Age-specific diastolic dysfunction improves prediction of symptomatic heart failure by Stage B heart failure

Abstract

Aims: We investigated whether addition of diastolic dysfunction (DD) and longitudinal strain (LS) to Stage B heart failure (SBHF) criteria (structural or systolic abnormality) improves prediction of symptomatic HF in participants of the SCReening Evaluation of the Evolution of New Heart Failure study, a self-selected population at increased cardiovascular disease risk recruited from members of a health insurance fund in Melbourne and Shepparton, Australia. Both American Society of Echocardiography and European Association of Cardiovascular Imaging (ASE/EACVI) criteria and age-specific Atherosclerosis Risk in Communities (ARIC) study criteria, for SBHF and DD, and ARIC criteria for abnormal LS, were examined.

Methods and results: Inclusion criteria were age ≥60 years with one or more of self-reported ischaemic or other heart disease, irregular or rapid heart rhythm, cerebrovascular disease, renal impairment, or treatment for hypertension or diabetes for ≥2 years. Exclusion criteria were known HF, or ejection fraction <50% or >mild valve abnormality detected on previous echocardiography or other imaging. Echocardiography was performed in 3190 participants who were followed for a median of 3.9 (interquartile range: 3.4, 4.5) years after echocardiography. Symptomatic HF was diagnosed in 139 participants at a median of 3.1 (interquartile range: 2.1, 3.9) years after echocardiography. ARIC structural, systolic, and diastolic abnormalities predicted HF in univariate and multivariable proportional hazards analyses, whereas ASE/EACVI structural and systolic, but not diastolic, abnormalities predicted HF. ARIC and ASE/EACVI SBHF criteria predicted HF with sensitivities of 81% and 55%, specificities of 39% and 76%, and C statistics of 0.60 (95% confidence interval: 0.57, 0.64) and 0.66 (0.61, 0.71), respectively. Adding ARIC DD to SBHF increased sensitivity to 94% with specificity of 24% and C statistic of 0.59 (0.57, 0.61), whereas addition of ASE/EACVI DD to SBHF increased sensitivity to 97% but reduced specificity to 9% and the C statistic to 0.52 (0.50, 0.54, P < 0.0001). Addition of LS to ARIC or ASE/EACVI SBHF criteria had minimal impact on prediction of HF.

Conclusions: Age-specific ARIC DD criteria, but not ASE/EACVI DD criteria, predicted symptomatic HF, and addition of age-specific ARIC DD criteria to ARIC SBHF criteria improved prediction of symptomatic HF in asymptomatic individuals with cardiovascular disease risk factors. Addition of LS to ASE/EACVI or ARIC SBHF criteria did not improve prediction of symptomatic HF.