Global Longitudinal Strain in Prediction of Incident Heart Failure
- In a cohort of 477 patients with risk factors for heart failure, abnormal LV global longitudinal strain (GLS) was associated with a higher risk of incident heart failure and death.
- In receiver-operating characteristic analysis, GLS was the only echocardiographic variable that significantly improved area under the curve for heart failure prediction when added to a purely clinical model (0.81 vs. 0.77, p = 0.04).
Among patients at high risk of developing clinical heart failure (HF), how does echocardiographic assessment of left ventricular (LV) global longitudinal strain (GLS) impact prognostication?
This single-center cohort study included cardiac inpatients ≥45 years of age with risk factors for HF, including diabetes mellitus, coronary artery disease, and hypertension. Patients with clinical HF and those with valve disease of moderate or greater severity were excluded. All subjects underwent transthoracic echocardiography with assessment of LV GLS by two-dimensional speckle tracking. Study outcomes were HF-related hospitalization and all-cause mortality.
A total of 477 patients (72% male, mean age 65 years, 49% with prior coronary revascularization) were included in the analysis. After a median follow-up of 9 years, 116 (26%) were hospitalized for HF or died. Among patients with normal GLS (absolute value >18%), 3% developed HF and 11% died. In patients with abnormal GLS (absolute value <16%), 19% developed HF and 29% died. In multivariable models, GLS and other echocardiographic variables such as LV ejection fraction, LV mass index, left atrial volume index, and mitral E/e’ all provided incremental value in predicting HF. However, in receiver-operating characteristic analysis, GLS was the only echocardiographic variable that significantly improved area under the curve for HF prediction when added to a purely clinical model (0.81 vs. 0.77, p = 0.04).
The authors concluded that LV GLS provides incremental value in prediction of incident HF, beyond traditional echocardiographic parameters.
A large body of literature has illustrated the prognostic value of LV GLS. GLS assessment can be challenging in daily practice because of limited image quality and time constraints. Even with semi-automated software, contour editing may be needed to ensure accurate endocardial tracking, and GLS values should not be reported if tracking is inadequate. Nonetheless, GLS can be very valuable in identifying subclinical LV systolic dysfunction, and the real-world utility of this parameter will likely increase as more laboratories become facile with image acquisition and analysis. Right ventricular free wall strain and left atrial reservoir strain assessment are also likely to become more prevalent in clinical practice now that automated software packages are available.
Clinical Topics: Cardiac Surgery, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Prevention, Atherosclerotic Disease (CAD/PAD), Cardiac Surgery and Arrhythmias, Cardiac Surgery and Heart Failure, Acute Heart Failure, Interventions and Coronary Artery Disease, Interventions and Imaging, Echocardiography/Ultrasound, Hypertension
Keywords: Coronary Artery Disease, Diabetes Mellitus, Diagnostic Imaging, Echocardiography, Heart Failure, Hypertension, Inpatients, Myocardial Revascularization, Risk Factors, Secondary Prevention, Stroke Volume
< Back to Listings