Log in to MyACC
Clinical and Echocardiographic Assessment of Heart Failure Risk
Quick Takes
- The ARIC-HF score adequately identifies patients at high- and low-risk for symptomatic HF in three Australian cohorts of patients with HF risk factors.
- Addition of echocardiographic parameters provides significant incremental value to ARIC-HF in patients classified as intermediate clinical risk.
- Specifically, normal GLS and diastolic function reclassify more than half of patients designated as intermediate-risk by ARIC-HF into a new low-risk category with an HF incidence of only 12%.
Study Questions:
How can echocardiographic parameters be used to refine the risk of heart failure (HF), as determined by a validated clinical risk score, ARIC-HF (Atherosclerosis Risk in Communities Heart Failure)?
Methods:
The same clinical and echocardiographic evaluation was performed in two groups: 1) a derivation group composed of 980 patients with HF risk factors from two different Australian trials, and 2) a validation group composed of 355 patients at risk for HF with normal or mildly reduced ejection fraction (EF) (≥40%). In the first group, inclusion criteria were age ≥65 years, hypertension (HTN), type 2 diabetes mellitus (DM), and body mass index (BMI) ≥30 kg/m2, and exclusion criteria were a prior diagnosis or treatment for HF, ≥ moderate valvular disease, known coronary artery disease (CAD), or ischemic heart disease. In the second group, inclusion criteria included HTN, DM, and CAD, while exclusion criteria included clinical HF, congenital heart disease, surgically reparable or significant valve disease, or terminal condition at the time of index admission.
For all patients, the ARIC-HF risk score was calculated based on age, smoking status, race, BMI, blood pressure, heart rate, ischemic heart disease, HTN, DM, functional status (based on DASI questionnaire), and N-terminal pro–B-type natriuretic peptide level. Patients were categorized by ARIC-HF as low (score <9%), intermediate (9-33%), or high (>33%) risk for developing HF at 4 years. Echocardiographic parameters obtained on all patients included global longitudinal strain (GLS), left ventricular hypertrophy (LVH), diastolic function (DD), and left atrial enlargement (LAVi >34 mL/m2). LV dysfunction was defined as GLS ≤ -16%, DD, or LVH.
Follow-up in the derivation group was by questionnaires, study physician review every 6-12 months for 2 years, and telephone contact up to a maximum of 7 years. Follow-up in the validation group was via the Centre for Victorian Data Linkage for a period of up to 10 years from recruitment (median follow-up, 9 years). A risk stratification algorithm based on clinical risk and echocardiographic markers of LV dysfunction was developed using a classification and regression tree (CART) analysis. The ability of echocardiographic parameters to reclassify patients risk stratified using ARIC-HF was also evaluated.
Results:
In the derivation group, 12% developed stage C HF over a median follow-up of 3.4 (interquartile range, 1.3-5.3) years, including 9%, 18%, and 73% of low-, intermediate-, and high-risk patients by ARIC-HF score. Integration of echocardiographic parameters was not deemed to be necessary in the high clinical risk group, but was found to have utility in low and intermediate clinical risk groups. Specifically, the presence of abnormal GLS and LVH in low-risk patients, albeit rare (only 2.2% of low clinical risk group), reclassified these patients as high-risk, with a HF incidence of 36%. Likewise, normal GLS and DD reclassified 61% of intermediate clinical risk patients as low-risk, with an HF incidence of only 12%. Abnormal GLS was the strongest independent predictor of HF (hazard ratio, 2.92; 95% confidence interval [CI], 1.95-4.37).
In the validation cohort, 11% developed HF over the course of 4.5 years, including 4%, 17%, and 39% of low, intermediate, and high clinical risk patients. In this group, echo parameters were only useful in the intermediate clinical risk group, reclassifying a comparable proportion of these patients into a low-risk group. In general, echo parameters provided significant incremental value to the ARIC-HF score in prediction of new HF admission with increase in C-statistic from 0.78 (95% CI, 0.71-0.84) to 0.83 (95% CI, 0.77-0.88; p = 0.027).
Conclusions:
In asymptomatic patients with HF risk factors but LVEF ≥40%, no known CAD, and no significant valvular disease, risk stratification with the ARIC-HF risk score was adequate to classify patients as low- or high-risk of incident HF. Echocardiographic parameters were of greatest utility in patients classified as intermediate-risk by the clinical risk score, with normal GLS and diastolic function reclassifying 61% of these patients into a low-risk group.
Perspective:
Risk stratification of asymptomatic patients with HF risk factors is a necessary first step in developing interventions to prevent or delay progression to overt HF. This manuscript finds that a previously validated clinical risk score, ARIC-HF, adequately identifies high- and low-risk patients from three Australian cohorts, albeit with less discriminative power than in previous reports. The primary finding is that echocardiographic parameters (normal GLS and diastolic function) effectively reclassify more than half of patients labeled intermediate-risk by the clinical score alone. At the same time, it is not clear that this new algorithm is ready for widespread implementation. ARIC-HF is not the only clinical risk score and may be inferior to others in both its performance and lack of sex- and race-specific risk estimates. Likewise, the algorithm does not incorporate biomarkers—in part because natriuretic peptide testing was incomplete in the derivation group. Finally, all three of the cohorts used in the study are somewhat dated, such that no patients were treated with newer guideline-directed medical therapies, such as angiotensin receptor/neprilysin inhibitors (ARNIs) or sodium-glucose cotransporter 2 (SGLT2) inhibitors, which may modify the relationship between risk factors and the likelihood of eventual HF.
Clinical Topics: Heart Failure and Cardiomyopathies, Noninvasive Imaging, Prevention, Valvular Heart Disease, Atherosclerotic Disease (CAD/PAD), Acute Heart Failure, Heart Failure and Cardiac Biomarkers, Echocardiography/Ultrasound, Hypertension
Keywords: Atherosclerosis, Body Mass Index, Coronary Artery Disease, Diabetes Mellitus, Type 2, Diagnostic Imaging, Echocardiography, Heart Failure, Heart Valve Diseases, Hypertension, Hypertrophy, Myocardial Ischemia, Natriuretic Peptide, Brain, Risk, Risk Assessment, Secondary Prevention, Stroke Volume, Ventricular Dysfunction, Left
< Back to Listings
