Optimizing Treatment of Diabetic Dyslipidemia

At an industry-sponsored symposium held in conjunction with the American Association of Clinical Endocrinologists’ 2003 Annual Meeting and Clinical Congress, three leaders in diabetes research and treatment presented new information on optimizing treatment in patients with diabetic dyslipidemia.

This program was supported by an unrestricted educational grant from KOS Pharmaceuticals, Inc.

Facing the Challenges of the Metabolic Syndrome and Diabetes

The American Diabetes Association considers diabetes a coronary heart disease (CHD) equivalent. This is based on three observations, according to Dr. Steven M. Haffner, MD, Professor, Department of Medicine at the University of Texas Health Sciences Center at San Antonio.

The first is that the risk of CHD is similar in diabetic subjects who do not have pre-existing heart disease as in non-diabetic subjects with heart disease. A large, multinational study showed both diabetes and CHD resulted in a three-fold increased risk, diabetes alone had a two-fold increase and those with just cardiovascular disease had 70% excess risk (Malmberg K, et al. Circulation. 2000; 102:1014).

Another study from Stockholm looked at patients with confirmed myocardial infarctions (MI) and no history of type 2 Diabetes. Each was given a glucose tolerance test. Two out of every three patients had some form of glucose intolerance (Norhammer A, et al. Lancet. 2002;359:2140) (Figure 1).

The second observation is that intensive glycemic control does not eliminate the excess risk of CHD in diabetic patients. Data from the United Kingdom Prospective Diabetes Study (UKPDS) showed a ten-fold increase in microvascular complications from hemoglobin A1C levels of 51¼2 to 11. (Adler AL, et al. BMJ. 2000;321:412)

“It is very unlikely that we can get most patients from a hemoglobin A1C of 11% to 5.5%,” said Dr. Haffner. “We cannot just rely on glycemic control,
although it is important.”

The San Antonio Heart Study followed a group with normal glucose tolerance until they developed type 2 diabetes. Those who converted had higher baseline triglycerides, lower HDLs, higher systolic blood pressures, higher fasting glucose as well as higher insulin concentrations (Haffner SM, et al. JAMA. 1990;263:2893; Stratton IM, et al. BMJ. 2000;321:405).

“This sounds remarkably like the metabolic syndrome, although it occurred long before the National Cholesterol Education Program (NCEP) definitions,” said Dr. Haffner.

Similar results were seen in the Nurses Health Study. Of special interest to Dr. Haffner was the finding that those with CHD prior to the onset of Type-2 diabetes still had a 2.8 fold increase in risk (Hu FB, et al. Diabetes Care. 2002; 25:1129)

“These studies are important because they show screening for diabetes in those with CHD existing prior to diabetes is not going to eliminate the increased risk,” said Dr. Haffner. “Only diabetes prevention strategies will be useful.”

Are lipid interventions to prevent CHD equally effective in diabetic and non-diabetic subjects? Type 2 diabetics do not have particularly high LDL cholesterol levels by conventional measures. They do tend to have smaller, denser and potentially more atherogenic LDL (Mykkanen L, et al. Atherosclerosis. 1991;88:153).

Baseline predictors of CHD in the UKPDS showed that the important predictors were found within the LDL and low HDL data, even when accounting for triglycerides and HDLs over time.

“This is what convinced us that LDL might be the first priority,” said Dr. Haffner.

Data from the Scandinavian Simvastatin Survival Study (4S) showed that simvastatin therapy was just as effective in those with impaired fasting glucose (IFG) as normal fasting glucose (Haffner SM, et al. BMJ.1998;316:823).

Newer data from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) study looked at 40,000 subjects involved in a sub-study. The randomization was pravastatin 40 mg versus placebo (ALLHAT Collaborative Research Group JAMA. 2002;288:2981).

“The effect size for major CHD of 9% was the same as the differential for total cholesterol,” said Dr. Haffner. “Many clinical trials suggest that for each percent you change LDL or total cholesterol, you get about a 1% risk differential with agents like the statins.”

There is evidence from the Helsinki Heart Study and the VA-HIT data that fibric acid lowers risk in those with low LDLs and low HDLs that is similar to lowering LDLs by about 25%. It was the same for diabetic and non-diabetic subjects (Koskinen P, et al. Diabetes Care. 1992;15:820; Rubins HB, et al. NEJM. 1999;314:410).

According to Dr. Haffner, the traditional NCEP view is to focus only on high LDL. With metabolic syndrome, they acknowledge that it may lead to CHD with or without the development of type 2 diabetes.

The National Health and Nutrition Examination Survey (NHANES) data suggests that 23% of people 20 years or older have metabolic syndrome. The incidence increases in older people (Ford ES, et al. JAMA.2002;287:356)

Ford also found that in those over 50, 29% had metabolic syndrome but not diabetes. This is almost twice as large as the number of people with type 2 diabetes. Eighty-six percent of people with diabetes also have the syndrome. Although complicated somewhat by small numbers, diabetics without metabolic syndrome have a prevalence of CHD indistinguishable from non-diabetics without the syndrome.

“Why is this?” asked Dr. Haffner. “Diabetics tend to have higher HDLs, lower triglycerides, and higher blood pressures. These are all risk factors for the metabolic syndrome.”

Event Reduction with Combination Therapy: Results from HATS

There are many challenges to the clinician when dealing with metabolic syndrome and diabetes. One approach that has come to the forefront is combining statins and nicotinic acid as outlined in the results of the HDL Atherosclerosis Treatment Study (HATS) (Brown BG, et al. NEJM. 2001;345:1583).

“The biostatisticians tell us that there is about 1% rise in the risk of a cardiovascular event for each 1% increase in LDL,” said B. Greg Brown, MD, Professor of Medicine, University of Washington School of Medicine in Seattle. “The statin studies have reinforced that idea, showing a 25% to 35% event rate reduction accompanying 25% to 35% reductions in LDL cholesterol.”

In addition, epidemiologic studies show that a 1% increase in HDL gives a slightly better than 1.3% reduction in risk. These are statistically independent relationships so a 30% reduction in LDL and a 30% increase in HDL would result in a roughly 60% reduction in risk.

Another predictor is LDL particle size. Data from the Quebec Cardiovascular Study have shown those with the lowest Apo B, an indication of low particle number, have half the risk of the population with higher numbers. However, patients at a given Apo B level with smaller, dense particles have a substantially greater risk (more than two-fold) when compared to those with larger, more buoyant LDL (St. Pierre AC, et al. Circulation. 2001;104:2295),

“Patients with diabetic dyslipidemia have low HDL, high triglycerides, and relatively normal LDL levels, although the LDL particles do tend to be smaller and denser,” said Dr. Brown. “Because of strong favorable effects on LDL, HDL, triglycerides and particle size, the statin/niacin combination would seem an ideal form of therapy, if it is safe and well-tolerated.”

In a historical review of the 8 major statin- or niacin-related angiographic trials, the average lesion in the control groups worsened by 3% over the 2- to 4-year course of these studies. In the groups treated with statins alone, there was a 25% to 40% LDL lowering and the average rate of progression was reduced by half (Brown BG, et al. J Int. Med.1997;241:283).

“It is clear that LDL is an important determinant of whether arteries get worse or better,” said Dr. Brown. “Further-more, these data support the idea that when LDL decrease is combined with substantial HDL increases, arterial obstruction is actually reversed.”

When comparing niacin with placebo in the Coronary Drug Project (CDP), after a six-year treatment period, there was a 28% mean reduction in non-fatal MI and 17% reduction in coronary death in nonfatal MI. Those with diabetes at baseline (FBS>126 mg/dl) had a higher MI and death rate over the 7 and 15 years of follow up. The benefits of therapy (overall 28% MI reduction and 11% mortality reduction) were slightly greater in the diabetic subjects.

In the CDP trial, where niacin monotherapy was compared with placebo, being diabetic did not impact on the outcome either favorably or unfavorably. This helped dispel the idea that niacin could not be given to diabetics (Canner, PL, et al. Circulation. 2002;106:II [abstr. 3138]).

The HATS coronary angiographic study was conducted to answer some of these questions. HATS enrolled 160 patients with low HDL, normal LDL, and clinical coronary disease. About half had a previous MI and half had previous percutaneous intervention. Only poorly controlled diabetics (fasting glucose (FG)>180) and those with previous bypass surgeries were excluded. Angio-grams were done at admission and at three-year follow-up. Treatment was simvastatin 10-20 mg and niacin 2-4 gm daily.

In this group, the LDL was lowered by 43% and Apo B similarly reduced. The HDL was increased 30% with simvastatin and niacin, but only 20% when antioxidants were added (Cheung M, et al. ATVB. 2001;21:1320).

“We originally thought that was a play of chance and not a real finding,” said Dr. Brown. “But on closer inspection, HDL-2 cholesterol, a cardioprotective fraction, rose 60% in those getting simvastatin and niacin but only 15% among those also given antioxidants. Thus the antioxidant vitamins appear to blunt the HDL-2 response to simvastatin-niacin.”

Furthermore, in HATS, those taking simvastatin/niacin changed their large buoyant LDL size profile from 26% Pattern A at baseline to 74% post-treatment, a substantial cardioprotective improvement (Krauss, R, et al. Unpublished data).

There appears to be no major insulin resistance issues. Among those classified as diabetics in the treatment arm, insulin levels increased slightly for 4-8 months, but were down to baseline by study’s end. Both fasting glucose and hemoglobin A1C levels in the simvastatin/niacin group were as good or better at the end of the study compared to baseline.

“The tendency of transient small glucose elevations to dissipate over time is related to several interventions,” Dr. Brown. “We were teaching them better dietary habits, involving them in exercise programs and, in some cases, increasing their diabetes medications.”

Progression of coronary disease was tracked over 3 years by measuring the worst stenosis in each of the nine proximal coronary artery segments. The severity of the stenosis was summed to get a total at two points in time. Differences represented changes in the proximal stenosis burden.

Proximal burden increased by 34% in those given placebo. The antioxidant-treated patients had a non-significant modest reduction in disease progression. Those receiving the statin/niacin combination had an actual 4% regression (p=0.0006 vs. placebo).

Among patients classified as having diabetes or IFG, those treated with placebo had a progression rate of 43%. This was much more rapid than the 20% seen in the non-diabetic/placebo group. Those treated with simvastatin in the diabetes/IFG group had a 75% reduction in the rate of progression when compared with diabetics in the placebo arm. There was a small, net regression among non-treated diabetics.

“Thus, whether or not you have diabetes/IFG, there is substantial benefit from this form of lipid therapy,” said Dr. Brown.

Finally those with metabolic syndrome in HATS were examined. About half of these patients met the NCEP-III criteria.

The rate of stenosis progression was three times greater in those with metabolic syndrome when compared to those without. Coronary death, nonfatal MI, stroke, and revascularization were reduced by 40% among metabolic syndrome patients given simvastatin/niacin combination.

“We have good evidence from the statin trials that LDL lowering is beneficial and from other trials that raising HDL is also beneficial,” said Dr. Brown. “The HATS data puts these two together and says that raising HDL and lowering LDL simultaneously results in substantially greater risk reduction, as predicted by the epidemiology of LDL and HDL.”

Alternatives to Statin Therapy in Diabetes: Long-Term Perspectives from the Coronary Drug Project

Two widely held perceptions exist. One suggests that niacin is not very effective for lipid management in patients with diabetes or insulin resistance. The other says it isn’t effective for reduction of coronary artery events. Eliot Brinton, MD, Chief, Section of Metab-olism, Endocrinology, and Nutrition at the Hayden VA Medical Center in Phoenix addressed those concerns focusing on data from the Coronary Drug Project (CDP).

The CDP was the first large clinical trial to look at lipid lowering’s impact on cardiac events. Originally reported in the mid-1970s, there was a 15-year follow up and, more recently, a re-analysis of the data. There were six study arms in the original trial where patients were given niacin, clofibrate, low-dose estrogen, high-dose estrogen, D-thyroxine or placebo. Both estrogen and D-thyroxine arms were discontinued. Over six years, all of the cardiovascular outcomes were reduced with niacin treatment although there was no significant reduction in total mortality (The Coronary Drug Project Research Group, JAMA.1975;231:360)

One of the more interesting findings to Dr. Brinton was the change in fasting glucose levels over the six-year study when broken out by allocation to the niacin or placebo arms. There was an elevated fasting glucose baseline in both groups. At one year, the niacin group had a modest but statistically significant increase in FG levels, which persisted throughout the study. The mean one-hour glucose testing showed the same pattern.

“What is very curious is that after stopping niacin, there is a striking falloff in glucose levels so it is actually lower in the niacin-treated group than placebo,” noted Dr. Brinton. “I’m not quite sure why, but we saw the same thing in the HATS data.”

Niacin had a positive impact on non-fatal MI independent of fasting glucose status. This remained true looking at postprandial or post-challenge glucose levels in that those who were either normal or with impaired glucose tolerance (IGT) benefited. The benefits were similarly distributed across glucose status in the 15-year follow up (Canner PL et al. Circulation. 2002; 106:II [abstract 3138].

“Looking at CHD event reduction, there is a trend indicating the diabetic patients may have gotten better than the entire group,” said Dr. Brinton. “These trends depend somewhat on how you subdivide the data. At worst they have an equal, and possibly greater, event reduction.”

Due to the recent emphasis on the metabolic syndrome as a risk factor for CHD, re-analysis of the CDP data was performed according to the presence or absence of the syndrome. Categorization was hampered by a lack of certain criteria for the syndrome such as waist circumference and, in most subjects, HDL-C. Instead of waist circumference, the researchers used body mass index (BMI) and assigned a BMI of >30 as the surrogate for waist greater than 40 inches. HDL was just removed as a criterion so that instead of needing 3 of 5 criteria to diagnose metabolic syndrome, 2 of the remaining 4 were considered sufficient.

The decrease in CHD or MI at 6 years, and the reduction in total mortality at 15 years, were found to be independent of the presence or absence of the metabolic syndrome, at least by this altered definition.

Interestingly, despite the evidence that niacin can enhance insulin resistance, there was no evidence that niacin caused or increased the frequency of metabolic syndrome. To the contrary, with niacin, there was a dramatic improvement in triglycerides and modest positive changes in BMI, which easily counterbalanced the tendency toward an increase in fasting glucose. Dr. Brinton feels it is safe to conclude that niacin does not cause a dramatic increase in metabolic syndrome and may even be protective.

“We started off with perceptions but where does reality lie?” asked Dr. Brinton. “Niacin is quite effective for the dyslipidemia of diabetes and insulin resistance and appears to effective in CHD event reduction in patients with those conditions. It is a viable alternative or adjunct to the statins in atheroprevention in diabetes and the metabolic
syndrome.”