Given the high prevalence of undesirable cholesterol levels in the U.S. population, it is recommended that all Americans adopt the phase I diet (Table 27.7):

The phase I diet usually does not require a dramatic alteration in eating habits and can be readily adopted by most persons. It is important to note that much of the polyunsaturated fat found in “low-cholesterol, low-fat” processed food products is usually partially hydrogenated, converting otherwise beneficial vegetable oils into the undesirable trans unsaturated configuration. Such processed food products should not be considered a substitute for saturated fat, but neither should saturated fat be considered a healthier alternative to such products and trigger reverting back to lard-based or tropical oil–based processed foods.

The phase II diet (Table 27.7) entails more effort because it goes beyond eliminating the obvious sources of fat and cholesterol. It is indicated for patients who do not achieve adequate results with a phase I diet and for patients at highest CHD risk (e.g., established CHD).

If just the phase I dietary interventions were widely implemented, the overall incidence of CHD in the population would probably drop significantly. The recommended percentages of fat intake in the diet must be translated into real menus and food recommendations if good compliance with these recommendations is to be realized. The counsel of a dietitian is often beneficial, particularly if a phase II diet is indicated. A good working knowledge of the fat content of common foods is essential for patient, family, and health care team (see Appendix III). A number of “heart-healthy” cookbooks are on the market to help patients in their food choices and preparation. The use of faddish cholesterol cookbooks should be discouraged because these often do not promote sustainable healthy eating habits. Increasingly, restaurants are offering low-fat menu choices, and patients should be encouraged to select them.

Preliminary data from prospective studies of omega-3 fish oil supplements are encouraging, but they are too limited to serve as the basis for dietary recommendations. Impairment of clotting has been noted with use of high doses.

Vitamins C, E, and beta-carotene (the so-called “antioxidant vitamins”) do not lower cholesterol levels, but they are capable of increasing LDL resistance to oxidative change and theoretically might reduce the risk of arterial wall injury. Data from small-scale human studies of vitamin E suggested a possible reduction in CHD risk, but early prospective, large-scale, randomized trials have so far failed to confirm a significant benefit; the same pertains to vitamin C. The typical dose of vitamin E used in these studies is 400 to 800 IU/d. There is no evidence that beta-carotene reduces CHD events, and some data suggest an association with lung neoplasms; its use for CHD prevention is not recommended. More data on vitamin E and C will be forthcoming and should further clarify their value, if any, in preventing coronary events.

One-half to one clove of garlic a day may produce a modest (5%) reduction in serum cholesterol, but powder and oil preparations have not demonstrated consistent significant benefit. Use of fiber preparations such as psyllium (10 g/d) also provides modest benefit, as does the soluble fiber in oat-containing cereals. Red yeast extracts contain a cholesterol synthesis inhibitor that is a member of the statin family and can lower LDL 10% to 15%. Alcohols contained in sugar cane (policanosols) can reduce LDL cholesterol by 20% to 30%, but data are needed on their safety and long-term efficacy.

Cost and cost-effectiveness are important considerations in selecting statin therapy because all these agents are expensive and use is likely to be measured in years. Although fluvastatin has usually been the least expensive and pravastatin and simvastatin the most expensive of the statins, cost often depends on specific pricing contracts with large payers, making generalizations difficult. Lovastatin is now available as a generic agent, and simvastatin will transition to that status in 2006, making it likely that some downward cost pressure will be imposed on the whole class. The statins differ principally in cost and potency, which often parallel one another. Cerivastatin was removed from the market in the summer of 2001 because of high adverse event report rates of myositis and rhabdomyolysis, particularly when used in combination with gemfibrozil. Simvastatin, atorvastatin, and rosuvastatin are the statins that give the greatest LDL-cholesterol reduction at U.S. Food and Drug Administration (FDA)–approved dosing levels. Within this group of three, the rank order of LDL-cholesterol–reducing activity is simva < atorva < rosuva. LDL reductions of 50% to 65% can be achieved with these drugs.

Although the major drawback to statin use is cost, cost savings from reduced rates of cardiac events, time lost from productive work, and premature death at least partly offset the high cost of treatment. Formal cost-effectiveness analyses are appearing in the literature. Data from the Scandinavian Simvastatin Survival Study examined cost per year of life gained and found statin therapy cost-effective in men and women over a broad spectrum of ages and cholesterol ranges. The cost per year of life gained was less than half of that for other cost-effective preventive measures such as mammography.

Starting dose for these agents is typically 10 to 20 mg/d, with the more potent drugs started at the lower dose. The maximum dose is currently 80 mg/d for lovastatin, simvastatin, fluvastatin, atorvastatin, and pravastatin. The maximum dose of rosuvastatin is 40 mg/d. The shorter-acting agents are best taken at night, the time of peak cholesterol synthesis, but all drugs work satisfactorily even if taken once daily in the morning.

Asymptomatic hepatocellular dysfunction manifested by an increase in serum levels of hepatocellular enzymes (e.g., aspartate and alanine aminotranferases) is among the most common adverse effects. Incidence ranges from about 3% for any elevation to less than 1% for elevations greater than three times the upper limit of normal. All such increases are reversible with cessation of therapy, which should be considered when liver enzyme levels continue to rise or reach more than three times the upper limit of normal. Transaminase monitoring is strongly recommended, with initial measurements made after 1 to 2 months of therapy, and follow-up measurements made at 6 months and 1 year. The lack of new onset of liver toxicity after 1 year of therapy has led many to advocate limited monitoring after 1 year.

Harmless elevations in muscle enzymes (e.g., creatine phosphokinase <10 times the upper limit of normal) occur in about 0.6% of cases and require no action in an asymptomatic patient. Myalgias without creatine phosphokinase elevations also occur with statin use, making routine monitoring of creatine phosphokinase of limited value. Myalgias are probably underreported in patients on statins, particularly in the elderly, because the symptom is frequently attributed to arthritis or muscle stiffness associated with the aging process. Patients should be questioned about increased muscle discomfort when placed on statins, and drug holidays may need to be instituted to clarify complaints. Symptomatic myositis with demonstrable muscle breakdown is much less common, but the risk is increased with high-dose therapy and when other lipid-lowering agents are used in conjunction with statin therapy. Statin monotherapy has been reported to cause rhabdomyolysis and renal failure, but it is more common in the setting of concurrent gemfibrozil use. Dual therapy with niacin can also increase myositis. Joint use of a statin and a fibrate (either gemfibrozil or fenofibrate) should be done only with great caution and probably only by a lipid specialist. Because of an FDA warning against this combination, pharmacists will often not fill a prescription that places a patient on both drugs without first calling the physician. The withdrawal of cerivastatin from the market in 2001 was prompted by many reports of rhabdomyolysis when the drug was used alone or, more commonly, in combination with a fibrate. This event will likely increase the vigilance of pharmacists in responding to combination-therapy prescriptions.

As noted earlier, initial concerns about an increase in non-CHD mortality (e.g., from cancer, suicide, or violence) were raised by early meta-analyses, but large-scale prospective studies of statins show no evidence for such adverse effects.

Although most experts believe that all of the statins will reduce coronary events in proportion to their efficacy in lowering LDL cholesterol, not all statins have been proven to do so. The best studies of clinical outcomes have been performed using simvastatin, pravastatin, and lovastatin, but atorvastatin also has been found to reduce cardiovascular event rates. Prospective studies with statins have not only demonstrated a reduction in coronary events, but they have also been shown to lower all-cause mortality. Patients who require modest reductions in LDL cholesterol (25% or less) can be started on any of the statins with the expectation that the reduction will be achieved. Those with marked LDL elevations and high overall CHD risk require more intensive therapy (>40% reductions) and are better served if started on the more potent statins.

A high-capacity, early route of niacin metabolism involves conjugation with glycine to produce nicotinuric acid, which produces flushing, the immediate and principle side effect that limits clinical use. Tolerance to flushing develops over time and allows slow advancement of dose. The bothersome side effect can also be lessened by starting with a low dose (e.g., 100 mg thrice daily), using an extended-release preparation (see later discussion), taking once-daily pretreatment with a nonsteroidal antiinflammatory drug (e.g., aspirin 325 mg, or ibuprofen 200 mg, 30 to 30 minutes before first dose), and taking niacin with meals.

Hepatotoxicity is the other major concern, related to nonconjugative metabolism of the drug to nicotinamide and pyrimidines. Risk of hepatotoxicity is greatest with use of sustained-release preparations (18 to 24 hours' duration), which are metabolized predominantly by the nonconjugative route. Risk is minimized by taking an intermediate (“extended”)-release formulation (8 to 10 hours' duration), which optimizes metabolism through both pathways. Periodic monitoring of liver function (e.g., aspartate aminotransferase determination) is required with niacin use.

Other adverse effects include exacerbation of gout and slight worsening of glucose intolerance, necessitating an occasional check of uric acid and glucose. Rashes, dry skin, and occasionally acanthosis nigricans may accompany niacin use. Lanolin cream helps the former, and prompt cessation clears the latter. Some patients experience gastrointestinal upset.

As a B-complex vitamin (nicotinic acid), niacin is available without prescription in immediate-release and sustained-release forms. Although least costly, these are associated with the highest frequency of flushing and hepatotoxicity, respectively. A prescription is required for the intermediate-release preparation (also referred to as extended-release niacin [e.g., Niaspan]), which is the best tolerated but most expensive formulation. Its use should be considered in persons who cannot tolerate the immediate-release preparation or in whom there is concern about risk of hepatoxicity. If an inexpensive, well-tolerated, immediate-release brand of niacin is found, the patient should stay with it; changing brands might not be as well tolerated.

The bile sequestrants are nonabsorbable resins whose major side effects are gastrointestinal—constipation, bloating, heartburn, and nausea. A high-fiber diet or psyllium supplement and use of these agents just before a meal will usually ameliorate the gastrointestinal upset. The potential to impede absorption of certain drugs (e.g., digoxin, thyroxin, warfarin, tetracycline, phenobarbital) necessitates that bile sequestrants not be taken until at least 1 hour after or 4 hours before these other drugs. In rare instances, steatorrhea and malabsorption of the fat-soluble vitamins (A, D, E, and K) can occur. The usual starting dose is one scoop of the powdered form of the drug (4 g of cholestyramine, 5 g of colestipol) in a large glass of water twice a day. Dose can be increased to a total of three scoops twice daily. A newer formulation of cholestyramine, which is significantly less gritty (Lo-Cholest), can be tried if the texture of the generic drug formulation should prove unacceptable. Colesevelam is given as six 625-mg tablets a day, given in once-a-day or twice-daily dosing.

The newest revisions recommend an optional LDL-cholesterol treatment threshold of 100 mg/dL, with a goal of less than 70 mg/dL.

The LDL-cholesterol treatment threshold has been lowered to 130 mg/dL, with a treatment goal of less than 100 mg/dL.

The threshold for persons in the lower-risk categories remains unchanged from previous recommendations. For this group the LDL-cholesterol cut-point continues to be 160 mg/dL.

Drug therapy is definitively recommended for LDL-cholesterol levels above 190 mg/dL and considered optional for those with values between 160 and 190 mg/dL.

Although epidemiologic data show a strong inverse relation between HDL level and CHD risk, there are no data yet from large-scale, randomized, prospective clinical trials showing that raising HDL cholesterol alone significantly reduces CHD mortality. However, treatment of patients with low HDL levels with statins does seem to lower CHD morbidity. Generally, healthy middle-aged and elderly persons with low HDL cholesterol and “normal” LDL cholesterol demonstrate a significant reduction in risk of a first acute major coronary event (e.g., myocardial infarction, unstable angina) when treated with a statin drug (e.g., the AFCAPS/TexCAPS trial). Such findings suggest that even persons with modest increases in CHD risk—as manifested by advancing age and an isolated low HDL cholesterol—might benefit from pharmacologic therapy. The mechanism of the benefit may be other than the effect on HDL cholesterol because in the AFCAPS/TexCAPS study, HDL cholesterol rose only 6%.

Lowering the threshold for drug therapy below that recommended by the new NCEP guidelines could be justified on the basis of recent outcome studies, but the cost-effectiveness of such an approach remains to be established. Because results are typically reported in terms of reduction in relative risk, the magnitude of the benefit to lower-risk patients may sometimes appear inflated. Absolute risk certainly becomes an issue in younger patient populations, where a 20% to 30% reduction in relative risk may represent only a modest clinical achievement (i.e., if the absolute risk of having a CHD event over the next 10 years is only 2%, a 20% reduction results in the absolute risk falling to 1.6%).