Right ventricular global longitudinal strain is associated with cardiovascular outcome in left ventricular non-compaction phenotype

Abstract Introduction Speckle-tracking echocardiography is an increasingly important tool for assessing left ventricular non-compaction (LVNC) phenotype. This study aims to analyse right (RV) and left (LV) ventricular global longitudinal strain (GLS) in patients with LVNC and explore their association with CV outcome during long-term follow-up. Methods From our prospective LVNC cohort study, 40 patients were identified with a comprehensive echocardiographic examination including strain analysis of both ventricles. Strain was determined using TomTec ImageArena v.4.6. A combined endpoint (CV event) was defined to include: atrial flutter/fibrillation, sustained ventricular arrhythmias, cardiac syncope, thromboembolic event, heart failure progression, aborted cardiac arrest, cardiovascular death or heart transplantation. Results Over a median follow-up duration of 8 [6–10] years, 23 (58%) of the patients experienced one or more CV event (event group); while the rest (17 [42%]) did not (no event group). Baseline clinical characteristics of the two groups were comparable except for a significantly older age (P<0.01) and a higher BMI (P=0.033) in the event group. RVGLS (Figure 1) was significantly lower in the event group (-13.9[-17.1 – -10.3]%) compared to no-event group (-17.3[-18.8 – -15.1]%, P=0.019). While LVGLS did not show a significant difference (P=0.112), LVEF was significantly lower (P=0.011) in the event group. Other echocardiographic parameters were comparable between the groups. RVGLS > -14.3% (sens. 61%, spec. 88%, AUC 72%; P = 0.010) was applied as cut-off to identify patients with a higher risk of CV events during follow-up. Kaplan-Meier curves revealed a higher probability of CV events when RVGLS was > -14.3% (ꭓ2 = 10.28, P = 0.001; Figure 2). Univariable regression models showed that RVGLS (HR 1.10; P = 0.013), LVGLS (HR 1.23; P = 0.045), and LVEF (HR 0.95; P = 0.005) were associated with an increased risk of CV events. The association of RVGLS (OR 1.19, P = 0.047) remained independent of LV GLS and LVEF in a multivariable logistic regression analysis. When RVGLS was included to the LV function model, a significant improvement was noted (ꭓ2 from 7.31 to 11.58; ANOVA P = 0.009). Conclusions RVGLS differentiated LVNC patients suffering a CV event from those who did not and provided incremental value to the association of LV function (LVEF and LVGLS) with an increased risk of CV events. These findings highlight the potential value of RVGLS for evaluating long-term outcome during follow-up of LVNC patients.