Australian researchers have reported in detail on the natural history of body fat and metabolic changes in people commencing antiretroviral therapy. Their findings challenge the notion that body fat changes could be linked to immune reconstitution, and show a strong association between lipoatrophy (fat loss) and cholesterol levels, both at baseline and after six months of therapy.
The findings are published in the May 2nd edition of AIDS, and comprise data on 40 individuals from St Vincent’s Hospital, Sydney, who began antiretroviral therapy between July 1997 and May 2000. The median age was 39.5 years, only 15% had a prior AIDS diagnosis, and the median CD4 cell count was 246 cells/mm3 at baseline (range 62-430). Median viral load was just above 100,000 copies/ml and median cholesterol was 4.25mmol/l.
Nineteen received a protease inhibitor-containing regimen and 21 an NNRTI-containing regimen, with ddI/d4T (13), d4T/3TC (13) and AZT/3TC (10) the most common components of the nucleoside analogue backbone.
Ninety per cent of subjects were followed for at least 96 weeks, and half were followed for 144 weeks.
Body fat changes (lipodystrophy)
Body composition changes were measured by DEXA at baseline and at weeks 12, 24, 48 and then every 48 weeks.
Individuals gained limb fat until week 24, and thereafter began to lose it. This pattern was also seen in the ACTG 384 study, which compared two nucleoside analogue backbones. By week 96 the median limb fat content in the Australian cohort was significantly lower than at baseline (48% of patients lost at least 1kg of limb fat by week 96). Participants lost a median of 13.6% of limb fat each year (IQR 0.9 - 26.3%).
Fat loss thus occurred after the period of greatest immune reconstitution, when viral activity had been largely suppressed (70% had viral load below 400 copies/ml by week 24, and 82% had viral load below 400 copies at the time of analysis). The authors argue that this temporal pattern undermines the case for an immune-mediated basis for lipoatrophy.
Fat loss was associated with d4T (stavudine) treatment at week 24, together with higher baseline viral load and higher lean body mass at week 24. d4T treatment was the only significant factor associated with fat loss by multivariate analysis (p
Fat loss was also significantly associated with higher baseline cholesterol levels (p
Central abdominal fat accumulation was associated with higher serum insulin concentrations at baseline and higher cholesterol levels at week 24. Those with greatest fat accumulation between baseline and week 24 also had significantly higher fat content across all parameters at week 24. Immune reconstitution had a positive association with fat accumulation in both the limbs and the trunk; those who experienced the greatest CD4 cell increase up to week 24 also gained the most fat. The median CD4 cell increase at week 24 was 126 cells/mm3, with a further median increase of 56 cells/mm3 between weeks 24 and 144.
People taking PI-containing regimens had significantly more central fat at week 24, whilst people taking NNRTI-containing regimens had significantly less fat by this point.
Metabolic changes
Changes in fasting insulin did not differ between the PI and NNRTI-treated groups, suggesting that insulin resistance was not responsible for the difference in central fat between the two groups. Hyperinsulinaemia developed after central fat accumulation, with only moderate changes detectable before week 96.
Total cholesterol and LDL cholesterol had risen significantly by week 12 and this elevation was sustained to week 144. However, triglyceride levels rose more slowly, and did not show a significant elevation above baseline until week 96.
Greater cholesterol increases were associated with lower baseline cholesterol, PI treatment and lower baseline lean body mass.
Further information on this website
Body fat and metabolic changes whilst on treatment - comprehensive overview
Mallon PWG et al. Prospective evaluation of the effects of antiretroviral therapy on body composition in HIV-1 infected men starting therapy. AIDS 17: 971-979, 2003.