Interventions for the management of lipodystrophy and metabolic complications

Simon Collins, HIV i-Base

Limited benefit was reported in results from several lipoatrophy trials using rosiglitazone, and pioglitazone. Both of these drugs are insulin sensitising drugs used for people with diabetes and have a good theoretical basis for why they should help reverse fat loss. Rosiglitazone has been studied more frequently, but reported mixed results. Although the largest and most publicised of these studies showed no benefit [1], other researchers proposed that continuing to use a thymidine analogue while using rosiglitazone, would have cancelled out the benefit. [2, 3] Results of two studies this week supported the negative affect of continuing d4T or AZT.

Kathleen Mulligan reported results from an American study looking at rosiglitazone and metformin, both of which are insulin sensitising drug – used in management of diabetes. In general, the changes linked to lipodystrophy – both fat loss and fat gain – are linked to the way that your body processes both fat and sugar.

Rosiglitazone increases limb fat; metformin has no beneficial effect on visceral fat

ACTG 5082 randomised 115 patients with hyperinsulinaemia and elevated waist/hip ratio, to rosiglitazone (4 mg once daily) and metformin (500 mg twice daily, increasing to 1000 mg twice daily) together and separately vs placebo.

Patients were 66% male, 67% Caucasian, median age 45, CD4 count 567 cells/mm3, and waist-to-hip ratio 1.01 (males) or 0.99 (females) with similar demographic and body composition characteristics were across groups at baseline.

Table 1: Main safety and efficacy differences

Active drug Met Rosi Met/Rosi None P value
[Placebo Rosy Met no pcb both]
N 26 27 25 27
D/C 12 3 4 8
Diarrhoea 65% 8% 52% 15% p<0.001
Δ VAT (%) -0.6% 0% -8% -7.3% NS
Δ SAT (%) -3.2% +3.2% 0% -0.8% NS
Δ Limb fat (%) -3.8% +4.8% +0.5% -8.3% p=0.034*

* Rosi vs PCB only

Both metformin and rosiglitazone significantly reduced insulin AUC compared to placebo (p=0.03). There were no significant differences between any group for changes in abdominal visceral adipose tissue (VAT) or sub-cutaneous fat (SAT). DEXA scans showed that limb fat increased significant compared to placebo (+4.8% vs -8.3%), thought the increase was not noticeable to patients. Weight decreased significantly in both metformin groups compared to placebo (p=0.03 and P=0.05 for metformin/placebo and metformin/rosiglitazone respectively).

Rosiglitazone also significant affected lipids, increasing LDL and reducing HDL compared to placebo. The people using 4mg rosiglitazone once a day, reported an increase in leg fat (approx +5% increase). Metformin didn’t have a beneficial effect in the study.

A lack of beneficial effect using metformin was also supported by a study from Rakhi Kohli and colleagues from Tufts University, Boston. They randomised 48 patients with lipodystrophy, defined by self-reported increase in abdominal girth and waist-hip ratio =0.95 in men and =0.85 in women, and normal glucose tolerance (fasting glucose =100 mg/dL), to the metformin (1500 mg daily, n = 25) or placebo (n = 23).

Of all participants, 56% were male and mean age was 42 years. 52% of participants received protease inhibitors. Mean CD4 count and percentage of participants in each group with an undetectable viral load was similar.

Metformin use over 24 weeks was associated with a decrease in appendicular fat mass, measured by DEXA, compared to placebo (–686.0 vs 161.0 kg, p = 0.03). After adjusting for age, height, and baseline appendicular fat mass, a trend towards decreased appendicular fat mass remained (–614.0 vs 95.3 kg, p = 0.12). However, there was no significant change in visceral adipose tissue measured by single CT slice after 24 weeks of metformin vs placebo (–22.2 vs –3.85 cm2, p = 0.17). Metformin did not significantly change triglycerides, LDL, or HDL after 24 weeks compared to placebo.

The study concluded that the trend toward reduction in appendicular fat mass was unexpected and that metformin should be used with caution in HIV-associated lipodystrophy, and if used, should be reserved for persons with impaired glucose tolerance.

Another study looking at pioglitazone presented new data, although a couple of years ago a study of 11 people reported a small benefit (increase in approximately 3% leg fat after 6 months. In a late breaker presentation, Willy Rosenbaum presented results from a French study that enrolled 130 people, randomising half to 30mg pioglitazone once daily and half to placebo. After a year, leg fat increased by an average of +0.38 kg in the pioglitazone group. There was an increase of +0.44kg in people not using d4T, with no effect on people who continued using d4T. This is a small amount of fat to recover, and was not generally noticed by people, but if the benefits continue for another year the results are likely to become more noticeable. Leg circumference increased by about 1.2 cm.

The lipid profile was not significantly different between the 2 groups at week 48 except for HDL cholesterol which was improved in the pioglitazone group (+0.08 mmol/L vs –0.08; p = 0.005). There were 16 serious adverse events, 10 in the pioglitazone group and 6 in the placebo group.

AACTG 5186: fish oil and fenofibrates for management of elevated triglycerides

Finally, John Gerber from the University of Colorado presented results from and AACTG study looking at fish oil and fenofibrates for management of elevated triglycerides.

ACTG5186 was an open-labeled prospective study examining the efficacy (defined as serum TG <200 mg/dL) of fish oil 3g (1500 mg elcosapentaeonoic acid + 910 mg docosahexaenoic acid twice daily + fenofibrate 160 mg once daily in subjects with incomplete serum triglyceride-lowering response to fish oil or fenofibrate alone.

100 patients on effective ART with fasting serum TG >400 mg/dL and normal LDL cholesterol 1:1 to fish oil or fenofibrate for 8 weeks (step 1). If serum TG was >200 mg/dL at week 8, the combination of fish oil + fenofibrate was given from week 10 to week 18 (step 2).

During step 1 both fish oil and fenofibrate decreased serum TG by a median of 46% and 58%, respectively (intent to treat, p = 0.039, from median baseline serum TG of 662 and 694 mg/dL respectively. 75 (90.4%) of the step 1 non-responders entered step 2. The combination of fish oil + fenofibrate further decreased serum TG and the response rate increased to 22.7% with no difference between fish oil and fenofibrate arms. The median decrease in serum TG from baseline to week 18 was 65% for subjects participating in step 2.

Only 1 subject discontinued fish oil and 3 subjects discontinued the fish oil + fenofibrate combination because of side effects. Fish oil had no significant effect on CD4 count, CD4%, or trough concentration of lopinavir, the most commonly used PI.


All effects of insulin sensitising agents reported on lipoatrophy were below a clinical/esthetical relevant threshold. Disappointment by patients and physicians may limit the use of these substances.

The effect of fenofibrate and fish oil was moderate according to the ACTG guidelines in this specifically selected group of patients. LDL-cholesterol increased under therapy. In addition, there is no study with clinical endpoints supporting the use of either a fibrate or fish oil.

A recent large trial showed no decrease in cardiovascular mortality in patients with metabolic syndrome treated with a fibrate. For this reason, the rational of an approach targeting primarily triglycerides is not supported by evidence-based medicine. Fish oil as a medical product (e. g. Omacor) is about as expensive as atovarstatin based on daily drug dose.


  1. Carr A, Workman C, Carey D, et al. No effect of rosiglitazone for treatment of HIV-1 lipoatrophy: randomised, double-blind, placebo-controlled trial. Lancet. 2004 Feb 7;363(9407):429-38.
  2. Mallon P, Unemori P, Bowen M, et al. Nucleoside reverse transcriptase inhibitors decrease mitochondrial and PPAR-gamma gene expression in adipose tissue after only 2 weeks in HIV-uninfected healthy adults. 11th Conference on Retroviruses and Opportunistic Infections. February 8-11, 2004. San Francisco. Abstract 76.
  3. Nucleosides reduce PPAR-gamma: a role for rosiglitazone without thymidine analogues? HIV Treatment Bulletin Vol6 No1, January 2005.
  4. Mulligan K, Yang Y, Koletar S et al. Effects of metformin and rosiglitazone on body composition in HIV-infected patients with hyperinsulinemia and elevated waist/hip ratio: a randomized, placebo-controlled trial. Abstract 147.
  5. Kohli R, Wanke C, Gorbach S et al. A randomised placebo-controlled trial of metformin for the treatment of HIV lipodystrophy. Abstract 148.
  6. Slama L, Lanoy E, Rosenbaum W et al. Effect of pioglitazone on HIV-1 related lipoatrophy: a randomised double-blind placebo-controlled trial (ANRS 113) with 130 patients . Abstract 151LB.
  7. Gerber J, Kitch D, Aberg J et al. The safety and efficacy of fish oil in combination with fenofibrate in subjects on ART with hypertriglyceridemia who had an incomplete response to either agent alone: results of ACTG A5186. Abstract 146.

Links to other websites are current at date of posting but not maintained.