HOSPITAL PRACTICE




Giving Triglycerides Their Due

MICHAEL MILLER
University of Maryland

CME
Often considered less important than other serum lipids, triglycerides are nevertheless an independent risk factor for coronary artery disease. Treatment may be indicated even when triglyceride levels are only mildly elevated and cholesterol readings are within the normal range. Detailed lipoprotein profiles, now clinically available, may guide decision making in such cases.



Dr. Miller is Director, Center for Preventive Cardiology, and Associate Professor of Medicine, University of Maryland School of Medicine, Baltimore.


Triglycerides have been well established as a risk factor for coronary artery disease (CAD) for several decades. As early as 1959, Margaret Albrink and E. B. Mann reported higher serum triglyceride levels in patients with CAD than in age-matched controls. For many years, however, triglycerides were overshadowed by other serum lipids. In particular, multivariate analysis suggested that triglycerides were less important than high-density lipoprotein (HDL). In 1996, however, a meta-analysis by John Hokanson and Melissa Austin confirmed that triglycerides are an independent risk factor for CAD. For each mmol/L increase in triglycerides--which translates into 88.5 mg/dL--the risk of CAD increases by 37% in women and 14% in men (Figure 1). All else being equal, a man with a triglyceride level of 300 mg/dL would have a risk of cardiovascular events roughly 28% higher than that of an otherwise comparable man who has a level of 100 mg/dL.

Relationship to Atherosclerosis

During the past 10 to 15 years, researchers have worked out the mechanisms by which triglycerides may contribute to atherosclerosis. To understand those mechanisms, one must realize that the concept of plasma lipids held by the general public--in which lipids are cleanly divided into HDL, low-density lipoprotein (LDL), and triglycerides--is simplistic and somewhat misleading. In fact, each of those categories comprises a variety of particles, which differ in size and often in the degree of risk they pose for CAD.

All lipoproteins contain both cholesterol and triglycerides, although LDL and HDL carry most of the cholesterol and chylomicrons and very-low-density lipoprotein (VLDL) carry most of the triglycerides. When triglyceride levels become elevated, the ratio of triglycerides to cholesterol in LDL and HDL may shift toward triglycerides and lead to reduced HDL concentrations.

Diet is key to the link between triglycerides and CAD. The free fatty acids that are processed from dietary fat are taken up into small intestinal cells (enterocytes), where they are reesterified into triglycerides and then packaged as chylomicrons within the lymphatic system. At the thoracic duct, the chylomicrons leave the lymphatic system and enter the venous circulation.

Certain small- and medium-chain fatty acids may enter directly into the portal circulation rather than being reesterified and processed as chylomicrons. That mechanism is of clinical significance in patients with a specific genetic disorder that causes very high triglyceride levels secondary to lipoprotein lipase deficiency. These patients have triglyceride levels in the range of 5,000 to 20,000 mg/dL. Treatment is to eliminate all fat in the diet except for essential fatty acids. Patients may use a commercially available preparation that contains fatty acids that are metabolized directly in the liver. Mechanisms of hypertriglyceridemia are shown in Table 1.


Table 1. Conditions That Can Cause
Elevated Triglyceride Levels
Genetic Disorders
Lipoprotein lipase abnormalities (e.g., familial chylomicronemia)
Apolipoprotein CII or CIII abnormalities
Apo E abnormalities (e.g., familial dyslipoproteinemia)
Familial combined hyperlipidemia (e.g., hyperapo B syndrome)
Familial hypertriglyceridemia

Metabolic Disorders
Nephrotic syndrome/renal failure
Hypothyroidism
Diabetes mellitus
Paraproteinemia

Medications
Bile acid resins
Estrogenic compounds
Steroids
Protease inhibitors
Retinoids
Beta-blockers

As chylomicrons enter the venous circulation, they are quickly hydrolyzed by lipoprotein lipase, an enzyme on the capillary endothelial surface. The hydrolysis produces free fatty acids that can be taken up by adipocytes or used by muscle cells for energy. It also results in a higher concentration of cholesterol ester:triglycerides per particle. These so-called remnant particles are believed to be atherogenic because they are readily taken up by macrophages; the lipid-laden macrophages then become foam cells, the progenitors of atherosclerotic lesions. Remnant particles can also be taken up by the endothelium, whereas chylomicrons are too large to penetrate the endothelial surface.

The other important type of triglyceride-rich lipoprotein (TRLP) is VLDL. These particles are produced in the liver and then enter the circulation, where they are hydrolyzed by lipoprotein lipase into smaller VLDL remnant particles. Ultimately, the particles may be converted to atherogenic intermediate-density lipoproteins or further metabolized to LDL.

All cells, and especially liver cells, express VLDL receptors. Monocytes, macrophages, and endothelial cells express a distinct receptor for TRLP and apolipoprotein B48 (apo B48). In healthy persons, VLDL does not bind to TRLP/apo B48 receptors. Patients with hypertriglyceridemia, however, have large VLDL that binds toTRLP/ apo B48 receptors. This results in rapid and massive lipid accumulation within monocytes and macrophages, transforming theminto foam cells, and within endothelial cells, impairing their ability to mediate fibrinolysis (Figure 2).

Myocardial infarction most often involves a combination of atherosclerosis and thrombosis, and it has been well documented for a number of years that hypertriglyceridemia can increase the risk of thrombus formation. VLDL activates the intrinsic coagulation pathway. In addition, VLDL enhances the transcription of plasminogen activator inhibitor, which inhibits the endogenous production of tissue plasminogen activator. A. Hamsten and colleagues showed in the mid-1980s that one important risk factor for myocardial infarction before age 45 is an enhanced concentration of plasminogen activator inhibitor.

The final postulated mechanism involves elevated concentrations of triglycerides and HDL and the production of small, dense LDL. This typically occurs in patients who have a genetic predisposition to hypertriglyceridemia or, more commonly, through the interaction of a genetic predisposition and a diet high in saturated fat. In these cases, elevated triglyceride levels may result from either overproduction of TRLP in the liver or inefficient hydrolysis. Excess triglyceride is transferred from VLDL and chylomicrons to lower-density lipoproteins, including LDL and HDL.

These triglyceride-enriched LDL and HDL particles are avidly degraded by hepatic lipases into very small, dense particles, resulting in cholesterol-depleted HDL. Cholesterol measurements will show a low level of HDL that may be less effective in transporting cholesterol from the periphery back to the liver.

Patients also have large quantities of small, dense LDL, which is believed to be more atherogenic than nascent LDL. As A. Chait and colleagues have shown, small, dense LDL particles are more susceptible to oxidation, after which they are more avidly incorporated by scavenger cells. The proportion of LDL particles that are small, dense, and proatherogenic versus large, buoyant, and less atherogenic may be compared by evaluating the corresponding triglyceride levels, as was shown by M. A. Austin and colleagues in 1990. For example, in a patient with a triglyceride level of about 180 mg/dL, 70% to 80% of the LDL will be small and dense; in contrast, if the triglyceride level is about 50 mg/dL, 90% of the LDL will be large, buoyant, and less atherogenic (Figure 3).

Studies such as those suggest that current triglyceride cutpoints may be set too high. According to the National Cholesterol Education Program (NCEP), a triglyceride level of less than 200 mg/dL is desirable, a level of 200 to 400 mg/dL is borderline high, and 400 to 1,000 mg/dL is high. The Baltimore Coronary Observation Long-Term Study (COLTS) recently suggested that a triglyceride level greater than 100 mg/dL may be abnormal. Whether and to what extent further refinement of triglyceride guidelines by the NCEP will occur is unknown. However, present "desirable" triglyceride cutpoints, established by the NCEP in 1993, are lower than previous ones. In 1984, the National Institutes of Health defined a desirable triglyceride level as less than 250 mg/dL. In addition, the American Heart Association and the American College of Cardiology recently endorsed a desirable triglyceride level of 150 mg/dL or less in women and diabetic patients.

Other Measurement Issues

Circulating HDL and LDL concentrations do not change significantly with food intake; triglyceride concentrations do. Whenever we eat a meal that contains fat, our triglyceride levels rise. The degree of the increase depends on the baseline level. For example, if a person who has a triglyceride level of about 50 mg/dL eats the classic fast-food meal of a hamburger, french fries, and milkshake, triglycerides may increase by 15% to 20%; the triglyceride level may rise to 70, 80, or perhaps 90 mg/dL but still remain within the normal range. In persons whose baseline levels are greater than 200 mg/dL, that high-fat meal could catapult the levels to 300, 400, or even 500 mg/dL, and in some cases, the elevation may be prolonged for several hours beyond the normal eight-hour clearance period.

By definition, the only way to obtain a true baseline triglyceride measurement is to draw the blood sample after the patient has fasted for at least 12 hours. In theory, blood samples for lipid measurement are always supposed to be drawn with the patient fasting, but in practice that principle is often given no more than lip service.

If the patient has not been fasting when the sample was drawn, many physicians will dismiss an elevated triglyceride reading as insignificant. As a rule, however, the lower the baseline level, the less the variability. We see the most variability in patients with a high baseline triglyceride level (i.e., 100 mg/dL or greater).

Another factor that affects triglyceride measurement is postural change; the level can vary by as much as 15%, depending on whether the patient was standing or supine when the blood was drawn, because of associated shifts in plasma volume. Yet another factor is laboratory variability, which can range between 5% and 10% on a daily basis. Finally, intraindividual variability can be another 5% to 10%. Those variations will not matter as much in patients who have a normal baseline triglyceride value as they will in those who have an elevated value, in whom the variability will tend to be much more dramatic.

Regardless of variations, patients with triglyceride levels less than 100 mg/dL clearly do not require treatment, and those with levels greater than 200 mg/dL do need treatment. About 80% of our patients have triglycerides between 100 and 250 mg/dL. Lipid measurement may show a normal total cholesterol, with a low HDL and a moderately elevated triglyceride level--for example, a total cholesterol of 197 mg/dL, an LDL of 120 mg/dL, an HDL of 35 mg/dL, and a triglyceride of 210 mg/dL. That pattern is not unusual; in fact, it is seen in about 20% to 30% of patients who have had a myocardial infarction. How to best manage these patients is an important issue.

Decision making in those cases may benefit from the use of a new test that became commercially available this year. This procedure uses nuclear magnetic resonance (NMR) spectroscopy to fractionate VLDL, LDL, and HDL concentrations into 15 different subfractions. It is completely automated and takes only about one minute to perform, since it measures all subfractions simultaneously. Two patients may have similar lipid levels, but if NMR spectroscopy shows that one has a much higher proportion of small, dense lipoprotein particles, that might tip the balance toward treatment, especially if there is a family history of CAD or other risk factors.

Causes and Complications

When lipid measurement shows an elevated triglyceride level, the first step in management is to determine the possible underlying causes. Genetic disorders, medications, and metabolic disorders all can contribute to hypertriglyceridemia. Among the most common causes is diabetes mellitus. In susceptible individuals, triglyceride levels can rise precipitously in the face of uncontrolled hyperglycemia.

A triglyceride level greater than 1,000 mg/dL may be causally associated with pancreatitis. Another manifestation of a very high triglyceride level (especially greater than 2,000 mg/dL) is the development of painless, nonpruritic skin lesions known as eruptive xanthomas--small, yellowish or yellowish orange papules that appear in clusters, especially on the elbows, buttocks, and thighs (Figure 4). The xanthomas may develop within several days in susceptible patients, such as those with diabetes and marked deterioration in blood glucose control. Controlling hyperglycemia generally reduces triglyceride levels as well, although triglyceride-lowering medication may also be necessary. Once the triglyceride level has been reduced to approximately the patient's baseline level, the xanthomas will resolve. Extreme hypertriglyceridemia may also cause lipemia retinalis. On fundoscopic examination, the retinal vessels will have a creamy-white appearance. Fortunately, vision remains unaffected.

Treatment

Dietary changes can be an effective means of lowering triglyceride levels. In fact, diet generally affects triglyceride levels much more readily than it does cholesterol levels. Switching to a diet that is low in saturated fat can reduce triglycerides by 20% to 30%, whereas the LDL level usually decreases by only 5% to 10%. Moreover, this response often occurs within two to four weeks and may be striking when accompanied by significant weight reduction.

The difficulty with dietary treatment of elevated triglyceride levels is that the important sources of triglycerides are standards of the American diet. Saturated fat, sweets (because glycerol serves as the chemical backbone of triglycerides), baked goods, fried foods--intake of all of these needs to be reduced.

My principal dietary recommendation for lowering triglyceride levels is to reduce total fat to less than 30% of caloric intake and to exchange saturated fat for monounsaturated and polyunsaturated fats. Omega-3 fatty acids are among the best sources to reduce triglyceride levels because they inhibit production of VLDL in the liver. Omega-3 fatty acids are found predominantly in fatty fish such as salmon. To achieve the recommended amount, however, one would need to eat about 10 ounces of salmon a day.

Fish-oil capsules can be purchased at health food stores, but they provide only about 300 mg of omega-3 fatty acids for each gram of oil, which translates into about eight capsules daily. A Norwegian company has filed for approval from the Food and Drug Administration of a very potent omega-3 supplement that would require taking only two to four capsules a day, and that may be available in a year or so.

It is also useful to restrict combinated intake of alcohol and fat. Alcohol alone seems to have a less impressive impact on triglyceride levels than previously believed, but concomitant intake of alcohol and fat may raise triglyceride levels markedly. Finally, aerobic exercise can reduce triglyceride levels by 10% to 20%, essentially by activating lipoprotein lipase and enhancing lipolytic activity on triglyceride-rich lipoproteins.

If conservative measures fail, I turn to drug therapy. There are three main classes of cholesterol-lowering drugs: niacin, HMG-CoA reductase inhibitors (statins), and fibrates.

Niacin lowers triglyceride levels by 10% to 30%. It can be given in either the immediate- or sustained-release form. Dosages of 1 to 2 gm daily are effective for increasing HDL levels and decreasing triglyceride levels. Higher dosages are usually required for significant LDL lowering.

All of the statin drugs lower triglyceride levels, but those that have the greatest effect on LDL levels (i.e., simvastatin and atorvastatin) also have the best triglyceride-lowering impact. The statins are most effective as treatment for hypertriglyceridemia when the baseline triglyceride concentration is especially high. If the baseline triglyceride is greater than 250 mg/dL, statin therapy may induce a 20% to 40% reduction. If the baseline is less than 150 mg/dL, the decrease will average 5% to 10%.

In patients with normal total cholesterol concentrations (<200 mg/dL), fibrates are useful for lowering triglyceride levels. Indeed, they are the most potent agents available, providing a reduction of 20% to 60%. I typically use gemfibrozil or fenofibrate. A recent large study of gemfibrozil, which is expected to be published in late 1999, suggests that this agent can reduce cardiovascular events in patients with CAD who have a relatively normal lipid profile. In that study, the average values were total cholesterol, 175 mg/dL; LDL, 111 mg/dL; HDL, 32 mg/dL; and triglycerides, 161 mg/dL. This is a group of patients that most clinicians traditionally would not treat, yet they had an impressive 22% reduction in cardiovascular events.

CME
Selected Reading

Austin MA et al: Atherogenic lipoprotein phenotype: A proposed genetic marker for coronary heart disease risk. Circulation 82:495, 1990

Chait A et al: Susceptibility of small, dense, low-density lipoproteins to oxidative modification in subjects with the atherogenic lipoprotein phenotype, pattern B. Am J Med 94:350, 1993

Gianturco SH, Bradley WA: Pathophysiology of triglyceride-rich lipoproteins in atherothrombosis: Cellular aspects. Clin Cardiol 22(Suppl II):7, 1999

Hamsten A et al: Increased plasma levels of a rapid inhibitor of tissue plasminogen activator in young survivors of myocardial infarction. N Engl J Med 313:1557, 1985

Hokanson JE, Austin MA: Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: A meta-analysis of population-based prospective studies. J Cardiovasc Risk 3:231, 1996

Miller M et al: Normal triglyceride levels and coronary artery disease events: The Baltimore Coronary Observational Long-Term Study (COLTS). J Am Coll Cardiol 31:1252, 1998

Otvos J: Measurement of triglyceride-rich lipoproteins by nuclear magnetic resonance spectroscopy. Clin Cardiol 22(Suppl II):21, 1999

Segal A et al: Hypertriglyceridaemia and vascular risk: Report of a meeting of physicians and scientists, University College of London Medical School. Lancet 342:781, 1993


HOME | CURRENT ISSUE | PAST ARTICLES | GENETICS SERIES
CAPSULE & COMMENT | CME | CLASSIFIED ADVERTISING | ABOUT US

McGraw-Hill Logo
Copyright (C) 1999. The McGraw-Hill Companies. All Rights Reserved

E-mail Privacy Notice