HIV infection or inflammatory changes associated with it may be responsible for fat accumulation and body fat redistribution, rather than HIV drugs. This is according to Dr Grace McComsey of Case Western University in Cleveland, Ohio who said that although the association of subcutaneous fat loss (lipoatrophy) with mitochondrial damage caused by certain HIV drugs was well-established (and most of the world no longer uses the drugs like stavudine that are most strongly associated with it), two decades of research had failed to establish a cause for the distinctive fat gain (lipohypertrophy), especially in the trunk and within the abdomen, seen in some people on antiretroviral therapy (ART).
Initially these fat gains were associated with protease inhibitors (PIs) but switching from PIs to non-nucleoside reverse transcriptase inhibitors (NNRTIs) or to integrase inhibitors did not reverse fat gains. One study had found greater fat gain in people taking boosted atazanavir rather than efavirenz but a general association with any drug or class of drugs had not been demonstrated. Given that untreated HIV infection usually results in weight loss, fat gains when people started ART, once it became available, may have understandably been associated with ART rather than HIV.
The study McComsey presented, compared changes in limb fat, trunk fat, visceral adipose tissue (central abdominal fat) and lean muscle mass in 1809 ART-naïve patients starting either of the two boosted PIs atazanavir or darunavir, or the integrase inhibitor raltegravir, all combined with tenofovir/emtricitabine (Truvada).
DEXA and CAT scans measured fat and muscle distribution at baseline and nearly two years (96 weeks) later. They were then tested for association with drug regimen, baseline HIV viral load, Framingham risk score (a measure of the likelihood of cardiovascular disease), and a number of hormones, cytokines (cell messenger chemicals) and markers of inflammation: leptin (higher in the obese) adiponectin (lower in the obese), D-dimer and C-reactive protein (inflammation indicators) and the cytokines/cytokine receptors IL-6, CD14 and CD163.
In terms of demographics, the study participants were aged on average 36, 90% were men and 44% were white (slightly more taking atazanavir). Their average pre-ART viral load was 34,150 copies/ml, and their average CD4 count 351 cells/mm3. Those taking atazanavir had slightly, but not significantly, more limb and trunk fat but not more visceral fat or muscle.
Limb, trunk and visceral fat all increased in the 96 weeks on ART. Limb fat increased by 15% (20% on raltegravir), trunk fat by 22% (16% on atazanavir and 29% on raltegravir) and visceral fat by 31% (the differences between drugs were not statistically significant). Lean muscle mass increased slightly by 2% in people on atazanavir or raltegravir and 1.2% in patients taking darunavir, and this was a statistically significant difference but probably does not reflect any real difference between the drugs.
Increases in visceral fat were associated with lower leptin and higher adiponectin, but this probably is effect rather than cause as adiponectin is secreted by fatty tissue. Subcutaneous limb fat gain was also associated with IL-6 and lean body mass gain with d-dimer and lower baseline CD4 – the latter not unexpected, as lean body mass falls with AIDS.
However by far the strongest association with fat gains was with high viral load – there were mean increases of at least 25% and up to 35% in both subcutaneous and visceral fat in people who had a baseline viral load over 100,000 copies/ml, whichever drug they were taking. In patients with a baseline viral load below 100,000 the fat gains were below 10% apart from a gain in visceral fat of about 14% in people taking raltegravir.
McComsey said that the fat gains observed were not necessarily those associated with health improvement due to the control of HIV. “A 30% gain in visceral adipose tissue in just two years is pretty bad,” she said. “Even limb fat increased by 1.5 kilos from a baseline of seven kilos (21%). There was an average increase in 3.0 to 3.5 in people’s BMI score.” Although people in general may be getting fatter, it does not happen as fast as this, she added. She added that even when adjusted for inflammatory markers, HIV viral load was still significantly associated with fat gain.