slice. At strain analysis, post-systolic shortening (PSS) was defined as a post-systolic peak strain exceeding systolic peak strain by >10%. Left ventricular mechanical dispersion (LVMD) was defined as the standard deviation of time to peak in longitudinal strain across 16 LV segments.
Results
Fifty-five DSP LP/P carriers (60% female, 27% proband, age 46 [ IQR 28, 57] years) were included. PSS was most prevalent in the basal segments (22/55 (40%)) with the highest prevalence in the basal-anteroseptal (12/55 (22%)) and basal-inferoseptal (15/55 (27%)) segments across all individuals (Figure 1). Strikingly, almost all individuals with PSS in mid segments (12/55 (22%)) also had PSS in basal segments (11/12 (92%)). In Figure 2, the presence of mid PSS was significantly associated with HF (7/12 (58%) vs. 2/43 (5%); p < 0.001), nsVT on Holter monitoring (6/7 (86%) vs. 8/34 (24%); p = 0.006), T-wave inversion in V6 (4/12 (33%) vs. 0/39 (0%); p = 0.002), T-wave inversion in V5 (4/12 (33%) vs. 1/39 (3%); p = 0.010), and DCM diagnosis (11/12 (92%) vs. 4/42 (10%); p < 0.001). Among the 48 (87%) individuals with available CMR, ring-like LGE was significantly associated with a higher LVMD (43.00 ms [36.00, 53.00] vs. 31.00 ms [22.50, 40.25]; p = 0.001).
Conclusion
PSS in DSP CMP is most common in basal segments, while almost all individuals with mid PSS also had basal PSS. Presence of mid PSS was significantly associated with abnormal repolarization, VA and HF. Ring-like LGE was significantly associated with higher LVMD.Figure 1.PSS localizationFor image description, please refer to the figure legend and surrounding text.Figure 2.Population characteristicsFor image description, please refer to the figure legend and surrounding text.