A small subset of patients with hypercholesterolemia have an inadequate lipid-lowering response to maximal diet and drug treatment and should be considered for additional therapy to come as close as possible to the therapeutic targets of the National Cholesterol Education Program (NCEP). The development of new classes of lipid-modifying drugs should increase the number of patients who achieve lipid goals. Most candidates for treatment beyond diet and drugs are patients with familial hypercholesterolemia (FH), including all individuals with homozygous FH and patients with heterozygous FH who do not attain NCEP goals.
Traditional treatment options alone or in combination include plasmapheresis, low-density lipoprotein apheresis (LDLA), portacaval shunt, liver transplantation, and partial ileal bypass (PIB) surgery. Plasmapheresis is a nonspecific extracorporeal procedure that removes all plasma proteins, including high-density lipoprotein (HDL), from the blood. A superior method that specifically lowers LDL cholesterol (LDL-c) in these patients is LDLA. Methods for performing LDLA include dextran sulfate cellulose adsorption, immunoadsorption, heparin-induced extracorporeal precipitation (HELP), and perfusion through whole blood–compatible columns. Time-averaged LDL lowering of 40% to 50% is achieved with weekly or biweekly LDLA therapy. The U.S. Food and Drug Administration (FDA) has approved LDLA for patients who, despite maximal tolerated diet and drug therapy, have an LDL-c concentration greater than 300 mg/dL without coronary artery disease (CAD) or an LDL-c concentration greater than 200 mg/dL with CAD. Systems that use dextran sulfate cellulose adsorption, such as the Liposorber System (Kaneka, Osaka, Japan) and the HELP Plasmat Futura System (B. Braun Medical Inc., Bethlehem, PA), are available in the United States. The gap between FDA-approved LDL-c levels for initiating LDLA and those levels established as clinical LDL-c targets has grown, and newer guidelines target LDL-c levels of 70 mg/dL or less in many individuals with CAD.
Surgical procedures for lowering LDL-c are more invasive than LDLA and should only be considered if LDLA therapy does not provide adequate LDL-c lowering or if it is technically difficult to perform. Portacaval shunts have been used to achieve LDL-c lowering of approximately 40% in patients with homozygous FH, but concerns about long-term hepatic and endocrine complications have limited its usefulness. Liver transplantation can achieve near-normal LDL-c levels but is associated with the need for long-term immunosuppressive therapy. Improvements in surgical techniques and immunosuppressive therapies have made liver transplantation an increasingly attractive option. PIB surgery has slowed the progression of CAD, but it has appreciable morbidity and is not appropriate for patients with homozygous FH. Gene transfer has been unsuccessfully tested in a small number of patients, and at present it is not a therapeutic option. LDLA is beneficial in patients with refractory hypercholesterolemia and CAD and is the therapy of choice for patients who require treatment beyond diet and drugs. The rapidity of clinical improvement in patients who receive LDLA suggests mechanisms in addition to the regression of plaque. Whole blood–compatible systems for LDLA are simpler than standard LDLA to perform. These systems have been used extensively in Europe but have not been introduced in the United States.
Definition of the Target Population
Because nondietary, nonpharmacologic therapy for hypercholesterolemia entails a major commitment from the patient and medical community, and in some instances substantial risk, clear guidelines for considering such therapy are necessary. All patients should have therapeutic lifestyle changes (TLCs) instituted and should be treated with maximum combination lipid-lowering drug therapy that includes, as tolerated, a 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitor, a bile acid binder, and nicotinic acid. The usefulness of ezetimibe in this population is less clear, but it has continued to be used for lack of better alternatives. Some patients will also be treated with promising experimental pharmacologic therapies such as mipomersen, an apolipoprotein B (ApoB) antisense drug. Criteria for additional therapy should include the degree of LDL-c elevation and whether the patient has CAD or multiple risk factors for CAD. The FDA has approved LDLA for the management of patients with CAD and LDL-c concentrations of more than 200 mg/dL or patients without CAD but an LDL-c concentration of more than 300 mg/dL. Individuals with extremely high triglyceride concentrations or secondary causes of hypercholesterolemia are generally not candidates for the therapies described in this chapter.
Description of the Patient Population
Homozygous Familial Hypercholesterolemia
The clearest indication for nondietary, nondrug therapy is in patients with homozygous FH. The classic form of this disorder is caused by the inheritance of two mutant genes at the LDL receptor locus. The clinical expression for classical homozygous FH occurs in approximately 1 person in 1 million. The defect in LDL receptor function causes a marked elevation in the plasma concentration of LDL-c, which typically exceeds 500 mg/dL but can reach as high as 1000 mg/dL. HDL cholesterol (HDL-c) concentrations tend to be substantially below normal. Clinical features include the presence of xanthomas, severe aortic root disease that includes aortic stenosis, and the premature onset of CAD. Angina pectoris, myocardial infarction (MI), or sudden death frequently occurs between the ages of 5 and 20 years. Mabuchi and colleagues followed 10 patients with homozygous FH for a period of approximately 14 years. During that time, 6 of the patients died from sudden death or heart failure at an average age of 26 years. Similar observations were reported in patients from South Africa.
The severity of the clinical expression depends to a great extent on the percentage of functioning LDL receptors. In the study by Goldstein and Brown of 57 homozygotes, more than one fourth of receptor-absent patients died before the age of 25 years compared with 1 of 26 individuals with residual LDL receptor activity. Because of the very high risk of premature CAD and the poor response to diet and drug therapy, all patients with homozygous FH require alternative therapy. It is possible that some individuals who phenotypically appear to have homozygous FH have one gene expressing a defective ApoB that will result in decreased clearance of LDL-c.
Patients with Low-Density Lipoprotein Cholesterol Concentrations of More than 200 mg/dL and Coronary Artery Disease
The majority of patients in this group will have the heterozygous form of FH. This disorder has a prevalence of approximately 1 in 500 persons and is typically manifested by the occurrence of premature CAD by the fifth decade for men and the sixth decade for women. Clinically similar syndromes are produced by defects in the LDL receptor and defective ApoB structure with impaired receptor binding. The presence of both CAD and elevated LDL-c concentrations greatly increases the risk of subsequent coronary events; therefore patients with CAD require intensive LDL-c concentration control. The reinfarction rate found in seven secondary prevention trials reviewed by Pekkanen and colleagues was approximately 6% annually compared with a 1% to 2% rate of first infarction in four primary prevention trials. Trials in patients with marked hypercholesterolemia have reported regression or a lower rate of progression of coronary lesions when the elevated LDL-c concentration is lowered with diet and drugs. Coupled with the impressive results from the trials that used the statins, a very persuasive case can be made for the aggressive treatment of subjects with FH to achieve LDL-c targets set forth by the NCEP.
Patients with Low-Density Lipoprotein Cholesterol Concentrations of More than 300 mg/dL Without Coronary Artery Disease
The decision whether to use nondietary, nondrug therapy for primary prevention in asymptomatic adults is never as easy as it is in homozygotes. The risk of premature CAD is most apparent in patients with FH because of the presence of lifelong elevated LDL-c concentrations. In addition, the presence of risk factors other than LDL-c helps determine which patients are most likely to develop CAD. For example, a lipoprotein A [Lp(a)] concentration of more than 20 mg/dL has been recognized as an independent risk factor in patients with FH. The use of noninvasive screening procedures for CAD, such as quantitation of coronary artery calcium, provides additional guidance for determining whether patients with elevated LDL-c will develop clinically significant disease.
Based on current FDA guidelines, it is reasonable to consider nondietary, nondrug therapy for primary prevention in patients with an LDL-c concentration of more than 300 mg/dL despite diet and maximal tolerated lipid-lowering drug therapy.
Extracorporeal Therapies for the Treatment of Severe Hypercholesterolemia
The first report of plasma exchange for FH was published by de Gennes and colleagues in 1967. In 1975 Thompson and colleagues described the use of plasmapheresis with an automated cell separator to treat patients with homozygous FH. Plasmapheresis therapy has been shown to improve the survival of treated homozygotes compared with their untreated control siblings.
The nonselectivity of plasmapheresis led to the development of LDLA for the specific removal of ApoB-containing lipoproteins from the blood. The use and clinical acceptance of the procedure have increased worldwide, including approval in 1996 by the FDA, and it is most useful in patients who respond incompletely to diet and lipid-lowering drug therapy because of drug intolerance or extremely high baseline LDL-c concentrations. The specific removal of ApoB-containing lipoproteins using heparin agarose beads was first described in 1976 by Lupien and colleagues. Methods for performing LDLA include HELP, columns that contain immobilized antibodies to LDL, and columns that contain dextran sulfate cellulose. Whole blood–compatible systems that use columns containing either a modified polyacrylate gel (DALI System; Fresenius AG, St. Wendel, Germany) or dextran sulfate (Liposorber-D, Kaneka) single- or double-column technologies have been developed and are available outside the United States.
LDLA will lower both LDL-c and Lp(a). An immunoadsorption system has been developed that specifically targets Lp(a) for patients in whom elevated Lp(a) is the primary problem, but it is not widely available.
Fortunately, venous access from an antecubital fossa vein is most often sufficient for LDLA because of the lower blood flow rates required (50 to 100 mL/min) compared with hemodialysis (400 mL/min). If access from antecubital fossa veins provides insufficient blood flow, the placement of a catheter, fistula, or graft may be necessary.
Some form of anticoagulation is necessary for all extracorporeal procedures. Heparin and acid citrate dextrose (ACD) are the anticoagulants most commonly used in extracorporeal therapies. Heparin is typically used for extracorporeal procedures that involve a membrane to separate whole blood into plasma and cells or in the whole blood–compatible system. Typically, a heparin bolus of approximately 30 to 60 U/kg is administered, followed by a continuous infusion of approximately 1000 U/h. ACD has the advantage of rapid metabolism and produces little residual effect after the procedure. Side effects as a result of ACD administration include symptoms of hypocalcemia, which may include perioral tingling, hypotension, or, very rarely, tetany.
Most forms of LDLA require the separation of blood into cells and plasma before the plasma is processed to specifically remove ApoB-containing lipoproteins. Whole blood–compatible systems use direct adsorption of lipoproteins on columns and thereby eliminate the requirement for cell separation. For systems that require cell separation, either a membrane or centrifuge is used to separate blood into plasma and cellular elements. Membrane separation of blood is simpler and requires less extracorporeal volume, but it is less efficient than centrifugal techniques.
Most patients are treated with LDLA at a treatment frequency of approximately every 2 weeks. This is based on the rebound of the LDL-c and on the observation that most patients prefer to be treated no more often than every other week. At a treatment frequency of every 2 weeks, reduction in pretreatment LDL-c level is usually stepwise until a new plateau is reached.
Cholesterol reduction can be quantified by measuring the achievement using either acute or time-averaged lowering ( Figure 27-1 ). The acute lowering is the difference between preprocedure and postprocedure lipid values as a percent of the initial value and is a function of the amount of plasma or blood processed during a single treatment. LDLA procedures process 1.5 to 2.0 plasma or blood volumes and reduce LDL-c concentrations short term by 70% to 80%. For a blood flow of 50 to 80 mL/min, it takes about 3 hours to process 1.5 to 2.0 plasma or blood volumes.
Although acute lowering is helpful in determining treatment efficiency, the time-averaged lipid value is a better indicator of the lipid concentration to which the patient’s arteries are exposed over an extended period. The time-averaged lowering is related to the treatment frequency and rate of rebound. The time-averaged lipid value is estimated by the formula 0.73 (pre–LDL-c − post–LDL-c) plus post LDL-c, or it can be directly determined by measuring LDL-c daily after a treatment. The time-averaged LDL-c lowering is usually between 40% and 50%.
LDLA has been reported to maintain or increase HDL-c concentrations over time. The mechanism for this effect is unknown, but the increase in HDL-c is seen most frequently in subjects with homozygous FH. In contrast, plasmapheresis lowers HDL-c through nonspecific depletion of all plasma constituents including HDL-c.
Low-Density Lipoprotein Apheresis Using Dextran Sulfate Cellulose Columns
Dextran sulfate cellulose selectively binds ApoB-containing lipoproteins on the basis of a charge attraction. Dextran sulfate cellulose columns were initially provided by the manufacturer as single, large-volume, nonregenerable columns that could be attached to any cell separator. Limited LDL-binding capacity resulted in the development of dual regenerable columns in a system that included a hollow-fiber primary cell separator ( Figure 27-2 ). Plasma is alternately perfused through each 150-mL column, permitting the regeneration of the off-line column with hypertonic sodium chloride solution; a computerized unit controls the process. After passing through the adsorbent column, the plasma is recombined with the cells and returned to the patient. The advantage of this system is the almost unlimited LDL-binding capacity as a result of the on-line regeneration of the columns. The columns are discarded after each treatment.
Low-Density Lipoprotein Apheresis Using Immunoadsorption Columns
Immunoadsorption for the performance of LDLA has been used to treat patients for approximately 25 years. Polyclonal monospecific or monoclonal antibodies to ApoB are immobilized on a support, typically sepharose beads, and are packed into glass columns. Each patient has two columns, which are reused during each procedure and reused from one procedure to the next because of the expense. Online regeneration of the columns is controlled by a column regeneration unit (see Figure 27-2 ). Typically, the columns are eluted with an acid solution, neutralized with a buffer solution, and then rinsed with saline. The procedure requires storage of the columns between treatments. All ApoB-containing lipoproteins are removed: LDL, very-low-density lipoprotein (VLDL), and Lp(a). Because the columns are reused multiple times, they must be monitored for loss of activity that may occur after several months. In addition, sensitization of the patient to small quantities of shed antibody has been demonstrated. Because of the cumbersome requirement of storing columns between uses, immunoadsorption LDLA is used less commonly than the other methods described in this chapter. A variant of this technique primarily removes circulating Lp(a) for patients in whom elevated Lp(a) levels are thought to be the primary problem.
Low-Density Lipoprotein Apheresis Using Heparin-Induced Extracorporeal Low-Density Lipoprotein Precipitation
HELP LDLA specifically removes LDL, VLDL, and Lp(a) while minimally affecting HDL. HELP differs from other procedures in that it removes a substantial quantity of fibrinogen. The technique is based on the precipitation of positively charged LDL and other β-lipoproteins when heparin is added at a pH just above 5.0 ( Figure 27-3 ). A few other plasma proteins precipitate to some extent with heparin, most notably fibrinogen. An anion exchange column removes excess heparin, and the plasma is treated with bicarbonate dialysis and ultrafiltration to return the pH to normal and remove excess fluid. The entire process is controlled by a microprocessor.
Low-Density Lipoprotein Apheresis Using Whole Blood–compatible Systems
Potential cost savings and simplicity of use make a whole blood–compatible system very attractive for the performance of LDLA. Two systems are most commonly used: the DALI System and Liposorber-D ( Figure 27-4 ). A comparison between the whole blood–compatible systems and classic techniques for performing LDLA is shown in Table 27-1 . The DALI system uses a modified polyacrylate gel immobilized on a solid support. The mechanism of ApoB binding is an electrostatic interaction between the positive charges of the ApoB and the negatively charged carbohydrate moieties on the polyanionic gel. The Liposorber-D system is based on binding of ApoB-containing lipoproteins to dextran sulfate. The matrix has been modified to permit the safe interaction between whole blood and the column.
|Setup time (approximate)||30 min||30 min||60 min||60 min||60 min|
|Plasma/blood||Whole blood||Whole blood||Plasma||Plasma||Plasma|
|Principle||Polyacrylate||Dextran sulfate||Dextran sulfate||Antibodies||Heparin precipitation|
The initial two-center study of the DALI procedure reported on 12 hypercholesterolemic patients treated once each with columns containing 480 mL polyacrylate gel. No side effects were observed. A follow-up study reported the use of this technology in the therapy of three hypercholesterolemic subjects with atherosclerosis. The acute reduction in lipids was 66% for LDL-c, 63% for Lp(a), and 29% for triglycerides. HDL and fibrinogen were lowered minimally. No clinically significant adverse events occurred, and the changes in biochemical and hematologic parameters were minor. The authors noted that the setup time for the procedure was only 30 minutes as opposed to 1 hour or more for other forms of LDLA. The clinical experience in 10 subjects treated with the Liposorber D system demonstrated an LDL cholesterol reduction of 62%. No significant side effects were observed. Recently a whole blood, dual-column system has been tested and was found better able to lower LDL-c acutely compared with the single-column technique (79.7% vs. 68.2%).
Risks of Low-Density Lipoprotein Apheresis
Adverse reactions as a result of LDLA have been few. The extracorporeal volume is well tolerated even in patients with severe CAD and in children as young as 3 years. Hypotension requiring infusion of saline occurs in less than 5% of treatments, and unusual side effects include angina, hemolysis, and allergic or anaphylactic reactions. In immunoadsorption treatments, possible causes of adverse reactions include complement activation and sensitization of the patient to column constituents. An anaphylactoid reaction during dextran sulfate LDLA has been described in patients taking angiotensin-converting enzyme (ACE) inhibitors. The mechanism is related to release of bradykinin during LDLA with concomitant decreased degradation of kinins by ACE inhibitors. The DALI system also activates the kallikrein-kinin system. The best approach is to switch patients from ACE inhibitors to angiotensin II receptor–blocking drugs before the initiation of LDLA. Treatments with immunoadsorption LDLA or the HELP system are not associated with this clinical problem. For systems utilizing whole blood–compatible columns, the long-term effects of whole blood interactions with the column, such as cell activation, are unknown.
Benefits of Low-Density Lipoprotein Apheresis
Regression of tendon xanthoma and improvement in CAD have been demonstrated in patients with severe hypercholesterolemia when the LDL-c concentration is lowered through LDLA protocols. In the LDL Apheresis Regression Study (LARS), angiographic evidence for regression of CAD was observed in 10 of 30 patients treated with LDLA and lipid-lowering drugs despite a baseline LDL concentration of 311 mg/dL. The HELP-LDL Apheresis Multicenter Study was a 2-year investigation in 51 patients treated with weekly LDLA and lipid-lowering drugs. Computer-assisted analysis of paired angiograms from 33 evaluable patients revealed that 23 patients had regression, 1 had little change, and 9 patients had progression. The German Multicenter LDL Apheresis Trial was a four-center, 3-year prospective trial of 32 patients with FH. All patients who had symptomatic CAD demonstrated improvements in their symptoms by the end of the study. Improvement in electrocardiographic (ECG) stress testing was demonstrated in 17 patients. Analysis of the paired angiograms did not reveal regression of disease, although definite progression of disease was observed in only five patients over 3 years.
An important study was the LDL-Apheresis Atherosclerosis Regression Study (LAARS), a prospective 2-year, randomized, single-center study in men with hypercholesterolemia and CAD. This trial demonstrated that the addition of biweekly LDLA to treatment with simvastatin improved regional myocardial perfusion and decreased myocardial ischemia; it is interesting to note that little change was seen in the coronary angiograms. Peripheral vascular disease was also found to improve in LAARS, as shown by standardized techniques including B-mode ultrasound.
Studies using intravascular ultrasound demonstrate that the angiogram may miss changes that occur in the blood vessel wall rather than in the blood vessel lumen. The Low-Density Lipoprotein-Apheresis Coronary Morphology and Reserve Trial (LACMART) used intravascular ultrasound to compare change in coronary artery disease in 18 patients with FH randomized to 1 year of either drug therapy or LDLA combined with drug therapy. Favorable changes in coronary artery minimal luminal diameter and plaque burden were observed in subjects who received LDLA plus drug compared with subjects receiving drug alone. LDLA with the Liposorber system decreased the rate of restenosis in patients with elevated Lp(a) concentrations who had undergone coronary angioplasty.
A multicenter study in the United States demonstrated the ability of LDLA to decrease the number of clinical events in patients with CAD. The rate of clinical events per 1000 patient-months decreased from 9.14 for the 5-year period before the initiation of LDLA to 4.72 during the LDLA period ( P = .037). An analysis of data from the U.S. LDLA Registry, established at the time of FDA approval of LDLA, confirmed the decrease in clinical events in a larger number of patients. In 1998, Mabuchi and colleagues reported long-term outcome data in a study of 130 heterozygous FH patients with CAD, in which LDLA combined with drug therapy produced a reduction in coronary events during a 6-year period. The rate of coronary events in the LDLA-plus-drug group (n = 87) was 10% compared with 36% in the drug-alone group (n = 43).
Benefits of Low-Density Lipoprotein Apheresis Beyond Reducing Low-Density Lipoprotein Concentrations
The rapidity of clinical responses in some patients receiving LDLA has suggested effects in addition to LDL lowering and regression of plaque. Several mechanisms may contribute to this observation, including decreased blood viscosity with improved blood rheology and downregulation of leukocyte and endothelial adhesion molecules. Tamai and colleagues measured forearm blood flow while infusing acetylcholine before and after a single LDLA treatment and found that vasodilation responses were significantly augmented. The reductions in LDL and oxidized LDL correlated with the increase in acetylcholine-induced vasodilation.