Metabolic and morphologic changes are prevalent among HIV-infected individuals treated with antiretroviral therapy (ART). The metabolic disturbances may contribute to a range of morbidities, most notably cardiovascular disease (CVD) and type 2 diabetes mellitus. Changes in lipids may be influenced by drug choice within each of the approved oral drug classes. The incidence of these treatment-associated adverse changes can be minimized through prudent drug selection at treatment initiation and proactive treatment modification.
Metabolic changes characteristically observed with HIV infection include low total and high-density lipoprotein (HDL) cholesterol and elevated triglycerides. After initiation of ART, HDL cholesterol may “correct” upward; however, other changes commonly associated with ART include the following:
- Elevated total and low-density lipoprotein (LDL) cholesterol
- Elevated very-low-density lipoprotein (VLDL) cholesterol and triglycerides
- Insulin resistance with hyperglycemia, particularly in susceptible individuals
- Morphologic changes (ie, visceral, breast, local fat accumulation, subcutaneous fat loss)
Visceral fat accumulation leads to increased waist circumference. When this condition is accompanied by insulin resistance, low HDL cholesterol, elevated triglycerides, and hypertension, this constellation of signs and symptoms commonly observed in the general population is referred to as the metabolic syndrome. This syndrome is recognized with increasing frequency within general medical practice, has an age- and gender-related prevalence, and is overrepresented in populations experiencing CVD and type 2 diabetes. Once established, the metabolic syndrome is a vicious cycle that is hard to break. Lipid accumulation in the liver, pancreas, skeletal muscle, and visceral sites attributable to abnormal peripheral storage contributes to insulin resistance, thereby further exacerbating problems with peripheral lipid storage and glucose handling. Thus, avoiding initial establishment of the syndrome may be particularly relevant to limiting the risk of CVD and type 2 diabetes mellitus.
This review discusses the consequences and management of risk for CVD and type 2 diabetes mellitus in persons with HIV infection and focuses on the role that choice of protease inhibitor (PI) may play in the observed changes in the metabolic parameters during ART that harbinger these events.
CARDIOVASCULAR COMPLICATIONS IN HIV-INFECTED PATIENTS
In the general population, CVD, which includes coronary heart disease, stroke, congestive cardiac failure, and hypertensive disease, is the leading cause of death in the United States and Europe.1,2 A number of primary or “traditional” risk factors for CVD are well established, and several additional risk factors have been described more recently. Factors associated with an increased risk of CVD development can be classified as modifiable and nonmodifiable. Modifiable traditional factors include smoking, elevated LDL and non-HDL cholesterol, obesity, sedentary lifestyle, and factors comprising the metabolic syndrome. Nonmodifiable factors include age, gender, and genetic predisposition.3,4 In addition, several emerging CVD risk factors have been identified in recent years, such as homocystinuria and elevated levels of C-reactive protein. For persons receiving ART, the risk for CVD has been suggested to be substantially greater than that for the general population,5 although establishing control populations that match HIV-infected individuals for multiple traditional (eg, smoking) and other risk factors (notably, cocaine/amphetamine use and inflammatory markers) is challenging. The risk of CVD increases with each year of ART exposure,6 specifically PI exposure but not nonnucleoside reverse transcriptase inhibitor (NNRTI) use. Further analyses of these data suggest that lipid changes associated with HIV therapy lie on the causal pathway for CVD.7 Cohort studies suggest that cardiovascular events substantially contribute to mortality in HIV-infected patients receiving successful ART.8 Cardiovascular events are the underlying cause of more than 10% of deaths in this population, and CVD consistently ranks in the top 4 or 5 leading causes of death in persons with HIV (typically after AIDS-related events, end-stage liver disease, and malignancies).9 The close relation between ART and CVD highlights the need for careful consideration of CVD risk in the long-term management of persons with HIV infection.
IMPORTANCE OF LIPIDS AND THE METABOLIC SYNDROME IN HIV-INFECTED PERSONS
The Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) Study Group conducted a prospective observational study to examine the risk of myocardial infarction (MI) in HIV-infected patients receiving ART. The study collected data on more than 23,000 patients, with a median age of 39 years, who were enrolled in 11 previously established cohorts in Europe, the United States, and Australia. The patient population had a number of traditional cardiovascular risk factors with high prevalence at baseline, including current smoker or former smoker (56.2%), dyslipidemia (45.9%), hypertension (7.2%), body mass index (BMI) >30 kg/m2 (4.8%), and diabetes mellitus (2.8%). Over a period of 1.6 years up through February 2002, 126 patients had an MI (incidence: 3.5 events per 1000 person-years).6
Follow-up through February 2005 showed that a total of 345 patients experienced an MI over 94,469 patient-years. The relative risk of MI increased 16% for every year of ART exposure. A similar increase in relative risk was observed per year of PI exposure but not per year of NNRTI exposure. Adjusting for other factors, the authors observed that the effect of PIs on the risk for MI was partly explained by dyslipidemia. Of note, the data also indicated that the decline in age-adjusted MI risk observed in recent years was, in part, explained by a reduction in dyslipidemia in the observed population.7
Risk calculators, such as the Framingham calculators, are typically used to assess the 10-year CVD risk in individuals in the general population who have 2 or more risk factors for CVD. Application of the Framingham tool to the D:A:D study population consistently underestimated the number of events observed in this study.10 This may be because some key factors common in the D:A:D study and the general treated HIV population, such as hypertriglyceridemia, insulin resistance, and increased waist circumference, are not included in the calculations. These data suggest that lower 10-year CVD risk calculations should be used to guide CVD interventions among HIV-treated patients than those risk calculations used for the general population.
The significance of the metabolic syndrome in the setting of HIV has recently been elucidated. The INITIO study, an open-label randomized trial to evaluate different therapeutic strategies for HIV-1 infection, included 881 HIV-infected treatment-naive adults randomized in an international multicenter trial to initiate therapy with stavudine (d4T) plus didanosine (ddI) with efavirenz (EFV), nelfinavir (NFV), or EFV plus NFV. The median follow-up was 192 weeks. Participants were 21% female and had a mean age of 38.7 years and a mean BMI of 23 kg/m2. Analyses regarding the onset of the metabolic syndrome were adjusted for risk factors, including age, gender, and smoking status. Two different definitions of the metabolic syndrome were used: the National Cholesterol Education Program Adult Treatment Panel (NCEP ATP) III definition (at least 3 of the following risk factors: waist circumference >88 cm [women] or >102 cm [men], blood pressure >130/85 mm Hg, triglycerides >1.7 mmol/L [151.3 mg/dL], fasting glucose >6.1 mmol/L [109.8 mg/dL], and HDL cholesterol <1.29 mmol/L [50.31 mg/dL] in women or <1.04 mmol/L [40.56 mg/dL] in men) and the International Diabetes Federation (IDF) definition, which requires an increased waist circumference (waist circumference >80 cm [women] or >94 cm [men]) plus at least 2 additional factors (blood pressure ≥130/85 mm Hg, triglycerides ≥1.7 mmol/L [151.3 mg/dL], fasting glucose ≥5.6 mmol/L [100.8 mg/dL], and HDL cholesterol <1.29 mmol/L [50.31 mg/dL] in women or <1.03 mmol/L [40.17 mg/dL] in men). At baseline, 11% and 9% of participants met the NCEP and IDF definitions, respectively, increasing to 32% and 22% after a median 192 weeks of follow-up. In a multivariate model, randomization to a PI-containing group and baseline BMI, baseline hip circumference, waist-hip ratio >0.9, baseline total and HDL cholesterol, triglycerides, and diastolic blood pressure were also associated with progression to the metabolic syndrome. New CVD events were noted in 21 individuals, and new type 2 diabetes mellitus was noted in 41 individuals. The presence of the metabolic syndrome at baseline or during the study by either definition was associated with an increased relative risk of CVD or diabetes. For example, the presence of the metabolic syndrome at baseline according to the NCEP definition was associated with a 2.58 (95% confidence interval [CI]: 0.99 to 6.70) and 4.37 (95% CI: 2.24 to 8.52) increase in risk for CVD and diabetes, respectively, and hazard ratios according to the IDF definition were 2.97 (range: 1.14-7.74) and 2.89 (range: 1.35-6.20). Hazard ratios for the presence of the metabolic syndrome were substantially greater than the contribution of individual components of the metabolic syndrome.11 These data indicate that the metabolic syndrome plays a similar role in persons with HIV infection as it does in the general population in identifying persons at highest CVD risk.
Taken together, data from the D:A:D and INITIO studies suggest that scrutiny of metabolic and morphologic changes in persons with HIV and the management of these changes are critical to reduce the risk of CVD and type 2 diabetes mellitus in this population. Because the impact of any intervention has a lag time and the benefits accrue over several years, the earlier that CVD risk is identified and interventions are introduced, the sooner the risk can be managed. It is important to consider, however, that risk reduction is relative rather than absolute. Thus, interventions implemented now in the HIV population only serve to slow the trajectory of a potential CVD epidemic among an aging population of HIV-infected persons, and the benefits for each individual cannot be clearly measured.
MANAGING CARDIOVASCULAR DISEASE RISK
Managing CVD risk involves assessing all potential CVD risk factors, such as lipids, glucose, blood pressure, smoking, diet, exercise, and weight.12 In the setting of HIV infection, many of the interventions for reducing the risk of CVD are comparable to those recommended for the general population. Therapeutic lifestyle changes, including smoking cessation; regular exercise; and adhering to a diet high in ω-3 fatty acids, low in total and saturated fats, high in fiber, and rich in fresh fruits and vegetables, comprise the basic initial intervention. A recent report of lifestyle management in persons with HIV infection and the metabolic syndrome suggests comparable benefits on waist circumference, hemoglobin A1c (HbA1c), and systolic blood pressure as those observed with similar intervention in the general population.13 Algorithms for the management of dyslipidemia in the setting of HIV consider 2 main treatment approaches after lifestyle interventions: treatment switching and addition of conventional lipid-lowering agents.12 Pharmacologic interventions with lipid-lowering drugs seem to work similarly in persons with HIV as in the general population, accounting for drug interactions that occur between some antiretrovirals and lipid-lowering agents.14 3-Hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins) are the most effective for treating hypercholesterolemia, and brush-border inhibitors such as ezetimibe add further benefits, whereas fibric acid derivatives (fibrates) and fish oil preparations are more effective for treating hypertriglyceridemia and raising HDL cholesterol. The NCEP ATP III guidelines provide direction for managing dyslipidemia, with targets adjusted according to risk.15 Treatment with lipid-lowering therapy, even in combination, rarely achieves NCEP-defined targets, however.16-20 This underlines the importance of treatment modification in bringing lipid levels closer to values at which conventional interventions may enable NCEP targets to be reached or further intervention postponed.
PROTEASE INHIBITOR CHOICE AND METABOLIC CHANGES
The initiation of ART generally leads to increases in lipids. These effects include a rise in total cholesterol and HDL and LDL cholesterol. This may, in part, be attributable to a correction of the lipid abnormalities associated with HIV progression. The changes in total and LDL cholesterol typically increase beyond premorbid levels, however.21 Studies in healthy volunteers indicate that some individual antiretroviral agents can raise lipids and trigger insulin resistance. These changes are most evident and most well documented with PIs. The remainder of this review focuses on the documented effects of PIs on metabolic parameters. Consideration should be given to the role played by partner agents, nucleoside reverse transcriptase inhibitors (NRTIs), and NNRTIs in influencing the effects on metabolic parameters observed in individuals receiving combination ART.
Lipid changes are most well documented with ritonavir, where the effect may be exposure dependent.22 Data on ritonavir-boosted PIs (PI/r) derived from comparative studies in treatment-initiating patients are emerging. Data from comparative studies in more experienced patients also provide useful information but may be less readily interpreted because of the influence of prior therapies and morphologic changes on lipids and insulin sensitivity. Although comparative studies of PI/r versus NFV have been reported, the use of NFV as an initial choice for ART is no longer recommended. This review therefore focuses on comparative data among PI/r regimens. Several ongoing studies or PI/r comparisons have not yet reported data, most notably atazanavir (ATV)/r versus lopinavir (LPV)/r, Comparison of antiviral efficacy and safety of atazanavir/ritonavir with lipinovir/ritonavir each in combination with fixed dose tenofovir-emtricitabine in HIV-1-infected treatment naive patients (CASTLE), darunavir (DRV)/r versus LPV/r, A phase III randomized, controlled open-label trial to investigate the antiviral activity, tolerability and safety of TMC 114/RTV in treatment-naive HIV-1-infected subjects (ARTEMIS), and saquinavir (SQV)/r versus ATV/r, Boosted atazanavir or saquinavir induced lipid changes (BASIC), underlining that data in this area are still evolving.
The influence of just 100 mg of ritonavir administered once daily on lipids is best demonstrated by the AI424-089 study, in which 400 mg of ATV administered once daily was compared with 300/100 mg of ATV/r administered once daily with extended-release d4T (d4T XR) plus lamivudine (3TC) in initial therapy. In the AI424-089 study, increases in cholesterol and fasting triglyceride levels were observed in both arms at 48 weeks, with the exception of fasting triglyceride levels in the ATV arm, which decreased by 3% but rose by 26% with ATV/r. In general, cholesterol increases were greater in the ATV/r arm than with ATV alone (15% versus 6% for total cholesterol and 23% versus 16% for fasting LDL cholesterol), whereas changes in HDL cholesterol were similar at 30% versus 29%, respectively.23
Data on changes in lipids may not only be influenced by ritonavir exposure and possibly by dosing interval (once daily vs. twice daily) but by NRTI partner, making comparison across studies challenging.
The use of 700/100 mg of fosamprenavir (FPV)/r administered twice daily versus 400/100 mg of LPV/r administered twice daily with abacavir (ABC) plus 3TC has similar effects on lipids (Fig. 1).24 Changes in lipids in an initial therapy study comparing 300/100 mg of ATV/r administered once daily and 1400/100 mg of FPV/r administered once daily with tenofovir (TDF) plus emtricitabine (FTC) through week 24 are shown in Figure 2 and suggest a similar impact on lipids with these 2 regimens, with the exception of rises in triglycerides with FPV/r.25 The once-daily dosing of FPV with only 100 mg of ritonavir is currently investigational, however, and data have only been reported in small populations of individuals starting therapy through week 24. The observed differences between the effects of FPV in these studies may relate to differences in background NRTIs and ritonavir dose as well as to the lower total exposure of FPV achieved with the once-daily regimen.
Recent data suggest that 1000/100 mg of SQV/r administered twice daily as initial therapy may have a more favorable impact on lipids than 400/100 mg of LPV/r administered twice daily.26 These are consistent with data from the comparative MaxCmin2 trial, which included treatment-naive and -experienced individuals and also found more favorable effects of SQV/r on lipids, particularly triglycerides, relative to LPV/r.27
Over 4 weeks of therapy in 10 and 12 patients, respectively, median changes in lipid parameters for 2000/100 mg of SQV/r administered once daily and 300/100 mg of ATV/r administered once daily with TDF plus FTC were as follows: total cholesterol (7.4% vs. 6.1%), fasting triglycerides (39.4% vs. 11.5%), HDL cholesterol (0.7% vs. 7.9%), and LDL cholesterol (0.7% vs. 4.0%), respectively.28 A larger comparative study of these PI/r is ongoing.
Comparative data of the 2 most widely prescribed PI regimens, ATV/r and LPV/r, in initial therapy regimens are pending. Ten-day data in healthy volunteers suggest that even over this brief exposure period (during which lipid changes are unlikely to be completely established), modest but significantly smaller increases in total and LDL cholesterol and triglycerides were observed with ATV/r (Fig. 3).29
Further comparison of these 2 PIs in treatment-experienced patients is derived from the BMS 045 study. Changes observed in this study, as shown in Figure 4, indicated that switching from a failing treatment regimen to ATV/r led to favorable changes in lipids, whereas LPV/r was associated with rises in lipids, especially triglycerides.30
Finally, although comparative data of DRV/r versus LPV/r in initial therapy are pending, lipid changes reported from the Performance of TMC 114/r when evaluated in treatment-experienced patients with PI resistance (POWER) 1 and 2 studies, in which 600/100 mg of DRV/r was administered twice daily, suggest that this PI/r was associated with similar or lesser changes in lipid parameters relative to those observed with the comparator PI/r used in these treatment-experienced patients, most typically 400/100 mg of LPV/r administered twice daily and 700/100 mg of FPV/r administered twice daily.31
The overall impression from these data is that, among the more commonly prescribed PI regimens, ATV/r has the least impact on lipids, SQV/r administered once daily and twice daily and FPV plus 100 mg of ritonavir administered once daily have a moderate impact on lipids, whereas FPV/r administered twice daily and LPV/r have the greatest impact, especially on triglycerides. Currently, data on DRV are insufficient to draw conclusions.
Insulin resistance is an established risk factor for CVD risk and type 2 diabetes mellitus and a potential entry point for developing the metabolic syndrome and visceral fat accumulation. Direct inhibition of the insulin-responsive glucose transporter GLUT-4 is thought to contribute to insulin resistance. In vitro studies demonstrated that, unlike indinavir (IDV) and ritonavir, ATV had little or no inhibitory activity against GLUT-4.32 Recent data suggest that modest but statistically significant differences in peripheral glucose disposal occur when healthy volunteers are given 400/100 mg of LPV/r administered twice daily versus 300/100 mg of ATV/r administered once daily.33 LPV/r resulted in greater (25%) declines in glucose disposal from baseline relative to an 18% decline with ATV/r.
Use of unboosted ATV at a dose of 400 mg administered once daily in healthy volunteers is not associated with reductions in peripheral glucose disposal.34 Additional studies in healthy volunteers have found that IDV, ritonavir, and IDV/r all have an unfavorable impact on peripheral glucose disposal, whereas these changes have not been observed at steady state with unboosted ATV or with single doses of amprenavir. The effects of ATV/r are modest (∼10% decline) and are intermediate for placebo and LPV/r (∼25% decline). Changes in peripheral glucose disposal in persons with HIV starting combination therapy have not been extensively investigated. Preliminary data from a study of TDF plus FTC with 300/100 mg of ATV/r administered once daily or 2000/100 mg of SQV/r administered once daily as initial therapy in persons with HIV infection suggest that substantial variation in changes in peripheral glucose disposal may be observed with both approaches, with no changes in peripheral glucose disposal with ATV/r and a small (∼10%) decline with SQV/r.28
Recent data on 9 individuals switching from IDV, IDV/r, or LPV/r to ATV/r indicated a 28% increase in peripheral glucose disposal, reflecting reduced insulin resistance after switching.35 A large ongoing study is evaluating the benefits of this switch in patients with increased waist circumference after anecdotal reports of benefit.
Although therapeutic lifestyle changes comprise the preferred initial approach to CVD risk management for individuals on established ART, treatment guidelines and some available clinical trials data suggest that switching certain antiretroviral drugs to agents with better lipid profiles, when feasible, can meet NCEP targets without the (immediate) need for further interventions (eg, statins) or can improve the chance of meeting target levels in conjunction with conventional interventions. For patients with available ART options, switching to agents with a lower potential to induce metabolic abnormalities may be considered. For highly treatment-experienced patients who have limited therapeutic options, the risk of virologic or immunologic failure must be weighed against the benefit of reducing risk of a vascular event in the distant future.
Replacement of a PI with an NNRTI or ABC has been shown to help manage lipid abnormalities. In the recent past, this has been the preferred strategy for individuals who initiated a PI-based regimen and who still have the NNRTI option available. These switches generally lead to a range of metabolic benefits, which include declines in total and LDL cholesterol and triglycerides and an increase in HDL cholesterol.
Switching from a boosted or unboosted PI to ATV has been associated with lipid benefits, particularly with regard to triglyceride and total and LDL cholesterol levels. Improvements in total and LDL cholesterol with this switch often exceed 15%, an effect similar in magnitude to use of a less potent statin. It should be noted that the effects of boosted ATV as a replacement for an alternative boosted PI are less well established but have been reported from cohort studies to be of a similar magnitude.
The largest study to investigate drug switching within the PI class is the Switching to atazanavir (SWAN) study. This was a comparative, multicenter, open-label, randomized study in which 407 individuals receiving successful treatment with a twice-daily PI-based regimen and with a viral load of <50 copies/mL were randomized 2:1 to replace their PI with an ATV-containing regimen (n = 274) or to continue on their current PI (n = 133). Participants were excluded if they had previously failed a PI-based regimen or demonstrated PI resistance. Unboosted ATV was dosed at 400 mg once daily for all individuals except those receiving TDF (9% of all individuals); in that case, the boosted dose of 300/100 mg of ATV/r administered once daily was used. A total of 54% of individuals entering the study were on a PI/r regimen, with 37% of this total receiving LPV/r. On average, participants had been receiving a PI for 40.3 months. The primary study endpoint was virologic failure, defined as a rebound in HIV-1 RNA to >50 copies/mL. Viral rebound was observed in 7% of individuals who switched to ATV compared with 16% in the continuation group (P < 0.01). This difference was largely driven by the subset of individuals who were receiving an unboosted PI at baseline.
Switching to ATV resulted in marked improvements in a number of lipid parameters. Most notably, non-HDL cholesterol declined by 18% with the switch to ATV compared with a decrease of 3% in those individuals who continued on their original PI (P < 0.0001). This resulted in a higher proportion of ATV recipients meeting NCEP target values for multiple lipid parameters compared with those who continued treatment with their original PI (Fig. 5).36
These data suggest that individuals receiving a boosted or unboosted PI who have not previously failed PI therapy could consider modifying their treatment regimen by switching to an ATV-based combination as a means of improving their lipid profile. The choice of boosted or unboosted ATV may depend on the agents to be coadministered with ATV. When TDF is currently being used, it is obligatory to boost ATV; in other circumstances, boosting may not be necessary. Many physicians feel more confident with boosting ATV, however, and the results of the AI424-089 study in treatment-naive individuals indicate that boosted ATV may have some efficacy advantages and improved pharmacokinetics is associated with fewer important mutations at failure and that the inclusion of ritonavir results in only limited negative effects on the lipid profile of ATV. Cohort studies of switching from a PI/r to ATV/r suggest similar benefits on lipids as those reported in the SWAN study. The applicability of the switch approach to individuals who have experienced prior failure or resistance to PIs has not been evaluated from an efficacy standpoint.
As HIV infection becomes a chronic manageable disease and members of the HIV-infected population age, we are likely to see an emergence of chronic prevalent diseases, including diabetes mellitus and CVD. CVD risk data from the largest and longest studied cohorts in the setting of HIV infection suggest that cumulative years of exposure to combination ART, specifically PI-based ART, increases the risk of MI. Clinicians need to implement preventative and preemptive measures within the HIV-infected population to slow the emergence of a potential CVD epidemic. To this end, health care professionals need to consider CVD risk assessment as a routine component of the medical care of individuals with HIV infection.
There are several measures appropriate to managing this long-term risk. The first includes the routine institution and consistent encouragement of therapeutic lifestyle changes, including smoking cessation, weight reduction, regular exercise, and adherence to a diet rich in fresh fruits and vegetables and ω-3 fatty acids.
Consideration should be given to the impact of specific antiretroviral agents on lipid levels and insulin resistance. Comparative data are helping to define the effects of different antiretroviral agents on lipid levels and insulin sensitivity, both of which may contribute to the establishment of the metabolic syndrome and to future CVD risk. Differences among PI regimens with regard to lipids and insulin resistance have been documented and should be used to guide treatment choice. For individuals who are successfully established on treatment regimens and who do not have resistance to a specific agent, treatment switching should be considered as the preferred initial management approach for dyslipidemia and glucose intolerance. Clinical trials data suggest that within-class switching of PIs can improve a number of lipid parameters and achieve NCEP targets without the need for further interventions.
Intervention with conventional lipid-lowering therapy, in accordance with the guidelines set out by the NCEP ATP III, is recommended for individuals who do not achieve applicable NCEP targets through treatment switching or for whom this option is not appropriate. The addition of lipid-lowering agents, including statins, ezetimibe, and fibrates, to ART leads to changes in lipids similar to those observed in the general population; however, the changes in HIV-infected individuals are typically not sufficient to achieve NCEP targets.
1. American Stroke Association and American Heart Association. Heart disease and stroke statistics-2006 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation
2. Sans S, Kesteloot H, Kromhout D. The burden of cardiovascular diseases mortality in Europe. Task Force of the European Society of Cardiology on Cardiovascular Mortality and Morbidity Statistics in Europe. Eur Heart J
3. Grundy SM, Pasternak R, Greenland P, et al. AHA/ACC scientific statement: assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: a statement for healthcare professionals from the American Heart Association and the American College of Cardiology. J Am Coll Cardiol
4. Hackam DG, Anand SS. Emerging risk factors for atherosclerotic vascular disease: a critical review of the evidence. JAMA
5. Savès M, Chêne G, Ducimetière P, et al. Risk factors for coronary heart disease in patients treated for human immunodeficiency virus infection compared with the general population. Clin Infect Dis
6. Friis-Moller N, Sabin CA, Weber R, et al. Combination antiretroviral therapy and the risk of myocardial infarction. N Engl J Med
7. Friis-Moller N, Reiss P, El-Sadr WM, et al. Exposure to PI and NNRTI and risk of myocardial infarction: results from the D:A:D study. Presented at: 13th Conference on Retroviruses and Opportunistic Infections; 2006; Denver.
8. Reisler RB, Han C, Burman WJ, et al. Grade 4 events are as important as AIDS events in the era of HAART. J Acquir Immune Defic Syndr
9. Sackoff JE, Hanna DB, Pfeiffer MR, et al. Causes of death among persons with AIDS in the era of highly active antiretroviral therapy: New York City. Ann Intern Med
10. Law MG, Friis-Moller N, El-Sadr WM, et al. The use of the Framingham equation to predict myocardial infarctions in HIV-infected patients: comparison with observed events in the D:A:D study. HIV Med
11. Wand H, Calmy A, Carey D, et al. Metabolic syndrome, cardiovascular disease, and type 2 diabetes mellitus after initiation of antiretroviral therapy in HIV-infected adults. Presented at: Eighth International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV; 2006; San Francisco.
12. Grinspoon S, Carr A. Cardiovascular risk and body-fat abnormalities in HIV-infected adults. N Engl J Med
13. Fitch K, Anderdson E, Hubbard J, et al. Effects of a lifestyle modification program in HIV-infected patients with metabolic syndrome. Presented at: Eighth International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV; 2006; San Francisco.
14. Dube MP, Stein JH, Aberg JA, et al. Guidelines for the evaluation and management of dyslipidemia in human immunodeficiency virus (HIV)-infected adults receiving antiretroviral therapy: recommendations of the HIV Medical Association of the Infectious Disease Society of America and the Adult AIDS Clinical Trials Group. Clin Infect Dis
15. National Cholesterol Education Program. Third report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). NIH publication no. 01-3670. Bethesda, MD: National Heart, Lung, and Blood Institute, National Institutes of Health, US Department of Health and Human Services; 2001.
16. Moyle GJ, Lloyd M, Reynolds B, et al. Dietary advice with or without pravastatin for the management of hypercholesterolaemia associated with protease inhibitor
17. Nutescu EA, Shapiro NL. Ezetimibe: a selective cholesterol absorption inhibitor. Pharmacotherapy
18. Wohl DA, Tien HC, Busby M, et al. Randomized study of the safety and efficacy of fish oil (omega-3 fatty acid) supplementation with dietary and exercise counseling for the treatment of antiretroviral therapy-associated hypertriglyceridemia. Clin Infect Dis
19. Hollowell S, Kaye KS, Townsend ML. Achievement of NCEP goals is less frequent among HIV-infected versus non-HIV-infected veterans with dyslipidemia. Presented at: 45th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy; 2005; Washington, DC.
20. Negredo E, Rey-Joly C, Puig J. Ezetimibe, a selective inhibitor of cholesterol absorption, as a new strategy for treatment of hypercholesterolemia secondary to antiretroviral therapy. Presented at: 45th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy; 2005; Washington, DC.
21. Riddler SA, Smit E, Cole SR, et al. Impact of HIV infection and HAART on serum lipids in men. JAMA
22. Purnell JQ, Zambon A, Knopp RH, et al. Effect of ritonavir on lipids and post-heparin lipase activities in normal subjects. AIDS
23. Malan N, Krantz E, David N, et al. Efficacy and safety of antiretroviral-based therapy in antiretroviral naive HIV-1 infected subjects, both with and without ritonavir: 48 week results from AI424-089. Presented at: 13th Conference on Retroviruses and Opportunistic Infections; 2006; Denver.
24. Eron J Jr, Yeni P, Gathe J Jr, et al. The KLEAN study of fosamprenavir-ritonavir versus lopinavir-ritonavir, each in combination with abacavir-lamivudine, for initial treatment of HIV infection over 48 weeks: a randomised non-inferiority trial. Lancet
25. Smith K, Weinberg WG, DeJesus E, et al. Efficacy and safety of once-daily boosted fosamprenavir (FPV/r) or atazanavir (ATV/r) with tenofovir (TDF)/emtricitabine (FTC) in antiretroviral-naive HIV-1 infected patients: 24-week results from COL103952 (ALERT). Presented at: 46th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy; 2006; San Francisco.
26. Slim J, Avihingsanon A, Ruxrungtham K, et al. Saquinavir/r (SQV/r) bid vs lopinavir (LPV/r) bid plus emtricitabine/tenofovir (FTC/TDF) qd in ARV-naive HIV-1 infected patients: GEMINI study. Presented at: Eighth International Congress on Drug Therapy in HIV Infection; 2006; Glasgow.
27. Dragsted UB, Gerstoft J, Youle M, et al. A randomized trial to evaluate lopinavir/ritonavir versus saquinavir/ritonavir in HIV-1-infected patients: the MaxCmin2 trial. Antivir Ther
28. Jackson A, Patel N, Lo G, et al. Effects of atazanavir or saquinavir once daily with ritonavir 100 mg and tenofovir/emtricitabine as initial therapy for HIV-1 infection on peripheral glucose disposal: a randomized open-label study. Presented at: 14th Conference on Retroviruses and Opportunistic Infections; 2007; Los Angeles.
29. Noor MA, Flint OP, Parker RA, et al. Evaluation of insulin sensitivity in healthy volunteers treated with low-dose ritonavir combined with atazanavir or lopinavir: a prospective, randomized study using hyperinsulinaemic, euglycaemic clamp and oral glucose tolerance testing. Antivir Ther
30. Johnson M, Grinsztejn B, Rodriguez C, et al. 96-week comparison of once-daily atazanavir/ritonavir and twice-daily lopinavir/ritonavir in patients with multiple virologic failures. AIDS
31. Lazzarin A, Quieroz-Telles F, Rockstroh J, et al. TMC114 provides durable viral load suppression in treatment-experienced patients: POWER 1 and 2 combined week 48 analysis. Presented at: XVI International AIDS Conference; 2006; Toronto.
32. Murata H, Hruz PW, Mueckler M. The mechanism of insulin resistance caused by HIV protease inhibitor
therapy. J Biol Chem
33. Noor MA, Flint OP, Maa JF, et al. Effects of atazanavir/ritonavir and lopinavir/ritonavir on glucose uptake and insulin sensitivity: demonstrable differences in vitro and clinically. AIDS
34. Noor MA, Parker RA, O'Mara E, et al. The effects of HIV protease inhibitors atazanavir and lopinavir/ritonavir on insulin sensitivity in HIV-seronegative healthy adults. AIDS
35. Busti AJ, Bedimo R, Margolis D, et al. Improvement in insulin sensitivity and dyslipidemia in protease inhibitor
-treated patients after switch to atazanavir/ritonavir (ATV/RTV): a prospective study using hyperinsulinemic, euglycemic clamp testing. Antivir Ther
36. Gatell JM. Efficacy and safety of atazanavir (ATV) based HAART in patients switched from a stable boosted/unboosted protease-inhibitor (PI) treatment. Presented at: European AIDS Clinical Society 10th European AIDS Conference; 2005; Dublin.