Objective: Heart failure (HF) is a leading cause of mortality especially in older populations. Early detection of high-risk individuals is imperative for primary prevention. The purpose of this study was to develop a HF risk model from a population without clinical cardiac disease.
Methods: The Multi-Ethnic Study of Atherosclerosis is a multicentre observational cohort study following 6814 subjects (mean age 62±10 years; 47% men) who were free of clinical cardiovascular disease at baseline. Median follow-up was 4.7 years. HF events developed in 176 participants. Cox proportional hazards models and regression coefficients were used to determine independent risk factors and generate a 5-year risk score for incident HF. Bootstrapping with bias correction was used for internal validation.
Results: Independent predictors for HF (HR, p value) were age (1.30 (1.10 to 1.50) per 10 years), male gender (2.27 (1.53 to 3.36)), current smoking (1.97 (1.15 to 3.36)), body mass index (1.40 (1.10 to 1.80) per 5 kg/m2), systolic blood pressure (1.10 (1.00 to 1.10) per 10 mm Hg), heart rate (1.30) (1.10 to 1.40) per 10 bpm), diabetes (2.27 (1.48 to 3.47)), N-terminal pro-B-type natriuretic peptide (NT proBNP) (2.48 (2.16 to 2.84) per unit log increment) and left ventricular mass index (1.40 (1.30 to 1.40) per 10 g/m2). A parsimonious model based on age, gender, body mass index, smoking status, systolic blood pressure, heart rate, diabetes and NT proBNP natriuretic peptide predicted incident HF risk with a c-statistic of 0.87.
Conclusions: A clinical algorithm based on risk factors readily available in the primary care setting can used to identify individuals with high likelihood of developing HF without pre-existing cardiac disease.