The association between fibrate use, change in high-density lipoprotein cholesterol, and the risk of cardiovascular disease: a retrospective chart review involving up to 8 years of follow-up.

BACKGROUND: Clinical trials have indicated that the use of fibric acid derivatives confers a benefit against cardiovascular disease (CVD) in selected populations. However, whether fibrates provide a CVD risk reduction independent of changes in the traditional lipoprotein fractions and other known CVD risk factors is not clear.

OBJECTIVE: This study examined whether the use of fibrate therapy in a general clinical setting provided cardiovascular benefits independent of changes in the traditional lipoprotein fractions.

METHODS: This was a matched, retrospective cohort study. From the electronic records of a large health maintenance organization in the northwestern United States, we identified a population that had newly initiated fibrate pharmacotherapy between January 1, 1996, and December 31, 2000. We then identified a comparator group of patients not using fibrates, matching them to fibrate users based on high-density lipoprotein cholesterol (HDL-C) and triglyceride (TG) levels, age, sex, and year of HDL-C and TG measurement. Subjects were followed until a CVD hospitalization, termination from the health plan, or December 31, 2003, whichever came first. We then used multivariate analysis accounting for differences in followup to identify predictors of CVD incidence.

RESULTS: The study population included 1722 matched pairs (56.6% male; mean [SD] age, 57.3 [11.1] years). The patients who had newly initiated fibrate pharmacotherapy had low baseline HDL-C levels (mean, 37.4 mg/dL) and very high TG levels (617 mg/dL). The 2 groups were similar overall, with the only significant differences between fibrate users and nonfibrate controls being a greater prevalence of diabetes (37.7% vs 34.3%, respectively; P=0.040) and greater use of angiotensin-converting enzyme inhibitors/angiotensin-receptor blockers (56.6% vs 51.6%, respectively; P=0.003), beta-blockers (53.7% vs 49.0%; P=0.006), calcium channel blockers (25.1% vs 20.9%; P=0.004), and niacin (11.7% vs 7.4%; P<0.001). Overall, CVD risk was 26% lower for every 5-mg/dL increment in HDL-C. After adjustment for age, sex, smoking history, diabetes, existing diagnosis of CVD, weight, systolic blood pressure, baseline HDL-C, change in HDL-C, total cholesterol, TG, and use of statins, niacin, and other CVD drugs, fibrate use did not confer an additional CVD risk reduction.

CONCLUSIONS: In this cohort with low baseline HDL-C levels and very high TG levels, fibrate use did not confer an independent CVD risk reduction after adjustment for CVD risk factors. Given the current obesity epidemic in the United States and the corresponding rise in the number of patients with the metabolic syndrome, the apparent risk reduction observed in association with higher HDL-C levels should not be ignored.