Secondary Logo

Journal Logo


HIV protease inhibitor substitution in patients with lipodystrophy: a randomized, controlled, open-label, multicentre study

Carr, Andrewa; Hudson, Jeffb; Chuah, Johnc; Mallal, Simond; Law, Matthewe; Hoy, Jennifere; Doong, Nicholasf; French, Martynd; Smith, Done; Cooper, David A.* for the PIILR study group

Author Information
  • Free



Lipodystrophy (peripheral lipoatrophy and central fat accumulation), dyslipidaemia and insulin resistance often complicate protease inhibitor (PI)-containing highly active antiretroviral therapy (HAART) of HIV infection [1–4]. The three largest prevalence surveys found that lipodystrophy was common overall (49–53% of about 4000 patients), more common in PI recipients and almost exclusive to patients receiving antiretroviral therapy [5–7]. Hyperlipidaemia and insulin resistance were also more common with PI therapy, and in particular in PI recipients with lipodystrophy [1,4,6,8,9]. Most PI can stimulate lysis and/or impair differentiation of adipocytes [10,11]. Because adipocyte dysfunction, lipoatrophy and lipomata can occur with nucleoside analogues in the absence of PI, the contribution of PI therapy to lipodystrophy remains ill-defined [12–16].

There are several implications of lipodystrophy. Its physical features can be distressing to patients, and can lead to reduced adherence to HAART and so to virological failure. Physical changes could also stigmatize patients. The associated metabolic features may place patients at an increased risk of future cardiovascular disease [17,18].

There is no proven treatment for lipodystrophy. One possible strategy is PI cessation but substitution with equipotent therapy would be required for therapeutic benefits to be preserved [19]. HAART including a non-nucleoside reverse transcriptase inhibitor (NN RTI) rather than a PI appears equipotent in drug-naive adults, but non-PI HAART in patients who have virologically failed nucleoside analogue therapy has not been studied extensively [20]. A single-centre, non-randomized study found that PI substitution with nevirapine lead to some metabolic improvements but no objective body composition assessments were performed [21]. We hypothesized that PI substitution with non-PI therapy in aviraemic patients would lead to improvement of lipodystrophy and its associated metabolic features, with continued suppression of HIV replication. A randomized, open-label, multicentre, 24-week study was undertaken to address these hypotheses.


Study population

Study eligibility criteria were documented HIV infection, age > 18 years, plasma HIV RNA load < 400 copies/ml for at least the preceding 6 months, stable PI-containing therapy for at least 8 weeks, no prior abacavir, NNRTI or adefovir therapy, development of lipodystrophy in the presence of PI therapy, no active AIDS-defining condition in the preceding 3 months, use of standard of care opportunistic infection prophylaxis, negative pregnancy test in women of child-bearing age, no use of chemotherapy, radiotherapy or immune modulators, and no ongoing alcohol or substance abuse. All patients provided written, informed consent after approval by each site's Research Ethics Committee. To accelerate recruitment, the protocol was subsequently modified to permit enrolment of subjects who had commenced a NNRTI as part of the PI-containing regimen that lead to a sustained plasma HIV RNA load of < 400 copies/ml.


In the absence of a validated definition, lipodystrophy was defined by patient report (standardized questionnaire) of peripheral lipoatrophy (face, arms, buttocks or legs) with or without central fat accumulation (abdomen, dorso-cervical fat pad) confirmed by physical examination [1]. Total body fat < 20% on dual-energy X-ray absorptiometry (DEXA) was used as a guide, but not as an entry criterion. Virological failure was defined as HIV RNA > 400 copies/ml plasma on two occasions more than 14 days apart in the absence of intercurrent illness. Normal glucose tolerance, impaired glucose tolerance and diabetes mellitus were diagnosed according to the criteria of the American Diabetes Association [22].


The study was designed to randomize 80 eligible patients. All patients continued current nucleoside analogue therapy. Switch group patients ceased PI(s), and commenced open-label abacavir (300 mg twice daily), nevirapine (200 mg daily for 2 weeks, then 200 mg twice daily), adefovir (60 mg daily, with l-carnitine supplementation 500 mg daily) at baseline, and hydroxyurea (500 mg twice daily) from week 4. This aggressive regimen was chosen to maximize the likelihood of continued viral suppression in this heavily pretreated patient population, most of whom had failed at least one nucleoside analogue regimen prior to commencing potent therapy that lead to undetectable plasma HIV RNA. At that time, adefovir was an investigational agent without well demonstrated long term toxicity, and hydroxyurea had been shown to exert antiretroviral activity in vitro. Patients continuing PI therapy were offered the non-protease switch regimen at week 24.

Randomization was performed centrally at the National Centre in HIV Epidemiology and Clinical Research, Sydney, Australia after confirmation of eligibility, and stratified by current PI type. Sixty percent of patients were randomized to switch PI therapy and 40% to continue PI therapy. The unequal randomization was based on the possibility that 20% of patients might cease switch therapy and recommence PI therapy by week 24. Reductions of nevirapine and abacavir doses were not permitted; half doses of hydroxyurea and adefovir were allowed after resolution of drug-related grade 2 adverse events. The protocol mandated cessation of switch drugs for grade 3–4 adverse events, and permitted substitution with efavirenz for nevirapine toxicity. Grade 1–2 adverse events were managed according to the treating physician's assessment. Drug substitution of non-study antiretroviral drugs was mandatory for recurrent grade 3 or 4 drug-related adverse events and was optional for persistent drug-related grade 1 to 2 adverse events or for virological failure.


Details of all prior antiretroviral therapy were recorded. After a 2 week screening period, patients were seen at weeks 0, 2 and 4, then every 4 weeks to week 24, including patients who ceased therapy but who agreed to follow-up. From week 24, switch group patients were reviewed every 12 weeks, and continue group patients who switched therapy at that time were reviewed at weeks 26, 28, 32, 36 and 48. Weight, waist and hip circumferences, physical examination, safety bloods (complete blood count, electrolytes, liver enzymes, urea, creatinine, creatine kinase, phosphate, amylase) and adverse events were assessed at every visit. Quality of life (EuroQol [23]) was assessed at baseline and week 24. CD4 T-lymphocyte counts, fasting total and high density lipoprotein (HDL) cholesterol, triglyceride, glucose, insulin and C-peptide levels and insulin resistance were determined as previously described [1,4]. Oral glucose tolerance tests were performed every 12 weeks for determination of 2 h post-glucose loading values. Real-time plasma HIV RNA monitoring was performed at each visit according to site availability using either the Roche Amplicor Monitor assay version 1.0 (Roche Diagnostics, Branchburg, New Jersey, USA; lower limit of detection 400 copies/ml plasma) or Chiron b-DNA assay version 3 (Chiron Corporation, Emeryville, CA, USA; lower limit of detection 50 copies/ml). Batched HIV RNA analysis of frozen plasma was performed after study completion as specified in the original protocol using the Roche Amplicor Version 1.5 (Roche Diagnostics; lower limit of detection 50 copies/ml plasma).

Total and regional body composition was quantified at screening, and at weeks 12, 24 and 48 by DEXA (Lunar DPXL; Madison, Wisconsin, USA) at a central reading site [1,4]. Intra-abdominal and extra-abdominal (i.e., subcutaneous) fat areas at the L4 vertebral level were measured at each study site by single-cut computerized tomography [2]. Quality assurance programs for both techniques were instituted prior to study commencement. Lipodystrophy presence and severity in the periphery (face, arms, legs and buttocks), centrally (abdomen and dorso-cervical spine) and overall were assessed by study subjects every 12 weeks using a standardized questionnaire, the results of which have been found to correlate with DEXA [4]. In each region, a score of 0 for nil, 1 for mild, 2 for moderate, or 3 for severe was assigned, for a maximum possible score of 12 peripherally, 6 centrally and 18 overall.

To assess if PI withdrawal might lead to normalization of metabolic parameters, week 24 data were compared with fasting metabolic data derived from the 103 men without lipodystrophy who had never received antiretroviral therapy and who had participated in the Australian Lipodystrophy Syndrome Prevalence Survey [6].


The primary study endpoints were mean change in total body fat and the proportion of patients with HIV RNA < 50 copies/ml plasma at week 24. Total body fat was chosen as a primary endpoint rather than changes in any regional fat parameter, as several studies have found that HIV lipodystrophy is associated with lower total body fat overall, and as such is not merely a redistribution of fat [1,4,13–15]. Secondary endpoints were changes in total body fat and HIV RNA at week 48, and drug-specific adverse events, changes in regional body composition, biochemical, lipid and glycaemic parameters, antiretroviral therapy, CD4 T-lymphocyte counts and quality of life at weeks 24 and 48.

The sample size of 80 subjects was chosen to have an 80% power to detect a 2 kg difference in mean total body fat between the two groups assuming that up to 20% of the switch group would cease the switch regimen and resume PI therapy. All analyses were by intention-to-treat and included all follow-up data on all patients who started their randomized treatment regardless of any subsequent treatment changes. In summarizing changes from baseline by nominal study weeks, a last-value-carried-forward approach was adopted for patients who were lost to the remainder of the follow-up period. For analysis of the percentage of patients with undetectable HIV RNA, the one patient lost to follow-up was assigned as a virological failure [24]. Comparisons between randomized treatment groups were made using logistic regression for binary endpoints, and using analysis of variance methods for continuous endpoints. Change from baseline were analysed by the Wilcoxon signed rank test.

Two secondary efficacy analyses were performed: first, by intention-to-treat but not carrying last values forward; and second, adjusting comparisons between randomized treatments for baseline values. These analyses gave qualitatively identical results and are not presented, except for triglycerides, for which the adjusted analysis is presented.



Eighty-one patients were screened and randomized between May and December 1998 (Fig. 1). One patient randomized to the continue group was lost to follow-up after week 0 and was excluded from all analyses. Three patients who were receiving a NNRTI but were otherwise eligible were randomized to the switch group. Baseline patient characteristics of the remaining 80 patients were generally similar between the groups although baseline triglycerides were lower in patients randomized to switch therapy (Table 1).

Fig. 1.
Fig. 1.:
Trial profile. No patient was lost to follow-up after week 2.
Table 1
Table 1:
Baseline patient characteristics.

Treatment outcomes, safety and quality of life

Overall, 31 (63%) switch group patients completed 24 weeks of randomized therapy (Table 2). Between weeks 24 and 48, however, 40 (80%) and 11 (22%) switch group patients ceased adefovir or hydroxyurea, respectively. Six (12%) patients ceased all switch drugs for virological failure (n = 3), nausea (n = 2), and grade 3 liver enzymes (n = 1). Two (4%) switch group patients ceased nevirapine and commenced efavirenz at week 12. Discontinuation of one or more switch drugs did not affect any study endpoint (data not shown). Only five (10%) switch group patients recommenced PI therapy (two at week 12, and one each at weeks 1, 24 and 40).

Table 2
Table 2:
Treatment outcomes and main adverse events in switch group patients.

Nine (18%) switch group patients and five (16%) continue group patients ceased background nucleoside therapy by week 24 (stavudine, five switch group and three continue group patients; lamivudine, three patients in each group; zidovudine, two continue group patients; and didanosine, one switch group patient). No continue group patient ceased PI therapy before week 24; nine (29%) patients in this group did not switch PI therapy at week 24. No patient in either group received lipid-lowering agents or commenced diabetic therapy during the study.

The most common adverse events associated with the switch regimen were proximal renal tubular dysfunction (hypophosphataemia, proteinuria, fatigue, weight loss and anorexia with adefovir), cytopenias and gastrointestinal intolerance (hydroxyurea) and hypersensitivity (abacavir and nevirapine). The next most common metabolic toxicity was a significant, but usually asymptomatic, grade 1 or 2 increase in liver enzymes (data not shown). Two switch group patients were hospitalized for bacterial pneumonia (both at week 25), and one switch group patient developed Hodgkin's disease at week 20. There was no other new AIDS-defining (Centers for Disease Control category C) illness in either group. One continue group patient suffered a non-fatal myocardial infarction at week 12.

Quality of life at baseline was similar between groups (Table 3). Quality of life was significantly greater in the switch group than in the continue group at week 24, with a significant relative decline in the continue group.

Table 3
Table 3:
Quality of life and patient-assessed lipodystrophy severity (all values are means).

Virological and immunologic responses

The proportions of patients with plasma HIV RNA < 50 copies/ml plasma at 24 weeks in the switch and continue groups were 98% and 83%, respectively (P = 0.061) and did not differ significantly over 48 weeks (Fig. 2a). Virological failure was not influenced by any baseline parameter. CD4 T-lymphocyte counts (Fig. 2b) and CD8 T-lymphocyte counts (data not shown) both declined significantly in the switch group as compared with the continue group after the addition of hydroxyurea.

Fig. 2.
Fig. 2.:
Plasma HIV RNA and CD4 T-lymphocyte counts. Percentages of patients with plasma HIV RNA < 50 copies/ml (a) and median changes from baseline in CD4 T-lymphocyte counts (b) in groups switching (solid line) or continuing (dotted line) PI therapy. Most continue group patients switched from PI therapy at week 24. P values are for comparisons at weeks 24 and 48.

Body composition


Mean changes in the following parameters were significantly greater in the switch group versus the continue group at week 24: total body fat (−1.6 and −0.4 kg, respectively;P = 0.006); weight (−2.8 and −0.5 kg, respectively;P < 0.001); and total lean mass (−1.1 and +0.3 kg, respectively;P < 0.001) (Fig. 3). Despite these greater objective declines in the switch group, overall patient-assessed lipodystrophy severity was perceived as improving relative to the continue group (Table 3). The decline in total body fat mass continued after week 24, despite recovery of lean mass and weight.

Fig. 3.
Fig. 3.:
Overall body composition. Mean changes from baseline in total fat mass (a) and total lean mass (b) in groups switching (solid line) or continuing (dotted line) PI therapy. Most continue group patients switched from PI therapy at week 24. P values are for comparisons at weeks 24 and 48.

Peripheral lipoatrophy

Significant greater reductions in limb fat mass and subcutaneous abdominal fat area were observed in the switch group than in the continue group (Fig. 4). This was not influenced by baseline peripheral fat mass or severity rating (data not shown). Despite these declines, lipoatrophy severity was perceived as improving significantly in the switch group relative to the continue group. These declines in peripheral fat persisted after week 24, despite recovery of lean mass and weight.

Fig. 4.
Fig. 4.:
Peripheral lipoatrophy. Mean changes from baseline in limb fat mass (a) and subcutaneous abdominal fat (SAT) area (b) in groups switching (solid line) or continuing (dotted line) PI therapy. Most continue group patients switched from PI therapy at week 24. P values are for comparisons at weeks 24 and 48.

Central fat accumulation

There was a greater decline in intra-abdominal fat area in the switch group as compared to the continue group at week 24, although this was not significant (11 cm2 and 3 cm2, respectively;P = 0.267). A significantly greater decline in intra-abdominal fat area occurred, however, in the switch group patients reporting moderate or severe abdominal fat accumulation at baseline (Fig. 5a). Intra-abdominal fat area did not alter significantly in the 40 patients reporting no or mild abdominal fat accumulation at baseline (Fig. 5b). The declines in intra-abdominal fat persisted at week 48, despite recovery of lean mass and weight. There were also significant relative declines in the switch group in waist (but not hip) circumference (Fig. 5c), truncal and central abdominal fat mass on DEXA (data not shown), and central lipodystrophy severity score.

Fig. 5.
Fig. 5.:
Central fat accumulation. Mean changes from baseline in intra-abdominal fat (VAT) area in patients reporting moderate or severe central obesity (a, n = 40) or no or mild central obesity (b, n = 40), and in waist circumference (c) in groups switching (solid line) or continuing (dotted line) PI therapy. Most continue group patients switched from PI therapy at week 24. P values are for comparisons at weeks 24 and 48.

Eight (10%) patients reported a dorso-cervical fat pad at baseline. There was no subjective change in the size of any dorso-cervical fat pad in either group.

Metabolic parameters


There were significant declines in fasting total cholesterol and triglyceride levels after 2 weeks of PI cessation, although total cholesterol appeared to take longer to decline maximally (Fig. 6). The mean differences between the groups at week 24 for total cholesterol and triglyceride were 1.24 mmol/l (P = 0.001) and 0.33 mmol/l (P = 0.002), respectively. These declines were independent of the baseline PI type or lipid levels (data not shown). Of note, both total cholesterol and triglyceride increased after week 24 in the switch group but remained significantly less at week 48 than at baseline (both groups, P = 0.004).

Fig. 6.
Fig. 6.:
Metabolic parameters. Median changes from baseline in total cholesterol (a), HDL cholesterol (b), and triglyceride (c) in groups switching (solid line) or continuing (dotted line) PI therapy. Most continue group patients switched from PI therapy at week 24. P values are for comparisons at weeks 24 and 48.

HDL cholesterol increased significantly in both groups by week 48 (switch group, 0.14 mmol/l, P = 0.002; continue group, 0.07 mmol/l, P = 0.023), but did not differ significantly between the groups at any time. Lipid levels at week 24 in the switch group were similar to those in the controls that had never received antiretroviral therapy (Table 4).

Table 4
Table 4:
Metabolic abnormalities in study subjects and controls.

Glycaemic parameters

Fasting glucose, insulin and C-peptide, estimated insulin resistance, and glucose values 2 h post-oral glucose loading did not change significantly in either group (data not shown), and appeared greater at week 24 than in men who had never received antiretroviral therapy.

Three switch group patients (who did not suffer significant nausea from their new regimen and so probably had no major dietary change) ceased (n = 2) or reduced (n = 1) insulin therapy of diabetes mellitus that had been diagnosed after the commencement of PI therapy. However, the proportions of patients with either impaired glucose tolerance or diabetes mellitus did not alter significantly in either group.


In predominantly lipoatrophic men with extensive prior antiretroviral therapy, PI substitution with alternative antiretroviral therapy resulted in improvements in intra-abdominal adiposity, cholesterol and triglycerides, maintained HIV suppression, but worsening peripheral lipoatrophy and lean body mass. Although diabetes improved in a few patients, other data suggested that insulin resistance persisted.

Body composition

There are several possible explanations for why peripheral lipoatrophy did not improve after PI cessation. First, lipoatrophy may have been mostly an effect of the type and duration of nucleoside analogue therapy at baseline, and may have worsened because of ongoing therapy as well as from additional toxicity of the new nucleoside analogues [12–16]. Plasma lactate and liver enzymes may be markers of nucleoside analogue lipoatrophy [14]. The present data are conflicting in this regard as, although lactate was not measured in this study, anion gap did not change during the study in either group (data not shown) but liver enzymes did increase with the switch regimen. Second, any lipoatrophy due to PI may be irreversible or take more than 12 months to reverse. Third, some lipoatrophy may have been secondary to nausea and vomiting induced by the switch regimen (mostly adefovir and hydroxyurea) although fat loss continued after weight and lean mass recovered in the switch group between weeks 24 and 48.

Intra-abdominal fat declined only in patients reporting moderate to severe central fat accumulation at baseline, but remained stable in those reporting no or mild central fat accumulation (it is unlikely that the post hoc visceral abdominal fat subanalysis led to a false positive result, as the results were highly significant after the use of tests for interaction [25]). This suggests that the decline in abdominal fat was due to PI withdrawal rather than initiation of the switch regimen. In turn, this result suggests that the switch drugs may be specifically toxic to peripheral adipocytes, as does the finding that lean mass increased after week 24 but peripheral fat mass continued to decline.

The dissociated peripheral and central fat responses support the hypothesis that lipodystrophy may represent more than one syndrome, but also implies that lipodystrophy is not a fat ‘redistribution’ phenomenon (in which case total body fat should stay constant and any reduction in central fat should be matched by an increase in peripheral fat). It has been suggested that lipodystrophy may be a feature of effective suppression of HIV replication, but the improvement of intra-abdominal fat accumulation in tandem with maintained HIV suppression suggests, at the minimum, that intra-abdominal fat accumulation is not due to effective HIV suppression.

Virological activity

The switch regimen was designed for maximal antiviral potency, given that patients had extensive prior nucleoside analogue therapy. It was suspected that baseline nucleoside analogue resistance would be common and predispose to virological failure of the switch regimen. It was estimated that 30% of subjects were resistant to lamivudine at baseline, having commenced lamivudine with no other nucleoside analogue at least 8 weeks prior to commencement of the HAART regimen that lead to undetectable plasma HIV RNA (data not shown [19]). As 96% and 59% of the switch group ceased adefovir and hydroxyurea by week 48, respectively, with minimal virological failure, switching to only abacavir and nevirapine may be as effective virologically but less toxic in a similar patient population.

Adverse events

Adverse events were common, but no toxicity unique to the combination of drugs in the switch regimen was observed. Hydroxyurea led to significant falls in CD4 and CD8 T-lymphocyte counts. Of uncertain associated significance, two patients were hospitalized for bacterial pneumonia and one patient developed Hodgkin's disease at week 20, although the latter's natural history suggests that it preceded the institution of hydroxyurea. Larger studies are required to show whether hydroxyurea is associated with increased rates of either infection or malignancy. The known renal and toxicity of adefovir did not appear to be ameliorated by a 60 mg/day dose [26].

Despite the high frequency of adverse effects associated with the switch regimen, both patients and physicians reported improvements in quality of life with switch therapy. Although reasons for this were not recorded, possibilities include the reduced frequency of dosing and independence of meals and fluid intake (as compared with indinavir), reduced pill burden (as compared with nelfinavir), and the resolution of mild, chronic PI side-effects such as nausea, indigestion, diarrhoea, flushing, peri-oral paraesthesiae, dry skin and lips, and nephrolithiasis [27].

Metabolic outcomes

Improvements in total cholesterol and triglycerides were rapid. From 12 weeks after PI cessation, levels were similar to those in PI-naive subjects. The increase in mean total cholesterol and triglycerides after week 24 in the switch group may have been a consequence of weight gain (and presumably increased food intake) following cessation of adefovir and hydroxyurea. It is less likely that this was caused by reinstitution of PI therapy, as this occurred in only 10% of switch group patients. HDL cholesterol increased significantly in both groups, suggesting that low HDL cholesterol is due to HIV replication or immune dysfunction rather than to PI therapy, and that this increase may be a result of ongoing suppression of HIV replication [28].

Insulin resistance can decrease in patients without lipodystrophy following PI cessation [29], and in centrally obese adults receiving metformin, that also lead to a decrease in intra-abdominal fat [30]. Insulin resistance persisted in the present study, but patients with non-HIV lipodystrophy are often insulin-resistant, and mice with congenital total lipodystrophy have insulin resistance that is reversed by leptin, a product of peripheral adipocytes [31–33]. Therefore, full reversal of insulin resistance may require normalization of peripheral fat mass. Insulin resistance, however, correlated significantly with baseline intra-abdominal fat area (r = 0.35;P = 0.005) but not baseline limb fat mass (r = 0.13;P = 0.319).

Other considerations

There are limitations of our study. The toxicity of the switch regimen may have affected the study's generalizability. After the withdrawal in most patients of adefovir and hydroxyurea between weeks 24 and 48, fat loss continued despite recovery of lean mass (Fig. 3). This suggests that switching to only abacavir and nevirapine may have lead to fat loss without lean tissue loss. Very few women and no children were studied, and almost all subjects were white. The study was not powered to evaluate any effect of cessation of a particular PI, nor of PI cessation on dorso-cervical fat pad or insulin sensitivity. The lipodystrophy severity questionnaire, although robust in two cross-sectional prevalence surveys [4,6], was clearly open to patient bias in the setting of an open-label, longitudinal study, and therefore will be most useful in blinded studies. Future open-label studies of lipodystrophy reversal should also include physician assessment of severity, ideally by clinicians blinded to patient therapy.


Lipodystrophy appears to have a multifactorial aetiology. PI therapy appears to be most associated with hyperlipidaemia and intra-abdominal fat accumulation and may be non-dominant in peripheral lipoatrophy. PI cessation may reduce somewhat the atherogenic risk of hyperlipidaemia in this patient population, but the persistent insulin resistance may continue to place patients at risk of cardiovascular disease. The effect of nucleoside analogue cessation on lipodystrophy is under investigation.


The authors acknowledge J. Miller for providing control data from the Australian Lipodystrophy Syndrome Prevalence Survey, and the time and commitment of the patients who participated in the study.


1. Carr A, Samaras K, Burton S. et al. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998, 12: F51–F58.
2. Miller KD, Jones E, Yanovski JA, Shankar R, Feuerstein I, Falloon J. Visceral abdominal-fat accumulation associated with use of indinavir. Lancet 1998, 351: 871–875.
3. Lo JC, Mulligan K, Tai VW, Algren H, Schambelan M. `Buffalo hump’ in men with HIV-1 infection. Lancet 1998, 351: 867–870.
4. Carr A, Samaras K, Thorisdottir A, Kaufmann G, Chisholm DJ, Cooper DA. Diagnosis, prediction and natural course of HIV protease inhibitor-associated lipodystrophy, hyperlipidaemia and diabetes mellitus. Lancet 1999, 353: 2893–2899.
5. Ward DJ, Delaney KM, Moorman AC et al.Description of lipodystrophy in HIV Outpatients Study (HOPS).First International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV. San Diego, June 1999 [abstract 14].
6. Miller JE, Emery S, French M, Baker D, Cooper DA. The Australian prevalence survey of lipodystrophy syndrome.Seventh Conference on Retroviruses and Opportunistic Infections. San Francisco, January 2000 [abstract 201].
7. Boubaker K, Sudre P, Flepp M et al.Hyperlactatemia and antiretroviral therapy in the Swiss HIV Cohort Study.Seventh Conference on Retroviruses and Opportunistic Infections. San Francisco, January 2000 [abstract 57].
8. Walli R, Herfort O, Michl GM. et al. Treatment with protease inhibitors associated with peripheral insulin resistance and impaired glucose tolerance in HIV-1-infected patients. AIDS 1998, 12: F167–F174.
9. Yarasheski KE, Tebas P, Sigmund C. et al. Insulin resistance in HIV protease inhibitor-associated diabetes. J Acquir Immune Defic Syndr 1999, 21: 209–216.
10. Zhang B, MacNaul K, Szalkowski D, Li Z, Berger J, Moller DE. Inhibition of adipocyte differentiation by HIV protease inhibitors. J Clin Endocrinol Metab 1999, 84: 4274–4277.
11. Lenhard J, Weiel JE, Paulik MA, Furfine ES. Stimulation of vitamin A1 acid signaling by the HIV protease inhibitor indinavir. Biochem Pharmacol 2000, 59: 1063–1068.
12. Gervasoni C, Ridolfo AL, Trifirò G. et al. Redistribution of body fat in HIV-infected women undergoing combined antiretroviral therapy. AIDS 1999, 13: 465–472.
13. Saint-Marc T, Partisani M, Poizot-Martin I. et al. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS 1999, 13: 1659–1667.
14. Carr A, Miller J, Law M, Cooper DA. A syndrome of lipoatrophy, lactic acidemia and liver dysfunction associated with HIV nucleoside analogue therapy: contribution to protease inhibitor-related lipodystrophy syndrome. AIDS 2000, 14: F25–F32.
15. Mallal SA, John M, Moore CB. et al. Contribution of nucleosides analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS 2000, 14: 1309–1316.
16. Parker RA, Meyers DS, Andrews Flint OP, Durham SK. Effects of nucleoside reverse transcriptase inhibitors on adipogenesis and adipocyte metabolism.Second International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV. Toronto, September 2000 [abstract O4].
17. Henry K, Melroe H, Huebsch J. et al. Severe premature coronary artery disease with protease inhibitors. Lancet 1998, 351: 1328.1328.
18. Behrens G, Schimdt H, Meyer D. et al. Vascular complications associated with HIV protease inhibitors. Lancet 1998, 351: 1958.1958.
19. Carpenter CC, Cooper DA, Fischl MA. et al. Antiretroviral therapy in adults: updated recommendations of the International AIDS Society-USA Panel. JAMA 2000, 283: 381–390.
20. Albrecht M, Katzenstein D, Bosch R, Liou S, Hammer S. ACTG364 – nelfinavir and/or efavirenz in combination with new NRTIs in nucleoside experienced subjects: week 48 ultrasensitive HIV RNA results.Seventh Conference on Retroviruses and Opportunistic Infections. San Francisco, January 2000 [abstract 531].
21. Martinez E, Conget I, Lozano L, Casamitjana R, Gatell JM. Reversion of metabolic abnormalities after switching from HIV-1 protease inhibitors to nevirapine. AIDS 1999, 13: 805–810.
22. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 1998, 21: S5–S19.
23. Kind P, Dolan P, Gudex C. et al. Variations in population health status: results from a United Kingdom national questionnaire survey. BMJ 1998, 316: 736–741.
24. The AVANTI Steering Committee. Analysis of HIV-1 clinical trials: statistical magic? Lancet 1999, 353: 2061–2064.
25. Assmann SF, Pocock SJ, Enos LE, Kasten LE. Subgroup analysis and other (mis)uses of baseline data in clinical trials. Lancet 2000, 355: 1064–1069.
26. Kahn J, Lagakos S, Wulfsohn M. et al. Efficacy and safety of adefovir dipivoxil with antiretroviral therapy: a randomized, controlled trial. JAMA 1999, 282: 2305–2312.
27. Carr A, Cooper DA. Adverse effects of antiretroviral therapy. Lancet 2000, 356: 1423–1430.
28. Grunfeld C, Pang M, Doerrler W, Shigenaga JK, Jensen P, Feingold KR. Lipids, lipoproteins, triglyceride clearance, and cytokines in human immunodeficiency virus infection and the acquired immunodeficiency syndrome. J Clin Endocrinol Metab 1992, 74: 1045–1052.
29. Rozenbaum W, Molina JM, Delfraissy JF, Bentata M, de Truchis P, Antoun Z. Improvement of lipodystrophy in HIV-1-infected subjects switching from 2 NRTI/PI to 2 NRTI/abacavir.Seventh Conference on Retroviruses and Opportunistic Infections. San Francisco, January 2000 [abstract 46].
30. Hadigan C, Corcoran C, Basgoz N. et al. Metformin in the treatment of HIV lipodystrophy syndrome: a randomized, controlled trial. JAMA 2000, 284: 472–477.
31. Shimomura I, Hammer RE, Ikemoto S, Brown MS, Goldstein JL. Leptin reverses insulin resistance and diabetes mellitus in mice with congenital lipodystrophy. Nature 1999, 401: 73–76.
32. Reitman ML, Mason MM, Moitra J. et al. Transgenic mice lacking white fat: models for understanding human lipoatrophic diabetes. Ann N Y Acad Sci 1999, 892: 289–296.
33. Seip M, Trygstad O. Generalized lipodystrophy, congenital and acquired. Acta Paediatr Suppl 1996, 413: 2–28.


Other investigators of the Protease Inhibitor Induced Lipodystrophy Reversal (PIILR) study were: P. Cunningham, H. Wood (HIV, Immunology and Infectious Disease Clinical Services Unit and Centre for Immunology, St Vincent's Hospital, Sydney); S. Jacobs, R. James, F. Clark (Gold Coast Sexual Health Clinic, Gold Coast District Hospital, Miami); N. Scull, J. Leung (Department of Clinical Immunology, Royal Perth Hospital); M. Bryant, C. McCormack, J. Roney, M. Plummer (Alfred Hospital, Melbourne); D. Baker (407 Doctors, Sydney); R. Finlayson (Taylor Square Private Clinic, Sydney); D. Austin, M. Bloch (Holdsworth House General Practice, Sydney); J. Anderson (Carlton Clinic, Melbourne), N. Roth (Prahran Market Clinic, Melbourne); and S. Emery, J. Groves, W. Lee, R. Munro, T. Sharkey (National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney).


HIV; lipodystrophy; hyperlipidaemia; insulin resistance; protease inhibitors

© 2001 Lippincott Williams & Wilkins, Inc.