Echocardiography, Natriuretic Peptides, and Risk for Incident Heart Failure in Older Adults: The Cardiovascular Health Study
Does echocardiography or N-terminal pro-B-type natriuretic peptide (NT-proBNP) improve risk stratification above that of the Health Aging and Body Composition (ABC) heart failure (HF) risk score?
This was a secondary analysis of the Cardiovascular Health Study, which recruited Medicare patients 65-100 years old with the original aim to identify factors related to the onset of coronary disease and stroke. In this analysis, patients without symptomatic HF who had a baseline echocardiogram (done within 2 years of enrollment), NT-proBNP values, and the necessary variables for ABC score calculation were included. The variables in the ABC score include age, history of coronary heart disease, smoking status, systolic blood pressure, heart rate, serum glucose, creatinine, albumin, and left ventricular hypertrophy by electrocardiogram. The primary outcome of interest was the change in risk stratification of incident HF upon adding NT-proBNP and/or “echocardiography score” (weighted sum of significant echo parameters) to ABC scores.
Of the 5,888 patients recruited between 1989 and 1990, 3,752 were included in the analysis. At baseline, 59% (2,228) of patients were low risk for HF (<5% 5-year predicted HF risk), 24% medium (5-10% predicted risk), 12% high (10-20% predicted risk), and 5% very high risk (>20% predicted risk) for incident HF. Patients with higher ABC risk scores had more baseline echocardiogram abnormalities, including a greater prevalence of left ventricular hypertrophy, left atrial enlargement, diastolic filling abnormalities, and lower ejection fractions (all p < 0.001). Similarly, NT-proBNP levels were higher in individuals with higher ABC scores (p < 0.001). Over 5 years of follow-up, 8% of individuals had an incident HF hospitalization. An elevated NT-BNP afforded (hazard ratio 1.6 [1.4-1.8] for natural log NT-BNP) increased risk of incident HF independent of ABC risk and led to a reclassification of ABC risk category in 11%, with an increase in C index from 0.75 to 0.77 (p < 0.004). On echo, diastolic filling abnormalities, an ejection fraction <45%, left atrial enlargement, and an elevated left ventricular mass were independently predictive of incident HF. When echocardiography was added to ABC risk stratification, C index improved from 0.75 to 0.77 (p < 0.001), and 11% of patients had risk recategorized. Reclassification was best in the medium-risk categories using NT-proBNP and echo data.
The authors concluded that the addition of echocardiography and NT-proBNP improves ABC risk stratification for incident HF.
Given the medical burden of HF and its associated economic and societal impact, risk stratification is helpful for identifying those at high risk for disease. The ABC risk score is a validated model for HF risk stratification. It is limited in its utility because it is not necessarily applicable to those under 65 years of age. It was also derived in an era that did not use current evidence-based interventions for HF management in asymptomatic patients. Information on clinical functional status is lacking in a group with a high prevalence of baseline echocardiography and NT-proBNP abnormalities, making the assumption that patients were truly asymptomatic from HF difficult. Data on medication use and baseline body mass index (which affects BNP), and information on overall mortality is also lacking. Finally, adding an echocardiography score to the ABC score may make validation of this tool difficult. If validated, this tool will be of use.
Keywords: Hypertrophy, Left Ventricular, Stroke, Coronary Disease, Blood Pressure, Creatinine, Heart Rate, Cost of Illness, Smoking, Heart Diseases, Natriuretic Peptides, Body Mass Index, Heart Failure, Cardiovascular Diseases, Hospitalization, United States, Natriuretic Peptide, Brain
< Back to Listings