Chylomicrons

Chylomicrons derive from dietary fat and carry triglycerides throughout the body. The major structural protein of chylomicrons is apolipoprotein (apo) B₄₈, a protein that is produced from apo B₁₀₀ RNA by a unique editing process employed by gut cells that generates a protein that is 48% of the length of apo B₁₀₀.⁸ Chylomicrons also contain other apoproteins, including members of the apo C and apo A family. These proteins can redistribute to other lipoprotein classes within serum as chylomicrons transfer their lipid content to cells and interacting lipoproteins. Chylomicrons have the lowest density of all lipoproteins and will float to the top of a plasma specimen left in a refrigerator overnight, forming a creamlike layer. Owing to its large size, the chylomicron does not readily enter the artery wall and is therefore thought not to be atherogenic, but the atherosclerotic role of the triglyceride-depleted chylomicron remnant remains controversial. Triglyceride makes up most of the chylomicron and is removed by the action of an enzyme that is bound to the surface of endothelial cells, lipoprotein lipase (LPL). Patients deficient in this enzyme or its apoprotein cofactor (apo CII) have very high serum triglyceride levels and are at increased risk for developing acute pancreatitis. Inasmuch as insulin is also a critical cofactor for LPL activity, the vast majority of patients presenting with hypertriglyceridemic pancreatitis have poorly controlled diabetes as the cause of their delayed chylomicron clearance.

Very Low-Density Lipoproteins

VLDLs are also triglyceride-rich, but their triglyceride content is lower and cholesterol content higher than that of chylomicrons. The protein composition of VLDL also differs from chylomicrons in that the major structural protein is full-length apo B (apo B₁₀₀) as opposed to the truncated apo B₄₈ form. Like chylomicrons, VLDLs are substrates for lipoprotein lipase–mediated triglyceride removal. Their function is to carry triglycerides synthesized in the liver and intestines to capillary beds in adipose tissue and muscle, where they are hydrolyzed to provide fatty acids that can be oxidized to produce adenosine triphosphate (ATP) for energy production. Alternatively, if not needed for energy production, they can be reesterified to glycerol and stored as fat. After removal of their triglyceride, VLDL remnants (called IDLs) can be further metabolized to LDL. VLDLs serve as acceptors of cholesterol transferred from HDL, accounting in part for the inverse relation between HDL cholesterol and VLDL triglyceride. This transfer process is mediated by an enzyme called cholesterol ester transfer protein (CETP). The most common lipid abnormality seen in type 2 diabetics is an elevated VLDL level, the causes of which will be discussed in the section on diabetic lipid abnormalities.

Low-Density Lipoproteins

LDLs are the major carriers of cholesterol in humans, responsible for supplying cholesterol to tissues with the highest sterol demands. LDLs are also the lipoproteins most clearly implicated in causing atherogenic plaque formation.⁹ Circulating LDL levels can be increased in persons who consume large amounts of saturated fat and/or cholesterol.¹⁰ LDL levels are also elevated in those who have genetic defects that affect LDL receptor function (familial hypercholesterolemia, PCSK9 mutations, autosomal recessive hypercholesterolemia), the structure of LDL’s apoprotein, apo B, or who have polygenic disorders affecting LDL metabolism.¹¹ Circulating LDLs are taken up by the endothelium lining the artery wall, traverse it, and are deposited and trapped in the arterial intima. There, they may undergo oxidation or other biochemical modification, be taken up by macrophages, and stimulate atherogenesis. The association of total serum cholesterol with CHD is predominantly a reflection of the role of LDL. Studies in patients with diabetes have produced somewhat conflicting data on the prevalence of increased LDL levels, as defined by general population norms, but the preponderance of the evidence indicates that diabetics have LDL levels that are similar to those seen in well-matched, nondiabetic control populations. Given diabetics’ greater risk of developing coronary artery disease (CAD), however, the National Cholesterol Education Program guidelines now define LDL levels above 100 mg/dL as being undesirable in diabetics. When judged by this stringent criterion, elevated LDL levels are extremely common in patients with diabetes. In addition, the LDL in diabetics is frequently smaller, denser, and more readily oxidized—properties that are associated with a higher degree of atherogenicity.^12,13

High-Density Lipoproteins

HDLs are the smallest of the lipoproteins; despite this, they carry a variety of apoproteins, including members of the apo A and C family. HDLs, unlike the other lipoproteins, are believed to function in humans primarily to return lipids from peripheral tissues to the liver and gut for excretion, rather than moving lipids from the gut organs to the periphery.¹⁴ This perspective on HDL function is an oversimplification, since cell culture experiments indicate that HDL can donate lipids as well as pick them up. It has proven to be quite difficult to quantitate the net movement of cholesterol carried in HDL in and out of peripheral tissues in whole animals. Further complicating the issue is the fact that many of the standard laboratory animals used in metabolic studies carry most of their serum cholesterol in HDL rather than LDL, diminishing the relevance of their lipoprotein metabolism to that of the LDL-rich human. Nevertheless, a considerable body of evidence suggests that HDL functions to protect tissues from unwanted accumulation of cholesterol. The mechanisms by which HDL can protect against atherosclerosis include its involvement in the reverse cholesterol transport pathway described earlier, as well as through contributions arising from its carriage of proteins that appear to have antiinflammatory and antioxidant properties.^15,16 The unesterified cholesterol from tissues that is transferred to HDL is esterified by the action of lecithin cholesterol acyltransferase (LCAT) and stored in the central core of HDL. This esterified cholesterol can be transferred back to lower-density lipoproteins by the action of cholesterol ester transfer protein, or it may be removed at the liver by the action of a plasma membrane receptor called scavenger receptor B-1 (SRB-1). A particularly effective reverse transport system is thought to explain, at least in part, the association of elevated HDL cholesterol levels with a reduced risk of developing CHD. Apolipoprotein AI (apo AI) is the major apoprotein of HDL, and its serum concentration also correlates inversely with the risk of CHD. Women have higher levels of HDL cholesterol than men, and this may partly explain the lower incidence of CHD in premenopausal women. Exercise increases HDL levels; obesity, hypertriglyceridemia, and smoking lower them. Patients with diabetes, having higher body mass indices and hypertriglyceridemia, typically have lower HDL cholesterol levels than do nondiabetics, contributing further to the atherogenic lipid profile of the diabetic.

Low-Density Lipoprotein Metabolism

There is still some dispute about whether mature LDL can be secreted from the liver rather than always being derived from the progressive removal of triglyceride from VLDL/IDL precursors.²⁹ The preponderance of data indicate that little if any LDL is directly secreted from the liver. Each LDL particle contains a single apo B molecule, which dictates that individuals heterozygous for apo B mutants will have two populations of LDL, one which carries a wild-type protein and one with a mutant protein. LDLs are utilized throughout the body as a source of exogenous cholesterol and are taken up by LDL receptors present at the plasma membrane. While LDL receptors are ubiquitous, the majority are expressed by the liver. The LDL receptor binds to both apo E– and apo B–containing lipoproteins. Because the affinity of the receptor is higher for apo E than for apo B, E-containing lipoproteins, such as IDL, can compete with LDL for uptake by the LDL receptor. However, there are other receptors that can bind apo E–containing lipoproteins, permitting their clearance but not that of LDL when LDL receptor activity is lost or diminished. Thus, mutations in the LDL receptor result in elevated LDL levels in the plasma, causing the autosomal co-dominant disorder, familial hypercholesterolemia, which is among the most common of Mendelian genetic disorders (see Genetics section later).³⁰ The elucidation of the cholesterol homeostatic mechanisms controlled by the LDL receptor constituted one of the major advances in our understanding of human physiology and disease.³¹ The internalization of LDL by the LDL receptor leads to trafficking of the internalized lipoprotein to the endosome, where after fusion with lysosomes, the sterol in the lipoprotein is released and shuttled to the endoplasmic reticulum (ER). A sterol-sensing system in the ER regulates the movement of a membrane-bound transcription factor, sterol response element binding protein (SREBP), from the ER to the nucleus.^32–37 SREBP works in concert with other transcriptional regulators to influence the expression of many of the cholesterol synthesis, metabolizing, and transport molecules, particularly in the liver. This elegant homeostatic system enables cells to reduce their de novo cholesterol synthesis when adequate sterol supplies are present and to increase synthesis and LDL receptor expression when cellular demands require more cholesterol.

High-Density Lipoprotein Metabolism

Although the relationship between elevated levels of HDL and lower rates of cardiovascular disease was recognized more than 50 years ago, the mechanism by which HDL protects against CHD is still only partially understood. HDL are initially produced as lipid-poor phospholipid discs containing apolipoprotein AI. Following interaction with the phospholipid/cholesterol transporter, apo AI, the particle acquires unesterified cholesterol from cellular stores.³⁸ For macrophages in particular, this cholesterol efflux pathway appears to be critical to their ability to unload the excess cholesterol they acquire by phagocytosis of dead cells and by their unregulated uptake of lipoproteins by scavenger receptor pathways.¹⁴ Cholesterol is stored in an intracellular lipid pool, having been esterified at its 3′OH position by the activity of acyl coenzyme A (CoA):cholesterol O-acyltransferase (ACAT).³⁹ A neutral cholesterol hydrolase cleaves the fatty acid from cholesterol ester to convert it back to free (unesterified) cholesterol when cellular signals call for the use or export of the sterol.⁴⁰ This free cholesterol can be shuttled to the plasma membrane, where it can be transported by the action of ABCA1. The interaction of lipid-poor nascent HDL with ABCA1 appears to involve direct binding of apo AI to the transporter, but the actual lipid transfer process is poorly understood.^41–43 Once the unesterified cholesterol transfers to the nascent HDL, it can be reesterified by the action of lecithin cholesterol acyltransferase (LCAT) and centralized in the lipid core of the growing HDL particle. This HDL particle can now interact with a second ABC transporter, ABCG1, and additional cholesterol can be acquired.⁴⁴ The role of ABCA1 is well established, since loss of its activity leads to Tangier disease (see Genetics section) and the absence or near absence of HDL in the plasma. ABCG1’s precise function is still being elucidated. Once the HDL has acquired sufficient cholesterol and esterified it, it can donate this lipid back to lower-density lipoproteins in the plasma, such as VLDL, through the activity of cholesterol ester transfer protein (CETP). Alternatively, HDL can interact with a cellular receptor, SR-BI, to selectively transfer the cholesterol ester to cells, including the liver, where it could excreted in the bile as cholesterol or one of its bile acid derivatives.⁴⁵ The protein component of HDL is cleared via the kidney, a process that appears to depend on the HDL being lipid depleted. The pathway described, moving cholesterol from cellular storage pools to HDL and ultimately to the liver, is termed the reverse cholesterol transport pathway. The activity of this pathway is postulated to explain at least in part the link between HDL levels and lower CHD risk. However, HDL also carries antioxidants and a host of proteins linked to complement and inflammation pathways that suggest it may be playing a more general role as a scavenger of molecules that stimulate inflammation. The complexity of its function has made targeting increases in HDL levels as a therapeutic objective challenging. Both animal and human data have established that not all methods of increasing HDL result in improvements in atherosclerosis.

Familial Hypercholesterolemia

Familial hypercholesterolemia (FH) is one of the most thoroughly studied and common genetic disorders of mankind. Approximately 1 in 500 individuals carry a mutation in one allele encoding the LDL receptor. These mutations results in defective clearance of LDL from the blood and a rise in serum total and LDL cholesterol levels. The elucidation of this defect and its associated cell biology led to insights into the homeostatic control of cholesterol metabolism that transformed the lipid field.³¹ Several hundred individual mutations in the LDL receptor have now been identified. The gene, located on chromosome 19 and spanning 45 kb, has 18 exons that encode a mature protein of 839 amino acids. The inheritance of the disorder is autosomal co-dominant. Heterozygous patients have LDL cholesterol levels in the 200 to 500 mg/dL range, and the rare homozygous FH patient typically has an LDL cholesterol above 500 mg/dL. Heterozygous patients commonly have tendon xanthomas and premature CAD, whereas these are universal in the untreated homozygous individual. Mutations in the LDL receptor have been described that affect multiple aspects of receptor biology, ranging from defective LDL binding to abnormal trafficking of the receptor to the plasma membrane. The effect of all of these mutations is to reduce LDL and IDL clearance from the blood. The latter effect results in an enhanced conversion of IDL to LDL. The net result is markedly elevated LDL cholesterol levels in the plasma and a strong predisposition to early CHD if the disorder is not treated. FH heterozygotes typically respond well to 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase inhibitors (statins), though additional therapies may be needed to reach desirable LDL levels. FH homozygotes respond poorly to reductase inhibitors and inevitably develop early atherosclerotic vascular disease if LDL apheresis or liver transplantation is not performed.

Familial Defective Apolipoprotein B

Familial defective apo B (FDB) is an autosomal co-dominant disorder caused by mutations near the carboxy terminus of apo B.^63,64 Apo B is a 4536-amino-acid protein encoded by a gene on chromosome 2 that is composed of 29 exons. The most commonly identified mutations in FDB affect an arginine at position 3500 that is either mutated to a glutamine or tryptophan. Most individuals with FDB are heterozygous for their mutation, and there is only one molecule of apo B per LDL particle, so the LDL in the serum of these individuals is a mixture of normal LDL and FDB LDL. The LDL containing the defective apo B does not bind normally to the LDL receptor and is therefore cleared more slowly. The explanation for the lower receptor affinity of LDL carrying the FDB mutations is not known with certainty. There is evidence, however, of a molecular interaction between the arginine at position 3500 and tryptophan at 4369 that appears to be necessary for the adoption of an apo B conformation recognized by the LDL receptor. Patients with FDB have elevations in their LDL levels that fall within the range of those seen in FH heterozygotes, but the mean is generally lower than the mean in FH cohorts. There is considerable variability in both genetic populations; however, making generalizations about the diagnostic value of this measurement not very useful. Tendon xanthomas are common but not universal, and premature CAD is prevalent. Treatment with reductase inhibitors or other standard LDL-lowering drugs is typically quite effective in FDB patients.

Autosomal Recessive Hypercholesterolemia

Studies of families in which LDL levels were inherited in an autosomal recessive pattern led to a successful genetic linkage analysis that mapped the genetic defect in these families to the short arm of chromosome 1. Candidate gene sequencing led to the identification of mutations in a gene now designated autosomal recessive hypercholesterolemia (ARH) and comprising nine exons and eight introns. Patients diagnosed with ARH to date have been clinically similar to patients who are homozygous for LDL receptor mutations in that they have markedly elevated LDL cholesterol levels (typically > 500 mg/dL), tendon xanthomas, and premature coronary atherosclerosis.⁶⁵ The autosomal recessive mode of inheritance is an important differentiating factor between ARH and familial hypercholesterolemia, as is greater LDL receptor activity, measured in cultured fibroblasts taken from the patients. The defect in ARH patients appears to affect liver cholesterol metabolism disproportionately, and current data indicate that the ARH gene product likely serves as an adaptor protein required for LDL receptor internalization via clathrin-coated pits.^66,67 ARH patients do respond to HMG CoA reductase inhibitors, but this treatment is usually inadequate to control their markedly elevated LDL cholesterol levels, making them candidates for LDL apheresis.

PCSK9 Mutations

Proprotein convertase subtilisin kexin 9 (PCSK9) is a 72-kD protease that is highly expressed in the liver and regulates the levels of LDL receptors acting at the plasma membrane.^68,69 Gain-of-function mutations in PCSK9 result in lower LDL receptor levels, whereas loss-of function mutations increase those levels. Serum LDL levels are inversely related to liver LDL receptor expression; the activity of the latter clears the former from the circulation. Thus, gain-of-function mutations are associated with higher serum LDL levels and increased risk of CHD, and loss-of-function mutations result in lower LDL levels and decreased CHD risk. The role of PCSK9 in lipid metabolism was originally recognized when it was discovered that specific heterozygous mutations in the protein were associated with autosomal dominant hypercholesterolemia. Approximately 1 in 50 African Americans has one of two nonsense mutations in PCSK9, causing LDL cholesterol levels to fall approximately 30%. Caucasians can carry a missense mutation that reduces LDL cholesterol by approximately 15%. Individuals with complete loss of PCSK9 activity have extraordinarily low levels of LDL-C, but appear to be otherwise normal. The mechanism by which PCSK9 regulates LDL receptor activity is still being elucidated, but it appears that the protease can bind to the receptor, targeting it for degradation in the lysosome. This action does not depend on the protein’s enzymatic activity, and exogenously administered PCSK9 can reduce LDL receptor expression and serum LDL-C levels. The mechanism by which PCSK9 works makes it an extremely attractive target for pharmaceutical intervention; therapeutics that block PCSK9 function are being actively pursued.

Tangier Disease

Tangier disease was first recognized in 1960 in a sibling pair living on Tangier Island in the Chesapeake Bay in Virginia. Enlarged, yellow-orange tonsils and little or no circulating HDL cholesterol are the classic findings in the disorder. Subsequently, cases have been identified in which the presenting symptom was a peripheral neuropathy. Patients may also have hepatosplenomegaly. Serum LDL and total cholesterol levels are usually quite low, while serum triglyceride values are moderately elevated. Tangier disease is an autosomal recessive disorder. The cause of Tangier disease was identified by several groups, following mapping of the gene defect to chromosome 9. Candidate gene analysis in the appropriate genetic interval identified mutations in an ATP-binding cassette (ABC) transporter as the cause. The transporter, now called ABCA1 (ATP-binding cassette, subfamily A, member 1), is a full-length ABC transporter transmembrane protein that is predicted to span the plasma membrane 12 times. ABCA1 is a 2261-amino-acid protein encoded by a gene spanning 50 exons. Approximately 50 mutations have been identified to date in the gene.⁷⁰ ABCA1 mediates the efflux of cholesterol from cholesterol-enriched cells when stimulated by the major apoprotein of HDL, apo AI.⁷¹ This activity is lost in Tangier patients and is reduced by approximately half in carriers of one abnormal ABCA1 allele. The mechanism of the movement of cholesterol from inside the cell to outside the cell has not been established. Individuals with Tangier disease and heterozygous carriers of ABCA1 mutations (a disorder called familial hypoalphalipoproteinemia [FHA]) both appear to have increased risk of premature coronary disease. There is no specific therapy for this disorder.

Lecithin Cholesterol Acyltransferase Deficiency

The esterification of free cholesterol in circulating lipoproteins is catalyzed by a plasma enzyme called lecithin cholesterol acyltransferase. Two clinically separable syndromes result from a deficiency of LCAT. Fish eye disease is due to a partial deficiency of LCAT, with patients presenting with dense corneal opacities and very low HDL cholesterol levels. Familial LCAT deficiency arises from a nearly completely absence of LCAT activity and produces a more severe syndrome characterized by corneal opacities, anemia, and proteinuric renal failure.⁷² The serum lipid and lipoprotein profile in the more severe disorder is characterized by normal or increased triglyceride levels, reduced LDL cholesterol values, and markedly diminished HDL levels. The gene encoding LCAT is located on chromosome 16 and is composed of 6 exons. Cleavage of a signal peptide of 24 amino acids converts the proenzyme from a 440-amino-acid precursor to the final 416-amino-acid glycoprotein that circulates in the plasma. When unesterified cholesterol from tissues is transferred to HDL, either by passive diffusion or by ABCA1-mediated lipid transport, LCAT’s activity esterifies the transferred cholesterol, trapping it in the HDL core. Since cholesterol ester is more hydrophobic than unesterified cholesterol, it is energetically unfavorable for the cholesterol ester to transfer back to the cell of origin. These steps of cholesterol transfer and esterification are the initial events in the reverse cholesterol transport pathway whereby cholesterol is moved from peripheral tissues back to the liver. Apo AI is the major activator of LCAT activity, accounting for the predominant effect of the enzyme deficiency on HDL levels. LCAT does, however, contribute to esterification of cholesterol in lower-density lipoproteins as well. Despite very low HDL levels, patients with either fish eye disease or familial LCAT deficiency do not seem to have a predilection for very early coronary atherosclerosis. The small number of patients with the disease, some of whom have been found to have CHD, make it difficult to determine if the risk of coronary atherosclerosis is substantially altered by the enzyme deficiency. There is no specific treatment for LCAT deficiency. Corneal and kidney transplantation are performed in these patients in order to ameliorate their major clinical disabilities.

Apolipoprotein AI Mutations

Apo AI is the major structural protein of HDL. The gene encoding apo AI is located on chromosome 11 and comprises 4 exons. Following cleavage of the signal and prohormone sequences, a 243-amino-acid mature protein is produced. The protein has multiple repeats of an amphipathic helical structure that enables it to interact with both lipid and aqueous environments. Mutations that cause profound alterations in apo AI structure or expression have been reported, though they are extremely rare.73–76 The individuals carrying these mutations have virtually no circulating HDL and typically develop early CHD. There are, however, a number of case reports of apo AI mutations that are not associated with early CHD. Corneal opacities and xanthomas have been documented in many but not all individuals with major apo AI mutations. Most patients harboring these mutations have been found to be homozygous for the gene defect, usually as a result of consanguinity. Heterozygotes carrying these mutations commonly have half-normal HDL cholesterol levels, although cases with greater reductions in HDL levels have been reported, suggesting that some mutations may exert a dominant-negative effect. HDL typically contains four apo A molecules per particle (either four apo AI or two apo AI and two apo AII proteins), so a heterozygous individual carrying an expressed apo AI mutant would have at least one mutant apo AI on most HDL particles. The atherosclerosis of individuals with structural mutations in apo AI appears to be much more pronounced than that seen in patients with either LCAT deficiency or Tangier disease. No specific therapy is available for this disorder, but aggressive LDL lipid-lowering therapy is justified.

Diet/Lifestyle Therapies

With the advent of very potent lipid-lowering therapies, the use of dietary approaches to controlling lipid levels has become deemphasized in much of the developed world. While drug treatments (especially for elevated LDL cholesterol levels) are very effective, the role of diet in promoting hyperlipidemia should not be forgotten. The dietary approach to the treatment of elevated LDL cholesterol levels focuses on reductions in total fat, saturated fat, partially hydrogenated unsaturated fatty acids, and dietary cholesterol. Substituting foods that provide polyunsaturated and monounsaturated fats in place of saturated and trans unsaturated fat is particularly important, whereas the value of reducing the total fat intake is less clear.¹⁰³ The 2008 clinical practice guidelines of the American Diabetes Association (ADA) recommend that saturated fat intake be restricted to less than 7% of calories, and that cholesterol intake not exceed 200 mg per day in patients with diabetes. These dietary recommendations mimic those proposed by the NCEP ATP III guidelines for individuals without diabetes who have established coronary disease. For nondiabetics without coronary disease, the NCEP liberalizes saturated fat intake to less than 10% of calories and cholesterol to under 300 mg/d. In both diets, it is recommended that trans fat intake be kept to a minimum. The adoption of a Mediterranean-style diet, with substitution of monounsaturated fats for saturated fat, has demonstrated benefits on LDL cholesterol reduction, insulin sensitivity, and endothelial reactivity.^104–106 The magnitude of the LDL reduction achieved with dietary interventions is quite variable and depends on the preintervention diet, as well as metabolic and genetic factors that influence diet-dependent lipid responses. LDL reductions of 5% to 20% are typical for most patients adopting the reduced fat and cholesterol diets recommended by the ADA and NCEP, but reductions of over 50% can occur in selected individuals.¹⁰⁷

For patients with VLDL elevations, which includes a large portion of the diabetic population, carbohydrate and alcohol consumption may be more important dietary factors to address than the intake of cholesterol or saturated fat. The current FDA-recommended daily allowance of carbohydrate is 130 g/d, but studies of long-term carbohydrate-restricted diets are limited. Recently, investigations comparing low-carbohydrate and low-fat diets have demonstrated that low-carbohydrate diets can produce as good or even modestly better weight reduction than can low-fat diets in 6- to 12-month timeframes.108–110 The lower-carbohydrate, higher-fat diets typically produce greater reductions in VLDL (serum triglycerides) and greater increases in HDL cholesterol than do the fat-restricted, higher-carbohydrate diets. On the low-carbohydrate, higher-fat diets, the LDL cholesterol level can vary significantly, and if a substantial amount of saturated fat is used in these diets, the rise could be undesirable.¹¹¹ Therefore, the use of low-carbohydrate diets should be accompanied by the use of monounsaturated and polyunsaturated fats as the major sources of fat in the diet.

Weight loss (if obese), aerobic exercise, and smoking cessation can increase HDL levels and contribute to the dietary lowering of lower-density lipoproteins and CHD risk. They also reduce CHD risk by decreasing blood pressure. In the Diabetes Prevention Program (DPP), subjects in the lifestyle intervention group reduced their fat intake to 28% of calories after 1 year (down from 34%), and most were able to maintain the goal of 150 minutes per week of moderate physical activity.^112,113 On this program, in which a body weight loss of 7% was targeted, subjects experienced less progression to diabetes and reductions in multiple cardiovascular risk factors, including dyslipidemia and hypertension.^114,115 Although the DPP was conducted in an academic environment with substantial investigative resources devoted to ensuring compliance with the program, more recent efforts modeled on DPP suggest that the results can be translated to community-based outreach programs.^116,117

The clinical trials data that demonstrate improvements in CAD outcomes as a result of dietary interventions alone have generally employed diets that restrict fat and/or cholesterol intake much more dramatically than the ADA and NCEP diets mentioned. In the St. Thomas’ Regression Study (STARS), cholesterol intake in men was reduced to 100 to 120 mg/d, and excess weight was addressed by prescribing a physical activity program. After 3 years, the group treated with a low-cholesterol diet had a slower rate of progression of coronary disease and a higher rate of regression as determined by angiography.¹¹⁸ The Lifestyle Heart Trial employed an even more stringent diet, reducing cholesterol intake to under 10 mg/d, combined with an exercise and behavioral modification program. This intervention led to regression of coronary lesions as measured by angiography and a reduction in symptoms of ischemia.¹¹⁹ This study was conducted in a very small number of highly motivated volunteers; it is unlikely that the interventions could be replicated in more typical clinical care settings. A longer-term follow-up of a small number of participants in this study demonstrated that further regression of coronary atherosclerosis could be obtained.¹²⁰ A study with potentially greater applicability than either STARS or the Lifestyle Heart Trial is the Lyon Diet Heart Study.^121–123 Conducted using a Mediterranean diet rich in alpha-linolenic acid, with participants who had experienced a prior myocardial infarction, this trial demonstrated improvements in its primary endpoint (myocardial infarction and death) as well as several CV and secondary endpoints. The Lyon Diet Heart Study results, combined with other investigations of the impact of similar diets on lipids and lipoproteins, provide substantial support for the utility of Mediterranean-style diets in the prevention and treatment of CHD.

CHD Outcomes With HMG CoA Reductase Inhibitors (Statins)

The current era of lipid-lowering therapy began in 1984 with the publication of the Lipid Research Clinics–Coronary Prevention Trial. A randomized, double-blind study employing cholestyramine compared to placebo, this trial demonstrated that an 8% reduction in total cholesterol yielded a 19% reduction in CHD death or myocardial infarction.^124,125 With the advent of statin therapy in 1987, much more effective LDL cholesterol-lowering agents became available. During the 1990s, a series of landmark cardiovascular prevention trials were completed using these agents. In 1994, the Scandinavian Simvastatin Survival Study (4S) established for the first time that a lipid-lowering agent could prolong overall survival in individuals with preexisting CHD.¹²⁶ Randomizing 4444 patients with a prior myocardial infarction or angina to simvastatin or placebo, the drug treatment cohort of 4S had baseline LDL cholesterol levels fall from approximately 190 mg/dL to 120 mg/dL. With a trial endpoint of death for 10% of the original study enrollees (444 study subjects), it took a little over 5 years to complete the study. The result was that 8% of the simvastatin patients died versus 12% of the placebo-treated individuals. This 30% reduction in mortality established statin therapy as an essential intervention in CHD patients with significant LDL elevations. In addition to the mortality benefit, 4S also produced substantial reductions in major coronary events, overall CHD death (19% versus 28%, a relative risk reduction of 42%) and cerebrovascular events (2.7% versus 4.3%). Unlike several of the pre-statin era cholesterol intervention studies, there were no increases in other causes of mortality to counterbalance the benefits on the cardiovascular system. When 4S was examined in greater detail and after longer follow-up to determine if the therapy had been confined to any subset of patients, these analyses concluded that the benefit was widespread.^127–129 Individuals at the highest risk appeared to benefit the most, particularly those with diabetes.¹³⁰ Though only 202 of the 4444 patients in 4S were diagnosed with diabetes, these individuals experienced relative risk reductions in mortality (43%) and major CHD events (55%) that exceeded those in the nondiabetic population. An equally impressive outcome was noted in individuals who in addition to their elevated LDL cholesterol levels had lipid phenotypes characteristic of the metabolic syndrome (low HDL cholesterol levels and high triglycerides). Study subjects with this lipid triad had the highest event rates among those receiving placebo (35.9%) and the greatest reduction in relative risk with treatment (0.48; 95% CI, 0.33 to 0.69).¹³¹

Subsequent studies have further refined our understanding of the benefits of statins in patients with and without preexisting CHD. The West of Scotland Coronary Prevention Group study (WESCOPS) enrolled hyperlipidemic men who had not experienced a prior myocardial infarct but who had high levels of other CHD risk factors. Pravastatin reduced LDL cholesterols from approximately 190 mg/dL to 140 mg/dL and, compared to the placebo-treated group, lowered CHD death and myocardial infarction by 30%.¹³²

A similar-magnitude improvement in a younger population of lower-risk patients was observed in the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS) when lovastatin was given to men and women with starting LDL-C levels of 150 mg/dL.¹³³ When compared to the placebo-treated patients, those receiving lovastatin reduced their LDL on average to 115 mg/dL and decreased their incidence of major coronary events by 30% to 40%. In the Cholesterol and Recurrent Events Trial (CARE), pravastatin was given to patients with myocardial infarctions whose LDL cholesterol levels were substantially below (mean of 139 mg/dL) those of the 4S and WESCOPS patients. With a primary endpoint of a fatal coronary event or a nonfatal myocardial infarction, the pravastatin-treated group had a 24% relative risk reduction compared to placebo-treated patients.¹³⁴ Stroke was reduced 31%. A post hoc analysis of CARE by the investigators who performed it led to the conclusion that reduction of LDL cholesterol in the range from 174 mg/dL to 125 mg/dL lowered the coronary event rate, but that further reductions below 125 mg/dL did not appear to confer additional benefit.¹³⁵ The value of LDL reductions to levels substantially below 125 mg/dL has been a topic of considerable controversy ever since CARE was published, but several subsequent studies have indicated that further reductions of LDL do confer benefit. For example, the Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) study group enrolled over 9000 patients with recent evidence of CHD and demonstrated benefits in overall mortality and CHD death using pravastatin.¹³⁶ When stratified by starting LDL cholesterol, the magnitude of the benefit appeared to be maintained even in those with starting levels below 116 mg/dL.^137–139 The Treating to New Targets (TNT) trial enrolled 10,000 patients with stable CHD and randomized treatment to high-dose (80 mg/d) or low-dose (10 mg/d) atorvastatin.¹⁴⁰ Pretreatment LDL-C levels were between 130 and 250 mg/dL and posttreatment LDL-C values were 77 and 101, respectively, in the high- and low-dose groups. The primary endpoint of the study, which was major cardiovascular events, declined 22% more in the high-dose group than the low-dose cohort, as did rates of myocardial infarction and stroke. However, overall mortality was not lower in the high-dose group, owing to an increase in noncardiovascular deaths. Cancer deaths accounted for most of these events, a finding not explicable on the basis of other studies using the same drug.

The Heart Protection Study (HPS), like several of its predecessor trials, attempted to examine the question of LDL-C lowering in patients with moderate hyperlipidemia. HPS was a much bigger trial, randomizing 20,536 patients to placebo or simvastatin (up to 40 mg/d). The inclusion criteria required evidence of prior cardiovascular disease and permitted a broad range of LDL-C levels, with a third of the patients having values less than 116 mg/dL. After average follow-up of 5.5 years, all-cause mortality was reduced in the statin-treated cohort, as was CHD death and major cardiovascular events.¹⁴¹ Notably, over 5000 diabetics were included in the study, and they also achieved a substantial relative risk reduction in myocardial infarction and stroke.^142,143 Particularly striking was the finding that the benefits of LDL-C lowering were produced in all three of the tertiles into which patients’ baseline LDL-C were stratified. Even those with starting LDL-C levels approximating the current NCEP target goal of 100 mg/dL (for those with CHD) had a benefit when their value was reduced substantially below that number. Both men and women and older and younger patients derived similar benefits from the statin treatment. As with HPS, the Collaborative Atorvastatin Diabetes Study (CARDS) demonstrated a significant benefit in diabetic patients with lower LDL-C levels (<160 mg/dL) randomized to the statin arm of a placebo-controlled trial.¹⁴⁴ This benefit appeared in those whose starting LDL-C levels were both above and below 120 mg/dL. After 4 years of follow-up, the rate of cardiovascular events dropped 37% (95% CI, −52% to −17%).

In addition to studies of their effects on major clinical outcomes, many of the statins have been used in trials assessing their effects on atherosclerosis progression or regression in the coronary arteries or the carotids. The Familial Atherosclerosis Treatment Study (FATS) compared three regimens that all contained a bile acid resin binder, colestipol, plus either placebo, niacin, or lovastatin. The lovastatin-treated cohort lowered its LDL by 46% and had more regression and less progression of their CAD, as assessed by angiography, than did the placebo-treated group. The niacin group’s response was equivalent to that seen with the lovastatin treatment.¹⁴⁵ The Regression Growth Evaluation Statin Study (REGRESS) utilized pravastatin in over 800 men with total cholesterol levels between 155 and 310 mg/dL to demonstrate delayed progression in CHD as well as a lower rate of progression of femoral and carotid atherosclerosis.^146,147 The REVERSAL trial compared atorvastatin (80 mg/d) to pravastatin (40 mg/d) in patients with CHD by assessing plaque volume using intravascular ultrasound measurements (IVUS). Atheroma volume declined slightly in the atorvastatin-treated group while increasing in the pravastatin cohort, the difference between the two being statistically significant.¹⁴⁸ ASTEROID employed an open-label treatment with rosuvastatin (40 mg/d) in 507 patients with CHD to lower LDL-C by 53% (from a mean of 130 to 61 mg/dL). With this intervention, atheroma volume, as measured by IVUS, was reduced. No control group was included in ASTEROID, making it difficult to assess what is otherwise an impressive reduction in coronary atherosclerosis.¹⁴⁹ Taken together, these statin studies provide impressive evidence that aggressive lipid lowering can reduce the rate of CAD progression in patients with established disease and may, with the most intensive therapy, even cause plaque regression.

In addition to their use in studies of patients with longstanding stable CHD, statins have been employed in the setting of acute or recent coronary syndromes to determine if more aggressive lipid-lowering might provide additional benefit in these settings. The Prove-It study, for example, compared individuals with starting LDL-C levels of approximately 110 mg/dL who were randomized to received pravastatin (40 mg/d) or atorvastatin (80 mg/d) after an acute coronary event.¹⁵⁰ The high-dose atorvastatin group lowered LDL-C levels to 62 mg/dL, whereas the mid-dose pravastatin cohort only reduced their LDL-C values to 95 mg/dL. A 16% reduction in coronary events was achieved on the more aggressive therapy. It seems probable that this study established that an LDL-C of 62 mg/dL is better than one of 95 mg/dL in patients experiencing an acute coronary event, but the more rigorous conclusion is that 80 mg of atorvastatin works better for CHD outcomes in this population than does 40 mg of pravastatin. Similarly, the Incremental Decrease in End Points through Aggressive Lipid-lowering study used an open-label/blinded-endpoint study design to randomize nearly 9000 patients with a past history of myocardial infarction to receive either atorvastatin (80 mg/d) or simvastatin (20 mg/d). After 24 weeks, the dose of simvastatin could be raised to 40 mg/d if the LDL-C remained greater than 190 mg/dL. The atorvastatin-treated cohort after nearly 5 years had a mean LDL-C that was lower than the simvastatin-treated cohort (81 versus 104 mg/dL), but no significant reduction in the primary endpoint of major coronary events was demonstrated. While there were reductions in some secondary endpoints on atorvastatin, no benefit on overall mortality was seen. Because these studies employed different drugs and achieved different LDL endpoints using those agents, it is not always easy to interpret the outcomes. Overall, the data do support aggressive lipid-lowering therapy in the acute coronary setting, but it is much less clear whether one drug is superior to another when nearly comparable LDL-C lowering is achieved.

In summary, the outcomes studies presented in this section demonstrate that LDL cholesterol reduction with multiple members of the statin class of drugs results in improvements in CHD events and in overall total mortality in patients with established coronary disease. These benefits are present or even magnified in individuals with diabetes or the metabolic syndrome. Lower rates of CHD events are also produced in patients without established CHD but who have substantial CHD risk factors. In lower-risk populations, statins clearly reduce coronary events, but data are less convincing for benefits on CHD death or overall mortality. These conclusions have been confirmed by several meta-analyses of clinical trials that employed a variety of statins.151–157 Although the precise goal of LDL-C reduction remains controversial, the NCEP guideline targets of progressively lower target LDLs for progressively higher levels of risk (i.e., <130, <100, and <70 mg/dL) are both reasonable and consistent with the findings of outcomes studies published to date.

Concerns about cholesterol lowering producing increased risks of cancer and violent deaths, which emerged in some of the early, pre-statin lipid-lowering trials, have largely been put to rest in the major prospective statin studies. In five of the largest such trials conducted to date, using several different members of the statin class of drugs and collectively involving tens of thousands of patients (4S, WESCOPS, AFCAPS/TexCAPS, LIPID, and HPS studies), no increase in rates of cancer or noncardiac deaths have been reproducibly observed. However, each class of drug therapies has important adverse effects that need to be taken into account when considering pharmacologic intervention. With the trend towards more aggressive lowering of LDL cholesterol, achieved through higher doses of statins and more complex multidrug regimens, comes an increase in the risk of drug side effects.

Whether statin therapy should be initiated in young individuals is a topic of debate with little data available to inform the discussion. Young men (age less than 35) and premenopausal women with no CHD risk factors other than hypercholesterolemia are typically at very low 10-year risks for developing CHD. An argument can be made to wait until the risk level escalates before initiating cholesterol-lowering therapies, inasmuch as the benefits of statin therapy are typically evident in less than 2 to 3 years. An alternative view cites the evidence that individuals with genetically driven low LDL cholesterols have extremely low rates of CHD and that one should try to mimic this lifelong low-cholesterol state with earlier and more aggressive drug therapy.⁶⁸ Both perspectives have merit. For now, this decision will likely balance the magnitude of the individual risk with preferences for more or less aggressive therapy on the part of both physician and patient. Physicians prescribing statins for young women, however, must not forget the potential teratogenicity of this class of drugs. Nonpharmacologic therapy, primarily focusing on improved diets and physical activity programs, should be adopted regardless of age. As widespread clinic use of statins moves into its third decade and the initially approved agents have become available in generic forms, the safety and cost issues associated with their use may recede in importance, and a greater consensus on earlier use of these remarkable drugs becomes possible. For now, the exercise of good clinical judgment is essential in these circumstances.

Lipid-Lowering Outcomes Trials With Agents Other Than Statins

As cited in the previous section, the current era of lipid-lowering was successfully initiated with a bile acid resin binder, cholestyramine.^124,125 Though achieving an LDL-C reduction of only 8%, CHD events were reduced 19%. Thus, it is the belief of most experts in the lipid field that LDL cholesterol lowering achieved by any means will result in reduced CHD events, provided the therapeutic approach does not have off-target consequences that would countervail the benefit of the LDL lowering. This view is supported by a wealth of data on the pathogenesis of atherosclerosis, as well as clinical trials of multiple modalities that lower LDL cholesterol, including nonpharmacologic approaches.^{1,2,120,158,159} At the current time, statin use supersedes that of all other LDL-C lowering agents as first-line monotherapy, because of the superior efficacy and the wealth of outcomes data establishing its benefit. Many patients, however, either cannot meet the LDL targets set by the NCEP guidelines or cannot achieve the levels obtained in the major outcomes trials using statins alone. Some patients have inadequate therapeutic responses to the drugs, and others cannot tolerate their adverse side effects. When that occurs, the use of other lipid-lowering agents becomes relevant. Moreover, some patients, particularly those with diabetes, may have very low HDL-C levels or dramatic elevations in lipoproteins other than LDL, for which statin therapy is much less effective. The following section addresses the current data on outcomes when these other, nonstatin therapies are employed.

Fibrates and CHD Outcomes

Unlike statin therapy, none of the other lipid-lowering medications have produced an overall mortality benefit in a randomized, prospective, double-blind trial. For the most part, this is because studies employing the agents were not powered for a primary outcome focused on mortality but instead targeted reductions in CHD events. With this endpoint, bile acid resins, gemfibrozil, and niacin have all been demonstrated in long-term prospective studies to reduce CHD and/or major CV events.124,125,160–164 Although not available in the United States, bezafibrate also showed promising results in a secondary prevention study.¹⁶⁵ Since fibrates (gemfibrozil, fenofibrate, bezafibrate, clofibrate, and ciprofibrate) are more effective than statins at lowering serum triglyceride/VLDL levels, they would appear to be particularly attractive agents for the treatment of the dyslipidemia associated with diabetes and the metabolic syndrome. Subgroup analyses of many studies employing fibrates support this hypothesis, with greater benefit shown in those with high triglycerides or low HDL.¹⁶⁶ Insofar as neither bezafibrate nor ciprofibrate are available in the United States, and clofibrate is no longer used because of a significantly higher mortality rate produced by the drug, these agents will not be discussed in this section. The clofibrate findings, however, have cast a long shadow over other fibrates because of the fear that similar biliary and gastrointestinal diseases might accompany their use.¹⁶⁷ To date, this fear seems not to have been realized, though the lack of overall mortality benefit with fibrates continues to raise suspicion.

Of the fibrates, then, only fenofibrate and gemfibrozil are prescribed in the United States. Gemfibrozil was used in two major trials, the Helsinki Heart Study (HHS) and the Veterans Affairs High-Density Lipoprotein Intervention Trial (VA-HIT), with both reporting substantial benefit from the drug. The HHS was a double-blind, placebo-controlled 5-year trial that targeted HDL-C raising and non-HDL cholesterol lowering.¹⁶⁰ A group of 4081 asymptomatic men with non-HDL-C levels ≥200 mg/dL were studied over a 5-year period, with approximately half receiving 600 mg twice daily of the fibrate and half on placebo. HDL-C, non-HDL-C, LDL-C, and triglycerides all improved in the drug-treated cohort, which experienced a 34% reduction in CHD. On drug treatment, HDL-C rose 11%, LDL-C fell 10%, and triglycerides dropped 35%. Subsequent subgroup analysis of this study showed that the greatest benefits were achieved in those with BMIs above 26 kg/m² or in individuals that had higher triglyceride levels or lower HDL-C values.^168,169 In the VA-HIT study, 2531 men with established CHD, LDL-C levels ≤140 mg/dL and HDL-C levels ≤40 mg/dL were treated with gemfibrozil (1200 mg/d) or placebo.¹⁶¹ Those who received the fibrate experienced a 22% reduction in the primary outcome of nonfatal myocardial infarction or CHD death. These benefits were associated with a 6% higher HDL-C level, a 31% lower serum triglyceride value, and no significant change in LDL-C. Subgroup analysis in this study again revealed that those benefiting the most from the drug were individuals whose HDL-C rose. Somewhat surprisingly, the improvement in triglyceride was not associated with an improved CHD outcome.¹⁷⁰ The serious adverse events seen with clofibrate were not reproduced in the gemfibrozil studies, though overall mortality benefits were also not demonstrated in either study.

The results of the HHS and VA-HIT studies convinced many lipid experts that fibrate treatment was beneficial, and these agents have been widely prescribed in the past decade to treat hypertriglyceridemia and low HDL-C levels. Because gemfibrozil was found to affect the metabolism of multiple statins,¹⁷¹ many lipid experts switched to using fenofibrate whenever a statin/fibrate combination therapy was deemed appropriate in order to reduce the likelihood of a serious myopathy. Thus, the publication of the Fenofibrate Intervention and Event Lowering in Diabetes Study (FIELD) in 2005 was an important event in defining lipid therapeutics.¹⁷² This trial randomized 9795 middle-aged and older patients with type 2 diabetes to placebo and micronized fenofibrate (200 mg/d). Patients qualified by having total cholesterol levels between 116 and 252 mg/dL and total/HDL-C ratios ≥4.0 or plasma triglycerides between 88 and 435 mg/dL. An unusual feature of the FIELD trial was that patients were permitted to be treated with other lipid-lowering agents at their own physician’s discretion, so many received simultaneous statin therapy. The primary outcome of the study was CHD events. Triglycerides were reduced by the fibrate by 29%, but HDL-C rose only 2%. While CHD events were reduced in the drug-treated group by 11%, this overall outcome was not statistically significant. More worrisome was the fact that the overall event rate represented the average of a positive and negative outcome, nonfatal infarction falling 24%, but CHD mortality rising by 19%. When total mortality was examined, it was non-significantly higher in the fenofibrate cohort. Overall, the FIELD study was widely perceived to have failed to deliver the expected benefits suggested by the earlier gemfibrozil studies. This outcome has led to vigorous and protracted debates about the utility of fibrate therapy in general and fenofibrate in particular. In a recent consensus statement from the ADA and the American College of Cardiology Foundation, the authors favor niacin over fibrates as an agent to combine with statins in treating combined lipid disorders in which VLDL or HDL-C levels are targeted for treatment.¹⁷³ Despite the epidemiologic evidence linking high serum triglycerides to CHD events, the data from fibrate intervention studies have not shown convincing correlated reductions in triglycerides with CHD benefit. As pointed out earlier in VA-HIT, the effect on HDL-C levels, though modest, correlated with the desirable CHD outcomes, while the marked triglyceride reduction of 31% did not. Thus, the triglyceride effect of fibrates may be most clinically valuable when treating severely hypertriglyceridemic patients (serum triglycerides > 500 mg/dL) in whom the goal is to prevent triglyceride-induced pancreatitis. Ongoing studies with fibrates may clarify their place in the prevention of CHD, but at present, the currently available agents do not appear to have a great enough therapeutic benefit in most CHD patients to warrant widespread use.

Three other classes of drugs are commonly used to treat patients with hyperlipidemia: (1) nicotinic acid or niacin, (2) cholesterol absorption inhibitors (bile acid resins and ezetimibe), and (3) omega-3 fatty acids or fish oils. Outcomes studies with these agents are quite limited. The decrease in CHD produced with niacin in the Coronary Drug Project, combined with long-term follow-up of those originally assigned the drug, provide substantial confidence in the therapeutic value of using this B vitamin.^163,174 There is also evidence for benefit when niacin is combined with a statin.^175,176 Niacin can lower triglycerides, raise HDL cholesterol, and reduce LDL cholesterol, so it can provide a very useful adjunct to statin therapy in treating patients with combined lipid disorders. Its side-effect profile and its potential for increasing insulin resistance are the major drawbacks to employing niacin more widely. Bile acid resins were the mainstays of lipid therapy prior to the advent of statins; multiple CHD outcomes and atherosclerosis progression studies were performed with various members of the class. Overall, despite only modest LDL-C lowering performance, these drugs yielded benefits on CHD and coronary atherosclerosis when given as monotherapy or in combination with niacin or statins.^{118,124,125,145,177} Because bile acid resins can interfere with the absorption of other drugs, cause substantial gastrointestinal bloating and constipation, and often substantially increase serum triglycerides, the development of much more selective cholesterol absorption inhibitors was highly desirable. Ezetimibe, which inhibits cholesterol absorption from the gut via blockage of enterocyte cholesterol trafficking pathways, rapidly replaced bile acid resins in the lipid therapeutic armamentarium because of its ease of use. Unfortunately, ezetimibe has yet to demonstrate any meaningful cardiovascular outcome beyond its nearly 20% LDL-C lowering. In its first major outcome study, the addition of ezetimibe to simvastatin provided no additional benefit beyond simvastatin alone on carotid intimal medical thickness assessed by carotid sonography.¹⁷⁸ This controversial study demonstrated substantial additional LDL-C reduction with ezetimibe (16.5% greater than the simvastatin alone), indicating that the drug was effective in its primary action in the cohort that received it. The short duration of treatment, the mild degree of carotid disease at the outset of the study, and the use of patients previously treated with statins for many years may have all combined to make it extremely challenging for this study to have shown a benefit. These caveats notwithstanding, the fact remains that ezetimibe has no major outcomes data to support its use at present. Given the effectiveness of bile acid resins at reducing CHD events, it would be extremely surprising for the more potent ezetimibe not to confer at least the same benefit, unless there is some unexpected consequence of inhibiting the gut cholesterol trafficking pathway, or the drug has an as-yet-unidentified off-target effect. Ongoing studies should be completed in the next few years that provide insights into these issues. At present, ezetimibe should be used as an adjunct to statin therapy only when the latter has failed to achieve the target LDL goal at the maximum tolerated statin dose. In patients who cannot tolerate a statin, ezetimibe can also be considered, but niacin would be the first preferred alternative therapy if appropriate.

The final category of hypolipidemic agents, omega-3 fatty acids, are moderately effective agents in lowering elevated serum triglycerides. At doses of 2 to 4 g/d, omega-3 fatty acids inhibit VLDL production while having little or no effect on HDL-C and LDL-C. There are many epidemiologic studies of fish consumption that suggest a cardiovascular benefit, but there are no major CHD outcomes studies in which these compounds have been used to treat hypertriglyceridemic patients. The best available evidence suggests that omega-3 fatty acids may reduce cardiac arrhythmic events but have a modest effect if any on coronary atherosclerosis.179–181

HMG-CoA Reductase Inhibitors (Statins)

These agents have become the first-line drug therapy for treatment of elevated LDL levels because of their effectiveness, patient acceptability, clinical outcomes data, and favorable safety record. They block the rate-limiting enzyme in cholesterol synthesis, HMG-CoA reductase. Serum LDL levels fall by 20% to 60%, depending on dose and specific agent. HDL levels generally stay the same or increase slightly on statins. Statins also influence thrombotic and inflammatory mechanisms, effects whose importance remains to be elucidated. Myalgias, with or without creatine kinase elevations, are the most common side effect of the class. Asymptomatic hepatocellular dysfunction, manifested by an increase in serum levels of transaminases, occurs in approximately 1% of patients taking the medications. Statins that are currently available in the United States and Canada are lovastatin, simvastatin, pravastatin, fluvastatin, atorvastatin, and rosuvastatin. The oldest approved statins—lovastatin, simvastatin, and pravastatin—are all available in generic forms and are typically the approved agents on cost-sensitive formularies. Atorvastatin and rosuvastatin are the two most potent statins, with LDL-C reductions of more than 50% routinely achieved in patients given the higher dosages of either agent. Table 16-5 summarizes some key information about the available statins.

Niacin

Niacin is a B vitamin that must be used in megadoses to lower lipids. The precise mechanism of action remains although uncertain, although recent data indicate that it could block VLDL synthesis by inhibiting the activity of hepatic DGAT-2.¹⁸² The discovery of receptors for niacin and the elucidation of the prostaglandin-mediated mechanism by which it evokes flushing have contributed to renewed interest in a drug that has been used to treat lipids for over 50 years.^183,184 A new medication that combines niacin with an inhibitor of this flushing pathway has been developed but not yet approved by the FDA. Niacin, at doses of 1000 mg and higher, typically lowers LDL and VLDL levels and raises HDL levels. It is the most effective HDL cholesterol–raising drug available. The clinical trial results presented in the outcomes section of this chapter provide sufficient safety and efficacy data to conclude that niacin is an effective anti-atherosclerotic agent, though there are no long-term, randomized prospective clinical trials establishing an overall mortality benefit with its use. Niacin’s principal disadvantages include a litany of side effects reflecting the large doses required to achieve effective lipid lowering. Niacin can exacerbate gout and diabetes, elevate liver enzymes, and produce rashes, nausea, and vomiting. The impact on diabetes with the long-acting, single daily dosing formulation of niacin (Niaspan) appears to be modest (less than 0.3 unit rise in glycated hemoglobin), so many lipid experts recommend its use in diabetics to treat their hypertriglyceridemia, low HDL-C levels, or as adjunctive therapy to statins for LDL-C lowering.^185,186 The acute vasodilation associated with flushing can occasionally lead to postural light-headedness, a reaction that aspirin can often mitigate. Niacin is available in both prescription and nonprescription formulations, but physicians should be cautious in recommending no-flush versions of the drug; these often lack any therapeutic efficacy.¹⁸⁷ Many patients can tolerate niacin if pretreated with aspirin and given slowly escalating doses, building weekly in 500-mg increments. Typical effective dosing requires from 1500 mg to 3000 mg/d, with side effects increasing as the dose escalates.

Cholesterol Absorption Inhibitors (Cholestyramine, Colestipol, Colesevelam, Ezetimibe)

These agents block intestinal cholesterol absorption. The bulk agents (cholestyramine, colestipol, and colesevelam) do this by interfering with micellar solubilization of cholesterol, whereas ezetimibe interacts with the Niemann-Pick C1-like 1 (NPC1L1) protein to block enterocyte cholesterol trafficking. LDL reduction is typically 15% to 20% with these agents. Gastrointestinal sides effects such as constipation, bloating, heartburn, and nausea are the major drawbacks to the use of the bulk agents. They are commonly used as an adjunct to statin therapy to achieve greater LDL reductions or when statins are not tolerated. In addition to interfering with the absorption of multiple drugs, the bulk agents can also cause hypertriglyceridemia. Effective dosing of the bulk agents requires 10 to 30 g/d, while ezetimibe is available only as a single 10 mg/d tablet.

Fibrates (Gemfibrozil and Fenofibrate)

These agents work by activating the peroxisome proliferator–activated receptor alpha (PPAR-α) nuclear transcription factor. The result is a decrease in VLDL synthesis and enhanced VLDL clearance, so these drugs are primarily effective in the treatment of elevated triglyceride levels. The drugs can be used in combination with statins to treat patients with combined VLDL and LDL disorders, but this therapy is associated with an increased likelihood of serious muscle breakdown leading to rhabdomyolytic renal failure. The available literature suggests that fenofibrate is less likely to produce the metabolic changes that precipitate statin-induced myopathy, but as outlined in the section of fibrate treatment outcomes, the data supporting outcomes benefits favor gemfibrozil. Gemfibrozil is typically given as 600-mg tablets twice daily, whereas fenofibrate is available in several different formulations (145 to 200 mg, depending on formulation) that are typically administered as a single pill per day. Side effects are rare but include abdominal discomfort.