Daniel J. Rader Helen H. Hobbs
Lipoproteins are complexes of lipids and proteins that are essential for the transport of cholesterol, triglycerides, and fat-soluble vitamins. Previously, lipoprotein disorders were the purview of specialized lipidologists, but the demonstration that lipid-lowering therapy significantly reduces the clinical complications of atherosclerotic cardiovascular disease (ASCVD) has brought the diagnosis and treatment of these disorders into the domain of the internist. The number of individuals who are candidates for lipid-lowering therapy has continued to increase. The development of safe, effective, and well-tolerated pharmacologic agents has greatly expanded the therapeutic armamentarium available to the physician to treat disorders of lipid metabolism. Therefore, the appropriate diagnosis and management of lipoprotein disorders is of critical importance in the practice of medicine. This chapter will review normal lipoprotein physiology, the pathophysiology of primary (inherited) disorders of lipoprotein metabolism, the diseases and environmental factors that cause secondary disorders of lipoprotein metabolism, and the practical approaches to their diagnosis and management.
LIPOPROTEIN CLASSIFICATION AND COMPOSITION
Lipoproteins are large macromolecular complexes that transport hydrophobic lipids (primarily triglycerides, cholesterol, and fat-soluble vitamins) through body fluids (plasma, interstitial fluid, and lymph) to and from tissues. Lipoproteins play an essential role in the absorption of dietary cholesterol, long-chain fatty acids, and fat-soluble vitamins; the transport of triglycerides, cholesterol, and fat-soluble vitamins from the liver to peripheral tissues; and the transport of cholesterol from peripheral tissues to the liver.
Lipoproteins contain a core of hydrophobic lipids (triglycerides and cholesteryl esters) surrounded by hydrophilic lipids (phospholipids, unesterified cholesterol) and proteins that interact with body fluids. The plasma lipoproteins are divided into five major classes based on their relative density (Fig. 31-1 and Table 31-1): chylomicrons, very low-density lipoproteins (VLDLs), intermediate-density lipoproteins (IDLs), low-density lipoproteins (LDLs), and high-density lipoproteins (HDLs). Each lipoprotein class comprises a family of particles that vary slightly in density, size, and protein composition. The density of a lipoprotein is determined by the amount of lipid per particle. HDL is the smallest and most dense lipoprotein, whereas chylomicrons and VLDLs are the largest and least dense lipoprotein particles. Most plasma triglyceride is transported in chylomicrons or VLDLs, and most plasma cholesterol is carried as cholesteryl esters in LDLs and HDLs.
The density and size distribution of the major classes of lipoprotein particles. Lipoproteins are classified by density and size, which are inversely related. VLDL, very low-density lipoprotein; IDL, intermediate-density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein.
MAJOR LIPOPROTEIN CLASSES
The proteins associated with lipoproteins, called apolipoproteins(Table 31-2), are required for the assembly, structure, and function of lipoproteins. Apolipoproteins activate enzymes important in lipoprotein metabolism and act as ligands for cell surface receptors. ApoA-I, which is synthesized in the liver and intestine, is found on virtually all HDL particles. ApoA-II is the second most abundant HDL apolipoprotein and is on approximately two-thirds of the HDL particles. ApoB is the major structural protein of chylomicrons, VLDLs, IDLs, and LDLs; one molecule of apoB, either apoB-48 (chylomicron) or apoB-100 (VLDL, IDL, or LDL), is present on each lipoprotein particle. The human liver synthesizes apoB-100, and the intestine makes apoB-48, which is derived from the same gene by mRNA editing. ApoE is present in multiple copies on chylomicrons, VLDL, and IDL, and it plays a critical role in the metabolism and clearance of triglyceride-rich particles. Three apolipoproteins of the C-series (apoC-I, apoC-II, and apoC-III) also participate in the metabolism of triglyceride-rich lipoproteins. ApoB is the only major apolipoprotein that does not transfer between lipoprotein particles. Some of the minor apolipoproteins are listed in Table 31-2.
TRANSPORT OF DIETARY LIPIDS (EXOGENOUS PATHWAY)
The exogenous pathway of lipoprotein metabolism permits efficient transport of dietary lipids (Fig. 31-2). Dietary triglycerides are hydrolyzed by lipases within the intestinal lumen and emulsified with bile acids to form micelles. Dietary cholesterol, fatty acids, and fat-soluble vitamins are absorbed in the proximal small intestine. Cholesterol and retinol are esterified (by the addition of a fatty acid) in the enterocyte to form cholesteryl esters and retinyl esters, respectively. Longer-chain fatty acids (>12 carbons) are incorporated into triglycerides and packaged with apoB-48, cholesteryl esters, retinyl esters, phospholipids, and cholesterol to form chylomicrons. Nascent chylomicrons are secreted into the intestinal lymph and delivered via the thoracic duct directly to the systemic circulation, where they are extensively processed by peripheral tissues before reaching the liver. The particles encounter lipoprotein lipase (LPL), which is anchored to a glycosylphosphatidylinositol-anchored protein, GPIHBP1, that is attached to the endothelial surfaces of capillaries in adipose tissue, heart, and skeletal muscle (Fig. 31-2). The triglycerides of chylomicrons are hydrolyzed by LPL, and free fatty acids are released. ApoC-II, which is transferred to circulating chylomicrons from HDL, acts as a required cofactor for LPL in this reaction. The released free fatty acids are taken up by adjacent myocytes or adipocytes and either oxidized to generate energy or reesterified and stored as triglyceride. Some of the released free fatty acids bind albumin before entering cells and are transported to other tissues, especially the liver. The chylomicron particle progressively shrinks in size as the hydrophobic core is hydrolyzed and the hydrophilic lipids (cholesterol and phospholipids) and apolipoproteins on the particle surface are transferred to HDL, creating chylomicron remnants. Chylomicron remnants are rapidly removed from the circulation by the liver through a process that requires apoE as a ligand for receptors in the liver. Consequently, few, if any, chylomicrons or chylomicron remnants are present in the blood after a 12-h fast, except in patients with disorders of chylomicron metabolism.
The exogenous and endogenous lipoprotein metabolic pathways. The exogenous pathway transports dietary lipids to the periphery and the liver. The endogenous pathway transports hepatic lipids to the periphery. LPL, lipoprotein lipase; FFA, free fatty acid; VLDL, very low-density lipoprotein; IDL, intermediate-density lipoprotein; LDL, low-density lipoprotein; LDLR, low-density lipoprotein receptor; HL, hepatic lipase.
TRANSPORT OF HEPATIC LIPIDS (ENDOGENOUS PATHWAY)
The endogenous pathway of lipoprotein metabolism refers to the secretion of apoB-containing lipoproteins from the liver and the metabolism of these triglyceride-rich particles in peripheral tissues (Fig. 31-2). VLDL particles resemble chylomicrons in protein composition but contain apoB-100 rather than apoB-48 and have a higher ratio of cholesterol to triglyceride (~1 mg of cholesterol for every 5 mg of triglyceride). The triglycerides of VLDL are derived predominantly from the esterification of long-chain fatty acids in the liver. The packaging of hepatic triglycerides with the other major components of the nascent VLDL particle (apoB-100, cholesteryl esters, phospholipids, and vitamin E) requires the action of the enzyme microsomal triglyceride transfer protein (MTP). After secretion into the plasma, VLDL acquires multiple copies of apoE and apolipoproteins of the C series by transfer from HDL. As with chylomicrons, the triglycerides of VLDL are hydrolyzed by LPL, especially in muscle, heart, and adipose tissue. After the VLDL remnants dissociate from LPL, they are referred to as IDLs, which contain roughly similar amounts of cholesterol and triglyceride. The liver removes approximately 40–60% of IDL by LDL receptor–mediated endocytosis via binding to apoE. The remainder of IDL is remodeled by hepatic lipase (HL) to form LDL. During this process, most of the triglyceride in the particle hydrolyzed, and all apolipoproteins except apoB-100 are transferred to other lipoproteins. The cholesterol in LDL accounts for more than one-half of the plasma cholesterol in most individuals. Approximately 70% of circulating LDL is cleared by LDL receptor–mediated endocytosis in the liver. Lipoprotein(a) [Lp(a)] is a lipoprotein similar to LDL in lipid and protein composition, but it contains an additional protein called apolipoprotein(a) [apo(a)]. Apo(a) is synthesized in the liver and attached to apoB-100 by a disulfide linkage. The major site of clearance of Lp(a) is the liver, but the uptake pathway is not known.
HDL METABOLISM AND REVERSE CHOLESTEROL TRANSPORT
All nucleated cells synthesize cholesterol, but only hepatocytes and enterocytes can effectively excrete cholesterol from the body, into either the bile or the gut lumen. In the liver, cholesterol is secreted into the bile, either directly or after conversion to bile acids. Cholesterol in peripheral cells is transported from the plasma membranes of peripheral cells to the liver and intestine by a process termed “reverse cholesterol transport” that is facilitated by HDL (Fig. 31-3).
HDL metabolism and reverse cholesterol transport. This pathway transports excess cholesterol from the periphery back to the liver for excretion in the bile. The liver and the intestine produce nascent HDLs. Free cholesterol is acquired from macrophages and other peripheral cells and esterified by LCAT, forming mature HDLs. HDL cholesterol can be selectively taken up by the liver via SR-BI (scavenger receptor class BI). Alternatively, HDL cholesteryl ester can be transferred by CETP from HDLs to VLDLs and chylomicrons, which can then be taken up by the liver. LCAT, lecithin-cholesterol acyltransferase; CETP, cholesteryl ester transfer protein; VLDL, very low-density lipoprotein; IDL, intermediate-density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein; LDLR, low-density lipoprotein receptor.
Nascent HDL particles are synthesized by the intestine and the liver. Newly secreted apoA-I rapidly acquires phospholipids and unesterified cholesterol from its site of synthesis (intestine or liver) via efflux promoted by the membrane protein ATP-binding cassette protein A1 (ABCA1). This process results in the formation of discoidal HDL particles, which then recruit additional unesterified cholesterol from the periphery. Within the HDL particle, the cholesterol is esterified by lecithin-cholesterol acyltransferase (LCAT), a plasma enzyme associated with HDL, and the more hydrophobic cholesteryl ester moves to the core of the HDL particle. As HDL acquires more cholesteryl ester it becomes spherical, and additional apolipoproteins and lipids are transferred to the particles from the surfaces of chylomicrons and VLDLs during lipolysis.
HDL cholesterol is transported to hepatocytes by both an indirect and a direct pathway. HDL cholesteryl esters can be transferred to apoB-containing lipoproteins in exchange for triglyceride by the cholesteryl ester transfer protein (CETP). The cholesteryl esters are then removed from the circulation by LDL receptor– mediated endocytosis. HDL cholesterol can also be taken up directly by hepatocytes via the scavenger receptor class B1 (SR-B1), a cell surface receptor that mediates the selective transfer of lipids to cells.
HDL particles undergo extensive remodeling within the plasma compartment by a variety of lipid transfer proteins and lipases. The phospholipid transfer protein (PLTP) has the net effect of transferring phospholipids from other lipoproteins to HDL or among different classes of HDL particles. After CETP- and PLTP-mediated lipid exchange, the triglyceride-enriched HDL becomes a much better substrate for HL, which hydrolyzes the triglycerides and phospholipids to generate smaller HDL particles. A related enzyme called endothelial lipase hydrolyzes HDL phospholipids, generating smaller HDL particles that are catabolized faster. Remodeling of HDL influences the metabolism, function, and plasma concentrations of HDL.
DISORDERS OF LIPOPROTEIN METABOLISM
Fredrickson and Levy classified hyperlipoproteinemias according to the type of lipoprotein particles that accumulate in the blood (Type I to Type V) (Table 31-3). A classification scheme based on the molecular etiology and pathophysiology of the lipoprotein disorders complements this system and forms the basis for this chapter. The identification and characterization of genes responsible for the genetic forms of hyperlipidemia have provided important molecular insights into the critical roles of structural apolipoproteins, enzymes, and receptors in lipid metabolism (Table 31-4).
FREDRICKSON CLASSIFICATION OF HYPERLIPOPROTEINEMIAS
PRIMARY HYPERLIPOPROTEINEMIAS CAUSED BY KNOWN SINGLE GENE MUTATIONS
PRIMARY DISORDERS OF ELEVATED ApoB-CONTAINING LIPOPROTEINS
A variety of genetic conditions are associated with the accumulation in plasma of specific classes of lipoprotein particles. In general, these can be divided into those causing elevated LDL-cholesterol (LDL-C) with normal triglycerides and those causing elevated triglycerides (Table 31-4).
Lipid disorders associated with elevated LDL-C and normal triglycerides
Familial hypercholesterolemia (FH)
FH is an autosomal codominant disorder characterized by elevated plasma levels of LDL-C with normal triglycerides, tendon xanthomas, and premature coronary atherosclerosis. FH is caused by a large number (>1000) mutations in the LDL receptor gene. It has a higher incidence in certain founder populations, such as Afrikaners, Christian Lebanese, and French Canadians. The elevated levels of LDL-C in FH are due to an increase in the production of LDL from IDL (since a portion of IDL is normally cleared by LDL receptor–mediated endocytosis) and a delayed removal of LDL from the blood. Individuals with two mutated LDL receptor alleles (FH homozygotes) have much higher LDL-C levels than those with one mutant allele (FH heterozygotes).
Homozygous FH occurs in approximately 1 in 1 million persons worldwide. Patients with homozygous FH can be classified into one of two groups based on the amount of LDL receptor activity measured in their skin fibroblasts: those patients with <2% of normal LDL receptor activity (receptor negative) and those patients with 2–25% of normal LDL receptor activity (receptor defective). Most patients with homozygous FH present in childhood with cutaneous xanthomas on the hands, wrists, elbows, knees, heels, or buttocks. Total cholesterol levels are usually >500 mg/dL and can be higher than 1000 mg/dL. The devastating complication of homozygous FH is accelerated atherosclerosis, which can result in disability and death in childhood. Atherosclerosis often develops first in the aortic root, where it can cause aortic valvular or supravalvular stenosis, and typically extends into the coronary ostia, which become stenotic. Children with homozygous FH often develop symptomatic coronary atherosclerosis before puberty; symptoms can be atypical, and sudden death is not uncommon. Untreated, receptor-negative patients with homozygous FH rarely survive beyond the second decade; patients with receptor-defective LDL receptor defects have a better prognosis but almost invariably develop clinically apparent atherosclerotic vascular disease by age 30, and often much sooner. Carotid and femoral disease develops later in life and is usually not clinically significant.
A careful family history should be taken, and plasma lipid levels should be measured in the parents and other first-degree relatives of patients with homozygous FH. The disease has >90% penetrance so both parents of FH homozygotes usually have hypercholesterolemia. The diagnosis of homozygous FH can be confirmed by obtaining a skin biopsy and measuring LDL receptor activity in cultured skin fibroblasts, or by quantifying the number of LDL receptors on the surfaces of lymphocytes using cell sorting technology. Molecular assays are also available to define the mutations in the LDL receptor by DNA sequencing. In selected populations where particular mutations predominate (e.g., Africaners and French Canadians), the common mutations can be screened for directly. Alternatively, the entire coding region needs to be sequenced for mutation detection because a large number of different LDL receptor mutations can cause disease. Ten to 15% of LDL receptor mutations are large deletions or insertions, which may be missed by routine DNA sequencing.
Combination therapy with an HMG-CoA reductase inhibitor and a second drug (cholesterol absorption inhibitor or bile acid sequestrant) sometimes reduces plasma LDL-C in those FH homozygotes who have residual LDL receptor activity, but patients with homozygous FH invariably require additional lipid-lowering therapy. Since the liver is quantitatively the most important tissue for removing circulating LDLs via the LDL receptor, liver transplantation is effective in decreasing plasma LDL-C levels in this disorder. Liver transplantation, however, is associated with substantial risks, including the requirement for long-term immunosuppression. The current treatment of choice for homozygous FH is LDL apheresis (a process by which the LDL particles are selectively removed from the circulation), which can promote regression of xanthomas and may slow the progression of atherosclerosis. Initiation of LDL apheresis should generally be delayed until approximately 5 years of age, except when evidence of atherosclerotic vascular disease is present.
Heterozygous FH is caused by the inheritance of one mutant LDL receptor allele and occurs in approximately 1 in 500 persons worldwide, making it one of the most common single-gene disorders. It is characterized by elevated plasma levels of LDL-C (usually 200–400 mg/dL) and normal levels of triglyceride. Patients with heterozygous FH have hypercholesterolemia from birth, and disease recognition is usually based on detection of hypercholesterolemia on routine screening, the appearance of tendon xanthomas, or the development of symptomatic ASCVD. Since the disease is codominant in inheritance, one parent and ~50% of the patient’s siblings usually also have hypercholesterolemia. The family history is frequently positive for premature ASCVD on one side of the family. Corneal arcus is common, and tendon xanthomas involving the dorsum of the hands, elbows, knees, and especially the Achilles tendons are present in ~75% of patients. The age of onset of ASCVD is highly variable and depends in part on the molecular defect in the LDL receptor gene and also on coexisting cardiac risk factors. FH heterozygotes with elevated plasma levels of Lp(a) appear to be at greater risk for cardiovascular complications. Untreated men with heterozygous FH have an ~50% chance of having a myocardial infarction before age 60 years. Although the age of onset of atherosclerotic heart disease is later in women with FH, coronary heart disease (CHD) is significantly more common in women with FH than in the general female population.
No definitive diagnostic test for heterozygous FH is available. Although FH heterozygotes tend to have reduced levels of LDL receptor function in skin fibroblasts, significant overlap with the LDL receptor activity levels in normal fibroblasts exists. Molecular assays are now available to identify mutations in the LDL receptor gene by DNA sequencing, but the clinical utility of pinpointing the mutation has not been demonstrated. The clinical diagnosis is usually not problematic, but it is critical that hypothyroidism, nephrotic syndrome, and obstructive liver disease be excluded before initiating therapy.
FH patients should be aggressively treated to lower plasma levels of LDL-C. Initiation of a low-cholesterol, low-fat diet is recommended, but heterozygous FH patients require lipid-lowering drug therapy. Statins are effective in heterozygous FH, but combination drug therapy with the addition of a cholesterol absorption inhibitor and/or bile acid sequestrant is frequently required, and the addition of nicotinic acid is sometimes needed. Heterozygous FH patients who cannot be adequately controlled on combination drug therapy are candidates for LDL apheresis.
Familial defective ApoB-100 (FDB)
FDB is a dominantly inherited disorder that clinically resembles heterozygous FH. The disease is rare in most populations except individuals of German descent, where the frequency can be as high as 1 in 1000. FDB is characterized by elevated plasma LDL-C levels with normal triglycerides, tendon xanthomas, and an increased incidence of premature ASCVD. FDB is caused by mutations in the LDL receptor–binding domain of apoB-100, most commonly due to a substitution of glutamine for arginine at position 3500. As a consequence of the mutation in apoB-100, LDL binds the LDL receptor with reduced affinity, and LDL is removed from the circulation at a reduced rate. Patients with FDB cannot be clinically distinguished from patients with heterozygous FH, although patients with FDB tend to have lower plasma levels of LDL-C than FH heterozygotes. The apoB-100 gene mutation can be detected directly, but genetic diagnosis is not currently encouraged since the recommended management of FDB and heterozygous FH is identical.
Autosomal dominant hypercholesterolemia due to mutations in PCSK9 (ADH-PCSK9 or ADH3)
ADH-PCSK9 is a rare autosomal dominant disorder caused by gain-of-function mutations in proprotein convertase subtilisin/kexin type 9 (PCSK9). PCSK9 is a secreted protein that binds to the LDL receptor, resulting in its degradation. Normally, after LDL binds to the receptor it is internalized along with the receptor. In the low pH of the endosome, LDL dissociates from the receptor and returns to the cell surface. The LDL is delivered to the lysosome. When PCSK9 binds the receptor, the complex is internalized and the receptor is redirected to the lysosome rather than to the cell surface. The missense mutations in PCSK9 that cause hypercholesterolemia enhance the activity of PCSK9. As a consequence, the number of hepatic LDL receptors is reduced. Patients with ADH-PCSK9 are indistinguishable clinically from patients with FH. Interestingly, loss-of-function mutations in PCSK9 cause low LDL-C levels (see later).
Autosomal recessive hypercholesterolemia (ARH)
ARH is a rare disorder (except in Sardinia, Italy) due to mutations in a protein (ARH, also called LDLR adaptor protein, LDLRAP) involved in LDL receptor–mediated endocytosis in the liver. In the absence of LDLRAP, LDL binds to the LDL receptor but the lipoprotein-receptor complex fails to be internalized. ARH, like homozygous FH, is characterized by hypercholesterolemia, tendon xanthomas, and premature coronary artery disease (CAD). The levels of plasma LDL-C tend to be intermediate between the levels present in FH homozygotes and FH heterozygotes, and CAD is not usually symptomatic until at least the third decade. LDL receptor function in cultured fibroblasts is normal or only modestly reduced in ARH, whereas LDL receptor function in lymphocytes and the liver is negligible. Unlike FH homozygotes, the hyperlipidemia responds partially to treatment with HMG-CoA reductase inhibitors, but these patients usually require LDL apheresis to lower plasma LDL-C to recommended levels.
Sitosterolemia is another rare autosomal recessive disease that can result in severe hypercholesterolemia, tendon xanthomas, and premature ASCVD. Sitosterolemia is caused by mutations in either of two members of the ATP-binding cassette (ABC) half transporter family, ABCG5 and ABCG8. These genes are expressed in enterocytes and hepatocytes. The proteins heterodimerize to form a functional complex that pumps plant sterols such as sitosterol and campesterol, and animal sterols, predominantly cholesterol, into the gut lumen and into the bile. In normal individuals, <5% of dietary plant sterols are absorbed by the proximal small intestine and delivered to the liver. Absorbed plant sterols are preferentially secreted into the bile and are maintained at very low levels. In sitosterolemia, the intestinal absorption of sterols is increased and biliary excretion of the sterols is reduced, resulting in increased plasma and tissue levels of both plant sterols and cholesterol.
Incorporation of plant sterols into cell membranes results in misshapen red blood cells and megathrombocytes that are visible on blood smear. Episodes of hemolysis are a distinctive clinical feature of this disease compared to other genetic forms of hypercholesterolemia.
Sitosterolemia is diagnosed by demonstrating an increase in the plasma level of sitosterol using gas chromatography. The hypercholesterolemia is unusually responsive to reductions in dietary cholesterol content and should be suspected in individuals who have a >40% reduction in plasma cholesterol level on a lowcholesterol diet. The hypercholesterolemia does not respond to HMG-CoA reductase inhibitors, whereas bile acid sequestrants and cholesterol-absorption inhibitors such as ezetimibe, are effective in reducing plasma sterol levels in these patients.
This condition is characterized by hypercholesterolemia due to elevated LDL-C with a normal plasma level of triglyceride in the absence of secondary causes of hypercholesterolemia. Plasma LDL-C levels are generally not as elevated as they are in FH and FDB. Family studies are useful to differentiate polygenic hypercholesterolemia from the single-gene disorders described above; one-half of the first-degree relatives of patients with FH and FDB are hypercholesterolemic, whereas <10% of first-degree relatives of patients with polygenic hypercholesterolemia have hypercholesterolemia. Treatment of polygenic hypercholesterolemia is identical to that of other forms of hypercholesterolemia.
Elevated plasma levels of lipoprotein(a)
Unlike the other major classes of lipoproteins, that have a normal distribution in the population, plasma levels of Lp(a) have a highly skewed distribution with levels varying over a 1000-fold range. Levels are strongly influenced by genetic factors, with individuals of African and South Asian descent having higher levels than those of European descent. Although it has been well documented that elevated levels of Lp(a) are associated with an increase in ASCVD, lowering plasma levels of Lp(a) has not been demonstrated to reduce cardiovascular risk.
Lipid disorders associated with elevated triglycerides
Familial chylomicronemia syndrome (Type I hyperlipoproteinemia; lipoprotein lipase, and ApoC-II deficiency)
As noted above, LPL is required for the hydrolysis of triglycerides in chylomicrons and VLDLs, and apoC-II is a cofactor for LPL (Fig. 31-2). Genetic deficiency or inactivity of either protein results in impaired lipolysis and profound elevations in plasma chylomicrons. These patients can also have elevated plasma levels of VLDL, but chylomicronemia predominates. The fasting plasma is turbid, and if left at 4°C (39.2°F) for a few hours, the chylomicrons float to the top and form a creamy supernatant. In these disorders, called familial chylomicronemia syndromes, fasting triglyceride levels are almost invariably >1000 mg/dL. Fasting cholesterol levels are also elevated but to a lesser degree.
LPL deficiency has autosomal recessive inheritance and has a frequency of approximately 1 in 1 million in the population. ApoC-II deficiency is also recessive in inheritance pattern and is even less common than LPL deficiency. Multiple different mutations in the LPL and apoC-II genes cause these diseases. Obligate LPL heterozygotes have normal or mild-to-moderate elevations in plasma triglyceride levels, whereas individuals heterozygous for mutation in apoC-II do not have hypertriglyceridemia.
Both LPL and apoC-II deficiency usually present in childhood with recurrent episodes of severe abdominal pain due to acute pancreatitis. On funduscopic examination, the retinal blood vessels are opalescent (lipemia retinalis). Eruptive xanthomas, which are small, yellowish-white papules, often appear in clusters on the back, buttocks, and extensor surfaces of the arms and legs. These typically painless skin lesions may become pruritic. Hepatosplenomegaly results from the uptake of circulating chylomicrons by reticuloendothelial cells in the liver and spleen. For unknown reasons, some patients with persistent and pronounced chylomicronemia never develop pancreatitis, eruptive xanthomas, or hepatosplenomegaly. Premature CHD is not generally a feature of familial chylomicronemia syndromes.
The diagnoses of LPL and apoC-II deficiency are established enzymatically in specialized laboratories by assaying triglyceride lipolytic activity in postheparin plasma. Blood is sampled after an IV heparin injection to release the endothelial-bound LPL. LPL activity is profoundly reduced in both LPL and apoC-II deficiency; in patients with apoC-II deficiency, it normalizes after the addition of normal plasma (providing a source of apoC-II). Molecular sequencing of the genes can be used to confirm the diagnosis.
The major therapeutic intervention in familial chylomicronemia syndromes is dietary fat restriction (to as little as 15 g/d) with fat-soluble vitamin supplementation. Consultation with a registered dietician familiar with this disorder is essential. Caloric supplementation with medium-chain triglycerides, which are absorbed directly into the portal circulation, can be useful but may be associated with hepatic fibrosis if used for prolonged periods. If dietary fat restriction alone is not successful in resolving the chylomicronemia, fish oils have been effective in some patients. In patients with apoC-II deficiency, apoC-II can be provided by infusing fresh-frozen plasma to resolve the chylomicronemia in the acute setting. Management of patients with familial chylomicronemia syndrome is particularly challenging during pregnancy when VLDL production is increased and may require plasmapheresis to remove the circulating chylomicrons.
Another apolipoprotein, apoA-V, circulates at much lower concentrations than the other major apolipoproteins. Individuals harboring mutations in both apoA-V alleles can present as adults with chylomicronemia. The exact mechanism of action of apoA-V is not known, but it appears to be required for the association of VLDL and chylomicrons with LPL.
After LPL is synthesized in adipocytes, myocytes, or other cells, it is transported across the vascular endothelium and is attached to a protein on the endothelial surface of capillaries called GPIHBP1. Homozygosity for mutations that interfere with GPIHBP1 synthesis or folding cause severe hypertriglyceridemia. The frequency of chylomicronemia due to mutations in GHIHBP1 has not been established but appears to be very rare.
Hepatic lipase deficiency
HL is a member of the same gene family as LPL and hydrolyzes triglycerides and phospholipids in remnant lipoproteins and HDLs. HL deficiency is a very rare autosomal recessive disorder characterized by elevated plasma levels of cholesterol and triglycerides (mixed hyperlipidemia) due to the accumulation of circulating lipoprotein remnants and either a normal or elevated plasma level of HDL-C. The diagnosis is confirmed by measuring HL activity in postheparin plasma. Due to the small number of patients with HL deficiency, the association of this genetic defect with ASCVD is not clearly known, but lipid-lowering therapy is recommended.
Familial dysbetalipoproteinemia (Type III hyperlipoproteinemia)
Like HL deficiency, familial dysbetalipoproteinemia (FDBL) (also known as Type III hyperlipoproteinemia or familial broad β disease) is characterized by a mixed hyperlipidemia due to the accumulation of remnant lipoprotein particles. ApoE is present in multiple copies on chylomicron and VLDL remnants and mediates their removal via hepatic lipoprotein receptors (Fig. 31-2). FDBL is due to genetic variations in apoE that interfere with its ability to bind lipoprotein receptors. The APOE gene is polymorphic in sequence, resulting in the expression of three common isoforms: apoE3, which is the most common; and apoE2 and apoE4, which both differ from apoE3 by a single amino acid. Although associated with slightly higher LDL-C levels and increased CHD risk, the apoE4 allele is not associated with FDBL. Patients with apoE4 have an increased incidence of late-onset Alzheimer’s disease. ApoE2 has a lower affinity for the LDL receptor; therefore, chylomicron and VLDL remnants containing apoE2 are removed from plasma at a slower rate. Individuals who are homozygous for the E2 allele (the E2/E2 genotype) comprise the most common subset of patients with FDBL.
Approximately 0.5% of the general population are apoE2/E2 homozygotes, but only a small minority of these individuals develop FDBL. In most cases, an additional, identifiable factor precipitates the development of hyperlipoproteinemia. The most common precipitating factors are a high-fat diet, diabetes mellitus, obesity, hypothyroidism, renal disease, HIV infection, estrogen deficiency, alcohol use, or certain drugs. Other mutations in apoE can cause a dominant form of FDBL where the hyperlipidemia is fully manifest in the heterozygous state, but these mutations are rare.
Patients with FDBL usually present in adulthood with incidental hyperlipidemia, xanthomas, premature coronary disease, or peripheral vascular disease. The disease seldom presents in women before menopause. Two distinctive types of xanthomas, tuberoeruptive and palmar, are seen in FDBL patients. Tuberoeruptive xanthomas begin as clusters of small papules on the elbows, knees, or buttocks and can grow to the size of small grapes. Palmar xanthomas (alternatively called xanthomata striata palmaris) are orange-yellow discolorations of the creases in the palms and wrists. In FDBL, in contrast to other disorders of elevated triglycerides, the plasma levels of cholesterol and triglyceride are often elevated to a similar degree and the level of HDL-C is usually normal rather than being low.
The traditional approaches to diagnosis of this disorder are lipoprotein electrophoresis (broad β band) or ultracentrifugation (ratio of VLDL-C to total plasma triglyceride >0.30). Protein methods (apoE phenotyping) or DNA-based methods (apoE genotyping) can be performed to confirm homozygosity for apoE2. However, absence of the apoE2/E2 genotype does not rule out the diagnosis of FDBL, since other mutations in apoE can cause this condition.
Since FDBL is associated with increased risk of premature ASCVD, it should be treated aggressively. Subjects with FDBL tend to have more peripheral vascular disease than is typically seen in FH. Other metabolic conditions that can worsen the hyperlipidemia (see earlier) should be aggressively treated. Patients with FDBL are typically very diet responsive and can respond favorably to weight reduction and to low-cholesterol, low-fat diets. Alcohol intake should be curtailed. HMGCoA reductase inhibitors, fibrates, and niacin are all generally effective in the treatment of FDBL, and sometimes combination drug therapy is required.
Familial hypertriglyceridemia (FHTG)
FHTG is a relatively common (~1 in 500) autosomal dominant disorder of unknown etiology characterized by moderately elevated plasma triglycerides accompanied by more modest elevations in cholesterol. Since the major class of lipoproteins elevated in this disorder is VLDL, patients with this disorder are often referred to as having Type IV hyperlipoproteinemia (Fredrickson classification, Table 31-3). The elevated plasma levels of VLDL are due to increased production of VLDL, impaired catabolism of VLDL, or a combination of these mechanisms. Some patients with FHTG have a more severe form of hyperlipidemia in which both VLDLs and chylomicrons are elevated (Type V hyperlipidemia), since these two classes of lipoproteins compete for the same lipolytic pathway. Increased intake of simple carbohydrates, obesity, insulin resistance, alcohol use, and estrogen treatment, all of which increase VLDL synthesis, can exacerbate this syndrome. FHTG appears not to be associated with increased risk of ASCVD in many families.
The diagnosis of FHTG is suggested by the triad of elevated levels of plasma triglycerides (250–1000 mg/dL), normal or only mildly increased cholesterol levels (<250 mg/dL), and reduced plasma levels of HDL-C. Plasma LDL-C levels are generally not increased and are often reduced due to defective metabolism of the triglyceride-rich particles. The identification of other first-degree relatives with hypertriglyceridemia is useful in making the diagnosis. FDBL and familial combined hyperlipidemia (FCHL) should also be ruled out since these two conditions are associated with a significantly increased risk of ASCVD. The plasma apoB levels are lower and the ratio of plasma triglyceride to cholesterol is higher in FHTG than in either FDBL or FCHL.
It is important to consider and rule out secondary causes of hypertriglyceridemia (Table 31-5) before making the diagnosis of FHTG. Lipid-lowering drug therapy can frequently be avoided with appropriate dietary and lifestyle changes. Patients with plasma triglyceride levels >500 mg/dL after a trial of diet and exercise should be considered for drug therapy to avoid the development of chylomicronemia and pancreatitis. Fibrate drugs or fish oils (omega 3 fatty acids) are reasonable first-line approaches for FHTG, and niacin can also be considered in this condition. For more moderate elevations in triglyceride levels (250–500 mg/dL), statins are effective at lowering triglyceride levels.
SECONDARY FORMS OF HYPERLIPIDEMIA
Familial combined hyperlipidemia (FCHL)
FCHL is generally characterized by moderate elevations in plasma levels of triglycerides (VLDL) and cholesterol (LDL) and reduced plasma levels of HDL-C. Approximately 20% of patients who develop CHD under age 60 have FCHL. The disease appears to be autosomal dominant with incomplete penetrance and affected family members typically have one of three possible phenotypes: (1) elevated plasma levels of LDL-C, (2) elevated plasma levels of triglycerides due to elevation in VLDL, or (3) elevated plasma levels of both LDL-C and triglyceride. A classic feature of FCHL is that the lipoprotein profile can switch among these three phenotypes in the same individual over time and may depend on factors such as diet, exercise, and weight. FCHL can manifest in childhood but is usually not fully expressed until adulthood. A cluster of other metabolic risk factors are often found in association with this hyperlipidemia, including obesity, glucose intolerance, insulin resistance, and hypertension (the so-called metabolic syndrome, Chap. 32). These patients do not develop xanthomas.
Patients with FCHL almost always have significantly elevated plasma levels of apoB. The levels of apoB are disproportionately high relative to the plasma LDL-C concentration, indicating the presence of small, dense LDL particles, which are characteristic of this syndrome. Hyperapobetalipoproteinemia, which has been used to describe the state of elevated plasma levels of apoB with normal plasma LDL-C levels, is probably a form of FCHL. Individuals with FCHL generally share the same metabolic defect, which is overproduction of VLDL by the liver. The molecular etiology of FCHL remains poorly understood, and it is likely that defects in several different genes can cause the phenotype of FCHL.
The presence of a mixed dyslipidemia (plasma triglyceride levels between 200 and 800 mg/dL and total cholesterol levels between 200 and 400 mg/dL, usually with HDL-C levels <40 mg/dL in men and <50 mg/dL in women) and a family history of hyperlipidemia and/or premature CHD strongly suggests the diagnosis of FCHL.
Individuals with FCHL should be treated aggressively due to significantly increased risk of premature CHD. Decreased dietary intake of saturated fat and simple carbohydrates, aerobic exercise, and weight loss can all have beneficial effects on the lipid profile. Patients with diabetes should be aggressively treated to maintain good glucose control. Most patients with FCHL require lipid-lowering drug therapy to reduce lipoprotein levels to the recommended range and reduce the high risk of ASCVD. Statins are effective in this condition, but many patients will require a second drug (cholesterol absorption inhibitor, niacin, fibrate, or fish oils) for optimal control of lipoprotein levels.
INHERITED CAUSES OF LOW LEVELS OF ApoB-CONTAINING LIPOPROTEINS
Familial hypobetalipoproteinemia (FHB)
Low plasma levels of LDL-C (the “β-lipoprotein”) with a genetic or inherited basis are referred to generically as familial hypobetalipoproteinemia. Traditionally, this term has been used to refer to the condition of low total cholesterol and LDL-C due to mutations in apoB, which represents the most common inherited form of hypocholesterolemia. Most of the mutations causing FHB interfere with the production of apoB, resulting in reduced secretion and/or accelerated catabolism of the protein. Individuals heterozygous for these mutations usually have LDL-C levels <80 mg/dL and may enjoy protection from ASCVD, though this has not been rigorously demonstrated. Some heterozygotes have elevated levels of hepatic triglycerides.
Mutations in both apoB alleles cause homozygous FHB, a disorder resembling abetalipoproteinemia (see later), although the neurologic findings tend to be less severe. Patients with homozygous hypobetalipoproteinemia can be distinguished from individuals with abetalipoproteinemia by measuring the levels of LDL-C in the parents, which are low in hypobetalipoproteinemia and normal in abetalipoproteinemia.
A phenocopy of FHB results from loss-of-function mutations in PCSK9. As reviewed earlier, PCSK9 normally promotes the degradation of the LDL receptor. Mutations that interfere with the synthesis of PCSK9, which are more common in individuals of African descent, result in increased LDL receptor activity and ~40% reduction in plasma level of LDL-C. A sequence variation of higher frequency (R46L) is found predominantly in individuals of European descent and is associated with a 15% reduction in LDL-C. Individuals with inactivating mutations are protected from developing CHD relative to those without these sequence variations, presumably due to having lower plasma cholesterol levels since birth.
The synthesis and secretion of apoB-containing lipoproteins in the enterocytes of the proximal small bowel and in the hepatocytes of the liver involve a complex series of events that coordinate the coupling of various lipids with apoB-48 and apoB-100, respectively. Abetalipoproteinemia is a rare autosomal recessive disease caused by loss-of-function mutations in the gene encoding microsomal triglyceride transfer protein (MTP), a protein that transfers lipids to nascent chylomicrons and VLDLs in the intestine and liver, respectively. Plasma levels of cholesterol and triglyceride are extremely low in this disorder, and chylomicrons, VLDLs, LDLs, and apoB are undetectable in plasma. The parents of patients with abetalipoproteinemia (obligate heterozygotes) have normal plasma lipid and apoB levels. Abetalipoproteinemia usually presents in early childhood with diarrhea and failure to thrive due to fat malabsorption. The initial neurologic manifestations are loss of deep-tendon reflexes, followed by decreased distal lower extremity vibratory and proprioceptive sense, dysmetria, ataxia, and the development of a spastic gait, often by the third or fourth decade. Patients with abetalipoproteinemia also develop a progressive pigmented retinopathy presenting with decreased night and color vision, followed by reductions in daytime visual acuity and ultimately progressing to near-blindness. The presence of spinocerebellar degeneration and pigmented retinopathy in this disease has resulted in some patients with abetalipoproteinemia being misdiagnosed as having Friedreich’s ataxia.
Most clinical manifestations of abetalipoproteinemia result from defects in the absorption and transport of fat-soluble vitamins. Vitamin E and retinyl esters are normally transported from enterocytes to the liver by chylomicrons, and vitamin E is dependent on VLDL for transport out of the liver and into the circulation. As a consequence of the inability of these patients to secrete apoB-containing particles, patients with abetalipoproteinemia are markedly deficient in vitamin E and are also mildly to moderately deficient in vitamins A and K. Patients with abetalipoproteinemia should be referred to specialized centers for confirmation of the diagnosis and appropriate therapy. Treatment consists of a lowfat, high-caloric, vitamin-enriched diet accompanied by large supplemental doses of vitamin E. It is imperative that treatment be initiated as soon as possible to help forestall development of neurologic sequelae, which can progress even with appropriate therapy. New therapies for this serious disease are needed.
GENETIC DISORDERS OF HDL METABOLISM
Mutations in genes encoding proteins that play critical roles in HDL synthesis and catabolism can result in both reductions and elevations in plasma levels of HDL-C. Unlike the genetic forms of hypercholesterolemia, which are invariably associated with premature coronary atherosclerosis, genetic forms of hypoalphalipoproteinemia (low HDL-C) are not always associated with accelerated atherosclerosis.
INHERITED CAUSES OF LOW LEVELS OF HDL-C
Gene deletions in the ApoAV-AI-CIII-AIV locus and coding mutations in ApoA-I
Complete genetic deficiency of apoA-I due to deletion of the apoA-I gene results in the virtual absence of HDL from the plasma. The genes encoding apoA-I, apoC-III, apoA-IV, and apoA-V are clustered together on chromosome 11, and some patients with no apoA-I have genomic deletions that include other genes in the cluster. ApoA-I is required for LCAT activity. In the absence of LCAT, free cholesterol levels increase in both plasma (not HDL) and in tissues. The free cholesterol can form deposits in the cornea and in the skin, resulting in corneal opacities and planar xanthomas. Premature CHD is a common feature of apoA-I deficiency, especially when additional genes in the complex are also deleted.
Missense and nonsense mutations in the apoA-I gene have been identified in some patients with low plasma levels of HDL-C (usually 15–30 mg/dL), but these are very rare causes of low HDL-C levels. Patients heterozygous for an Arg173Cys substitution in APOAI (so-called apoA-IMilano) have very low plasma levels of HDL due to impaired LCAT activation and rapid catabolism of the mutant apolipoprotein and yet have no increased risk of premature CHD. Most other individuals with low plasma HDL-C levels due to missense mutations in apoA-I do not appear to have premature CHD. A few selected missense mutations in apoA-I and apoA-II promote the formation of amyloid fibrils causing systemic amyloidosis.
Tangier disease (ABCA1 deficiency)
Tangier disease is a very rare autosomal codominant form of extremely low plasma HDL-C caused by mutations in the gene encoding ABCA1, a cellular transporter that facilitates efflux of unesterified cholesterol and phospholipids from cells to apoA-I (Fig. 31-3). ABCA1 in the liver and intestine rapidly lipidates the apoA-I secreted from these tissues. In the absence of ABCA1, the nascent, poorly lipidated apoA-I is immediately cleared from the circulation. Thus, patients with Tangier disease have extremely low circulating plasma levels of HDL-C (<5 mg/dL) and apoA-I (<5 mg/dL). Cholesterol accumulates in the reticuloendothelial system of these patients, resulting in hepatosplenomegaly and pathognomonic enlarged, grayish yellow or orange tonsils. An intermittent peripheral neuropathy (mono-neuritis multiplex) or a sphingomyelia-like neurologic disorder can also be seen in this disorder. Tangier disease is probably associated with some increased risk of premature atherosclerotic disease, although the association is not as robust as might be anticipated, given the very low levels of HDL-C and apoA-I in these patients. Patients with Tangier disease also have low plasma levels of LDL-C, which may attenuate the atherosclerotic risk. Obligate heterozygotes for ABCA1 mutations have moderately reduced plasma HDL-C levels (15–30 mg/dL) but their risk of premature CHD remains uncertain. ABCA1 mutations appear to be the cause of low HDL-C in a minority of individuals.
This very rare autosomal recessive disorder is caused by mutations in LCAT, an enzyme synthesized in the liver and secreted into the plasma, where it circulates associated with lipoproteins (Fig. 31-3). As reviewed earlier, the enzyme is activated by apoA-I and mediates the esterification of cholesterol to form cholesteryl esters. Consequently, in LCAT deficiency the proportion of free cholesterol in circulating lipoproteins is greatly increased (from ~25% to >70% of total plasma cholesterol). Lack of normal cholesterol esterification impairs formation of mature HDL particles, resulting in the rapid catabolism of circulating apoA-I. Two genetic forms of LCAT deficiency have been described in humans: complete deficiency (also called classic LCAT deficiency) and partial deficiency (also called fish-eye disease). Progressive corneal opacification due to the deposition of free cholesterol in the cornea, very low plasma levels of HDL-C (usually <10 mg/dL), and variable hypertriglyceridemia are characteristic of both disorders. In partial LCAT deficiency, there are no other known clinical sequelae. In contrast, patients with complete LCAT deficiency have hemolytic anemia and progressive renal insufficiency that eventually leads to end-stage renal disease (ESRD). Remarkably, despite the extremely low plasma levels of HDL-C and apoA-I, premature ASCVD is not a consistent feature of either LCAT deficiency or fisheye disease. The diagnosis can be confirmed in a specialized laboratory by assaying plasma LCAT activity or by sequencing the LCAT gene.
Low plasma levels of HDL-C (the “alpha lipoprotein”) is referred to as hypoalphalipoproteinemia. Primary hypoalphalipoproteinemia is defined as a plasma HDL-C level below the tenth percentile in the setting of relatively normal cholesterol and triglyceride levels, no apparent secondary causes of low plasma HDL-C, and no clinical signs of LCAT deficiency or Tangier disease. This syndrome is often referred to as isolated low HDL. A family history of low HDL-C facilitates the diagnosis of an inherited condition, which usually follows an autosomal dominant pattern. The metabolic etiology of this disease appears to be primarily accelerated catabolism of HDL and its apolipoproteins. Some of these patients may have ABCA1 mutations and therefore technically have heterozygous Tangier disease. Several kindreds with primary hypoalphalipoproteinemia have been described in association with an increased incidence of premature CHD, although this is not an invariant association. Association of hypoalphalipoproteinemia with premature CHD may depend on the specific nature of the gene defect or the underlying metabolic defect responsible for the low plasma HDL-C level.
INHERITED CAUSES OF HIGH LEVELS OF HDL-C
Loss-of-function mutations in both alleles of the gene encoding CETP cause substantially elevated HDL-C levels (usually >150 mg/dL). As noted earlier, CETP facilitates the transfer of cholesteryl esters from HDL to apoB-containing lipoproteins (Fig. 31-3). The absence of this transfer results in an increase in the cholesteryl ester content of HDL and a reduction in plasma levels of LDL-C. The large, cholesterol-rich HDL particles circulating in these patients are cleared at a reduced rate. CETP deficiency was first diagnosed in Japanese persons and is rare outside of Japan. The relationship of CETP deficiency to ASCVD remains unresolved. Heterozygotes for CETP deficiency have only modestly elevated HDL-C levels. Based on the phenotype of high HDL-C in CETP deficiency, pharmacologic inhibition of CETP is under development as a new therapeutic approach to both raise HDL-C levels and lower LDL-C levels, but whether it will reduce risk of ASCVD remains to be determined.
The condition of high plasma levels of HDL-C is referred to as hyperalphalipoproteinemia and is defined as a plasma HDL-C level above the ninetieth percentile. This trait runs in families, and outside of Japan it is unlikely to be due to CETP deficiency. Most, but not all, persons with this condition appear to have a reduced risk of CHD and increased longevity. Recent evidence is consistent with mutations in endothelial lipase contributing to this phenotype in some cases.
SECONDARY DISORDERS OF LIPOPROTEIN METABOLISM
Significant changes in plasma levels of lipoproteins are seen in a variety of diseases. It is crucial that secondary causes of dyslipidemias (Table 31-5) are considered prior to initiation of lipid-lowering therapy.
Obesity is frequently accompanied by dyslipidemia. The increase in adipocyte mass and accompanying decreased insulin sensitivity associated with obesity has multiple effects on lipid metabolism. More free fatty acids are delivered from the expanded adipose tissue to the liver, where they are reesterified in hepatocytes to form triglycerides, which are packaged into VLDLs for secretion into the circulation. The increased insulin levels promote fatty acid synthesis in the liver. Increased dietary intake of simple carbohydrates also drives hepatic production of VLDLs, resulting in elevations in VLDL and/or LDL in some obese subjects. Plasma levels of HDL-C tend to be low in obesity, due in part to reduced lipolysis. Weight loss is often associated with reductions in plasma levels of circulating apoB-containing lipoproteins and increases in the plasma levels of HDL-C.
Patients with type I diabetes mellitus generally do not have hyperlipidemia if they remain under good glycemic control. Diabetic ketoacidosis is frequently accompanied by hypertriglyceridemia due to an increased hepatic influx of free fatty acids from adipose tissue. Patients with type II diabetes mellitus are usually dyslipidemic, even when under relatively good glycemic control. The high levels of insulin and insulin resistance associated with type II diabetes has multiple effects on fat metabolism: (1) a decrease in LPL activity resulting in reduced catabolism of chylomicrons and VLDLs, (2) an increase in the release of free fatty acid from the adipose tissue, (3) an increase in fatty acid synthesis in the liver, and (4) an increase in hepatic VLDL production. Patients with type II diabetes mellitus have several lipid abnormalities, including elevated plasma triglycerides (due to increased VLDL and lipoprotein remnants), elevated levels of dense LDL, and decreased plasma levels of HDL-C. In some diabetic patients, especially those with a genetic defect in lipid metabolism, the triglycerides can be extremely elevated, resulting in the development of pancreatitis. Elevated plasma LDL-C levels usually are not a feature of diabetes mellitus and suggest the presence of an underlying lipoprotein abnormality or may indicate the development of diabetic nephropathy.
Lipodystrophy is associated with profound insulin resistance and elevated plasma levels of VLDL and chylomicrons that can be especially difficult to control. Those with congenital generalized lipodystrophy have absence of subcutaneous fat associated with muscle hypertrophy and hepatic steatosis; some of these patients have been treated successfully with leptin. Partial lipodystrophy can present with dyslipidemia and the diagnosis should be entertained in patients with variations in body fat distribution, particularly increased truncal fat accompanied by reduced fat in the buttocks and extremities.
Hypothyroidism is associated with elevated plasma LDL-C levels due primarily to a reduction in hepatic LDL receptor function and delayed clearance of LDL. Conversely, plasma levels of LDL-C are often reduced in the hyperthyroid patient. Hypothyroid patients also frequently have increased levels of circulating IDL, and some patients with hypothyroidism also have mild hypertriglyceridemia. Because hypothyroidism is often subtle and therefore easily overlooked, all patients presenting with elevated plasma levels of LDL-C, IDL, or triglycerides should be screened for hypothyroidism. Thyroid replacement therapy usually ameliorates the hypercholesterolemia; if not, the patient probably has a primary lipoprotein disorder and may require lipid-lowering drug therapy.
Nephrotic syndrome is often associated with pronounced hyperlipoproteinemia, which is usually mixed but can manifest as hypercholesterolemia or hypertriglyceridemia. The hyperlipidemia of nephrotic syndrome appears to be due to a combination of increased hepatic production and decreased clearance of VLDLs, with increased LDL production. Effective treatment of the underlying renal disease normalizes the lipid profile, but most patients with chronic nephrotic syndrome require lipid-lowering drug therapy.
ESRD is often associated with mild hypertriglyceridemia (<300 mg/dL) due to the accumulation of VLDLs and remnant lipoproteins in the circulation. Triglyceride lipolysis and remnant clearance are both reduced in patients with renal failure. Because the risk of ASCVD is increased in ESRD subjects with hyperlipidemia, they should probably be aggressively treated with lipid-lowering agents, even though there is inadequate data at present to indicate that this population benefits from LDL-lowering therapy.
Patients with renal transplants usually have increased lipid levels due to the effect of the drugs required for immunosuppression (cyclosporine and glucocorticoids) and present a difficult management problem since HMG-CoA reductase inhibitors must be used cautiously in these patients.
Because the liver is the principal site of formation and clearance of lipoproteins, it is not surprising that liver diseases can affect plasma lipid levels in a variety of ways. Hepatitis due to infection, drugs, or alcohol is often associated with increased VLDL synthesis and mild to moderate hypertriglyceridemia. Severe hepatitis and liver failure are associated with dramatic reductions in plasma cholesterol and triglycerides due to reduced lipoprotein biosynthetic capacity. Cholestasis is associated with hypercholesterolemia, which can be very severe. A major pathway by which cholesterol is excreted from the body is via secretion into bile, either directly or after conversion to bile acids, and cholestasis blocks this critical excretory pathway. In cholestasis, free cholesterol, coupled with phospholipids, is secreted into the plasma as a constituent of a lamellar particle called LP-X. The particles can deposit in skinfolds, producing lesions resembling those seen in patients with FDBL (xanthomata strata palmaris). Planar and eruptive xanthomas can also be seen in patients with cholestasis.
Regular alcohol consumption has a variable effect on plasma lipid levels. The most common effect of alcohol is to increase plasma triglyceride levels. Alcohol consumption stimulates hepatic secretion of VLDL, possibly by inhibiting the hepatic oxidation of free fatty acids, which then promote hepatic triglyceride synthesis and VLDL secretion. The usual lipoprotein pattern seen with alcohol consumption is Type IV (increased VLDLs), but persons with an underlying primary lipid disorder may develop severe hypertriglyceridemia (Type V) if they drink alcohol. Regular alcohol use also raises plasma levels of HDL-C.
Estrogen administration is associated with increased VLDL and HDL synthesis, resulting in elevated plasma levels of both triglycerides and HDL-C. This lipoprotein pattern is distinctive since the levels of plasma triglyceride and HDL-C are typically inversely related. Plasma triglyceride levels should be monitored when birth control pills or postmenopausal estrogen therapy is initiated to ensure that the increase in VLDL production does not lead to severe hypertriglyceridemia. Use of low-dose preparations of estrogen or the estrogen patch can minimize the effect of exogenous estrogen on lipids.
Lysosomal storage diseases
Cholesteryl ester storage disease (due to deficiency in lysosomal acid lipase) and glycogen storage diseases such as von Gierke’s disease (caused by mutations in glucose-6-phosphatase) are rare causes of secondary hyperlipidemias.
Glucocorticoid excess is associated with increased VLDL synthesis and hypertriglyceridemia. Patients with Cushing’s syndrome can also have mild elevations in plasma levels of LDL-C.
Many drugs have an impact on lipid metabolism and can result in significant alterations in the lipoprotein profile (Table 31-5).
(See also Chaps. 2 and 32) Guidelines for the screening and management of lipid disorders have been provided by an expert Adult Treatment Panel (ATP) convened by the National Cholesterol Education Program (NCEP) of the National Heart, Lung, and Blood Institute. The NCEP ATPIII guidelines published in 2001 recommend that all adults older than age 20 years should have plasma levels of cholesterol, triglyceride, LDL-C, and HDL-C measured after a 12-h overnight fast. In most clinical laboratories, the total cholesterol and triglycerides in the plasma are measured enzymatically, and then the cholesterol in the supernatant is measured after precipitation of apoB-containing lipoproteins to determine the HDL-C. The LDL-C is estimated using the following equation:
LDL-C = total cholesterol – (triglycerides/5) – HDL-C
(The VLDL-C is estimated by dividing the plasma triglyceride by 5, reflecting the ratio of cholesterol to triglyceride in VLDL particles.) This formula is reasonably accurate if test results are obtained on fasting plasma and if the triglyceride level does not exceed ~200 mg/dL; by convention it cannot be used if the triglyceride level is >400 mg/dL. The accurate determination of LDL-C levels in patients with triglyceride levels >200 mg/dL requires application of ultracentrifugation techniques or other direct assays for LDL-C. If the triglyceride level is >200 mg/dL, the guidelines recommend that the “non-HDL-C” be calculated by simple subtraction of HDL-C from the total cholesterol and that this be considered a secondary target of therapy. Further evaluation and treatment is based primarily on the plasma LDL-C and non-HDL-C levels as well as assessment of overall cardiovascular risk.
The critical first step in managing a lipid disorder is to determine the class or classes of lipoproteins that are increased or decreased in the patient. The Fredrickson classification scheme for hyperlipoproteinemias (Table 31-3), though less commonly used now than in the past, can be helpful in this regard. Once the hyperlipidemia is accurately classified, efforts should be directed to rule out any possible secondary causes of the hyperlipidemia (Table 31-5). Although many patients with hyperlipidemia have a primary or genetic cause of their lipid disorder, secondary factors frequently contribute to the hyperlipidemia. A fasting glucose should be obtained in the initial workup of all subjects with an elevated triglyceride level. Nephrotic syndrome and chronic renal insufficiency should be excluded by obtaining urine protein and serum creatinine. Liver function tests should be performed to rule out hepatitis and cholestasis. Hypothyroidism should be ruled out by measuring serum TSH. Patients with hyperlipidemia, especially hypertriglyceridemia, who drink alcohol should be encouraged to decrease their intake. Sedentary lifestyle, obesity, and smoking are all associated with low HDL-C levels, and patients should be counseled about these issues.
Once secondary causes for the elevated lipoprotein levels have been ruled out, attempts should be made to diagnose the primary lipid disorder since the underlying etiology has a significant effect on the risk of developing CHD, on the response to drug therapy, and on the management of other family members. Often, determining the correct diagnosis requires a detailed family medical history and, in some cases, lipid analyses in family members.
If the fasting plasma triglyceride level is >1000 mg/dL, the patient almost always has chylomicronemia and either has Type I or Type V hyperlipoproteinemia (Table 31-3). The plasma triglyceride to cholesterol ratio helps distinguish between these two possibilities and is higher in Type I than Type V hyperlipoproteinemia. If the patient has Type I hyperlipoproteinemia, a postheparin lipolytic assay should be performed to determine if the patient has LPL or apoC-II deficiency. Type V is a much more frequent form of chylomicronemia in the adult patient. Often treatment of secondary factors contributing to the hyperlipidemia (diet, obesity, glucose intolerance, alcohol ingestion, estrogen therapy) will change a Type V into a Type IV pattern, reducing the risk of developing acute pancreatitis.
If the levels of LDL-C are very high (greater than a 95th percentile), it is likely the patient has a genetic form of hyperlipidemia. The presence of severe hypercholesterolemia, tendon xanthomas, and an autosomal dominant pattern of inheritance are consistent with the diagnosis of either FH, FDB, or ADH-PCSK9. At the present time, there is no compelling reason to perform molecular studies to further refine the molecular diagnosis, since the treatment of FH and FDB is identical. Recessive forms of severe hypercholesterolemia are rare and if the patient with severe hypercholesterolemia has parents with normal cholesterol levels, sitosterolemia should be considered; a clue to the diagnosis of sitosterolemia is the greater than expected response of the hypercholesterolemia to reductions in dietary cholesterol content or to treatment with either a cholesterol absorption inhibitor (ezetimibe) or to bile acid resins. Patients with more moderate hypercholesterolemia that does not segregate in families as a monogenic trait are likely to have polygenic hypercholesterolemia.
The most common error in the diagnosis and treatment of lipid disorders is in patients with a mixed hyperlipidemia without chylomicronemia. Elevations in the plasma levels of both cholesterol and triglycerides are seen in patients with increased plasma levels of IDL (Type III) and of LDL and VLDL (Type IIB) and in patients with increased levels of VLDL (Type IV). The ratio of triglyceride to cholesterol is higher in Type IV than the other two disorders. The plasma levels of apoB are highest in Type IIB. A beta quantification to determine the VLDL-C/triglyceride ratio in plasma (see discussion of FDBL) or a direct measurement of the plasma LDL-C should be performed at least once prior to initiation of lipid-lowering therapy to determine if the hyperlipidemia is due to the accumulation of remnants or to an increase in both LDL and VLDL.
TREATMENT Lipoprotein Disorders
CLINICAL EVIDENCE THAT TREATMENT OF DYSLIPIDEMIA REDUCES RISK OF CHD
Observational Data Multiple epidemiologic studies have demonstrated a strong relationship between plasma levels of LDL-C and CHD. A direct connection between plasma cholesterol levels and the atherosclerotic process was made in humans when aortic fatty streaks in young persons were shown to be strongly correlated with serum cholesterol levels. The elucidation of homozygous familial hypercholesterolemia was proof that high plasma levels of LDL-C alone are sufficient to cause CAD. Moreover, PCSK9 deficiency proves that having a lifelong reduction in plasma level of LDL-C is associated with a marked reduction in cardiovascular risk.
Clinical Trials: LDL-C Reduction Early clinical trials of cholesterol (mostly LDL-C) reduction utilized niacin, bile acid sequestrants, and even the surgical approach of partial ileal bypass to reduce serum cholesterol levels. Although most of these early studies found a small but significant reduction in cardiac events, no decrease in total mortality was seen. The discovery of more potent and well-tolerated cholesterol-lowering agents, namely HMG-CoA reductase inhibitors (statins), ushered in a series of large cholesterol reduction trials that unequivocally established the benefit of cholesterol reduction. The first of these studies was the Scandinavian Simvastatin Survival Study (4S) in which hypercholesterolemic men with CHD who were treated with simvastatin had a reduction in major coronary events of 44% and a reduction in total mortality of 30%. These impressive results were followed by additional studies using statins. The consistency of results of these studies is remarkable. They demonstrated statins to be effective in primary as well as secondary prevention, in women as well as men, in elderly as well as middle-aged individuals, and in patients with only modestly elevated LDL-C levels as well as those with severe hypercholesterolemia. In general, these studies demonstrated that a 1% reduction in LDL-C level is associated with a reduction in coronary events of a similar magnitude, and an ~40 mg/dL reduction in LDL-C is associated with an ~22% reduction in coronary events.
More recent studies have enrolled subjects with average or subaverage plasma LDL-C levels and have involved targeting the on-treatment LDL-C to even lower levels. For example, the Heart Protection Study (HPS) included 20,536 men and women, ages 40–80 years, who had either established ASCVD or were at high risk for the development of CHD (primarily diabetes); the only lipid entry criterion was a total plasma cholesterol level of >135 mg/dL. Treatment with simvastatin for an average of 5 years resulted in a 24% reduction in major coronary events and a highly significant 13% reduction in all-cause mortality. Importantly, the relative benefit of statin therapy was similar across tertiles of baseline LDL-C, and even the large subgroup of individuals with an LDL-C <100 mg/dL at baseline experienced significant benefit from therapy. This study demonstrated that statin therapy is beneficial in high-risk subjects, even if the baseline LDL-C level is below the currently recommended targeted goal; it also helped to shift the emphasis from simply treating elevated cholesterol to treating patients at high risk of CHD. Additional large-scale clinical trials have expanded on these findings and confirmed that individuals with other cardiovascular risk factors (hypertension, diabetes) benefit from LDL-lowering therapy even when the initial LDL-C level is only modestly elevated. The JUPITER trial was a primary prevention trial in subjects without CHD and with LDL-C <130 mg/dL but with an elevated plasma level of C-reactive protein (CRP). Treatment with rosuvastatin reduced LDL-C by an average of 50% and significantly reduced cardiovascular events, further extending the indication for statin therapy in primary prevention.
Further studies have compared different statin regimens to show that greater reductions in LDL-C levels with treatment are associated with a greater reduction in major cardiovascular events. Based on several of these studies, a white paper was issued by the NCEP in 2004 establishing an “optional” LDL-C goal of <70 mg/dL in high-risk patients with CHD and of <100 mg/dL in very-high-risk patients without known CHD. These optional targets have been widely embraced, and clinical practice is clearly evolving to treating CHD and high-risk patients more aggressively for LDL reduction.
Clinical Trials: The Triglyceride-HDL Axis Abnormalities of the triglyceride high-density lipoprotein (TG-HDL) axis are common in patients with CHD, although data supporting pharmacologic intervention in the TG-HDL axis is less compelling than data supporting LDL-C reduction. Fibric acid derivatives (fibrates), nicotinic acid (niacin), and omega 3 fatty acids (fish oils) are the primary agents currently available to lower plasma triglyceride levels and increase plasma levels of HDL-C. Fibrates have been used as lipid-lowering drugs for several decades and are more effective in reducing plasma triglyceride levels and relatively less effective in increasing plasma HDL-C levels. The results of clinical trials using fibrates have been mixed. Some studies such as the Helsinki Heart Study (HHS) and the Veteran Affairs High-Density Lipoprotein Cholesterol Intervention Trial (VA-HIT) demonstrated a significant reduction in non-fatal myocardial infarction and coronary death with gemfibrozil therapy. However, the Bezafibrate Infarction Prevention (BIP) trial of bezafibrate vs. placebo in CHD patients with low HDL-C failed to demonstrate a statistically significant reduction in coronary events, the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) trial of fenofibrate in patients with type 2 diabetes failed to show a significant reduction in its primary endpoint of nonfatal myocardial infarction and coronary death, and the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study of fenofibrate vs. placebo added to simvastatin in patients with type 2 diabetes failed to show a significant reduction in its primary endpoint of major acute cardiovascular events. In each of these studies, the subgroup with elevated baseline triglycerides suggested benefit.
While niacin is the most effective HDL-raising drug currently available, it has not been tested for its ability to reduce cardiovascular risk in subjects with low plasma levels of HDL-C. The AIM-HIGH and HPS2-THRIVE trials are ongoing studies of the effect of niacin added to baseline statin therapy in patients with CHD and low HDL-C. Finally, while low-dose fish oils have been shown to reduce cardiovascular events, higher doses that reduce triglyceride levels have not been tested for their ability to reduce cardiovascular events. Definitive proof that treating the TG-HDL axis reduces cardiovascular events is likely to come from new therapies that are more effective at specifically targeting VLDL and/or HDL particles.
CLINICAL APPROACH TO LIPID-MODIFYING THERAPY The major goal of lipid-modifying therapy in most patients with disorders of lipid metabolism is to prevent ASCVD and its complications. Management of lipid disorders should be based on clinical trial data demonstrating that treatment reduces cardiovascular morbidity and mortality, although reasonable extrapolation of these data to specific subgroups is sometimes required. Clearly, elevated plasma levels of LDL-C are strongly associated with increased risk of ASCVD, and treatment to lower the levels of plasma LDL-C decreases the risk of clinical cardiovascular events in both secondary and primary prevention. Although the proportional benefit accrued from reducing plasma LDL-C appears to be similar over the entire range of LDL-C values, the absolute risk reduction depends on the baseline level of cardiovascular risk. The treatment guidelines developed by NCEP ATPIII and the 2004 white paper incorporate these principles. As noted above, abnormalities in the TG-HDL axis (elevated triglyceride, low HDL-C, or both) are commonly seen in patients with CHD or who are at high risk for developing it, but clinical trial data supporting the treatment of these abnormalities is much less compelling, and the pharmacologic tools for their management are more limited. Importantly, the NCEP ATPIII guidelines promote the use of the “non-HDL-C” as a secondary target of therapy in patients with triglyceride levels >200 mg/dL. The goals for non-HDL-C are 30 mg/dL higher than the goals for LDL-C. Thus, many patients with abnormalities of the TG-HDL axis require additional therapy for reduction of non-HDL-C to recommended goals.
NONPHARMACOLOGIC TREATMENT Diet Dietary modification is an important component in the management of dyslipidemia. The physician should assess the content of the patient’s diet and provide suggestions for dietary modifications. In the patient with elevated LDL-C, dietary saturated fat and cholesterol should be restricted. For individuals with hypertriglyceridemia, the intake of simple carbohydrates should be curtailed. For severe hypertriglyceridemia (>1000 mg/dL), restriction of total fat intake is critical. The most widely used diet to lower the LDL-C level is the “Step I diet” developed by the American Heart Association. Most patients have a relatively modest (<10%) decrease in plasma levels of LDL-C on a Step I diet in the absence of any associated weight loss. Almost all persons experience a decrease in plasma HDL-C levels with a reduction in the amount of total and saturated fat in their diet.
Foods and Additives Certain foods and dietary additives are associated with modest reductions in plasma cholesterol levels. Plant stanol and sterol esters are available in a variety of foods, such as spreads, salad dressings, and snack bars. Plant sterol and sterol esters interfere with cholesterol absorption and reduce plasma LDL-C levels by ~10% when taken three times per day. The addition to the diet of psyllium, soy protein, or Chinese red yeast rice (which contains lovastatin) can have modest cholesterol-lowering effects. No controlled studies have been performed in which several of these nonpharmacologic options have been combined to address their additive or synergistic effects.
Weight Loss and Exercise The treatment of obesity, if present, can have a favorable impact on plasma lipid levels and should be actively encouraged. Plasma triglyceride and LDL-C levels tend to fall and HDL-C levels tend to increase in obese subjects after weight reduction. Regular aerobic exercise can also have a positive effect on lipids, in large measure due to the associated weight reduction. Aerobic exercise has a very modest elevating effect on plasma levels of HDL-C in most individuals but also has cardiovascular benefits that extend beyond the effects on plasma lipid levels.
PHARMACOLOGIC TREATMENT The decision to use drug therapy depends on the level of cardiovascular risk. Drug therapy for hypercholesterolemia in patients with established CHD is well supported by clinical trial data, as reviewed above. Even patients with CHD or risk factors who have “average” LDL-C levels benefit from treatment. Drug treatment to lower LDL-C levels in patients with CHD is also highly cost-effective. Patients with diabetes mellitus without known CHD have similar cardiovascular risk to those without diabetes but with preexisting CHD. The NCEP ATPIII guidelines recommended estimating absolute risk of a cardiovascular event over 10 years using a scoring system based on the Framingham Heart Study database. Patients with a 10-year absolute CHD risk of >20% are considered “CHD risk equivalents” to be treated as aggressively as patients with existing CHD. Current NCEP ATPIII guidelines call for drug therapy to reduce LDL-C to <100 mg/dL in patients with established CHD, other ASCVD (aortic aneurysm, peripheral vascular disease, or cerebrovascular disease), diabetes mellitus, or CHD risk equivalents; and “optionally” to reduce LDL-C to <70 mg/dL in high-risk CHD patients. Based on these guidelines, virtually all CHD and CHD risk-equivalent patients require cholesterol-lowering drug therapy. Moderate-risk patients with two or more risk factors and a 10-year absolute risk of 10–20% should be treated to a goal LDL-C of <130 mg/dL or “optionally” to LDL-C <100 mg/dL.
Although helpful to consider 10-year absolute risk in making clinical decisions about lipid-altering drug therapy, there are situations where 10-year risk is low but lifetime risk is very high and therefore treatment is indicated. A typical example would be a young adult with heterozygous FH and an LDL-C >220 mg/dL. Despite a very low 10-year absolute risk, every such patient should be treated with drug therapy to reduce lifetime risk. Indeed, all patients with markedly elevated plasma levels of LDL-C levels (>190 mg/dL) should be strongly considered for drug therapy even if their 10-year absolute CHD risk is not elevated. The decision of whether to initiate drug treatment in individuals with plasma LDL-C levels between 130 and 190 mg/dL remains controversial and depends on both 10-year and lifetime risk. Although it is desirable to avoid drug treatment in patients who are unlikely to develop CHD, a high proportion of patients who eventually develop CHD have plasma LDL-C levels within this range. The presence of other risk factors such as a low plasma level of HDL-C (<40 mg/dL) or the diagnosis of the metabolic syndrome would argue in favor of drug therapy (Chap. 32). Other laboratory tests such as an elevated plasma level of apoB, Lp(a), or high-sensitivity C-reactive protein, may assist in the identification of high-risk individuals who should be considered for drug therapy when their LDL-C is in a “gray zone.”
Drug treatment is also indicated in patients with triglycerides >500 mg/dL who have been screened and treated for secondary causes of hypertriglyceridemia. The goal is to reduce fasting plasma triglycerides to below 500 mg/dL to prevent the risk of acute pancreatitis. When triglycerides are 200–500 mg/dL, the decision to use drug therapy depends on the risk of the patient developing chylomicronemia and an assessment of cardiovascular risk. Most major clinical endpoint trials with statins have excluded persons with triglyceride levels >350–450 mg/dL, and there are therefore few data regarding the effectiveness of statins in reducing cardiovascular risk in persons with hypertriglyceridemia. More data are needed regarding the relative effectiveness of statins, fibrates, niacin, and fish oils for reducing cardiovascular risk in this setting. Combination therapy is often required for optimal control of mixed dyslipidemia.
HMG-CoA Reductase Inhibitors (Statins)
HMG-CoA reductase is a key enzyme in cholesterol biosynthesis, and inhibition of this enzyme decreases cholesterol synthesis. By inhibiting cholesterol biosynthesis, statins lead to increased hepatic LDL receptor activity as a counterregulatory mechanism and thus accelerated clearance of circulating LDL, resulting in a dose-dependent reduction in plasma levels of LDL-C. The magnitude of LDL lowering associated with statin treatment varies widely among individuals, but once a patient is on a statin, the doubling of the statin dose produces an ~6% further reduction in the level of plasma LDL-C. The statins currently available differ in their LDL-C reducing potency (Table 31-6). Currently, there is no convincing evidence that any of the different statins confer an advantage that is independent of the effect on LDL-C. Statins also reduce plasma triglycerides in a dose-dependent fashion, which is roughly proportional to their LDL-C–lowering effects (if the triglycerides are <400 mg/dL). Statins have a modest HDL-raising effect (5–10%) that is not generally dose-dependent.
SUMMARY OF THE MAJOR DRUGS USED FOR THE TREATMENT OF HYPERLIPIDEMIA
Statins are well tolerated and can be taken in tablet form once a day. Potential side effects include dyspepsia, headaches, fatigue, and muscle or joint pains. Severe myopathy and even rhabdomyolysis occur rarely with statin treatment. The risk of statin-associated myopathy is increased by the presence of older age, frailty, renal insufficiency, and coadministration of drugs that interfere with the metabolism of statins such as erythromycin and related antibiotics, antifungal agents, immunosuppressive drugs, and fibric acid derivatives (particularly gemfibrozil). Severe myopathy can usually be avoided by careful patient selection, avoidance of interacting drugs, and instructing the patient to contact the physician immediately in the event of unexplained muscle pain. In the event of muscle symptoms, the plasma creatine kinase (CK) level should be obtained to document the myopathy. Serum CK levels need not be monitored on a routine basis in patients taking statins, as an elevated CK in the absence of symptoms does not predict the development of myopathy and does not necessarily suggest the need for discontinuing the drug.
Another consequence of statin therapy can be elevation in liver transaminases (alanine [ALT] and aspartate [AST]). They should be checked before starting therapy, at 2–3 months, and then annually. Substantial (greater than three times the upper limit of normal) elevation in transaminases is relatively rare and mild-to-moderate (one to three times normal) elevation in transaminases in the absence of symptoms need not mandate discontinuing the medication. Severe clinical hepatitis associated with statins is exceedingly rare, and the trend is toward less frequent monitoring of transaminases in patients taking statins. The statin-associated elevation in liver enzymes resolves upon discontinuation of the medication.
Statins appear to be remarkably safe. Meta-analyses of large randomized controlled clinical trials with statins do not suggest an increase in any major noncardiac diseases. Statins are the drug class of choice for LDL-C reduction and are by far the most widely used class of lipid-lowering drugs.
Cholesterol Absorption Inhibitors Cholesterol within the lumen of the small intestine is derived from the diet (about one-third) and the bile (about two-thirds) and is actively absorbed by the enterocyte through a process that involves the protein NPC1L1. Ezetimibe (Table 31-6) is a cholesterol absorption inhibitor that binds directly to and inhibits NPC1L1 and blocks the intestinal absorption of cholesterol. Ezetimibe (10 mg) inhibits cholesterol absorption by almost 60%, resulting in a reduction in delivery of dietary sterols in the liver and an increase in hepatic LDL receptor expression. The mean reduction in plasma LDL-C on ezetimibe (10 mg) is 18%, and the effect is additive when used in combination with a statin. Effects on triglyceride and HDL-C levels are negligible, and no cardiovascular outcome data have been reported. When used in combination with a statin, monitoring of liver transaminases is recommended. The only role for ezetimibe in monotherapy is in patients who do not tolerate statins; the drug is often added to a statin in patients who require further LDL-C reduction.
Bile Acid Sequestrants (Resins) Bile acid sequestrants bind bile acids in the intestine and promote their excretion rather than reabsorption in the ileum. To maintain the bile acid pool size, the liver diverts cholesterol to bile acid synthesis. The decreased hepatic intracellular cholesterol content results in upregulation of the LDL receptor and enhanced LDL clearance from the plasma. Bile acid sequestrants, including cholestyramine, colestipol, and colesevelam (Table 31-6), primarily reduce plasma LDL-C levels but can cause an increase in plasma triglycerides. Therefore, patients with hypertriglyceridemia should not be treated with bile acid–binding resins. Cholestyramine and colestipol are insoluble resins that must be suspended in liquids. Colesevelam is available as tablets but generally requires up to six to seven tablets per day for effective LDL-C lowering. Most side effects of resins are limited to the gastrointestinal tract and include bloating and constipation. Since bile acid sequestrants are not systemically absorbed, they are very safe and the cholesterol-lowering drug of choice in children and in women of childbearing age who are lactating, pregnant, or could become pregnant. They are effective in combination with statins as well as in combination with ezetimibe and are particularly useful with one or both of these drugs for difficult-to-treat patients or those with statin intolerance.
Nicotinic Acid (Niacin) Nicotinic acid, or niacin, is a B-complex vitamin that has been used as a lipid-modifying agent for more than five decades. Niacin reduces the flux of nonesterified fatty acids (NEFAs) to the liver, which is thought to be the mechanism for reduced hepatic triglyceride synthesis and VLDL secretion. Recently, a nicotinic acid receptor (GPR109A) was discovered that suppresses release of NEFA by adipose tissue, thus mediating the effect of niacin on NEFA suppression. Niacin reduces plasma triglyceride and LDL-C levels and raises the plasma concentration of HDL-C (Table 31-6), but it appears that these effects may not be mediated solely by GPR109A. Niacin is also the only currently available lipid-lowering drug that significantly reduces plasma levels of Lp(a) (up to 40%). If properly prescribed and monitored, niacin is a safe and effective lipid-lowering agent.
The most frequent side effect of niacin is cutaneous flushing, which is mediated by activating GPR109A in the skin, leading to local generation of prostaglandin D2 (PGD2) and prostaglandin E2. Flushing can be reduced by formulations that slow the absorption and by taking aspirin prior to dosing. A product is available in Europe that blocks the receptor for PGD2 and attenuates flushing. There is rapid tachyphylaxis to the flushing. Niacin therapy is generally started at lower doses and gradually titrated up to higher doses. Immediaterelease crystalline niacin is generally administered three times per day, over-the-counter sustained-release niacin is taken twice a day, and a prescription form of extended-release niacin is taken once a day. Mild elevations in transaminases occur in up to 15% of patients treated with any form of niacin, and on occasion these elevations may require stopping the medication. Niacin potentiates the effect of warfarin, and these two drugs should be prescribed together with caution. Acanthosis nigricans, a dark-colored coarse skin lesion, and maculopathy are infrequent side effects of niacin. Niacin is contraindicated in patients with peptic ulcer disease and can exacerbate the symptoms of esophageal reflux. It can also raise plasma levels of uric acid and precipitate gouty attacks in susceptible patients.
Niacin can raise fasting plasma glucose levels. A study in type 2 diabetics found only a slight increase in fasting glucose and no significant change in HbA1c level with niacin treatment. Low-dose niacin can be used effectively to reduce plasma triglyceride levels and increase HDL-C without adversely impacting on glycemic control. Thus, niacin can be used in diabetic patients, but every effort should be made to optimize the diabetes management before initiating niacin. Glucose should be carefully monitored in nondiabetic patients with impaired fasting glucose after initiation of niacin therapy.
Successful therapy with niacin requires careful education and motivation on the part of the patient. Its advantages are its low cost and long-term safety. It is the most effective drug currently available for raising HDL-C levels. It is particularly useful in patients with combined hyperlipidemia and low plasma levels of HDL-C and is effective in combination with statins. Outcome data are somewhat limited with niacin, but two clinical trials assessing the benefits of adding niacin to a statin in high-risk patients with low HDL-C are currently ongoing.
Fibric Acid Derivatives (Fibrates) Fibric acid derivatives are agonists of PPARα, a nuclear receptor involved in the regulation of lipid metabolism. Fibrates stimulate LPL activity (enhancing triglyceride hydrolysis), reduce apoC-III synthesis (enhancing lipoprotein remnant clearance), promote beta-oxidation of fatty acids, and may reduce VLDL triglyceride production. Fibrates are the most effective drugs available for reducing triglyceride levels and also raise HDL-C levels modestly (Table 31-6). They have variable effects on LDL-C and in hypertriglyceridemic patients can sometimes be associated with increases in plasma LDL-C levels.
Fibrates are generally very well tolerated. The most common side effect is dyspepsia. Myopathy and hepatitis occur rarely in the absence of other lipid-lowering agents. Fibrates promote cholesterol secretion into bile and are associated with an increased risk of gallstones. Fibrates can raise creatinine and should be used with caution in patients with chronic kidney disease. Importantly, fibrates can potentiate the effect of warfarin and certain oral hypoglycemic agents, so the anticoagulation status and plasma glucose levels should be closely monitored in patients on these agents.
Fibrates are useful and are a reasonable consideration for first-line therapy in patients with severe hypertriglyceridemia (>500 mg/dL) to prevent pancreatitis. Their role in patients with moderate hypertriglyceridemia (200–500 mg/dL) is to promote reduction in non-HDL-C levels, but outcome data regarding their effects on coronary events in this setting remains mixed. In patients with a triglyceride level <500 mg/dL, the role of fibrates is primarily in combination with statins in selected patients with mixed dyslipidemia. In this setting, the risk of myopathy can be minimized with appropriate patient and drug selection and must be carefully weighed against the clinical benefit of the therapy.
Omega 3 Fatty Acids (Fish Oils) N-3 polyunsaturated fatty acids (n-3 PUFAs) are present in high concentration in fish and in flaxseeds. The most widely used n-3 PUFAs for the treatment of hyperlipidemias are the two active molecules in fish oil: eicosapentaenoic acid (EPA) and decohexanoic acid (DHA). N-3 PUFAs have been concentrated into tablets and in doses of 3–4 g/d are effective at lowering fasting triglyceride levels. Fish oils can cause an increase in plasma LDL-C levels in some patients. Fish oil supplements can be used in combination with fibrates, niacin, or statins to treat hypertriglyceridemia. In general, fish oils are well tolerated and appear to be safe, at least at doses up to 3–4 g. Although fish oil administration is associated with a prolongation in the bleeding time, no increase in bleeding has been seen in clinical trials. A lower dose of omega 3 (about 1 g) has been associated with reduction in cardiovascular events in CHD patients and is used by some clinicians for this purpose.
Combination Drug Therapy Combination drug therapy is frequently used for (1) patients unable to reach LDL-C and non-HDL-C goals on statin monotherapy, (2) patients with combined elevated LDL-C and abnormalities of the TG-HDL axis, and (3) patients with severe hypertriglyceridemia who do not achieve non-HDL-C goal on a fibrate or on fish oils alone. When LDL-C and non-HDL-C goals are not achieved on statin mono-therapy, a cholesterol absorption inhibitor or bile acid sequestrant can be added to the drug regimen. Combination of niacin with a statin is an attractive option for high-risk patients who do not attain their target LDL-C level on statin monotherapy and have a low HDL-C level. Conversely, in high-risk patients on statin therapy who have an elevated plasma triglyceride level, addition of a fibrate or fish oils is a reasonable consideration.
Severely hypertriglyceridemic patients treated first with a fibrate often fail to reach LDL-C and non-HDL-C goals and are therefore candidates for addition of a statin. Coadministration of statins and fibrates has obvious appeal in patients with combined hyperlipidemia, but no clinical trial has assessed the effectiveness of a statin-fibrate combination compared with either a statin or a fibrate alone in reducing cardiovascular events. The long-term safety of the statin-fibrate combination is not known. Since coadministration of statins and fibrates is associated with an increased incidence of severe myopathy and rhabdomyolysis, patients treated with this combination must be carefully counseled and monitored. This combination of drugs should be used cautiously in patients with underlying renal or hepatic insufficiency; in the elderly, frail, and chronically ill; and in those on multiple medications.
OTHER APPROACHES Occasionally, patients cannot tolerate any of the existing lipid-lowering drugs at doses required for adequate control of their lipid levels. A larger group of patients, most of whom have genetic lipid disorders, remain significantly hypercholesterolemic despite combination drug therapy. These patients are at high risk for the development or progression of CHD and clinical CHD events. The preferred option for management of patients with severe refractory hypercholesterolemia is LDL apheresis. In this process, the patient’s plasma is passed over a column that selectively removes the LDL, and the LDL-depleted plasma is returned to the patient. Patients on maximally tolerated combination drug therapy who have CHD and a plasma LDL-C level >200 mg/dL or no CHD and a plasma LDL-C level >300 mg/dL are candidates for every-other-week LDL apheresis and should be referred to a specialized lipid center.
MANAGEMENT OF LOW HDL-C Severely reduced plasma levels of HDL-C (<20 mg/dL) accompanied by triglycerides <400 mg/dL usually indicate the presence of a genetic disorder such as a mutation in apoA-I, LCAT deficiency, or Tangier disease. HDL-C levels <20 mg/dL are common in the setting of severe hypertriglyceridemia, in which case the primary focus should be on the management of the triglycerides. HDL-C levels <20 mg/dL also occur in individuals using anabolic steroids. Secondary causes of more moderate reductions in plasma HDL (20–40 mg/dL) should be considered (Table 31-5). Smoking should be discontinued, obese persons should be encouraged to lose weight, sedentary persons should be encouraged to exercise, and diabetes should be optimally controlled. When possible, medications associated with reduced plasma levels of HDL-C should be discontinued. The presence of an isolated low plasma level of HDL-C in a patient with a borderline plasma level of LDL-C should prompt consideration of LDL-lowering drug therapy in high-risk individuals. Statins increase plasma levels of HDL-C only modestly (~5–10%). Fibrates also have only a modest effect on plasma HDL-C levels (increasing levels ~5–15%), except in patients with coexisting hypertriglyceridemia, where the effect on HDL levels can be greater. Niacin is the most effective HDL-C–raising therapeutic agent available and can increase plasma HDL-C by up to ~30%, although some patients fail to achieve clinically important increases in HDL-C levels from niacin therapy.
The issue of whether pharmacologic intervention should be used to specifically raise HDL-C levels has not been adequately addressed in clinical trials. In persons with established CHD and low HDL-C levels whose plasma LDL-C levels are at or below the goal, it may be reasonable to initiate therapy (with a fibrate or niacin) directed specifically at reducing plasma triglyceride levels and raising the level of plasma HDL-C. More data are required before broad recommendations are made to use drug therapy to specifically raise HDL-C levels to prevent cardiovascular events. New HDL-raising approaches are under development that may help to address this important issue.
Management of Elevated Levels of Lp(a)
High levels of Lp(a) are associated with increased risk of ASCVD. Genetic studies suggest that this association is causal, but there is no evidence that reducing plasma Lp(a) levels reduces cardiovascular risk. Until such studies are performed, the major therapeutic approach to patients with high plasma levels of Lp(a) and established CAD is to aggressively lower plasma levels of LDL-C. Niacin is the only drug currently available that lowers Lp(a), and might be considered as an addition to a statin in a very-high-risk patient with elevated Lp(a).