Use of Fenofibrate in the Management of Protease Inhibitor-Associated Lipid Abnormalities

James C. Thomas, M.D., Maria F. Lopes-Virella, M.D., Victor E. Del Bene, M.D., Joli D. Cerveny, Pharm.D., Kelly B. Taylor, B.S.N., Departments of Medicine (Drs. Thomas, Lopes-Virella, and DelBene, and Ms. Taylor) and Pharmacy Practice (Dr. Cerveny), Medical University of South Carolina, Charleston, South Carolina; Laura S. McWhorter, Pharm.D., the Department of Pharmacy Practice, University of Utah Health Science Center, Salt Lake City, Utah; Nanette C. Bultemeier, Pharm.D., Department of Pharmacy Practice, Oregon State University, Portland, Oregon

Pharmacotherapy. 2000;20(6) 

Abstract and Introduction

Human immunodeficiency virus (HIV) protease inhibitors are associated with several metabolic abnormalities including hypercholesterolemia and hypertriglyceridemia. Fenofibrate is a new lipid-lowering agent for adults with very high triglyceride levels that was administered to two HIV-positive patients who were taking protease inhibitors and developed hypertriglyceridemia. Starting dosages were 134 and 201 mg/day, and were increased to 268 mg/day in both patients. Triglyceride levels decreased from 1450 to 337 mg/dl (76.8%) and from 1985 to 322 mg/dl (83.8%), respectively, after 10 months of therapy. High-density lipoprotein levels increased in both patients.

The combination of human immunodeficiency virus (HIV) protease inhibitors and reverse transcriptase inhibitors is recommended as standard antiretroviral therapy.[1] Protease inhibitors confer virologic, immunologic, clinical, and survival benefits.[2] Therapy is begun early and aggressively, with most patients treated with combinations of three or four drugs, including a protease inhibitor. However, all protease inhibitors may be associated with development of metabolic abnormalities, including hypercholesterolemia, hyper-triglyceridemia, insulin resistance, glucose intolerance or diabetes mellitus, and lipodystrophies.[3-5]

In non-HIV-infected persons, hyper-triglyceridemia greater than 1000 mg/dl is associated with an increased risk of pancreatitis.[6] Patients with HIV infection may be at higher risk for pancreatitis because of drug toxicities, immunodeficiency, and protease inhibitor-associated hypertriglyceridemia.[7] In addition, hypertriglyceridemia and hypercholesterolemia are associated with increased risk of coronary artery disease (CAD).[8,9] In patients with HIV infection, this increased risk may be due to a high rate of concomitant chlamydial and cytomegalovirus infections, both of which are associated with CAD and acute myocardial infarction.[10] Isolated cases of pancreatitis and cardiovascular disease in patients with lipid changes secondary to protease inhibitors also have been reported.[11-14]

The number of lipid-lowering prescriptions is increasing for persons taking protease inhibitors with elevations in cholesterol and triglycerides. National Cholesterol Education Program Adult Treatment Panel II (NCEP ATP-II) guidelines do not specifically address protease inhibitor-induced lipid changes, although cholesterol-lowering agents have been administered in this setting.[9,15-18]

Due to potential complications of hyper-lipidemia, HIV-infected patients taking protease inhibitors could benefit from additional treatment options. Micronized fenofibrate (Tricor) is a fibric acid derivative approved by the Food and Drug Administration for the treatment of Fredrickson types IV and V hyperlipidemia. The literature also supports the drug in Fredrickson types IIa, IIb, and III hyperlipidemia.[19]

Case Reports

A 50-year-old woman diagnosed as HIV-positive 5 years earlier was referred to the lipid clinic from the infectious disease clinic because of an elevated triglyceride level. Her medical history was significant for acute pancreatitis 3 years earlier when her triglyceride level was 1760 mg/dl while taking saquinavir (replaced with ritonavir). Her only other medical problem was chronic epigastric pain for 8 months managed with a histamine H2 antagonist. Risk factors for CAD included a positive family history for premature CAD, a current cigarette smoker, and a low high-density lipoprotein (HDL) level (< 35 mg/dl). The patient did not drink alcohol and had not previously taken lipid-lowering drugs. She was taking indinavir, didanosine, and stavudine for HIV.

Physical examination was normal, without evidence of bruits, xanthomas, or xanthelasmas. Fasting total cholesterol was 272 mg/dl, HDL 25 mg/dl, and triglycerides 1450 mg/dl. Lipoprotein electrophoresis showed pre-beta 62%, b 24%, a 12%, and chylomicrons 2%, which is most consistent with an endogenous hyper-triglyceridemia with secondary chylomicronemia syndrome. Liver function tests were slightly elevated (alanine aminotransferase 57 IU/L, aspartate aminotransferase 59 IU/L), fasting serum glucose was 84 mg/dl, and serum creatinine was 1.2 mg/dl. The patient was instructed to consume a low-saturated fat/low-sucrose diet and was given an exercise program. She was prescribed fenofibrate 134 mg/day for 2 weeks, later increased to 201 mg/day because of a persistent high triglyceride level.

A nuclear magnetic resonance (NMR) spectroscopy lipid profile performed after the patient had been taking fenofibrate for 3 months showed total cholesterol 209 mg/dl, HDL 29 mg/dl, low-density lipoprotein (LDL) 129 mg/dl, and triglycerides 350 mg/dl. The LDL size was 20.0 nm, phenotype B, and LDL particles 1697 nmol/L. The predominant LDL subclasses were small 83 mg/dl and large 46 mg/dl. Most HDLs were small (HDL3) 22 mg/dl and large (HDL2) 7 mg/dl. The very low-density lipoprotein (VLDL) subclasses showed 107 mg/dl large, 205 mg/dl intermediate, and 5 mg/dl small. Four months later total cholesterol decreased to 208 mg/dl and triglycerides decreased to 421 mg/dl, and the patient ceased having abdominal pain. Fenofibrate was increased to 268 mg/day because of elevated triglycerides. After 10 months of treatment, total cholesterol was 194 mg/dl, HDL 31 mg/dl, triglycerides 337 mg/dl, and LDL 96 mg/dl (Table 1). Liver function tests remained normal and creatine kinase increased slightly, from 164 to 202 IU/L. The CD4 counts remained stable during therapy: 860 cells/mm3 baseline, 865 cells/mm3 mid-study, and 845 cells/mm3 end of the study.

A 36-year-old man diagnosed as HIV-positive 2 years earlier was referred to the lipid clinic because of elevated triglycerides. His medical history was significant for hypertriglyceridemia (treated in the past with gemfibrozil and niacin), hypertension, obesity, and mild elevation in transaminases. His family history was positive for type 2 diabetes mellitus and hypertriglyceridemia. Zidovudine-lamivudine and nelfinavir were begun several months after he was diagnosed with HIV.

His fasting baseline total cholesterol was 226 mg/dl, triglycerides 1282 mg/dl, and serum glucose 122 mg/dl. At week 4 of antiretroviral therapy, triglycerides had increased to 2263 mg/dl and fasting serum glucose was 217 mg/dl. When first seen in the lipid clinic, total cholesterol was 175 mg/dl, triglycerides 496 mg/dl, HDL 21 mg/dl, serum glucose 205 mg/dl, hemoglobin A1c (HgbA1c) 10.4%, creatinine 1.0 mg/dl, and lipase 183 IU/L; liver enzymes were slightly elevated. The patient was started on an 1800-calorie diet with 25% of calories as fat and a reduction in simple sugars and refined carbohydrates, an exercise program, and metformin 500 mg twice/day. Four and one-half months later total cholesterol was 263 mg/dl, triglycerides 1985 mg/dl, HDL 12 mg/dl, and HgbA1c 7.2%.

Fenofibrate 67 mg at lunch and 134 mg at dinner was prescribed. After 4 months of therapy with metformin and fenofibrate, total cholesterol was 202 mg/dl, triglycerides 1136 mg/dl, and HDL 17mg/dl. The HgbA1c decreased to 7.0% and creatine kinase decreased from 186 to 137 mg/dl. Fenofibrate was increased to 268 mg/day and metformin was maintained at 500 mg twice/day. At 10 months, total cholesterol was 237 mg/dl, HDL 29 mg/dl, LDL 144 mg/dl, and triglycerides 322 mg/dl (Table 1). Liver function tests and creatine kinase were normal; HgbA1c was 5.1%. The CD4 counts were 738 cells/mm3 at baseline, 857 cells/mm3 midstudy, and 828 cells/mm3 at study completion.


Fenofibrate achieved reductions in triglycerides of 76.8% and 83.8%, respectively, in two patients with hypertriglyceridemia secondary to protease inhibitor therapy. In both patients, total cholesterol levels decreased and HDL levels increased. The patients required higher dosages of fenofibrate than recommended due to very high triglyceride levels. They experienced no significant changes in liver function tests or creatine kinase after 10 months of therapy.

Patients with HIV infection may have several lipid abnormalities including increases in serum triglyceride and decreases in total cholesterol, HDL, and LDL levels. A decrease in HDL occurs first, followed by a decrease in serum cholesterol.[20] Since the introduction of protease inhibitors, lipid patterns have changed, with an increase in the frequency of hyperlipidemia with elevations in total cholesterol, VLDL, LDL, triglyceride, and lipoprotein(a) [Lp(a)] values and a decrease in HDL levels. Impaired glucose tolerance, diabetes, insulin resistance, and lipodystrophies also are associated with protease inhibitors.[4,21] Hyperlipidemia, glucose intolerance, or diabetes may occur within weeks of starting the agents.[3,4]

The prevalence of hyperlipidemia in HIV-infected patients taking protease inhibitors was reported to be up to 74%, compared with 28% in those not taking the drugs. Baseline triglyceride levels were above 180 mg/dl in 50% of 116 HIV-1-infected patients receiving at least one protease inhibitor, compared with 22% of 45 protease inhibitor-naïve patients. After 21 months of follow-up, triglycerides remained elevated in 50% and 26%, respectively.[21] Others reported that 27 (71%) of 38 HIV-infected patients treated with the agents had hyperlipidemia, compared with 24% of protease inhibitor-naïve patients.[22]

The effect of ritonavir on plasma lipids was evaluated in healthy volunteers in a double-blind, placebo-controlled study.[23] After 14 days of treatment, 11 patients taking ritonavir had increases in total cholesterol from 169 ± 10.8 to 208.8 ± 15.9 mg/dl (p<0.001) and triglycerides from 109.8 ± 12.4 to 270 ± 6.0 mg/dl (p<0.01) compared with no change from baseline in 8 subjects receiving placebo. The increase in total cholesterol in the ritonavir group resulted from increases in VLDL and intermediate-density lipoprotein, but not LDL. The lower HDL was a result of a decrease in HDL3 levels. The effect on lipids occurred within 2 weeks of therapy.

Protease inhibitors also seem to affect Lp(a), another atherogenic lipoprotein. Eleven patients with levels above 20 mg/dl had an increase in Lp(a) from 43.3 to 67.2 mg/dl (p=0.03), which was greater (p=0.05) than the change in seven protease inhibitor-naïve subjects with levels above 20 mg/dl (24.7 to 24.6 mg/dl, NS).[24] In another study, Lp(a) was above 30 mg/dl in 16% of protease inhibitor-treated patients who were hyperlipidemic.[22]

Small LDL subclass (phenotype B) has been demonstrated in patients with acquired immunodeficiency syndrome (AIDS). A 2.5-fold increase in small LDL was reported in patients with AIDS compared with healthy controls.[25] The presence of large LDL (phenotype A) was decreased by 39%. Subjects with phenotype B had marked increases in plasma triglyceride levels. It is not known if protease inhibitor-induced hyperlipidemia is also associated with small LDL subclass. Fifty-two percent of LDL in our patient no. 1 was small LDL as measured by NMR lipid subclass analysis. Elevated triglycerides are associated with increases in small LDL.[26] Increased small LDL is related to a 3.6-fold increase in the risk of CAD.[27]

Protease inhibitors affect glucose metabolism. In 24 HIV-infected patients treated with these agents, 46% had normal glucose tolerance, 37.5% had diabetes, and 16.7% had impaired glucose tolerance.[28] All patients with impaired glucose tolerance had peripheral insulin resistance. Impaired glucose tolerance was reported in 16% and diabetes in 7% of 113 protease inhibitor-treated patients.[21] Forty-six percent of HIV-infected patients receiving protease inhibitors had impaired glucose tolerance and 13% diabetes, compared with protease inhibitor-naïve subjects, of whom 24% had glucose intolerance and none had diabetes.[22]

Elevated plasma cholesterol levels have been noted with all protease inhibitors. Increases in serum lipids occurred frequently with ritonavir and ritonavir-saquinavir combination and were less common with indinavir, nelfinavir, and saquinavir.[12,23,24,28]

In some patients, as in both of ours, elevations in cholesterol and triglycerides may be extreme. In 133 patients taking protease inhibitors with elevated total cholesterol, LDL, or triglycerides as defined by NCEP APT-II guidelines, those receiving ritonavir and saquinavir were significantly more likely to have elevated lipid levels (p<0.001) than patients treated with indinavir or nelfinavir).[18]

Ninety-three HIV-infected adults receiving protease inhibitors were compared with 28 not taking the drugs.[24] Ritonavir caused a more pronounced increase in plasma cholesterol in the former group than in the latter (p<0.001) compared with indinavir (p=0.03) or nelfinavir (p=0.01). Ritonavir, but not indinavir or nelfinavir, was associated with a marked elevation in plasma triglycerides (p=0.002). Plasma HDL levels remained unchanged. The combination of ritonavir or nelfinavir with saquinavir did not further elevate plasma lipid levels.

In 15 patients, baseline total cholesterol of 200 ± 47 mg/dl (range 100-318 mg/dl) increased to 363 mg/dl and baseline triglycerides of 220 ± 192mg/dl (range 48-826 mg/dl) increased to 1000 mg/dl after starting triple-drug therapy, including at least one protease inhibitor (ritonavir 14, ritonavir-saquinavir 1).[17] One patient developed diabetes after taking ritonavir.

In 116 HIV-1-infected patients, triglyceride levels were significantly higher in those receiving ritonavir and saquinavir than in those receiving indinavir, nelfinavir, or nelfinavir-saquinavir combination (496, 248, 239.1, and 301.1 mg/dl, respectively, p<0.05).[21] Sixty-seven patients taking protease inhibitors had a significant increase in triglyceride and total cholesterol levels of 113 mg/dl and 37 mg/dl, respectively. A subgroup analysis of those with abnormal oral glucose tolerance had triglyceride values of 441 mg/dl versus 191 mg/dl in protease inhibitor-treated patients with normal glucose tolerance.[28]

Substantial increases in serum cholesterol and triglyceride levels were reported in a double-blind, placebo-controlled trial of four dosages of ritonavir: 300, 400, 500, and 600 mg twice/day.[3] Compared with subjects taking placebo, patients taking the two highest dosages of ritonavir had average cholesterol increases of 30-40% and triglyceride increases of 200-300%.

The mechanism of protease inhibitor-induced metabolic abnormalities is unknown. It is known that the drugs inhibit the cytochrome P450 (CYP) system.[29] A pathogenic model was hypothesized to explain part of the relationship between HIV-1 protease inhibitors and hyperlipidemia. The 12-amino acid genetic sequence of the catalytic region of HIV-1 protease was compared with all mammalian protein and genome sequences in gene libraries. Approximately 60% homology was found between the catalytic region of HIV-1 protease and regions within two proteins that regulate lipid metabolism, cytoplasmic retinoic-acid binding protein type 1 (CRABP-1) and LDL receptor-related protein (LRP). Protease inhibitors may bind CRABP-1, inhibiting retinoic acid activation, which may increase apoptosis of peripheral adipocytes, resulting in reduced triglyceride storage and increased lipid release into the circulation. Protease inhibitor binding to LRP would impair hepatic chylomicron uptake and triglyceride clearance by the endothelial LRP-lipoprotein lipase complex. The resulting hyperlipidemia may contribute to central fat deposition, insulin resistance, and, in susceptible individuals, type 2 diabetes.[30]

Insulin resistance may be the abnormality leading to glucose intolerance and/or diabetes. One group studied 67 HIV-infected patients treated with protease inhibitors, 13 protease inhibitor-naïve patients, and 18 HIV-negative controls for insulin sensitivity with intravenous insulin tolerance test.[28] Patients taking protease inhibitors had significantly decreased insulin sensitivity compared with protease inhibitor-naïve patients (median 75 and 156 µmol/L/min, respectively, p<0.001). In 24 protease inhibitor-treated patients who had a glucose tolerance test, all 13 with impaired glucose tolerance or diabetes had insulin resistance, as did 4 of 11 with normal glucose tolerance. Seventeen (70.8%) of the 24 had insulin resistance.

Because of the relatively high frequency of lipid abnormalities, we recommend that all patients who receive protease inhibitors be screened with a lipid panel before and during early stages (about 30 days) of therapy. A complete lipid panel including total cholesterol, LDL, HDL, and triglycerides is recommended since total cholesterol levels may be normal or even low.

The risk:benefit ratio in treating HIV-infected patients with elevated lipids is not known. Currently, no evidence suggests that treatment improves CAD outcomes. It does seem prudent to treat those with very high triglyceride levels to prevent pancreatitis and those with established CAD who are at high risk. Also, patients with several risk factors for CAD such as diabetes and hypertension are candidates for lipid-lowering therapy.

Guidelines for managing lipid abnormalities in patients with HIV infection have yet to be developed. Patients with hypercholesterolemia should be managed according to NCEP ATP-II, which stratifies treatment decisions and treatment goals based on LDL levels and risk factors for CAD.[9,31] Therapy should be individualized for the specific abnormality. The most common lipid abnormalities in the general population are Fredrickson types II ( type IIa, elevated LDL; type IIb, elevated LDL and triglycerides) and IV-V (primary elevated triglycerides). Fredrickson's classification is based on lipoprotein patterns associated with elevations in cholesterol and/or triglycerides, and disregards HDL.[32]

Of 27 HIV hyperlipidemic patients receiving protease inhibitors classified according to lipoprotein phenotype and following Fredrickson's classification, 18% had type IIa, 36% type IIb, 37% type IV, and 7% type V abnormalities.[22] This distribution is similar to that in patients with hyperlipidemia in the United States.[32]

Therapy of the lipid disorder is based on levels of LDL and/or triglycerides and the number of CAD risk factors. The goal LDL is less than 160 mg/dl for those with one or no risk factors and below 130 mg/dl for those with two or more risk factors. If CAD is present, target LDL is below 100 mg/dl. Normal triglycerides are less than 200 mg/dl, with 200-400 mg/dl considered borderline and greater than 400 mg/dl high. Triglyceride levels above 1000 mg/dl increase the risk for pancreatitis.[9]

Diet and exercise are recommended as initial therapy if lipid values are only modestly elevated. A low-fat (< 30% of total calories) and a low-saturated fat (< 10%) diet is recommended for patients with Fredrickson types IIa and IIb abnormalities, with addition of a low-sucrose diet for those with types IV-V and IIb disorders who have elevated triglycerides. Weight management and an exercise program are appropriate in all patients with lipid disorders and are essential in those who have elevated triglycerides.[9] However, dietary intervention lowers LDL by only 6-10%, and very high triglyceride levels are very difficult to manage with diet alone. In a study of 20 HIV-infected patients receiving protease inhibitors initially managed with diet and exercise, 12 were judged treatment failures and were treated with lipid-lowering agents.[18]

Lipid-lowering drugs are necessary if diet and exercise fail to control LDL or triglyceride values. The choice should be based not only on traditional considerations but also on aspects that are specific to the HIV-infected population. These patients often receive a number of other agents that have potentially severe adverse effects and interactions. For example, niacin, particularly sustained-release preparations, can cause hepatotoxicity and gastrointestinal distress, and many agents taken by these patients are also associated with similar effects.[15] In addition, there is concern for hyperglycemia with niacin, to which these patients may be predisposed.

Hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors are recommended for treatment of Fredrickson types IIa and IIb abnormalities. Care is recommended when considering one of these drugs in patients who are taking protease inhibitors since most inhibit the CYP enzyme, particularly CYP3A4. Lovastatin, simvastatin, and atorvastatin are metabolized by CYP3A4.[33] Atorvastatin caused no side effects in HIV-infected patients taking protease inhibitors.[18] Pravastatin is not metabolized by CYP3A4, and fluvastatin and cerivastatin have minor CYP3A4 metabolism and may be considered.[33] Studies with pravastatin, fluvastatin, and cerivastatin have not been reported in this population.

All protease inhibitors are metabolized by and are inhibitors of CYP3A4, with ritonavir being the most potent inhibitor.[29] Although drug-drug interactions have not been reported between HMG-CoA reductase inhibitors and protease inhibitors, the potential does exist for inhibition of metabolism of the former by the latter, with resultant toxicity. Monotherapy and the combination of a fibric acid and HMG-CoA reductase inhibitor are associated with rhabdomyolysis, elevated creatine kinase, and myoglobinuria, but rarely lead to acute renal failure. The rate of myopathic complications with the combination is estimated at 3-5%, and risk factors include renal impairments, age above 70 years, debilitation, and high HMG-CoA reductase inhibitor dosages. Monitoring the creatine kinase level is recommended.[19] The exact mechanism underlying this interaction and subsequent myotoxicity is unknown. Both classes are also associated with elevations in liver transaminases.

Fibric acids are the drugs of choice for treating HIV-infected patients with primary hyper-triglyceridemia who take protease inhibitors. Gemfibrozil and fenofibrate effectively lower elevated triglyceride concentrations. Micronized fenofibrate was approved in the United States in February 1998 as adjunctive therapy to diet for treatment of adults with hypertriglyceridemia (Fredrickson types IV, V) who are at risk for pancreatitis. It is rapidly hydrolyzed by esterases to the active metabolite, fenofibric acid; no unchanged drug is detected in plasma. Fenofibric acid is primarily conjugated with glucuronic acid and excreted in urine. A small amount is reduced at the carbonyl moiety to a benzhydrol metabolite, which in turn is conjugated with glucuronic acid and excreted in urine.[34] Interactions have not been reported between fenofibrate and protease inhibitors, and the potential appears to be low based on the route of metabolism.

Fenofibrate, like gemfibrozil, inhibits triglyceride synthesis and increases lipoprotein lipase activity, leading to both reduction in hepatic secretion of VLDL and increase in VLDL catabolism. As a consequence, plasma triglyceride levels are substantially decreased. Also during catabolism of VLDL, transfer of cholesterol from VLDL to HDL occurs, resulting in an increase in serum HDL levels. The distribution of the LDL subfraction is modified by fenofibrate to larger and less dense particles that are less atherogenic and have a higher binding affinity for cellular LDL receptors.[19,34] Fenofibrate is available in 67-mg capsules and recently in 200-mg capsules. The recommended starting dosage is 67 mg/day with increases determined by the degree of hypertriglyceridemia. With very high triglyceride levels (> 1000 mg/dl) a starting dosage of 200 mg/day is recommended. Fenofibrate administered daily, twice/day, and 3 times/day lowered triglycerides 40-55% in patients with Fredrickson types IIb, IV, and V hyperlipidemia.[19]

Fibrates increase the hypoprothrombinemic effects of oral anticoagulants, and the international normalized ratio may have to be monitored. Because cyclosporine can produce nephrotoxicity with decreases in creatinine clearance and rises in serum creatinine, and because renal excretion is the primary elimination route of fibrate drugs, it is possible that an interaction will result in deterioration.[34]

For patients who have elevations of both LDL and triglycerides, a combination of a fibric acid and HMG-CoA reductase inhibitor may be necessary. This combination increases the risk of myopathy and elevated liver function tests, and only one study reported its application in HIV-infected patients taking protease inhibitors.[18] Bile acid sequestrants are not recommended in these patients because they may increase triglyceride levels. In addition, the multiple-dose schedule required with protease inhibitors would be difficult, since other drugs must be taken 1 hour before or 4 hours after taking bile acid sequestrants.

Management of protease inhibitor-induced hyperlipidemia in patients with diabetes merits special consideration. The American Diabetes Association recently modified NCEP ATP-II guidelines for treatment of these patients. Specifically, the goal LDL level should be less than 100 mg/dl.[35] Glycemic control must be ideal to lower triglycerides. The HgbA1c levels should be between 5% and 6.5%, and levels above 7% are associated with elevations in triglycerides. Diet therapy should always be started if LDL is above 100 mg/dl. Drug therapy should be started for LDL levels above 100 mg/dl for patients with established vascular disease. For those without established vascular disease, drug treatment should be started at LDL levels of 130 mg/dl or higher.[35]

Treatment of protease inhibitor-induced lipid abnormalities has not been extensively studied, and existing trials involved small numbers of patients. Gemfibrozil resulted in an 83% reduction in median triglycerides and no change in total cholesterol in eight patients.[16] Atorvastatin 10-20 mg/day in 15 patients with hyperlipidemia secondary to a protease inhibitor-containing regimen (ritonavir 14 patients, ritonavir-saquinavir 1) produced a 25% reduction in total cholesterol and 35% reduction in triglycerides within 3 months.[17] No myopathy was observed; one patient had elevated liver function tests that returned to baseline in 3 months.

Gemfibrozil, atorvastatin, and a combination of the two were studied in three groups of HIV-infected patients taking protease inhibitors who developed hyperlipidemia.[18] In 25 patients, gemfibrozil 600 mg twice/day for a mean of 6 months reduced baseline total cholesterol from 316.6 to 251 mg/dl (-32%, p=0.002) and tri-glycerides from baseline of 1345.1 to 574.2 mg/dl (-57%, p=0.01). In 10 patients, atorvastatin 10-40 mg/day alone over a mean of 5.3 months reduced total cholesterol from 270.3 to 220.1 mg/dl (-19%, p=0.004) and triglycerides from 274.3 to 212.4 mg/dl (-21%, p=1.13). The combination of gemfibrozil 600 mg twice/day and atorvastatin 10-40 mg/day in 19 patients caused a decrease over 6.5 months from baseline total cholesterol of 312.7 to 220.1 mg/dl (-30%, p=0.004) and triglycerides from 1354 to 539.8 mg/dl (-60%, p=0.01). No patients experienced myopathy or elevated creatine kinase or liver enzymes.

Based on recent findings of differences in protease inhibitors and degree of elevated triglycerides, consideration may be given to changing the protease inhibitor. Ritonavir seems to be most commonly associated with lipid abnormalities. Changing it to nelfinavir or indinavir in seven hyperlipidemic subjects decreased total cholesterol from 297.3 to 247.1 mg/dl (p=0.004) and triglycerides from 827.4 ± 227.4 to 310.6 ± 61.1 mg/dl (p=0.02) within 2 months after the change.[36] Thus ritonavir-associated lipid disorders are partly reversible over a limited period of time. However, the decision to change therapy has to take into account the entire HIV regimen, and an HIV specialist should be consulted.

Hyperlipidemia is recognized as a consequence of uncontrolled HIV infection and is increasingly seen in individuals taking protease inhibitors. Metabolic abnormalities associated with the agents, including hypercholesterolemia, hypertriglyceridemia, low HDL, and increased Lp(a), may contribute to increased risk of CAD.[22] The extent of atherosclerotic risk of these patients must be taken in to consideration, including the magnitude of dyslipidemia and number of other atherosclerotic risk factors such as hypertension, diabetes, tobacco use, obesity, age, sedentary lifestyle, and family history of CAD. Therefore, management of dyslipidemias should aggressively address additional modifiable risk factors for CAD with interventions for weight management if obese, smoking cessation, and control of diabetes and hypertension if present.

Until recently, the long-term outcome for people with HIV infection and AIDS was so discouraging that concerns regarding chronic health conditions may have been of secondary importance. However, the success of therapy has decreased the frequency of opportunistic infections and other complications, and increased survival and quality of life. Moreover, attention to prevention and treatment of other diseases such as atherosclerosis and its sequelae and prevention of pancreatitis has become increasingly important.

Supported in part by Abbott Pharmaceuticals, Abbott Park, Illinois.


  1. Centers for Disease Control and Prevention. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. MMWR 1998;47(RR-5):1-82.

  2. Palella FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med 1998;338(13):853-966.

  3. Danner SA, Carr A, Leonard JM, et al. A short-term study of the safety, pharmacokinetics and efficacy of ritonavir, an inhibitor of HIV-1 protease. N Engl J Med 1995;333:1528-33.

  4. Carr A, Samaras K, Burton S, et al. A syndrome of peripheral lipodystrophy, hyperlipidemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998;12:F51-8.

  5. Viraben R, Aquilina C. Indinavir-associated lipodystrophy. AIDS 1998;12:F37-9.

  6. Toskes PP. Hyperlipidemic pancreatitis. Gastroenterol Clin North Am 1990;19(4):783-91.

  7. Dassopoulos T, Ehrenpreis ED. Acute pancreatitis in human immunodeficiency virus-infected patients. Am J Med 1999;107:78-84.

  8. Miller M. The epidemiology of triglyceride as a coronary artery disease risk factor. Clin Cardiol 1999;22(suppl II):II-1-6.

  9. National Cholesterol Education Panel. Detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel II). Circulation 1994;90:1329-1445.

  10. Danesh J, Collins R. Chronic infections and coronary artery disease: is there a link? Lancet 1997;350:430-6.

  11. Perry RC, Cushing HE, Deeg MA, Prince MJ. Ritonavir, triglycerides, and pancreatitis. Clin Infect Dis 1999;28:161-2.

  12. Di Perri G, Del Bravo P, Concia E. HIV-protease inhibitors. N Engl J Med 1998;339(11):773-4.

  13. Henry K, Melroe H, Huebsch J, et al. Severe premature coronary artery disease with protease inhibitors [letter]. Lancet 1998;351:1328.

  14. McBride M, Chin MTM, Lee SM, Kelly B. Lipid lowering therapy in patients with HIV infection. Lancet 1998;352: 1782-3.

  15. Echevarria KL, Hardin TC, Smith JA. Hyperlipidemia associated with protease inhibitor therapy. Ann Pharmacother 1999;33:859-64.

  16. Hewitt RG, Shelton MJ, Esch LD. Gemfibrozil effectively lowers protease inhibitor-associated hypertriglyceridemia in HIV-1-positive patients. AIDS 1999;13(7):868-9.

  17. Murillas J, Martin T, Ramos A, Portero JL. Atorvastatin for protease inhibitor-related hyperlipidemia. AIDS 1999;13(11):1424-5.

  18. Henry K, Melroe H, Huebesch J, et al. Atorvastatin and gemfibrozil for protease-inhibitor-related lipid abnormalities. Lancet 1998;352:1031-2.

  19. Guay DR. Micronized fenofibrate: a new fibric acid hypolipidemic agent. Ann Pharmacother 1999;33:1083-1103.

  20. Constans J, Pellegrin JL, Peuchant E, et al. Plasma lipids in HIV-infected patients: a prospective study in 95 patients. Eur J Clin Invest 1994;24:416-20.

  21. Carr A, Sampras F, Thorisdottir A, et al. Diagnosis, prediction, and natural course of HIV-1 protease-inhibitor associated lipodystrophy, hyperlipidemia, and diabetes mellitus; a cohort study. Lancet 1999;353:2093-9.

  22. Behrens G, Dejam A, Schmidt H, et al. Impaired glucose tolerance, beta cell function and lipid metabolism in HIV patients under treatment with protease inhibitors. AIDS 1999;13:F63-70.

  23. Purnell JQ, Zambon A, Knopp RH, et al. Effect of ritonavir on lipids and post-heparin lipase activities in normal subjects. AIDS 2000;14:51-7.

  24. Periard D, Telenti A, Sudre P, et al. Atherogenic dyslipidemia in HIV-infected individuals treated with protease inhibitors. Circulation 1999;100(7):700-5.

  25. Feingold KR, Krauss RM, Pang M, et al. The hyper-triglyceridemia of acquired immunodeficiency syndrome is associated with an increased prevalence of low density lipoprotein subclass pattern B. J Clin Endocrinol Metab 1993;76(6):1423-7.

  26. McNamara J, Jenner J, Li Z, et al. Change in LDL particle size is associated with change in plasma triglyceride concentration. Arterioscler Thromb 1992;12:1284-90.

  27. Lamarche B, Tchernof A, Moorjani S, et al. Small dense low density lipoprotein particles as a predicator of the risk of ischemic heart disease in men. Circulation 1997;95(1):69-75.

  28. Walli R, Herfort O, Michl G, et al. Treatment with protease inhibitors associated with peripheral insulin resistance and impaired glucose tolerance in HIV-1-infected patients. AIDS 1998;12(15):F167-73.

  29. Eagling VA, Back DJ, Barry MG. Differential inhibition of cytochrome P450 isoforms by the protease inhibitors, ritonavir, saquinavir, and indinavir. J Clin Pharmacol 1997;44:190-4.

  30. Carr A, Samavas K, Chisolm DJ, Cooper DA. Pathogenesis of HIV-1-protease inhibitor-associated peripheral lipodystrophy, hyperlipidaemia, and insulin resistance. Lancet 1998; 351:1981-3.

  31. Kaul DR, Cinti SK, Carver PL, Kazanjian PH. HIV protease inhibitors: advances in therapy and adverse reactions, including metabolic complications. Pharmacotherapy 1999;19(3):281-98.

  32. Gotto AM, Pownall HJ. Manual of lipid disorders; reducing the risk for coronary heart disease, 2nd ed. Baltimore: Williams & Wilkins, 1999.

  33. Wierzbicki AS, Reynolds TM, Crook MA, et al. Lipid lowering therapy in patients with HIV infection. Lancet 1998;352:1782.

  34. Abbott Laboratories. Tricor (fenofibrate) package insert. North Chicago, IL; June 1999.

  35. American Diabetes Association. Clinical practice recom-mendations 1999. Diabetes Care 1999;22(suppl 1):S56-9.

  36. Martinez E, Lazanol L, Congent I, et al. Reversion of lipodystrophy after switching HIV-1 protease inhibitors to nevirapine [abstr]. Presented at the 6th Conference on Retroviruses and Opportunistic Infections, Chicago, January 31-February 4, 1999.