Prognostic value of left ventricular global longitudinal strain for adverse outcomes in patients with left ventricular thrombus and preserved left ventricular ejection fraction
W C Ng, A S T Leow, B Y Q Tan, L L L Yeo, W K F Kong, K K Poh, R C C Wong, P Chai, T C Yeo, M Y Y Chan, C H SiaAbstract
Background
Left ventricular thrombus (LVT) is associated with high thromboembolic and mortality risk, and is typically diagnosed post-myocardial infarction or in patients with reduced left ventricular ejection fraction (LVEF). The clinical trajectory of LVT patients with preserved LVEF is less well-studied, hence our study aimed to characterise this cohort of patients and to evaluate the prognostic significance of left ventricular global longitudinal strain (LVGLS), a marker of subclinical systolic dysfunction not captured by LVEF.
Methods
This was a retrospective cohort study of 135 patients with transthoracic echocardiographically (TTE) confirmed LVT and LVEF ≥40% at a tertiary centre between 2011 and 2021. Patients were stratified by LVGLS on initial TTE into abnormal (less negative than -16%; n=108) and normal (more negative than -16%; n=27) groups for analysis. The primary endpoint was a composite outcome including stroke/transient ischaemic attack, acute myocardial infarction, acute decompensated heart failure, and all-cause mortality on follow-up.
Results
Among 135 LVT patients with preserved ejection fraction (mean age 58.4 ± 13.9 years, 83.0% male), 20.0% (n = 27) had abnormal LVGLS of less negative than -16%. Baseline demographics, co-morbidities, LVT aetiology and characteristics on TTE, and medical therapy including anticoagulation were similar across both groups. Patients with abnormal LVGLS had lower initial LVEF (44.7 ± 5.4% vs. 49.3 ± 8.7%, p <0.001), and encountered higher incidence of composite outcome (47.2% [n = 56] vs. 18.5% [n = 5], p = 0.007). Kaplan-Meier analysis demonstrated significantly lower event-free survival for composite outcome in the abnormal LVGLS group (log-rank p=0.021). On Cox proportional hazards regression model, less negative LVGLS (per %) was independently associated with higher hazard of composite outcome (aHR 1.10, 95% CI 1.01 – 1.21, p = 0.033) after adjusting for age, sex, ischaemic heart disease, haemoglobin, and estimated glomerular filtration rate, across a mean follow-up duration of 5.2 ± 3.4 years. LVGLS was not associated with all-cause mortality or LVT resolution.
Conclusion
Abnormal LVGLS was associated with higher risk of adverse cardiovascular events in LVT patients with LVEF ≥40%, and provided incremental prognostic information beyond LVEF, aiding risk stratification.Table 1