The Evaluation and Management of Hyperlipidemia In HIV Infection
Reported for NATAP by Carl J. Fichtenbaum, M.D., Associate Professor of Clinical Medicine, University of Cincinnati College of Medicine, Division of Infectious Diseases, University of Cincinnati College of Medicine

ICAAC Report from Mike Norton, PA, Greenwich House, New York City. ICAAC, Tuesday, September 19, 2000.   Symposium titled: Lipodystrophy and metabolic complications of HIV.

There were 4 separate presentations during this 2.5 hour session.  William Lewis from Emory University was the first speaker.  He made a compelling case for nucleoside induced mitochondrial dysfunction/destruction. 

Steven Grinspoon of Harvard presented glucose dysregulation.  He noted that the changes being seen in treated patients are hyperinsulinemia and that this is often a prelude to diabetes.  Typically one observes at the beginning of glucose dyregulation a normal fasting glucose but an abnormal 2 hour glucose after a 75gram glucose challenge.  He called this state compensated normoglycemia.  The emerging belief is that protease inhibitors contribute something to glucose dysregulation.  Still it needs to be noted that fat accumulation around/in the abdomen also contributes to insulin resistance, and the cause of the truncal obesity is not so closely aligned with any one class of therapies.

Dr. Grinspoon stated that a fasting insulin level is the best test to identify emerging/developing glucose dysregulation.  But that widespread testing at this time is probably not warranted outside of the research setting because there isnŪt consensus on treatment.  Options include:  metformin, metformin with rosiglitizone, switching from a PI to an NNRTI or all Nuc regimen if the patient has either of these as an available option.  One option he ruled out was the use of growth hormone.  Dr. Grinspoon commented that using growth hormone in this setting is the wrong use of that product and can lead to DM-2 (Diabetes).

There is an upcoming ACTG trial looking at rosiglitizone and metformin in patients who have glucose dysregulation or truncal obestity.  This remains an important issue because there is a lipid independent risk of coronary artery disease associated with hyperinsulinemia.  At a minimum, patients with these signs and symptoms should do cardiovascular exercise and modify their diets.  

    Report from Jules Levin: At the Lipodystrophy Workshop and again at ICAAC both in Toronto in September 2000, Mattthias Egger (Department of Social Medicine and MRC Health Services Research Collaboration at the University of Bristol, UK) presented oral talk about the risk of cardiovascular complications attributable to HAART. He reported data from studies in non HIV-infected persons being studied for heart disease. He reported risk factors related to heart disease in these non-HIV infected individuals. He suggests HAART may accelerate the development of these risk factors therefore possibly  accelerated potential for heart disease in HIV infected individuals. In other words, individuals on HAART are developing these risk factors at an earlier age than people in the studies Egger described. Therefore, it seems reasonable that risk for developing cardiovascular complications may be accelerated in persons on HAART who have some of these risk factoirs. But it was pointed out that we don't have data yet on whether the development of these risk factors will in fact lead to accelerated cardiovascular complications in HIV infected individuals. Again, it was also pointed out that it is reasonable to assume that risk for heart disease is accelerated until we have data. The risk factors increasing risk for heart disease listed by Egger were:


This suggests that these considerations may affect when to begin therapy. A patient's family history and current lifestyle and diet should possibly be considered in deciding when to begin therapy. Egger was also very careful to point out that  it is crucial to contrast the adverse effects of cardiovascular risk with the benefits of HAART.

Very relevant to this is information in the report below written for NATAP from Carl Fichtenbaum. He points out that anti-lipid medications have mixed results in HIV infected individuals:

"the best treatment may not be clear.  Another problem is that the safety and effectiveness of drug treatments are largely untested in HIV-infected patients.  Finally, it is not known whether these fat disorders will result in an accelerated risk of progression of CVD.  However, there is no reason to believe that elevations in LDL and triglycerides will not confer at least a similar risk of CVD progression as in the general population÷÷ There are few prospective studies of lipid-lowering therapy in persons with HIV infection÷÷...  We recently reviewed our experience with treating hyperlipidemia in 23 patients with different medications.  The initial choice of lipid lowering therapy was a statin in 91% (71% received atorvastatin).  The average decline in cholesterol was 60%, similar to results reported in non-HIV infected populations.  However, the triglycerides increased in 43% of patients and less than half of our patients achieved NCEP recommended goals after 12 weeks of therapy.  These results must be interpreted cautiously as many patients were taking relatively low doses of Lipitor or Pravachol.  Nevertheless, these results are similar to other studies of lipid-lowering therapy in HIV infected patients÷÷.. These studies suggest that statins and fibrates may be of limited benefit in persons with protease inhibitor associated hyperlipidemia (see text from full article).  

This is the start of Dr. Fichtenbaum's article:

Metabolic complications of human immunodeficiency virus (HIV) infection and potent antiretroviral therapy (PART) have been increasingly recognized over the past two years [1-4].  Among the chief concerns are the occurrence of hyperlipidemia (elevated cholesterol and fats in the bloodstream) and the potential risk of developing cardiovascular disease (CVD).  Disorders of cholesterol/fats metabolism and elevated levels of certain fats have been strongly linked to the development of CVD (e.g., heart attacks) [5].  Higher levels of high-density lipoproteins (HDL), also known as žgood cholesterolÓ, are associated with a lower risk of CVD.  Conversely, elevations of low-density lipoproteins (LDL), also known as žbad cholesterolÓ, are associated with a greater risk of CVD [5].  Recent evidence suggests that elevated triglyceride levels, another type of fat, may also increase the risk for developing CVD [6-7]. Cardiovascular disease is a leading cause of illness and death in the United States and Western European Countries [5,8].

Is heart disease a problem for persons with HIV Infection?

Heart disease has been described in HIV infection prior to the use of PART. In the pre-protease inhibitor era, the most common heart problems were congestive heart failure (a build up of fluid in the lungs and legs), cardiomyopathy (failure of the heart to pump correctly), endocarditis (bacterial infection of the heart valves) and ventricular arrhythmias (an abnormal heart beat) [9-11].  Premature atherosclerosis (hardening of the arteries of the heart) was rarely reported.  Recently, Dr. Keith Henry reported heart attacks in two young people with few obvious risk factors for CVD [12]. Several others have also reported heart attacks in persons with HIV infection [13-14]. We have also recently reported a study of 16 persons with HIV infection who had heart attacks or chest pain resulting from hardening of the arteries of the heart.  Many of these persons underwent heart surgery to correct blockages to blood flow.  Thus, clearly persons with HIV infection can develop heart disease.

Fat redistribution disorders (lipodystrophy), diabetes mellitus (elevated blood sugars), insulin resistance (the hormone that lowers blood sugar levels), and elevated fat levels (cholesterol, LDL and triglycerides) have all been reported in persons with HIV infection [1-4, 15-19].  In the past few years, researchers have begun to describe different types of fat abnormalities in the bloodstream of persons using PART (Table 1).  In the Swiss HIV Cohort Study, the proportion of patients with a total cholesterol of greater than 240 mg/dL increased from 7% to 44% with ritonavir use, from 12% to 35% with indinavir use and from 5% to 33% with nelfinavir [18]. Significant elevations in LDL were observed in patients taking ritonavir and nelfinavir [18]. The available data indicates that elevations in the serum lipid levels are common in persons using PART.  The impact of these abnormalities on long-term cardiovascular health is not known but clearly concerning. 

How do I know if I have elevated lipids (fat levels)?

In 1993, the National Cholesterol Education Project (NCEP) released guidelines for the detection, evaluation and treatment of hyperlipidemia [20]. Recommendations are based upon whether a patient has heart disease, has risk factors for developing heart disease and analysis of fat levels in the bloodstream.  Major risk factors for CVD besides elevated fat levels include: age greater than or equal to 45 years (men) or greater than or equal to 55 years (women); a history of premature CVD in the family; cigarette smoking; high blood pressure; and diabetes mellitus. The NCEP guidelines are based upon information collected from numerous studies and that demonstrate that risk reduction improves oneŪs chances of living longer free of heart disease. 

Specific NCEP recommendations for persons with elevated fat levels in the bloodstream include asking questions about other risk factors for heart disease, obtaining fractionated lipid levels (LDL, HDL and triglyceride levels) and counseling people about their diet and exercise.  Recommendations are based upon how high the total cholesterol, HDL and LDL levels are and the presence of other risk factors for heart disease.  For example, if oneŪs total cholesterol is 200-239 mg/dL and the HDL is greater than or equal to 35mg/dL then a low fat diet, exercise and risk factor reduction are recommended.

NCEP recommends a fasting lipoprotein analysis (9-12 hours without eating or drinking except water and medications) for persons with greater than or equal to 2 other risk factors for CVD, a total cholesterol is less than 240 mg/dL and an HDL less than 35 mg/dL; or if the total cholesterol is greater than or equal to 240 mg/dL.  Medical intervention is definitely recommended for persons with greater than or equal to 2 CVD risk factors and an LDL greater than  130 mg/dL OR for anyone whose LDL is greater than or equal to 160 mg/dL. 

The NCEP guidelines stress the importance of the cholesterol, HDL and LDL but do not address elevations in triglycerides.  Some experts have recommended further evaluation and treatment based upon elevations in triglycerides [21].  This is particularly important to HIV-infected persons who may have significant elevations in triglycerides.  These guidelines are currently under review and changes in the goals, based upon more recent studies, may be forthcoming.

What do I do if I have elevated lipids?

The initial medical treatment consists of a step I diet recommendation (daily intake of saturated fat equal to 8-10%; less than or equal to 30% of calories from total fat; and less than 300 mg of cholesterol/day).  Step II diet modification is recommended if the above fails (daily intake of saturated fat = 7% and less than 200 mg of cholesterol/day).  Typically means avoiding fried foods or those high in saturated fats (meats and diary products).  Increased physical activity and weight reduction are suggested.  Patients should see a dietician and learn about low fat diets.  If the total cholesterol or LDL remains elevated despite diet and exercise, then drug therapy should be considered.

There are a number of classes of medications available for the treatment of hyperlipidemia.  Drug classes available for the treatment of hyperlipidemia including 3-hydroxy-3-methyl glutaryl CoA (HMG CoA) reductase inhibitors also known as žstatinsÓ, fibrates, bile acid binding resins and nicotinic acid.

HMG CoA Reductase Inhibitors (statins)

Statins competitively inhibit an enzyme responsible for the production of fat [22].  They also cause more receptors (doorways to cells) to form in the liver to clear LDL from the bloodstream and decrease the amount of a dangerous type of LDL.  All of these effects help blood vessels from developing fatty deposits. [22].  Statins also play a role in improving endothelial (the lining of blood vessels) function that is independent of its cholesterol lowering effects [23,24].

There are six FDA approved drugs in this class:

All share the characteristic of decreasing LDL, increasing HDL, lowering total cholesterol and triglycerides, some being more effective than others [19].  Lipitor is the most potent followed by Zocor and then Pravachol.  Lescol and Mevacor are rarely used.  All statins except Pravachol are cleared primarily through the liver (cytochrome p450 system) by the some of the same enzymes that clear protease inhibitors and non-nucleoside reverse transcriptase inhibitors like Efavirenz (Sustiva) [24,25]. 

Thus, there is a risk of a drug-drug interaction.  The most common side effects include bloating in the stomach and diarrhea.  Less commonly there can be elevations of liver enzymes and rarely, muscle inflammation or breakdown that can be life threatening.


Fibrates work primarily by decreasing serum triglyceride levels [26].  Clofibrate (Atromid, generic also available), gemfibrozil (Lopid), and fenofibrate (Tricor) are approved for use in the United States. They work by increasing the removal and breakdown of free fatty acids in muscle and liver.  This, in turn, leads to lower levels of triglycerides in the blood stream [25,26].  Treatment with fibrates in persons with high triglycerides (greater than 700mg/dl) can sometimes increase the levels of LDL [25,26].  Fibrates are cleared mainly through the kidneys and do not use the same enzymes as protease inhibitors or statins.  They can cause adverse effects, ranging from mild self-limited skin rash to more serious complications such as muscle inflammation or muscle breakdown.  Sometimes mild elevations of liver enzymes and trouble with getting erections can occur. These drugs are mainly indicated for high triglyceride levels or low HDL levels or as part of combination therapy for persons with hard to treat hyperlipidemia [25,26].

Nicotinic Acid (Niacin)

Nicotinic acid, or vitamin B3, inhibits the liverŪs production of different types of fats.  It is a very effective drug that lowers LDL and triglycerides and raises HDL by as much as 30% [24,25].  However, its usefulness is limited by two main side effects. It commonly induces flushing, itching, headache and gastrointestinal upset.  It is also associated with a significant risk of liver damage. Other less common side effects include elevations in uric acid (associated with gout) and muscle inflammation [25].

Bile Acid Binding Resins

Bile acid binding resins lead to lower levels of LDL by blocking the normal actions of bile acids [25]. These drugs include cholestyramine and colestipol (Colestid∆). These drugs are not utilized anymore as primary agents and tend to be used as part of a combination regimen. Side effects are generally limited to abdominal bloating [25].  The main concern is drug-drug interactions, especially those involving binding to certain drugs in the intestines. This includes protease inhibitors and nucleoside reverse transcriptase inhibitors.  Bile acid binding resins are limited in their usefulness in elevated triglyceride states [25].

Controversies in the management of hyperlipidemia in HIV infection

The best way to treat elevated fat levels in HIV-infected persons is unclear. Several problems exist. For example, the reasons why fat elevations occur are not known.  Inherited tendencies to have high fat levels, diets high in fat, and changes in the way blood vessels function have all been implicated in high lipid levels [27].  Several investigators have also described elevated fat levels associated with the use of protease inhibitors [18-19, 28]. Mitochondrial toxicity from nucleoside analogues has also been implicated as a possible cause of elevated fats in HIV-infected persons [29]. Until the reasons are known, the best treatment may not be clear.  Another problem is that the safety and effectiveness of drug treatments are largely untested in HIV-infected patients.  Finally, it is not known whether these fat disorders will result in an accelerated risk of progression of CVD.  However, there is no reason to believe that elevations in LDL and triglycerides will not confer at least a similar risk of CVD progression as in the general population.

What drugs should not be used to treat hyperlipidemia?

There are some basic therapeutic assumptions that can be made in the absence of data from clinical trials.  Bile acid binding resins bind polar drugs thus there is the potential for decreasing the absorption of nucleoside analogues (e.g., d4T, zidovudine) and protease inhibitors. Given the relative ineffectiveness of these agents in treating elevations in triglycerides, their potential for drug-drug interactions and their intestinal side effects, these drugs should probably be avoided in persons with HIV infection. Nicotinic acid also has significant adverse effects often resulting in inflammation of the liver.  This drug should probably be avoided until there is more data available on its safety.

Statins have been used with variable success in limited studies of persons with HIV infection.  Some statins have the potential for drug-drug interactions with antiretroviral agents and other drugs often used (e.g., itraconazole). A recent pharmacokinetics study suggested that Zocor∆ (and probably Mevacor because it is metabolized similarly) should not be used in persons taking ritonavir and saquinavir [30].  Simvastatin acid levels increased 30-fold in subjects taking ritonavir and saquinavir.  Similarly, Carr recently reported the same problem with Zocor when using Lopinavir/ritonavir (Kaletra) [31].  Hsyu also reported similar findings with Zocor and nelfinavir [32].  It is likely that the same effect would be observed with indinavir.  Two studies have also documented increases in the levels of total active Lipitor concentrations from 74-79% with the use of different protease inhibitors [30-32].  In addition, two cases of rhabdomyolysis (a condition where muscle breaks down) were recently reported with the use of Zocor and Lipitor, respectively, in HIV infected patients using protease inhibitors [33].  Thus, Zocor and Mevacor should be avoided and Lipitor should be used cautiously in patients using protease inhibitors.

Experience with Statins and Fibrates in the treatment of hyperlipidemia in HIV infection

There are few prospective studies of lipid-lowering therapy in persons with HIV infection [34-38].  Based upon NCEP guidelines, some clinicians have elected to treat fat disorders in persons with HIV infection with diet, exercise and medications.  We recently reviewed our experience with treating hyperlipidemia in 23 patients with different medications.  The initial choice of lipid lowering therapy was a statin in 91% (71% received atorvastatin).  The average decline in cholesterol was 60%, similar to results reported in non-HIV infected populations.  However, the triglycerides increased in 43% of patients and less than half of our patients achieved NCEP recommended goals after 12 weeks of therapy.  These results must be interpreted cautiously as many patients were taking relatively low doses of Lipitor∆ or Pravachol∆.  Nevertheless, these results are similar to other studies of lipid-lowering therapy in HIV infected patients.  For example, Dr. Keith Henry reported the use of diet, exercise, Lopid∆ and Lipitor∆ in 44 patients with HIV infection and hyperlipidemia [37].  Initially, 20 subjects were treated with diet and exercise (mean in cholesterol = 11%; mean in triglycerides = 21%) and 60% failed.  Seventy-six percent of subjects (n=25) treated with Lopid∆ alone failed.  Ten subjects were treated with Lipitor∆ 10-40 mg/day (mean in TC=19%; mean in TG=21%).  Nineteen subjects were treated with Lipitor∆ plus Lopid∆ (mean    in TC=30%; mean    in TG=60%).  Dr. Graham Moyle recently reported a small, randomized, open-label trial of diet versus diet plus Pravachol∆[38].  In 13 subjects treated with diet plus Pravachol∆, the average cholesterol decreased by 42 mg/dL after 24 weeks while the triglyceride levels increased by 18 mg/dL.  These studies suggest that statins and fibrates may be of limited benefit in persons with protease inhibitor associated hyperlipidemia. 

Another approach has been to switch antiretroviral treatment to non-protease inhibitor based regimens.  There have been several recent switch studies in patients with undetectable levels of HIV.  In persons where protease inhibitors were switched to nevirapine (Virammune∆), fat levels usually return to pre-protease inhibitor levels within 12 weeks [39-42].  In most studies, approximately 10% of patients who switch develop measurable levels of HIV after 48 weeks.  The long-term safety of this approach has not been established.


Evaluation of total cholesterol should be undertaken in all persons with HIV infection utilizing NCEP guidelines.  Fasting lipoprotein analysis (specific fat levels) may be necessary in persons with elevated lipid levels.  Other risk factors (e.g., cigarette smoking, high blood pressure, diabetes mellitus) for CVD should be identified.  Patients should also be routinely asked about whether they are experiencing chest pain, shortness of breath or other symptoms consistent with insufficient blood flow to the heart.

Initial interventions should focus on overall CVD risk reduction through aggressive education. Smoking cessation should be encouraged with referral to smoking cessation programs and the use of medications (e.g., nicotine supplements, bupropion and counseling). Dietary intervention should occur at the same time.  A dietician may be helpful in identifying problems.  A graded cardiovascular exercise program is optimal.  Patients should be warned about the signs and symptoms of heart disease. In patients at high risk for CVD (e.g., family history of premature CVD, high blood pressure, etc.) consideration should be given to adding an aspirin 81 mg each day to their medication regimen.

Patients with significant elevations in fats who fail diet and exercise should consider taking medication.  Participation in clinical studies should be encouraged.  There is a large AIDS Clinical Trials Group (ACTG) study that is about to begin and will compare treatment with Pravachol∆ and Tricor∆.  This study (ACTG A5087) will be available in more than 30 cities across the United States.  The choice of the initial medication should either be a statin or a fibrate. Statins are probably the logical first choice if the primary problem is elevation in the LDL cholesterol without significant elevations in triglycerides.  It is reassuring to note that statins do not appear to significantly alter protease inhibitor levels [31-32, 43].  Lipitor∆ and Baycol∆ may be used with caution.  Pravachol∆ is probably safe but it is generally less potent than Lipitor∆.  Therapy should be initiated with lower doses and increased as clinically indicated (Lipitor∆ 10mg, Pravachol∆ 20 mg and Baycol∆ at 0.3 mg daily).  Careful monitoring should be undertaken for signs of muscle disease and inflammation of the liver. Follow up fasting lipid levels should be done 6-8 weeks after initiation of therapy to determine the response and allow dose adjustments.  Lipitor∆ should probably not be dosed higher than 40 mg/day in patients taking drugs that inhibit its metabolism (e.g., protease inhibitors).

Lopid∆ or Tricor∆ are reasonable first line agents for the treatment of hypertriglyceridemia.  Some doctors may be unfamiliar with the use of these medications and may wish to seek the advice of lipid specialists before initiating treatment for isolated elevations in triglyceride levels.  The standard dose of Tricor∆ is 67 mg of the micronised preparation daily.  The usual dose of Lopid∆ is 600 mg twice daily 30 minutes before meals.

The use of combination therapy with a statin and a fibrate should be attempted cautiously because of the potential for additive toxicity. This should be reserved for patients who fail single drug treatment at recommended doses.  The advice of a lipid specialist may be valuable before using combination therapy.  An alternative for patients who fail specific lipid lowering therapy is to switch to a non-nucleoside analogue based regimen.  Nevirapine may be the preferred agent because of the possible lipid elevating effects of efavirenz (Sustiva∆).  Switching should only be undertaken if HIV infection is under good virologic control, there is no prior history of failure of non-nucleoside agents and the patient understands and accepts the risk of loss of virologic control.

see References

Table 1 Ů Selected studies of hyperlipidemia in persons with HIV infection

Reference Population Cholesterol TG HDL LDL


HIV- (n=42)
155 mg/dL 185 mg/dL

100  mg/dL
112 mg/dL

37 mg/dL
43 mg/dL
98 mg/dL
119 mg/dL
Feingold16 HIV+(n=94)
HIV- (n=42)
153 mg/dL
196 mg/dL
 95  mg/dL
100 mg/dL
30  mg/dL
44 mg/dL
 82  mg/dL
129 mg/dL
Behrens17 HIV+(n=94)
HIV- (n=42)
Potent antiretroviral
therapy era
175 mg/dL
257 mg/dL
162  mg/dL
451 mg/dL
35 mg/dL
36 mg/dL
114 mg/dL
163 mg/dL

Baseline - Followups


Ritonavir (n=46)
Indinavir (n=26)
Nelfinavir (n=21)

PI na‘ve (n=28)

178 mg/dL to  255 mg/dL
189 mg/dL
to 220 mg/dL
162 mg/dL
to 208 mg/dL
181 mg/dL
to 185 mg/dL

159 mg/dL to 318 mg/dL
195 mg/dL
to 177 mg/dL
186 mg/dL
to 195 mg/dL
133 mg/dL
to 124 mg/dL

39 mg/dL to 39 mg/dL
35 mg/dL
to 39 mg/dL
35 mg/dL
to 46 mg/dL
46 mg/dL
to 42 mg/dL

108 mg/dL to 162 mg/dL
120 mg/dL to
151 mg/dL    
  96 mg/dL to 139 mg/dL

140 mg/dL to
120 mg/dL

TG=triglyceride, HDL=high-density lipoprotein, LDL=low-density lipoprotein.  In the study by Periard, individuals were followed for several months for repeat values.