Fibrate: The Treatment of Choice for Preventing Coronary Heart Disease in Diabetic Patients?
This report was reviewed for medical and scientific accuracy by Louis L. Battey, MD, SACC , Cardiology of Georgia.
It is firmly established that diabetes is associated with a marked increase in coronary heart disease (CHD). Studies have shown that patients with type 2 diabetes are characterized by a "diabetic dyslipidemia" of high triglyceride (TG) levels and decreased high density lipoprotein cholesterol (HDL-C). Because fibrates have been shown to be highly effective in increasing HDL-C, these agents may be more important than low density lipoprotein cholesterol (LDL-C)- lowering agents (e.g., statins) in this patient population. Results from the DAIS (Diabetes Atherosclerosis Intervention Study) trial demonstrate that fenofibrate significantly delays the progression of atherosclerosis in patients with type 2 diabetes, even if lipid levels are in a range considered. A subanalysis from VA-HIT (Veterans Affairs HDL Cholesterol Intervention Trial) showed that gemfibrozil significantly reduced cardiovascular (CV) events in patients with diabetes, hyperinsulinemia (HI), or both, but had a minimal impact on patients with neither condition. The VA-HIT investigators stated that virtually all of the benefit of fibrate treatment in VA-HIT can be attributed to the inclusion of patients with low HDL among whom the prevalence of metabolic syndrome-also known as "Syndrome X," which involves high levels of triglycerides, low levels of HDL, high blood pressure, and glucose intolerance-was high.
The DAIS Trial: Fibrate Reduces Atherosclerosis Progression
According to George Steiner, MD, Professor of Medicine and Physiology, University of Toronto, Ontario, the DAIS trial was designed specifically to determine by quantitative angiography whether correcting lipid abnormalities with fenofibrate in type 2 diabetes alters the progression or regression of CHD. "It is because we wanted to know both progression and regression questions that we had as one of our entry criteria to have at least one minimal lesion visible on coronary angiogram," Dr. Steiner told conference delegates.
DAIS was a 3-year, double-blind, randomized, placebo-controlled study on the use of 200 mg/day micronized fenofibrate. The study enrolled 418 patients with type 2 diabetes whose mean glycemic control (HbA1c) was 7.53 ± 1.19%. "This was not a study directed at glycemic control, it was a study directed at lipids," Dr. Steiner commented.
Baseline fasting glucose levels were slightly higher in the placebo group. However, Dr. Steiner noted, "when we looked at the final data, this difference had no impact on the final conclusions. The patients were in good glycemic control at baseline."
Lipid levels were similar between both groups and were typical of values observed in other diabetes trials, such as the United Kingdom Prospective Diabetes Study (UKPDS). Angiographic parameters included a measure of diffuse disease (mean segment diameter near the lesion) and 2 measures of focal disease (minimum lumen diameter and % diameter stenosis at the lesion). Reporting on the angiographic outcomes, Dr. Steiner said, "there was deterioration in both arms but less deterioration in those taking fenofibrate."
Commenting on the angiographic data, Dr. Steiner said, "fenofibrate is associated with a 40-42% less progression of focal coronary disease. It is also associated with a similar trend in parameters indicating diffuse disease. This is a pattern similar to that seen in other angiographic studies being conducted in nondiabetic patients." Although DAIS was not powered to examine clinical endpoints, Dr. Steiner showed delegates that there was a lower mortality, lower myocardial infarction (MI) rate, less angioplasty, and less coronary artery bypass grafting (CABG) in patients taking fenofibrate. "The only parameter in which there was no difference was in hospitalized angina-the softest and most subjective on the clinical endpoints," he stated.
On the combined clinical endpoints, there was a 23% reduction with fenofibrate treatment. "This was not significant, but, again, we were not designed to look at clinical endpoints," Dr. Steiner cautioned. Nevertheless, this statistic is comparable with the reduction in cardiovascular (CV) events found in diabetic subpopulations in several secondary intervention statin trials.
Commenting on the clinical implications of this trial, Dr. Steiner said, "First of all, in men and women with type 2 diabetes, lipid levels should be tested annually; secondly, these abnormalities must be treated, even if they are very mild; thirdly, if a healthy lifestyle and good glycemic control do not correct lipid abnormalities by themselves, then drugs should be added. Drugs should be appropriate to the nature of the lipid abnormality, but as you saw in the typical lipid abnormalities of type 2 diabetes, fenofibrate was extremely effective."
The VA-HIT Trial: Fibrate Increases HDL, Decreases Cardiovascular Events
The VA-HIT trial was a multicenter, double-blind, placebo-controlled trial of men with CHD who had low HDL-C levels (mean 0.80 mmol/L) and "low-risk" levels of LDL-C (mean 2.77 mmol/L) in which the effectiveness of gemfibrozil in reducing the risk of major CV events was examined.
According to Dr. Sandor Robins, Professor of Medicine, Boston University School of Medicine, Massachusetts, "We demonstrated [in VA-HIT] that treatment with the fibric acid gemfibrozil in a standard dose resulted in a significant 22% reduction in the 5-year incidence of non-fatal MI or CHD death, in conjunction with an increase in HDL-C, a decrease in triglycerides (TG), and no change in LDL-C." Perhaps because VA-HIT was specifically seeking individuals with low HDL-C levels, it was inevitable that those with associated conditions would be enrolled- 25% of the subjects had diabetes and the high risk metabolic syndrome (i.e. 63% had a body mass index [BMI] ≥27; 53% had a waist circumference ≥100 cm; mean glucose was 115 mg/dL, 32% had a fasting insulin > 30 µU/mL (defines the 90th percentile in Framingham study), 25% had a history of diabetes, and 57% had a history of hypertension. Importantly, the high risk associated with this syndrome is illustrated in the nearly 22% 5-year CV event rate observed in the VA-HIT placebo group.
In addition to the significant 22% decrease in nonfatal MI/CHD death seen with gemfibrozil, CHD events with active treatment were (1) inversely correlated with HDL-C achieved, (2) reduced by 17% with a 0.13 mmol/L increase in HDL-C, (3) maximally reduced by maximum increases in HDL-C, (4) reduced independently of on-trial TG or LDL-C, and (5) reduced especially in high risk patients with diabetes or hyperinsulinemia (HI).
"CHD events in VA-HIT were more greatly reduced with gemfibrozil more than with placebo at the same on-trial values of HDL-C, which is consistent with other benefits of fibrates and implies that other kinds of therapy or lifestyle changes that increase HDL-C may not necessarily result in a similar CHD benefit," Dr. Robins said.
Greater Benefit in Diabetic Patients
Diabetic patients in VA-HIT were similar to those in the Cholesterol and Recurrent Events (CARE) study on pravastatin (Pravachol®) with regard to glucose levels, TG levels, and total cholesterol/HDL-C ratios. However, diabetics in VA-HIT had lower levels of HDL-C (0.80 vs. 0.98 mmol/L) and lower levels of LDL-C (2.77 vs. 3.52 mmol/L) than those in CARE.
In terms of treatment influences on CHD events in diabetic patients versus non- diabetic patients, an interesting difference between the two groups emerges. Although non-diabetic patients in VA-HIT and CARE received a similar benefit from fibrate or statin treatment (absolute risk reduction [ARR] = 3.3% vs. 3.1%, respectively), diabetic patients in VA-HIT received a greater benefit with gemfibrozil (ARR = 7.6%) than did diabetic patients treated with pravastatin in CARE [ARR = 2.7%]. VA-HIT also showed that a smaller number of patients needed treatment to prevent 1 event as compared to 2 statin trials (VA-HIT: 13, CARE: 37, LIPID: 28). Dr. Robins explained, "compared to other large lipid intervention trials with clinical endpoints, patients in VA-HIT had lower HDL-C, a lower total cholesterol and LDL-C at entry, were more overweight with increased abdominal obesity, more diabetes, and more hypertension, and reflect the present day profile of a sizeable number of men-with CHD at high risk for recurrent CHD events-not yet shown to be benefited by statin therapy."
Although the focus of VA-HIT was the treatment of patients with low HDL-C, the trial shows that fibrate therapy has a special role in the treatment of a high risk category of patients with type 2 diabetes, low HDL-C, abdominal obesity, and insulin resistance, who also have diabetic dyslipidemia and type 2 diabetes.
In addition to diabetics, 6% of patients enrolled in VA-HIT had glucose levels in the diabetes range, and an additional 16% did not have diabetes but did have HI. "These three subgroups of patients made up almost 50% of our VA-HIT participants," Dr. Robins noted.
Importantly, patients with diabetes but without HI and patients without diabetes but with HI had comparable CV event rates.
Most strikingly, Dr. Robins presented preliminary data showing that of the 1,334 patients enrolled in VA-HIT without diabetes or HI, more patients in the placebo group than in the gemfibrozil group developed either new diabetes or HI by 1 year. Also, more patients in the placebo group than in the gemfibrozil group developed new CHD with newly diagnosed diabetes or HI.
Dr. Robins commented, "Even without HI or diabetes at entry into this trial, subjects were at high risk for developing diabetes or HI during the course of this trial and, consequently, should benefit from fibrate therapy." He concluded, "Fibrate therapy should be used in all patients with CHD, a low HDL, a low LDL, and evidence of a metabolic syndrome to reduce the incidence of major CHD events. Fibrate therapy may even delay the onset of HI and diabetes."