• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

A Review of the Aetiology of Dyslipidaemia and Hyperlipidaemia in Patients With HIV/Discussion

Posted on: Saturday, 29 October 2005, 03:01 CDT

By Moyle, Graeme; Gazzard, Brian; Churchill, Duncan; Baily, Guy; Et al

Graeme Moyle MD MBBS

St Stephen's Centre, Chelsea & Westminster Hospital, London, UK

Definitions of lipodystrophy

There are two broad approaches to the definition of lipodystrophy. The first is exemplified by the case definition developed by Andrew Carr's group, supported by the European Agency for the Evaluation of Medicinal Products (EMEA) and the US Food and Drug Administration (FDA), according to which lipodystrophy is moderate to severe lipoatrophy in at least one area of the body.1 A series of tests that included glucose, lipids and peripheral fat mass on dual-energy X-ray absorptiometry (DEXA) scanning, as well as mitochondrial factors, provided an objective definition with sensitivity and specificity of less than 80%.

The second approach is exemplified by the definition due to Carl Grunfeld, evolving within the Fat Redistribution and Metabolic Change in HIV Infection (FRAM) study, according to which lipoatrophy is a unique phenomenon that occurs in people with HIV and that everything else (ie visceral fat accumulation and abnormal lipids) is the metabolic syndrome.2 The metabolic syndrome occurs in the general population, and HIV clinics are faced with individuals with lipoatrophy upon which is superimposed the additional problem of the metabolic syndrome. Lipoatrophy may predispose patients to the metabolic syndrome. All other issues of mitochondrial toxicity and bone abnormalities can be considered peripheral 'noise' that is irrelevant to the issue: for example, peripheral neuropathy is not clearly associated with the presence of lipoatrophy, although the risk factors with respect to drugs may have similarities.

Patterns of fat loss

DEXA and computed tomography (CT) scans from the FRAM study that looked at objective assessments of fat mass showed no association between the presence of visceral adiposity and the presence of peripheral lipoatrophy.2 Low amounts of fat in the periphery were also associated with small amounts of subcutaneous fat in the abdomen, so subcutaneous fat loss is a generalized phenomenon that occurs in the arms, legs, face and trunk, and the central accumulation of fat is not necessarily associated with that.

The FRAM study also showed that people tend to lose more fat in the legs than in the arms and that there is a relative maintainance of fat centrally, so that part of the aspect of abnormal body shape is a relative rather than an absolute phenomenon.3 More fat is lost in the periphery, so patients appear to have a larger trunk, even if they have not accumulated fat at that site. All sites had less fat than would be seen in an HIV-negative American population, although differences in ethnicity may be important. The FRAM study included more Caucasians and Hispanic patients than the control population (which included more African-Americans).2

Risk factors for lipodystrophy

Non-drug-related risk factors

Some risk factors for lipodystrophy apart from drugs have been identified within cohort studies and potentially in prospective studies: these include host and genetic factors. The HIV Outpatient Study (HOPS) of about 350 people showed that those who had never had a CD4 count less than 350 and in whom CD4 counts remained above 350 over the 23 months between surveys had only a 3% chance of developing lipoatrophy over that time.3 Those who had a low CD4 count that did not recover during the study had a 10-fold greater risk of developing lipoatrophy, independent of drug treatment.3 One criticism of HOPS is that people were included in the incidence study only if they had not had lipoatrophy at the time of first survey (which was a prevalence survey). Drug factors may have fallen out of this analysis because people who were taking riskier drug combinations were excluded from the incidence study as they already had lipoatrophy at the time of enrolment.

Data from Western Australian cohorts showed that polymorphisms in a tumour necrosis gene associated with increased production of tumour necrosis factor approximately doubled the risk of developing lipoatrophy over time.4,5

Some factors were not used in the calculation of cardiovascular risks, including insulin resistance, cholesterol particle size, endothelial dysfunction, C-reactive protein and waist circumference.

Lipoatrophy is associated with insulin resistance. Lipoatrophic mice have severe dyslipidaemia and insulin resistance, but if fat is transplanted back into such mice, they recover normal lipidaemia and normal glycaemia irrespective of whether the fat is transplanted evenly or in a lump on their back.6 In human models for lipoatrophy, such as the Berardinelli-Seip syndrome and the Dunnigan variety of familial partial Iipodystrophy, severe dyslipidaemia and insulin resistance are seen.

Drug-related risk factors

The most objective study to look at the development of fat loss over time was the ACTG 384-5055S sub-study, which looked at DEXA scanning of subcutaneous fat on a 16-week basis in a subset of patients randomized to different nucleoside reverse transcriptase inhibitor (NRTI) backbones (either stavudine/didanosine or zidovudine/lamivudine), in addition to a different third drug (either nelfinavir or efavirenz).7 The baseline characteristics were similar, although the limb fat mass in patients entering the cohort was slightly lower than in the control population. These participants may already have had low fat mass at baseline because of their progressive HIV disease. Over time after starting antiretroviral therapy, an improvement in limb fat mass was seen; importantly, this peaked at week 16 and then declined in both groups over time, although the decline was more rapid in the group who received stavudine/didanosine relative to the group who received zidovudine/lamivudine (Figure 1). As the error bars show, considerable variation was seen within that effect. Some individuals who took the 'riskier' NRTI backbone had no fat loss over 80 weeks, while others who took the 'safer' NRTI backbone still had substantial amounts of fat loss over that period of time. Whether that variation is bimodal or normal, whether the results show a spectrum of this phenomenon over time and whether some patients are particularly predisposed to or protected against fat loss over time is uncertain. A recent study certainly suggests a bimodal distribution in the effect of mitochondrial DNA in peripheral blood mononuclear cells with stavudine and didanosine over time.8 ACTG 384 also showed that those who were randomized to a protease inhibitor (PI) as the third drug, regardless of the backbone drugs, lost fat. Fat loss in the nelfinavir group was more rapid, was significantly below baseline at the two later timepoints and was significantly lower than in those randomized to a non-nucleoside reverse transcriptase inhibitor (NNRTI).

Figure 1 Median percentage change in limb fat from baseline by nucleoside reverse transcriptase inhibitors.7 [dagger]P<0.05 within groups. *P<0.05 between groups

Thus, two drug risk factors have emerged in terms of the rate of fat loss:

* choice of NRTI backbone

* choice of NNRTI relative to the chosen PI.

PIs Apart from atazanavir, all PIs tested in vitro have a dose- dependent effect on reducing GLUT-4 translocation,9 probably by directly attaching to the GLUT-4 receptor. This leads to peripheral insulin resistance, as GLUT-4 is responsible for glucose uptake in fat and muscle cells in the periphery. In mice in which GLUT-4 is deleted, lipoatrophy syndrome develops in the periphery,9 but mouse fat cells are more dependent on glucose for fat storage than human cells, so that model may not be perfect for the human situation. Clinically, even a single dose of a PI such as indinavir given to healthy volunteers is enough to induce reduced peripheral removal of glucose.10 In addition, some very complex studies in volunteer patients have shown a reduced response to glucose load at the β- cell level in the presence of PIs.11 GLUT-2 is inhibited by PIs to a lesser degree than GLUT-4, so for a given glucose load, a lesser output of insulin occurs. Another group has shown that reduced activity of the peroxisome proliferator-activated receptor (PPAR)- γ at a nuclear level further contributes to insulin resistance peripherally.12 The impact on GLUT receptors is not reversed by glitazones in vitro, whereas glitazones do rescue the PPAR-γ effect in that regard.

PIs, particularly ritonavir, lead to an increased output of lipids from the liver.13 This effect is improved by the use of statins, and potentially by fibrates, and it has been shown that patients who receive NRTIs and PIs simultaneously have increased basal lipolysis of peripheral fat cells, as well as reduced fat uptake.14

Metabolic syndrome

In this syndrome, a vicious cycle is seen in terms of lipids: increased hepatic output, reduced peripheral uptake and increased peripheral output of fat all contribute to dyslipidaemia. The cycling of fat leads to hypermetabolism. People with lipodystrophy have metabolic rates about 10% higher than age-, sex- and weight- matched controls, and this may make it more difficult for them to maintain their weight.15 The fat stored in the liver creates hepatic steatosis, which may also be present because of mitochondrial toxicity. Fat is also deposited in the muscles, whic\h then enhances the insulin resistance problem, thus establishing a self- perpetuating metabolic syndrome.

PIs

In vitro studies that looked at hepatocyte cell lines and different PIs showed a hierarchy of effect and a dose-dependent effect with these drugs: higher exposures to ritonavir led to increased cholesterol output from these cells and higher exposures to nelfinavir increased triglyceride output in a dose-dependent way.16 In contrast, atazanavir does not seem to cause such effects.

Ritonavir, given at a dose of 500 mg twice daily to healthy volunteers, produces prompt rises in total cholesterol and triglyceride concentrations, with cholesterol rising about 25% and triglycerides about 200-fold.13 Although baseline and 3-month values are relevant, lipids should continue to be measured over time as in some patients they may rise progressively. This may be a hepatic peripheral lipolysis effect, with a development of lipoatrophy over time that then starts to contribute to insulin resistance and dyslipidaemia.

PIs and NNRTIs

An interaction effect between PIs and NNRTIs seems to make dyslipidemia worse. An effect of therapy may also interplay with the presence of lipoatrophy. In studies with lopinavir/ritonavir in naive patients, when either efavirenz or nevirapine are introduced, dyslipidaemia worsens.17

Ritonavir (500 mg twice daily) induces dyslipidaemia in healthy volunteers in just 14 days of dosing (Figure 2). In the MaxCmin1 study, in which patients received ritonavir 100 mg twice daily in combination with indinavir or saquinavir, more modest percentage changes from baseline were seen in the saquinavir group at weeks 4 and 48.18 The changes were fairly consistent over time in terms of total cholesterol, low-density lipoprotein (LDL) and triglyceride levels.

Data presented to the FDA advisory group on the BMS 045 study showed that atazanavir, even when boosted with ritonavir, led to modest improvements in total cholesterol.19,20 The effect was similar to that with atazanavir alone. Patients experienced with PI treatment who switched to lopinavir/ritonavir had about a 5% increase in total cholesterol and a marginal increase in triglycerides. The atazanavir group did not seem to have improvements in their triglycerides, which remained stable over time, whereas modest improvements were seen in the atazanavir/ saquinavir group.

Figure 2 Effect of ritonavir on total cholesterol and triglycerides in healthy people.25 *P<0.001 versus baseline. [dagger]P<0.01 versus 14 days' placebo. [double dagger]P<0.01 versus baseline

Another study showed changes in lipid profiles in those taking an NNRTI and a PI simultaneously, a lipid shift was seen in those patients who received indinavir and efavirenz together. This group had double the rises in cholesterol and triglyceride as also did the groups who received either drug alone.21

Effects of nudeoside analogues

Two studies have suggested differences between zidovudine and stavudine, with the latter being associated with greater hypertriglyceridaemia and hypercholesterolaemia than the former.10 Results at 48 weeks from another study showed an effect of tenofovir relative to stavudine, the former being associated with less hypercholesterolaemia; at 96 weeks, statistically greater increases in HDL were also seen (Figure 3).22 The same study also showed that patients who were failing on antiretroviral therapy in whom tenofovir was added had modest improvements in dyslipidemia; this suggests a benign antilipid effect of tenofovir rather than a dangerous effect of stavudine.

NNRTI effects

Among the NNRTIs, the ratio of total cholesterol to high-density lipoprotein (HDL) tended to improve in patients who received nevirapine but remained stable in patients who received efavirenz.23 In the 2NN study,23 the total cholesterol-to-HDL ratio improved substantially and statistically significantly in nevirapine groups but was unchanged over time in the efavirenz group, although when the figures were entered into a risk calculator they produced only very modest changes in the absolute risk of heart disease. With all groups, drugs had an initial effect on the total cholesterol-to-HDL ratio, but the effect with efavirenz seemed to wane over time, whereas the effect with nevirapine was relatively sustained. Nevirapine had a neutral effect on triglycerides, and modest elevations were seen in the efavirenz group. In this study, the NRTI backbone was stavudine/lamivudine, and it would be interesting to have seen the lipid effects in this study with a backbone of tenofovir.

Figure 3 Mean change (95% confidence interval) in fasting lipid profile through week 48 in patients taking stavudine/lamivudine/ efavirenz or tenofovir/lamivudine/efavirenz.22 *P<0.0001

Atherogenesis and cardiovascular risk

The pathogenesis of cardiovascular disease involves LDL being taken up into the subendothelial space to form a fatty streak in macrophages (this is probably reversible if HDL levels can be increased). This is followed by formation of a plaque. The instability of plaques is the important issue in terms of driving cardiovascular events, and some factors that contribute to this include ageing, smoking, diabetes, insulin resistance, VLDL concentrations, hypertension, chronic inflammatory states and coagulability states. It is possible to intervene:

* insulin sensitizers may be useful in improving insulin sensitivity and in the management of diabetes

* statins can reduce LDL concentrations

* fibrates can reduce VLDL

* angiotensin-converting enzyme (ACE) inhibitors have been shown to improve the hypercoagulability state associated with these events, and in terms of blood pressure and increases in HDL this is valuable with respect to plaque stability as well as reverse cholesterol transport

* increases in HDL can be achieved with nevirapine and, to a lesser degree, with efavirenz.

Other areas are relevant: carotid medial thickening, brachial reactivity and flow-mediated dilatation. One study showed that patients who received PIs had reduced flow-mediated dilatation in response to the standard tests relative to HIV-negative, untreated controls.24 Another study showed that this is related to PIs interfering with production of the nitric oxide responsible for flow- mediated dilatation;25 such interference may raise the issues not only of cardiovascular disease but also of why some individuals report erectile dysfunction in the HIV setting.

How do we know whether patients will have increased rates of myocardial infarction? A lag exists before familial hyperlipidaemia develops, and an excess of mortality is seen 3-5 years later. Large studies over long periods of time are needed.

A French study that compared a study population with the national database Monica showed a time effect: in patients who had been taking PIs for more than 30 months, the relative risk was statistically significantly increased relative to the general French population.26 In a number of cohorts that have reported on this issue, smoking was more prevalent in the HIV population relative to the general population,27 which is likely to be greatly increasing the risk in this population.

Other databases have shown that the chances of being admitted to hospital with a cardiovascular diagnosis were not increased in relation to whether or not these patients were on therapy. Physicians trying to get patients with HIV into hospital, who simply ticked the 'cardiovascular disease' box, might be the reason for this finding.28

DAD is a large study taking place across Europe, Australia and North America involving 23,000 people, mostly men, with an average age approaching 40 years.29 About a quarter of the participants have a prior diagnosis of AIDS and a medium viral load that is undetectable. This study found that the chances of having a World Health Organization (WHO)-defined myocardial infarction (irrespective of the inclusion of remote or improbable myocardial infarctions by the WHO definitions) increased per year of antiretroviral exposure. The study did not break down the results by specific drugs within the cohort, but the event rate was very modest over time, and most cardiovascular events occurred in male smokers with an average age of 47 years. Traditional risk factors were also observed within the cohort -i.e. older age, male sex, prior history of cardiovascular disease and smoking. In terms of baseline effects, cholesterol was significant, as was diabetes mellitus. An apparent protective effect of lipodystrophy was investigated, defined and dismissed recently.29

The Cochrane Library has assessed studies that have looked at the Mediterranean diet and showed level 1A evidence that increasing omega-3 fatty acids through oily fish, canola or rapeseed or flax oil, increasing fruit and vegetable intake, and reducing total fats are beneficial and essentially risk-free interventions.30

Management of dyslipidaemia

When managing dyslipidaemia in the clinic, drug-related risk factors for lipoatrophy and dyslipidaemia should be modified. The first considerations are whether the patient is NNRTI-naive, is taking a PI and has the option of switching within that class. Anticipated improvements from treatment switches include the following: insulin sensitivity improves, HDL tends to improve (particularly with the NNRTI switches) and total cholesterol improves (particularly with nevirapine and abacavir). A randomized study showed that the total cholesterol-to-HDL ratio improved similarly with all the drugs and that the NNRTIs have the greatest effect on insulin sensitivity and triglycerides, although a switch to abacavir is also associated with improvements in insulin sensitivity (Figure 4).31 Concentrations of HDL increase more with the nevirapine switch than the abacavir switch, while increases with efavirenz are intermediate. The other switch option is the switch from a PI to atazanavir. The switch from nelfinavir to atazana\vir has been confirmed as virologically safe, with no virological failures but improvements in total cholesterol, HDL, LDL and triglycerides; improvements in insulin sensitivity as a result of making this switch have also been reported.32 A large, fully recruited study is looking at switching from other PI regimens to atazanavir.

The main approaches to treating hypertriglyceridaemia are to reduce smoking and the consumption of refined carbohydrates, increase exercise, and reduce the amount of ritonavir being taken. Early development studies for ritonavir showed a dose-dependent effect of ritonavir in that regard,33 but most patients only take 100 mg twice daily, so the dose cannot reasonably be reduced further. In patients who present with severe triglyceridaemia, measurement of apolipoprotein E and a glucose tolerance test confirm whether the hypertriglyceridaemia is being driven by insulin resistance, in which case an insulin sensitizer may manage more of the problem other than intervention with a fibrate.

Gemfibrozil has been shown to have approximately the same effect on hypertriglyceridaemia as in individuals with endogenous dyslipidaemia.34 When treating hypercholesterolaemia, drug interaction issues rule out the use of simvastatin; fluvastatin may be useful, although there is the potential for interaction with nelfinavir and ritonavir through the CYP2CP pathway. Atorvastatin levels are increased about 70% in the presence of 400 mg ritonavir.35 Drug interaction studies have not been performed at other doses to see whether the effect is more modest, but it is likely to decrease with effect. Interaction studies looking at the impact of NNRTIs on statins are limited, so it is not known how much the level of atorvastatin, for example, is reduced by the presence of efavirenz or nevirapine.

Figure 4 Metabolic changes in patients switching from protease inhibitor to abacavir, efavirenz or nevirapine.31 TC, total cholesterol; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TC, triglycerides. *P<0.005

A study that compared pravastatin with dietary advice showed two effects.36 Firstly, pravastatin reduced the total cholesterol by approximately the same amount as in studies in the general population, which showed that over about 6 years in a population of Scottish men, the risk of primary cardiovascular disease can be reduced37 Secondly, patients given a statin do not follow the dietary advice given, tending to rely on their drug to do all the work, so a better effect might have been seen if patients on statins had been more adherent to dietary advice. Another randomized study looked at the achievement of National Cholesterol Education Program (NCEP) outcomes with fenofibrate and pravastatin.38 LDL targets are more likely to be achieved with pravastatin, and HDL and triglyceride targets with a fibrate, but neither type of drug was particularly effective at achieving all guideline targets. This perhaps underlines the need to consider drug switching before intervening with drugs that are modestly effective in the 20% of patients who take HAART and who develop significant lipid abnormalities.

Conclusion

The available data show that the prevalence of dyslipidaemia is substantial in people receiving therapy for HIV, particularly in those patients treated with approved PIs. NNRTIs are known to have an effect -it is not just the protease inhibitors that do so. Several in vitro mechanisms have been shown in healthy volunteers, so the presence of HIV infection is not important. Initial therapy needs to consider the risks and benefits of different drugs with respect to cardiovascular effects. When choosing drugs for patients who are already treatment-experienced, our hand is often forced towards particular drugs because of effectiveness, in which case cardiovascular risk steps into the background if control of viral load is a concern. Diet and exercise interventions should be routine, because they are effectively no-risk interventions that demonstrably are effective in secondary prevention; for example, a Mediterranean diet is approximately as effective as aspirin or a statin according to the Cochrane systematic review of diet.30 In clinical practice, we should follow the NCEP guidelines for the use of fibrates and statins, with consideration given to drug interactions, and the potential of new antiretroviral drugs to manage that problem should be exploited.

References

1 Carr A, Emery S, Law M, et al. An objective case of definition of lipodystrophy in HIV-infected adults: a case-control study. Lancet 2003;361:726-35

2 Tien PC, Grunfeld C. What is HlV-associated lipodystrophy? Defining fat distribution changes in HIV infection. Curr Opin Infect Dis 2004;17:27-32

3 Lichtenstein KA, Delaney KM, Armon C, et al. Incidence of and risk factors for lipoatrophy (abnormal fat loss) in ambulatory HIV- 1-infected patients. J Acquir Immune Defic Syndr 2003;32:48-56

4 Nolan D. Metabolic complications associated with HIV protease inhibitor therapy. Drugs 2003;63:2555-74

5 Martin AM, Nolan D, Gaudieri S, et al. Pharmacogenetics of antiretroviral therapy: genetic variaton of response and toxicity. Pharmacogenomics 2004;5:643-55

6 Reitman ML, Gavrilova O. A-ZIP/F-1 mice lacking white fat: a model for understanding lipoatrophic diabetes. Int J Obes Relat Metab Disord 2000;24:S11-14

7 Dub MP, Zackin R, Tebas P, et al. Prospective study of regional body composition in antiretroviral-nave subjects randomized to receive zidovudine+lamivudine or didanosine+stavudine combined with nelfinavir, efavirenz, or both: A5005S, a substudy of ACTG 384. Presented at the 4th International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV, San Diego, September 2002. Antiviral Ther 2002;7:L18 Abst 27

8 Cote HC, Brumme ZL, Craib KJ, et al. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. N Engl J Med 2002;346:811-20

9 Murata H, Hruz PW, Mueckler M. The mechanism of insulin resistance caused by HIV protease inhibitor therapy. J Biol Chem 2000;275:20251-4

10 Noor MA, Seneviratne T, Aweeka FT, et al. Indinavir acutely inhibits insulin-stimulated glucose disposal in humans: a randomized, placebo-controlled study. AIDS 2002;16:F1-8

11 Noor MA, Lo JC, Mulligan K, et al. Metabolic effects of indinavir in healthy sero-negative men. AIDS 2001; 15:F11-18

12 Bastard JP, Caron M, Vidal H, et al. Association between altered expression of adipogenic factor SREBP1 in lipoatrophic adipose tissue from HIV-1-infected patients and abnormal adipocyte differentiation and insulin resistance. Lancet 2002;359:1026-31

13 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

14 Sekhar RV, Jahoor F, White AC, et al. Metabolic basis of HIV- lipdystrophy syndrome. Am J Physiol Endocrinol Metab 2002;283:E332- 7

15 Kosmiski LA, Kuritzkes DR, Lichenstein KA, et al. Fat redistribution and metabolic changes are strongly correlated and energy expenditure is increased in the HIV lipodystrophy syndrome. AIDS 2001;15:1993-2000

16 Parker RA, Meyers DS, Andrews BA, et al. Effects of nucleoside reverse transcriptase inhibitors and HIV protease inhibitors on adipogenesis and adipocyte metabolism. Presented at 2nd International Workshop on Adverse Drug Interactions and Lipodystrophy in HIV, Toronto, September 2000: Abst 04

17 Murphy RL, Brun S, Hicks C, et al. ABT-378/ritonavir plus stavudine and lamivudine for the treatment of antiretroviral-nave adults with HIV-1 infection: 48-week results. AIDS 2001;15:F1-9

18 Dragsted UB, Gerstoft J, Pedersen C, et al. Randomized trial to evaluate indinaivr/ritonavir versus saquinavir/ritonavir in human immunodeficiency virus typle 1-infected patients: the MaxCmin1 trial. J Infect Dis 2003;188:635-42

19 Squires K, Lazzarin A, Gatell JM, et al. Comparison of once- daily atazanavir with efavirenz, each in combination with fixed- dose zidovudine and lamivudine, as initial therapy for patients infected with HIV. J Acquir Immune Defic Syndr 2004;36:1011-19

20 Cahn PE, Gatell JM, Squires K, et al. Atazanavir - a once- daily HIV protease inhibitor that does not cause dyslipidemia in newly treated patients: results from two randomized clinical trials. J Int Assoc Physicians AIDS Care 2004;3:92-8

21 Staszewski S, Morales-Ramirez J, Tashima KT, et al. Efavirenz plus zidovudine and lamivudine, efavirenz plus indinavir, and indinavir plus zidovudine and lamivudine in the treatment of HIV-1 infection in adults. Study 006 team. N Engl J Med 1999;341:1865-73

22 Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-nave patients: a 3-year randomized trial. JAMA 2004;292:191-201

23 van Leth F, Phanuphak P, Ruxrungtham K, et al. Comparison of first-line antiretroviral therapy with regimens including nevirapine, efavirenz, or both drugs, plus stavudine and lamivudine: a randomized open-label trial, the 2NN study. Lancet 2004;363:1253- 63

24 Stein JH, Klein MA, Bellehumeur JL, et al. Use of human immunodeficiency virus-1 protease inhibitors is associated with atherogenic lipoprotein changes and endothelial dysfunction. Circulation 2001;104:257-62

25 Dub MP, Shankar S, Vanderluitgaren JM, et al. Effect of indinavir (IDV) monotherapy on endothelial function in men without HIV infection. Presented at the 9th Conference on Retroviruses and Opportunistic Infections, Seattle, February 2002: Abst LB 10

26 Mary-Krause M, Cotte L, Simon A, et al. Increased risk of myocardial infarction with duration of protease inhibitor therapy in HIV-infected men. AIDS 2003;17: 2479-66

27 Mooser V. Atherosclerosis and HIV in the highly active antiretroviral therapy era: towards an epidemic of cardiovascular disease? AIDS 2003;17:S65-9

28 Klein D, Hurley LB, Quesenberry Jr CP, Sidney S. Do protease inhibitors increase the risk for coronary hea\rt disease in patients with HIV-1 infection? J Acquir Immune Defic Syndr 2002;30:471-7

29 Friis-Moller N, Sabin CA, Weber R, et al. Combination antiretroviral therapy and the risk of myocardial infarction. N Engl J Med 2003;349:1993-2003

30 Hooper L, Summerbell CD, Higgins JP, et al. Dietary fat intake and prevention of cardiovascular disease: systematic review. BMJ 2001;322:757-63

31 Martinez E, Arnaiz JA, Podzamezer D, et al Substitution of nevirapine, efavirenz, or abacavir for protease inhibitors in patients with human immunodeficiency virus infection. N Engl J Med 2003;349:1036-46

32 Murphy R, Ledesma E, et al. A nelfinavir (NFV) to atazanavir (ATV) switch correlates with lipid improvements at 12 weeks: results from BMS AI424-044, for the AI424-044 International Study Team. Presented at the 6th International Conference on Drug Therapy in HIV Infection, Glasgow, November 2002

33 Markowitz M, Saag M, Powderly WG, et al. A preliminary study of ritonavir, and inhibitor of HIV-1 protease, to treat HIV-1 infection. N Engl J Med 1995;333:1534-9

34 Miller J, Brown D, Amin J, et al. A randomized, double-blind study of gemfibrozil for the treatment of protease inhibitor- associated hypertriglyceridaemia. AIDS 2002; 16:2195-200

35 Fichtenbaum CJ, Gerber JG, Rosenkranz SL, et al. Pharmacokinetic interactions between protease inhibitors and statins in HIV seronegative volunteers: ACTG Study A5047. AIDS 2002;16:569- 77

36 Moyle GJ, Lloyd M, Reynolds B, et al. Dietary advice with or without pravastatin for the management of hypercholesterolaemia associated with protease inhibitor therapy. AIDS 2001;15:1503-8

37 Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995;333:1301-7

38 Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of The 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). JAMA 2001;285:2486-97

Discussion

BRIAN GAZZARD: You discussed data that showed that 10% of 16- year-olds have heart disease.

GRAEME MOYLE: That is correct - in the British population, the prevalence of cardiovascular disease in young people is probably increasing.1

BRIAN GAZZARD: I do not believe those data. I think that differences in prevalence are present between men and women in the relevant age groups (30-60 years), and those differences might translate into a different relative risk for those age groups.

BRIAN GAZZARD: The data for GLUT-4 in a sense are very tantalizing, in that there is a mechanistic explanation for why protease inhibitors produce a problem.

GRAEME MOYLE: The indinavir study2 has been repeated for the different protease inhibitors. Although the effect of dyslipidaemia was seen with lopinavir/ritonavir, the impact on insulin resistance was much more modest. Even though the curves for inhibition of GLUT- 4 in vitro for indinavir, ritonavir, lopinavir and nelfinavir approximately follow each other, the impact at the clinical level may be different. The effect is more modest in studies of insulin- resistant patients with, for example, lopinavir/ ritonavir relative to indinavir, even though the dyslipidaemia tends to be greater with lopinavir/ritonavir than with indinavir in the same population. This suggests that insulin resistance is not the whole driver of the problem.

DUNCAN CHURCHILL: The effect of protease inhibitors on flow- mediated vascular dilatation has been studied in people with familial hypercholesterolaemia and children; children have impaired flow-mediated vascular dilatation, but they usually do not develop coronary heart disease for some decades.

GRAEME MOYLE: That may be a red herring in the story. The data on carotid plaque formation may be stronger, and some data support an increase in carotid plaque growth over 1 year in follow-up.2 One study strongly showed a faster growth rate of plaques,3 while another suggested that plaques were larger in people with HIV than age- and sex-matched controls.4

BRIAN GAZZARD: Excitement currently surrounds NRTI-sparing regimes. Two sets of data with indinavir do not support such regimens.5

GRAEME MOYLE: One of the problems with that switch study was that even though there was modest fat recovery in patients who went on to a protease inhibitor plus efavirenz and came off nucleosides, dyslipidaemia was bad in many patients and a number had to start lipid-lowering therapy.5 That, in part, nicely shows the disconnection between lipoatrophy and dyslipidaemia. Dyslipidaemia is one of the problems with sparing NRTIs, and obviously atazanavir is not a candidate in that setting until we have at least more clear interaction data.

BRIAN GAZZARD: Do you think efavirenz and indinavir are the problem, or do the NRTIs not matter as much as we might think?

GRAEME MOYLE: I believe that multiple mechanisms are going to be relevant in some people, so, at a clinical level, in someone taking stavudine and a protease inhibitor with significant dyslipidaemia, I would look to switch them from both drugs to tenofovir if that is an option, because the most important factor may be family history or genetics.

BRIAN GAZZARD: Is aiming for less lipoatrophy and less lipid abnormality with a new protease inhibitor combination likely to be successful from the sort of data we have already?

GRAEME MOYLE: It depends what you mean by failure - from a patient standpoint, avoiding lipoatrophy is more critical than worrying about an uncertain increase in the 10-year risk of cardiovascular disease.

BRIAN GAZZARD: I tend to agree with you, but that does point out something very important: lipoatrophy is unique and the other issues are peripheral and relate to changes in lifestyle and infection risks.

GUY BAILY: The data are very conflicting as they seem to imply that if you lost peripheral fat you would be left with a big stomach, but that is not what we see in the clinic - unlike buffalo humps, which are a real phenomenon.

GRAEME MOYLE: The superimposition of wasting is often probably the reason why the CD4 count was important in the HOPS study: once a person has become wasted, lipoatrophy becomes more rapidly evident because you only need a little fat loss in a specific area.6

DUNCAN CAMPBELL: I think it was a cohort effect: in 1997, many more people had central adiposity than now.

GRAEME MOYLE: I was surprised that the EMEA suggested that wasting in HIV had vanished, because it is very evident in the clinic that people are underweight and losing weight.

BRIAN GAZZARD: Is loss of all subcutaneous fat harmful? The people I see seem to have enough critical fat left - they have lost weight and they have normal lipids.

GRAEME MOYLE: I think it is probably minor in terms of a contribution to dyslipidaemia. Loss of fat is thought to contribute to about one-third of peripheral insulin sensitivity, but the rest is contributed by muscle. The problem with dyslipidaemia is that if muscle wasting occurs then insulin resistance will develop. Muscle is lost in wasting, but in terms of lipoatrophy, patients certainly have less muscle in their buttocks by clinical observation.

SIMON COLLFNS: How specific is visceral fat to HIV? Distended fat from the inside rather than more subcutaneous fat seems to be very HIV-specific. The FRAM data are convincing in terms of the importance of fat loss,7 but I think that when you look qualitatively at what is going on, the percentage of people who have pushed out distended fat from inside the abdomen will be low.

GRAEME MOYLE: Metabolic syndrome in the general population is accompanied by visceral adiposity.

BRIAN GAZZARD: One problem in all studies is that untreated HIV- positive people are relatively protected from the metabolic syndrome, but once they are treated, they develop metabolic syndrome.

GRAEME MOYLE: The FRAM data also quite nicely show that people with HIV have fat loss regardless of whether their doctor thinks they have lipodystrophy.7 This means that the doctor simply did not observe lipodystrophy in the control population - so they were not a control population but rather a group of people with an incompletely expressed syndrome that may or may not have developed. Some of the people who were controls at our site, 2 years on, now have clinical lipoatrophy.

SIMON COLLINS: Is there a way of introducing more sensitive monitoring for lipodystrophy to allow identification of people who have problems and will or will not progress.

BRIAN GAZZARD: If you are worried about lipodystrophy, you could change them to a regimen that is less likely to produce the problem.

MARGARET JOHNSON: If we could identify such people at an early stage, they would be ideal to take part in a trial. If we could identify them and randomize them to different regimens, we would know which regimens are likely to be more beneficial.

GRAEME MOYLE: The difficulty is that choice of drug often is driven more by resistance and effectiveness than by concern about lipoatrophy.

MARGARET JOHNSON: The DAD study 8 did not collect any information on recreational drug use, but when you look at a population of young patients coming into casualty with chest pain, a lot of this could be due to cocaine use. Amphetamines are another drug of concern; ecstasy is often mixed with amphetamines, so amphetamine consumption may be higher than cocaine consumption.

DUNCAN CHURCHILL: Mega trials are looking at folic acid supplementation and, if it is shown to be effective, then everyone will use it in secondary prevention. I am interested in homocysteine in HIV because it is very cheap, simple, given once a day and non- toxic.

DEVI NAIR: No endpoint data are available on homocysteine in the general population. It will be an interesting study looking at homocysteine levels in H\IV patients. I always note the megaloblastic picture and assume that patients have low levels of vitamin B12 and folate. Does metabolism of the drug interfere with vitamin B12 and folate levels?

GRAEME MOYLE: That effect is caused by delayed maturation of red blood cells in the bone marrow.

BRIAN GAZZARD: There is an effect on vitamin B12: vitamin B12 levels diminish as HIV disease progresses.

GRAEME MOYLE: We are uncertain whether it is because the vitamins are acute phase reactants or whether it is an effect on total body vitamin levels. It is also related to enteropathy, which corrects very rapidly after drug withdrawal.

DEVI NAIR: If vitamin B12 and folate are not abnormal then homocysteine will not help to manage these patients. DUNCAN CHURCHILL: Do physiological levels of folic acid lower your homocysteine concentrations?

DEVI NAIR: Whatever dose of folic acid is given (400 g/ 1 mg or 5 mg), homocysteine levels will only reduce by 25%. For general cardiovascular patients with high homocysteine levels, I aim for a level of folic acid of 10 ng/mL and of homocysteine of below 10 mol/ L. I have not looked at folic acid and vitamin B12 in HIV patients. I tend not to check homocysteine levels regularly, because the tests are very expensive and we do not have any evidence to support their use in risk prevention in this group.

References

1 Egger M, Junghans C, Friis-Moller N, Lundgren JD. Highly active antiretroviral therapy and coronary heart disease: the need for perspective. AIDS 2001;15:S193-201

2 Noor MA, Lo JC, Mulligan K, et al. Metabolic effects of indinavir in healthy HIV-seronegative men. AIDS 2001;15:F11-18

3 Maggi P, et al. Premature lesions of the carotid vessels in HIV- 1 infected patients treated with protease inhibitors. AIDS 2000;14:F123-8

4 Maggi P Lillo A, Perilli F, et al. PREVALEAT Group. Colour- Doppler ultrasonography of carotid vessels in patients treated with antiretroviral therapy: a comparative study. AIDS 2004;18:1023-8

5 Boyd MA, Aarnoutse RE, Ruxrungthan K, et al. Pharmacokinetics of indinavir/ritonavir (800/100 mg) in combination with efavirenz (600 mg) in HIV-1-infected subjects. J Acquir Immune Defic Syndr 2003;34:134-9

6 Lichtenstein KA, Delaney KM, Armon C, et al. Incidence of and risk factors for lipoatrophy (abnormal fat loss) in ambulatory HIV- 1-infected patients. J Acquir Immune Defic Syncir 2003;32:48-56

7 Tien PC, Grunfeld C. What is HIV-associated lipodystrophy? Defining fat distribution changes in HIV infection. Curr Opin Infect Dis 2004;17:27-32

8 Friis-Moller N, Sabin CA, Weber R, et al. Combination antiretroviral therapy and the risk of myocardial infarction. N Engl J Med 2003;349:1993-2003

Dr Guy Baily

Ambrose King Centre

Royal London Hospital

Whitechapel

London E1 1BB

UK

Mr Simon Collins

HIV i-Base, 3rd Floor East

Thrale House

London SE1 1UN

UK

Professor Brian Gazzard

St Stephen's Centre

Chelsea & Westminster Hospital

London SW10 9TH

UK

Dr Margaret Johnson

Department of Thoracic Medicine

Royal Free Hospital

London NW3 2QG

UK

Dr Graeme Moyle

St Stephen's Centre

Chelsea & Westminster Hospital

London SW10 9TH

UK

Dr Devi Nair

Department of Clinical Biochemistry

Royal Free Hospital

London NW3 2QG

UK

Copyright Royal Society of Medicine Press Ltd. Oct 2005

Source: International Journal of STD & AIDS

More News in this Category