HIV lipodystrophy is characterized by peripheral, subcutaneous lipoatrophy and a relative increase in visceral adipose tissue (VAT), and is commonly associated with dyslipidaemia, insulin resistance and, to a lesser extent, type 2 diabetes [1–3]. Lipodystrophy was first noted with protease inhibitor-based antiretroviral therapy (ART). Some protease inhibitors, including ritonavir-boosted lopinavir (lopinavir/r), can adversely affect the differentiation, function and mass of subcutaneous adipocytes in vitro, but mostly at supratherapeutic concentrations [4–6].
Thymidine nucleoside reverse transcriptase inhibitors (tNRTI) may play a greater role in the development of lipoatrophy than protease inhibitors. Switching a tNRTI to lopinavir/r increased peripheral fat mass . The tNRTI stavudine was a stronger risk factor for lipoatrophy than the protease inhibitor, indinavir . Lastly, randomized trials found variable effects of three protease inhibitors on lipoatrophy by measuring limb fat mass: relative to efavirenz, nelfinavir caused more lipoatrophy, unboosted atazanavir was similar, whereas a larger, longer study found less lipoatrophy with lopinavir/r [8–10].
The effect of protease inhibitors on VAT was studied once, in a trial in which patients were randomized to continue or switch their protease inhibitor therapy . The switch drugs, adefovir and hydroxyurea, caused significant anorexia, and so it is unclear whether the reduced VAT observed was due to protease inhibitor cessation or to anorexia-induced weight loss. VAT has not been evaluated in patients initiating ART.
Case reports, cross-sectional studies and short-term studies of healthy volunteers suggest that some protease inhibitors, including low-dose ritonavir, may acutely induce insulin resistance and type 2 diabetes, but it is unclear if these effects persist for more than the 2 to 4 weeks over which healthy volunteers were studied [12,13].
Adiponectin levels are low in adults with HIV lipodystrophy [14–17]. Paradoxically, adiponectin levels increased after 4 weeks of protease inhibitor exposure in healthy adults . Leptin levels increased over the first 24 weeks of tNRTI-containing therapy and greater increases were associated with a higher risk of subsequent lipoatrophy .
The effects on peripheral fat, VAT, insulin sensitivity and adipocytokines of initial ART including a ritonavir-boosted protease inhibitor but excluding a tNRTI are unstudied. We studied these effects with tipranavir/r or lopinavir/r, both with tenofovir + lamivudine, in antiretroviral-naive adults.
Trial 1182.33 was a randomized, open-label, three-arm trial of initial ART (NCT00144105). The trial evaluated three regimens including tipranavir (TPV 500 mg twice a day), pharmacokinetically boosted with ritonavir (either 200 mg twice a day or 100 mg twice a day), and lopinavir (LPV/r 400/100 mg twice a day). The TPV/r100 dose was chosen for potential tolerance advantages compared with higher doses. In the licensed TPV/r200 group, RTV was dosed at 100 mg twice a day for the first 2 weeks. LPV/r was the control as it is the most commonly used protease inhibitor in previously untreated adults. All groups received the dual-NRTI backbone of tenofovir (TDF; 300 mg daily) and lamivudine (3TC; 300 mg daily), which was chosen for its efficacy, tolerance, once-daily dosing, and low-capacity for lipoatrophy [20,21]. Randomization was stratified by screening CD4+ lymphocyte count (> or ≤200 cells/mm3).
The 1182.33 metabolic substudy was conducted in Spain (five centres), Argentina and Australia (four centres each), Brazil, Romania, and Russia (two centres each), and Canada (one centre). It examined the effects of TPV/r and LPV/r on body composition and metabolic abnormalities. To minimize recruitment bias, participating sites offered enrolment to all trial participants until the substudy was fully recruited.
The main protocol and metabolic substudy protocol were approved by the responsible Independent Ethics Committees and regulatory agencies. Written, informed consent was obtained from each participant.
Key eligibility criteria were: HIV-1 infection, age at least 18 years, 7 days or less of ART ever, CD4+ lymphocyte count less than 500 cells/mm3, HIV-1 viral load at least 5000 copies/ml plasma, and grade 1 or less hepatic transaminases. Individuals were ineligible if they had an active AIDS illness, were pregnant, planning pregnancy or breast-feeding, using investigational or immunomodulatory drugs or concomitant drugs that might significantly reduce plasma levels of study drug, or were thought likely to survive less than 12 months. There was no metabolic substudy-specific entry criterion.
Dual-energy x-ray absorptiometry (DEXA)
Total and regional body composition was measured by DEXA at baseline and week 48 with manual cuts used to define regional fat depots . Only one DEXA instrument was used to image each participant. Bio-Imaging (Leiden, The Netherlands) was the core laboratory for image collection, processing, quality control, analysis, data queries and archiving. All DEXA were reviewed by two technologists. Instrument quality control was performed at each scanning site to ensure consistent scan acquisition and analysis across imaging centres and participants. Instrument standardization and crosscalibration were accomplished before site inclusion and at regular intervals through the use of one Bona Fide Variable Composition Phantom (Bio-Imaging); this phantom was scanned in different configurations to mimic a range of percentage fat values, and was also used to assess the accuracy and precision of DEXA body composition measurements.
Computed tomography (CT) was performed at baseline and week 48 at the middle of the L4 vertebral body to measure VAT and subcutaneous adipose tissue (SAT; ref. ). Methods, equipment and imaging parameters (field of view, matrix, slice thickness and spacing/gap) were standardized. All scans were digitized and reviewed at Bio-Imaging for completeness and adherence to protocol.
CT data were interpreted by one radiologist not affiliated with the study and blinded to study treatments and patient data. All radiologist image assessments were captured electronically and read on three occasions.
Other measures of body composition
The following were recorded at baseline and week 48: waist and hip circumferences ; lipodystrophy prevalence and severity were objectively measured using the lipodystrophy case definition and derived severity grading scale ; and subjective gain and/or loss in body shape in seven regions were recorded independently by participants and physicians using a specific questionnaire .
Blood for metabolic assessments was collected every 12 weeks from patients who fasted for at least 12 h, were well hydrated and rested for at least 30 min. Glucose, high-density lipoprotein (HDL) cholesterol, total cholesterol, and triglycerides were measured every 12 weeks. Adiponectin, insulin and leptin were measured every 24 weeks. A 75 g oral glucose tolerance test (OGTT) for fasting and 2-h glucose and insulin levels was performed at baseline and week 48. All metabolic indices were measured by Covance Central Laboratory Services (Indianapolis, USA; Geneva, Switzerland; Sydney, Australia).
Concentrations of triglycerides, total cholesterol, HDL cholesterol and glucose were determined using standard enzymatic methods (Roche Diagnostics, Indianapolis, Indiana, USA). Plasma insulin was assayed using the Access Ultrasensitive Insulin method (Beckman Instruments, Fullerton, California, USA) with a detection limit of 0.03 μIU/ml. Adiponectin and leptin were analysed by quantitative sandwich enzyme immunoassay (R&D Systems, Minneapolis, Minnesota, USA) with a detection limit of 3.9 ng/ml (coefficient of variation (CV) <15%) for adiponectin and 7.8 pg/ml for leptin.
The Homeostasis Model Assessment (HOMA; ref. ) was used to estimate insulin resistance (HOMA-R) and β-cell insulin secretion (HOMA-B). Each index was determined as follows:
Equation (Uncited)Image Tools
The parent trial was intended to run for 3 years and the substudy for 96 weeks. The parent trial was prematurely terminated after an interim analysis of the week-48 primary endpoint data, where it was found that TPV/r100 was slightly less effective than LPV/r, whereas TPV/r200 had similar efficacy but was somewhat less well tolerated . As a consequence of premature trial termination, few patients in the substudy had 96-week data and so only the week-48 data were analysed.
The primary analysis was the comparison of the change in limb fat mass from baseline at week 48 between the LPV/r group and each TPV/r group using the Wilcoxon rank-sum test. Comparison was not made between the two TPV/r groups so as to minimize the number of comparisons. A Wilcoxon rank-sum test adjusting for CD4+ stratum (>200 versus ≤200 cells/mm3) was used as a sensitivity analysis. A Kruskal–Wallis test for limb fat mass at baseline among three treatment groups was performed to explore whether it was necessary to adjust for baseline limb fat mass in the primary analysis.
A SD of 0.8 kg and a difference of 0.7 kg for limb fat measured by DEXA were used to estimate effect size (nQuery Advisor 4.0), which was used to estimate the sample size for the Wilcoxon rank-sum test under the normal distribution assumption . Thirty participants per group would achieve 80% power for a two-sided Wilcoxon rank-sum test for continuous data with the significance level 0.025 chosen to adjust for multiplicity. Fifty participants per group were to be recruited to allow for dropout of up to 40%.
Changes in body composition and metabolic endpoint values from baseline at week 48 were also analysed using the Wilcoxon rank-sum test.
In several ad hoc analyses, factors associated with changes in limb fat mass, VAT and adiponectin levels were explored using general linear models with backward selection to identify potential predictors for change. Other than the treatment variable, which was always included in the final model, the initial model contained age, sex, and baseline body mass index (BMI), CD4+ lymphocyte count, HIV viral load, insulin, adiponectin, leptin, total cholesterol, triglyceride, VAT and limb fat. To test the significance of within-group changes to week 48 for limb fat, lipid and glycaemic parameters, adiponectin and leptin, a Wilcoxon signed-rank test was performed. All analyses were done with on-treatment values.
Of 179 patients screened for substudy participation, 163 were randomized, of which 140 (86%) had either DEXA or CT data at baseline and week 48, and 136 (83%) had data for both measurements (Fig. 1). Baseline characteristics of participants are well balanced among the three groups (Table 1). The LPV/r group had more patients with CD4+ more than 200 cells/ml (67%) compared with the TPV/r groups (TPV/r 100: 54%; TPV/r200: 52%).
No substudy participant changed protease inhibitor therapy or was diagnosed with diabetes through week 48. Four of 140 (3%) participants, two in each TPV/r group, received lipid-lowering therapy (gemfibrozil) when on-study.
Baseline median limb fat mass in the TPV/r groups was non-significantly higher than in the LPV/r group (Kruskal–Wallis test, P = 0.91). The median increase in limb fat mass at week 48 was higher with LPV/r (1.17 kg) than with TPV/r200 (0.83 kg) or TPV/r100 (0.41 kg), although the difference was not significant (Table 2). After adjusting for baseline CD4+ stratum, the median limb fat increase with LPV/r was marginally higher than with TPV/r100 (P = 0.03) or TPV/r200 (P = 0.05). The within-group change in limb fat mass from baseline to week 48 was significant for all groups: TPV/r200 (P = 0.0024), TPV/r100 (P = 0.012) and LPV/r (P < 0.0001).
Generally, the median of each subcutaneous fat variable and hip circumference increased in all three groups. Overall, higher increases were observed with LPV/r. There were significant differences between both TPV/r groups and the LPV/r group with respect to the increase from baseline for trunk fat mass and percentage, and between the TPV/r100 and LPV/r groups with respect to the increase from baseline for arm, leg, limb and total fat percentage, but not for hip circumference. The increase in SAT for LPV/r group was higher than for either TPV/r group though this was not statistically significant. The sensitivity analysis for these variables adjusted for baseline CD4+ stratum showed similar results (data not shown).
The two factors significantly associated with a greater increase in limb fat mass were a higher baseline HIV viral load (a 1.0 log10 higher baseline viral load was associated with 0.6 kg greater increase in limb fat; P = 0.02) and a lower baseline insulin 2 h after an OGTT (a 1 μU/ml lower insulin value was associated with 0.008 kg greater increase in limb fat; P = 0.037).
Visceral and truncal fat
VAT did not increase but rather declined modestly in all groups, despite the increases in limb fat mass and weight. The most prominent change was observed for the TPV/r200 group (P = 0.04 for comparison with LPV/r group).
Notably, the change in VAT correlated positively with change in limb fat mass (r = 0.25, P = 0.004). Baseline factors associated with a greater decline in VAT were higher VAT (each 10 cm2 higher VAT was associated with 1.7 cm2 greater decrease in VAT; P < 0.0001) and higher insulin levels 2 h after an OGTT (each 1 μU/ml higher insulin level was associated with 0.14 cm2 greater decrease in VAT; P = 0.014).
Waist circumference increased most with LPV/r, although none of the between-group differences was significant.
Overall body composition
The increases in BMI were comparable across groups. Overall body fat distribution did not appear to change significantly, with minimal changes in trunk:limb fat percentage, VAT:SAT, and waist:hip ratios. Nevertheless, there was a significant increase in waist:hip ratio in the LPV/r group relative to the TPV/r200 group. Also of note, the total fat and total fat percentage increase for LPV/r group was higher than for either TPV/r group.
When analysed using the Lipodystrophy Case Definition, there was a relatively small change in lipodystrophy severity score in all three groups. The prevalence of lipodystrophy increased by 11% with LPV/r compared with an 8% decrease with TPV/r200 (P = 0.006) and a 2% increase with TPV/r100 (P = 0.20).
Subjective changes in body composition
Physicians and participants identified only small changes in fat accumulation or fat wasting to week 48 in any group, with most finding no limb fat gain or loss (data not shown).
At baseline, approximately 90% of the physicians found normal waist size compared with approximately 74–87% of participants for the three treatment groups. On the basis of physician assessment, the proportion of participants with no gain in waist size decreased to 80% in the TPV/r100, 65% in the TPV/r200, and 60% in the LPV/r group on physical examination, and to 74, 65 and 52%, respectively, on participant report after 48 weeks.
Changes over 48 weeks in fasting glucose, insulin, HOMA-IR and HOMA-IS, and 2-h glucose and 2-h insulin values were modest and not significantly different between groups (Table 3). For all three treatments, the within-group changes at week 48 were not significant, except for a significant decline in fasting insulin in the TPV/r100 group (P = 0.012) and a significant (P = 0.003) increase in fasting glucose with LPV/r (although both insulin and HOMA-IR declined slightly with LPV/r).
Plasma adiponectin levels increased significantly (P < 0.0001) within all groups from baseline, but significantly more in both TPV/r groups compared with LPV/r (Table 3). Nevertheless, median values in all groups at week 48 remained within the reference range.
Factors associated with the change in adiponectin levels were sex, baseline triglyceride levels and treatment with TPV/r. The greater increase in adiponectin levels was associated significantly with TPV/r100 (P = 0.015) or TPV/r200 (P < 0.0001) treatment compared with LPV/r. Women had a 4030 ng/ml greater increase in adiponectin than men (P = 0.004). Each 10 mg/dl higher baseline triglyceride level was also associated with a 155 ng/ml greater decrease in adiponectin (P = 0.005).
Leptin concentrations increased with LPV/r and TPV/r200 but the increase in the LPV/r group was significant compared with the observed decrease in the TPV/r100 group (P = 0.015). A strong correlation between the changes in leptin and limb fat mass was observed (r = 0.67; P < 0.0001). The within-group change for leptin from baseline to week 48 was significant only for LPV/r (P = 0.001).
Total cholesterol and triglycerides increased in all groups, although the effects were greater in both TPV/r groups. The increase in HDL cholesterol was comparable in all groups. The ratio of total cholesterol:HDL cholesterol increased in both TPV groups (+0.3) but declined slightly (−0.1) in the LPV/r group. The within-group changes to week 48 were significant for all three parameters and all three treatments (P < 0.0001).
Over 48 weeks, TPV/r or LPV/r, each with TDF–3TC, increased limb fat mass, did not increase visceral abdominal fat, and was not associated with significant onset of insulin resistance or type 2 diabetes mellitus. A positive correlation was found between changes in limb fat and VAT, as would be expected in a wasted person who gains weight.
The median increases in limb fat were somewhat lower in both TPV/r groups, especially after adjustment for baseline CD4+ lymphocyte counts as patients on LPV/r had a higher CD4+ count at baseline. Even though the LPV/r group had lower limb fat at baseline, the week-48 values were very similar for all treatments. Therefore, it is possible that the greater increase in limb fat seen in the LPV/r group after 48 weeks may be due to greater HIV-related fat wasting prior to baseline in these patients. The findings for limb fat were similar to those of recent, unpublished studies that evaluated initial ART that included a ritonavir-boosted protease inhibitor but excluded a NRTI [26,27]. Nevertheless, our data should be interpreted cautiously, as limb fat was only measured at week 48. It is possible that limb fat may fall with more prolonged exposure, although treatment with LPV/r without a tNRTI for up to 24 months was also not associated with an overall loss of limb fat .
One possible reason for the subcutaneous fat outcome in the present study is the use of low-dose ritonavir. In a randomized trial in which ritonavir-boosted atazanavir (AZV/r) was compared with unboosted AZV over 96 weeks, there was less lipoatrophy with AZV/r . In keeping with these data, ritonavir can promote adipogenesis in vitro .
The outcomes for VAT were unexpected, given the longstanding perception that protease inhibitor therapy is associated with intra-abdominal fat accumulation. The data should be interpreted with care, however, as VAT was only measured once after baseline. Nevertheless, previous prospective studies found that central abdominal fat increased over 24 weeks and thereafter remained stable for up to 3 years in patients receiving therapy including a tNRTI, whereas we observed a decline in VAT .
There was no evidence that any regimen induced insulin resistance, including when assessed by OGTT, from the first postbaseline assessment at week 12 through week 48. This does not exclude the possibility that protease inhibitors might transiently induce insulin resistance or induce sustained insulin resistance in a minority of patients, but it does suggest that any acute effects seen with some protease inhibitors and with low-dose ritonavir are transient in most patients. The lack of insulin resistance seen in the present study may also stem from the lack of lipoatrophy through week 48, as both congenital and HIV-associated lipodystrophy are strongly associated with insulin resistance. Another likely reason is the avoidance of tNRTI therapy, which can rapidly induce insulin resistance before any measurable change in body fat [27,31].
As we did not observe limb fat loss, VAT increase or insulin resistance, the present data suggest that patients initiating TPV/r or LPV/r-based ART and who do not become lipodystrophic, will not develop insulin resistance. This possibility requires further evaluation.
Plasma adiponectin levels increased in all groups, particularly with TPV/r and in women. This finding is somewhat unexpected as adiponectin levels generally fall with weight gain . Recent studies have shown, however, that adiponectin levels are lower in HIV-infected patients than in healthy controls and that the levels decrease further with non-NRTI-based ART [16,33,34]. The increase we observed may reflect improvement in a perturbation induced by HIV infection; this seems unlikely, however, as the antiretroviral effects of the three regimens were very similar. Another possibility is that the increase was a compensatory response to antiretroviral-induced insulin resistance, in which case the increase appeared to be effective as insulin sensitivity did not appreciably change. Four weeks of LPV/r or indinavir in healthy volunteers increased plasma adiponectin and lipid levels, but insulin resistance was only observed with indinavir . The increases in adiponectin levels we observed may, therefore, be a compensatory response to the dyslipidaemic effects of both regimens rather than to insulin resistance. This may explain the significant association between baseline triglyceride levels and change in adiponectin levels at week 48. This possibility is supported by a study that showed adiponectin administered to mice receiving ritonavir significantly reduced lipid levels . Lastly, it is important to note that adiponectin levels tended to remain within the physiological range. The clinical implications of these changes, therefore, are uncertain.
Total cholesterol and triglyceride levels increased more in the TPV/r200 group than in the LPV/r group, however, the ratio of total to HDL cholesterol, a predictor of cardiovascular disease , does not suggest a clinically meaningful difference between these protease inhibitors.
Although there was a relatively small change in the lipodystrophy severity score in all three groups, patients in the LPV/r group had an increase in lipodystrophy prevalence compared with both TPV/r groups.
Our study has limitations. It only recruited adults and mostly white men. We only recorded body composition at two timepoints, and so a rise followed by a fall in limb fat mass that has been observed with tNRTI-based regimens remains a possibility, although treatment with lipoatrophic drugs such as stavudine and zidovudine for 1 year generally reduces limb fat mass to levels that are lower than at baseline, whereas in the present study limb fat was significantly higher. Lastly, we did not study adults with abnormal glucose tolerance, who might be less able to compensate for any drug-induced metabolic disturbance.
Our study suggests that TPV/r or LPV/r, coadministered with TDF-3TC, does not incur a significant risk of diabetes and do not significantly induce visceral fat accumulation, a complication that is both disfiguring and associated with cardiovascular disease. Lastly, the lack of lipoatrophy observed with these combinations will make these options attractive to patients, for whom lipoatrophy can be stigmatizing and can lead to poor ART adherence.
We would like to thank all sub-study investigators and staff, and especially the participants for their time.
Author contributions: The substudy was designed by Andrew Carr and Ricardo Chaves. The statistical analyses were performed by Armin Ritzhaupt and Wei Zhang. All authors contributed to the analysis plan and the manuscript.
Disclosure statement: Andrew Carr has received research funding from Abbott, Merck and Roche; consultancy fees from Bristol-Myers Squibb, Gilead Sciences, GlaxoSmithKline, Merck and Roche; lecture sponsorships from Abbott, Boehringer-Ingelheim, Bristol-Myers Squibb, Gilead Sciences, GlaxoSmithKline, and Merck; and has served on advisory boards for Abbott, Bristol-Myers Squibb, GlaxoSmithKline, Merck and Roche.
Armin Ritzhaupt, Wei Zhang, and Ricardo Chaves were all employees of Boehringer-Ingelheim, which wholly funded the study. Roberto Zajdenverg declares no conflict of interest.
Cassy Workman has served on advisory boards for Abbott, GlaxoSmithKline, Gilead, Janssen-Cilag, Merck and Roche; and has received speaker fees from Abbott, GlaxoSmithKline, Gilead, Janssen-Cilag, and Roche.
Pedro Cahn has served as advisor or speaker for: Avexa, Abbott, BMS, Boehringer-Ingelheim, GlaxoSmithKline, Merck Sharp & Dohme, Pfizer, Pharmasset, Schering Plough, and Tibotec. Jose M. Gatell has received research grants or honoraria for lectures or advisory boards from: Abbott, Boehringer-Ingelheim, Bristol-Myers Squibb, Gilead, GlaxoSmithKline, Janssen-Cilag, Merck, Roche, Tibotec, and Virco.
Data presented previously at the 9th International Workshop on Adverse Events and Lipodystrophy in HIV, Sydney, July, 2007 and published as abstract 7.
1. Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm DJ, Cooper DA. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998; 12:F51–F58.
2. Grinspoon SK, Carr A. Cardiovascular risk and body fat abnormalities in HIV-infected adults. N Engl J Med 2005; 352:48–62.
3. Mallon PWG, Miller J, Cooper DA, Carr A. Prospective evaluation of the effects of antiretroviral therapy on body composition in HIV-1-infected men starting therapy. AIDS 2003; 17:971–979.
4. Caron M, Auclair M, Vigouroux C, Glorian M, Forest C, Capeau J. The HIV protease inhibitor indinavir impairs sterol regulatory element-binding protein-1 intranuclear localization, inhibits preadipocyte differentiation, and induces insulin resistance. Diabetes 2001; 50:1378–1388.
5. Kim RJ, Wilson CG, Wabitsch M, Lazar MA, Steppan CM. HIV protease inhibitor-specific alterations in human adipocyte differentiation and metabolism. Obesity 2006; 14:994–1002.
6. Jones SP, Janneh O, Back DJ, Pirmohamed M. Altered adipokine response in murine 3T3-F442A adipocytes treated with protease inhibitors and nucleoside reverse transcriptase inhibitors. Antivir Ther 2005; 10:207–213.
7. Murphy R, Zhang J, Hafner R, et al. Peripheral and visceral fat changes following a treatment switch to a nonthymidine analogue or nucleoside-sparing regimen in patients with peripheral lipoatrophy: 48-week final results of ACTG5110, a prospective, randomized, multicenter clinical trial [abstract]. Program and abstracts of the 13th Conference on Retroviruses and Opportunistic Infections. Boston, USA; 2006; p. 755.
8. Dubé MP, Parker RA, Tebas P, Grinspoon SK, Zackin RA, Robbins GK, et al. Glucose metabolism, lipid, and body fat changes in antiretroviral-naive subjects randomized to nelfinavir or efavirenz plus dual nucleosides. AIDS 2005; 19:1807–1818.
9. Jemsek JG, Arathoon E, Arlotti M, Perez C, Sosa N, Pokrovskiy V, et al. Body fat and other metabolic effects of atazanavir and efavirenz, each administered in combination with zidovudine plus lamivudine, in antiretroviral-naive HIV-infected patients. Clin Infect Dis 2006; 42:273–280.
10. Haubrich R, Riddler S, DiRienzo G, et al. Metabolic outcomes of ACTG5142; A prospective, randomised, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for initial treatment of HIV-1 infection [abstract]. Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections. Los Angeles, California; 2007. p. 38.
11. Carr A, Hudson J, Chuah J, et al. HIV protease inhibitor substitution in patients with lipodystrophy: a randomised, multicentre, open-label study. AIDS 2001; 15:1811–1822.
12. Lee GA, Lo JC, Aweeka F, Schwarz JM, Mulligan K, Schambelan M, Grunfeld C. Single-dose lopinavir-ritonavir acutely inhibits insulin-mediated glucose disposal in healthy volunteers. Clin Infect Dis 2006; 43:658–660.
13. Noor MA, Flint OP, Maa JF, Parker RA. Effects of atazanavir/ritonavir and lopinavir/ritonavir on glucose uptake and insulin sensitivity: demonstrable differences in vitro and clinically. AIDS 2006; 20:1813–1821.
14. Kosmiski L, Kuritzkes D, Lichtenstein K, Eckel R. Adipocyte-derived hormone levels in HIV lipodystrophy. Antiviral Ther 2003; 8:9–15.
15. Petit JM, Duong M, Masson D, Buisson M, Duvillard L, Bour JB, et al. Serum adiponectin and metabolic parameters in HIV-I-infected patients after substitution of nevirapine for protease inhibitors. Eur J Clin Invest 2004; 34:569–575.
16. Lindegaard B, Keller P, Bruunsgaard H, Gerstoft J, Pedersen BK. Low plasma level of adiponectin is associated with stavudine treatment and lipodystrophy in HIV-infected patients. Clin Exp Immunol 2004; 135:273–279.
17. Samaras K, Wand H, Law M, Emery S, Cooper DA, Carr A. Prevalence of metabolic syndrome in HIV-infected patients receiving highly active antiretroviral therapy using International Diabetes Foundation and Adult Treatment Panel III criteria: associations with insulin resistance, disturbed body fat compartmentalization, elevated C-reactive protein and hypoadiponectinemia. Diabetes Care 2007; 30:113–119.
18. Lee GA, Mafong DD, Noor MA, et al. HIV protease inhibitors increase adiponectin levels in HIV-negative men. J Acquir Immune Defic Syndr 2004; 36:645–647.
19. Calmy A, Carey D, Mallon PWG, et al. Predictors of lipoatrophy and visceral fat accumulation in men initiating antiretroviral therapy. Antiviral Ther 2005.
20. Gallant JE, Staszewski S, Pozniak AL, DeJesus E, Suleiman JM, Miller MD, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA 2004; 292:191–201.
21. Pozniak AL, Gallant JE, Dejesus E, Arribas JR, Gazzard B, Campo RE, et al. Tenofovir disoproxil fumarate, emtricitabine, and efavirenz versus fixed-dose zidovudine/lamivudine and efavirenz in antiretroviral-naive patients: virologic, immunologic, and morphologic changes: a 96-week analysis. J Acquir Immune Defic Syndr 2006; 43:535–540.
22. Carr A, Law M, HIV Lipodystrophy Case Definition Study Group. An objective lipodystrophy severity grading scale using the lipodystrophy case definition score. J Acquir Immun Defic Syndr 2003; 33:571–576.
23. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and B-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28:412–419.
24. Cooper DA, Zajdenverg R, Ruxrungtham K, Chavez RL. Efficacy and safety of two doses of tipranavir/r versus lopinavir/r-based therapy in antiretroviral naïve patients: results of BI1182.33 trial [abstract]. In: Abstracts of the 8th International Congress on Drug Therapy in HIV Infection. November 12–16, 2006, Glasgow, Scotland; p. PL13.4.
25. Noether GE. Sample size determination for some common nonparametric tests. J Am Stat Association 1987; 82:645–647.
26. Wohl D, Lancaster T, DeJesus E, et al. Determination of body composition changes by total body DEXA after 48 weeks of treatment with once-daily fosamprenavir (FPV) boosted with two different doses of ritonavir(r) plus abacavir (ABC)/lamivudine(3TC): COL100758 [abstract TUPEB080]. In: Abstracts of the 4th IAS Conference on AIDS Pathogenesis. Treatment and Prevention, Sydney, Australia.
27. van Vonderen MGA, Blümer RME, Hassink E, et al. ZDV/3TC persistently contributes to peripheral insulin resistance by a body composition-independent mechanism demonstrated by repeated clamp studies during 2 years of first line ART with ZDV/3TC/LPV/r. Antiviral Ther 2007; 12 Suppl 2:L13.
28. Cameron DW, da Silva B, Arribas J, et al. Significant sparing of peripheral lipoatrophy by HIV treatment with LPV/r + ZDV/3TC induction followed by LPV/R monotherapy compared with EFV + ZDV/3TC [Abstract 44LB]. Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections. Los Angeles, California; 2007.
29. Malan N, JKrantz E, David N, Mathew M, Rightmire A, Witz V, et al. Efficacy and safety of atazanavir-based therapy in antiretroviral-naïve HIV-1-infected subjects, both with and without ritonavir: 96-week results from A1424-089 [abstract WEPEB024]. In: Abstracts of the 4th IAS Conference on HIV Pathogenesis, treatment and Prevention. Sydney, Australia; 2007.
30. Nguyen AT, Gagnon A, Angel JB, Sorisky A. Ritonavir increases the level of active ADD-1/SREBP-1 protein during adipogenesis. AIDS 2000; 14:2467–2473.
31. Fleischman A, Johnsen S, Systrom DM, Hrovat M, Farrar CT, Frontera W, et al. Effects of a nucleoside reverse transcriptase inhibitor, stavudine, on glucose disposal and mitochondrial function in muscle of healthy adults. Am J Physiol Endocrinol Metab 2007; 292:E1666–E1673. Epub 2007 Feb 6.
32. Modan-Moses D, Stein D, Pariente C, Yaroslavsky A, Ram A, Faigin M, et al. Modulation of adiponectin and leptin during refeeding of female anorexia nervosa patients. J Clin Endocrinol Metab 2007; 92:1843–1847. Epub 2007 Feb 27.
33. Das S, Shahmanesh M, Stolinski M, Shojaee-Moradie F, Jefferson W, Jackson NC, et al. In treatment-naïve and antiretroviral-treated subjects with HIV, reduced plasma adiponectin is associated with a reduced fractional clearance rate of VLDL, IDL and LDL apolipoprotein B-100. Diabetologia 2006; 49:538–542.
34. Mynarcik DC, Combs T, McNurlan MA, Scherer PE, Komaroff E, Gelato MC. Adiponectin and leptin levels in HIV-infected subjects with insulin resistance and body fat redistribution. J Acquir Immune Defic Syndr 2002; 31:514–520.
35. Xu A, Yin S, Wong L, Chan KW, Lam KS. Adiponectin ameliorates dyslipidemia induced by the human immunodeficiency virus protease inhibitor ritonavir in mice. Endocrinology 2004; 145:487–494.
36. DAD Study Group. Class of antiretroviral drug and the risk of myocardial infarction. N Engl J Med 2007; 356:1723–1735.
© 2008 Lippincott Williams & Wilkins, Inc.