Abstract 8654: Low In-Treatment HDL Cholesterol Levels Strongly Predict New Heart Failure in Hypertensive Patients: The LIFE Study

Background: Although low HDL levels predict incident heart failure (HF) in a population-based study, whether low HDL is associated with increased risk of new HF in hypertensive patients is unclear. Methods: Incident HF was examined in relation to in-treatment HDL levels in 8445 hypertensive patients with no history of HF with baseline HDL levels randomly assigned to losartan- or atenolol-based treatment. In-treatment HDL at each year of testing was categorized into the lowest quartile vs upper 3 quartiles according to baseline HDL levels or examined as a continuous variable with hazard ratios calculated for each SD of the baseline mean lower HDL (0.44 mmol/l). Results: During 4.7±1.1 years follow-up, there were 264 new HF hospitalizations (3.1%). In univariate Cox analyses, compared with HDL ≥1.21 mmol/l, in-treatment HDL <1.21 was associated with a 117% greater risk of new HF. In parallel analyses which used in-treatment HDL as a continuos variable, each 1 SD lower HDL was associated with a 45% higher HF risk. In multivariate Cox analyses adjusting for randomized treatment, age, sex, body mass index, prevalent and history of atrial fibrillation and diabetes, history of MI, ischemic heart disease, stroke, peripheral vascular disease, smoking status, baseline serum creatinine, glucose, urine albumin/creatinine ratio as standard risk factors and for incident MI, in-treatment diastolic and systolic pressure, heart rate, QRS duration, Cornell product and Sokolow-Lyon voltage criteria for left ventricular hypertrophy and non-HDL cholesterol levels as time-varying covariates, the lowest quartile of in-treatment HDL remained associated with a 102% greater risk and, in alternative analyses, each 1 SD lower HDL treated as a continuous variable with a 38% higher risk of incident HF. Conclusions: Lower in-treatment HDL is an independent predictor of new HF in hypertensive patients, even after adjusting for other potential risk factors and treatment effects.