New Developments in Fat Redistribution andMetabolic Disorders Associated with HAART

The 7th National Conference on Retroviruses and Opportunistic Infections devoted a significant part of its oral and poster presentations to lipodystrophy, which is more accurately called HIV-related adipose redistribution syndrome (HARS). This disorder, a possible long-term complication of anti-HIV therapy, has become one of the leading obstacles in the management of HIV disease. Fear of it dissuades newly infected persons from starting treatment, and leads others who have been on therapy to interrupt or stop treatment altogether.

The syndrome of fat redistribution has multiple characteristics. These include body habitus changes such as deposition of fat in the dorsocervical area (buffalo humps) and around the torso and waist (truncal adiposity); enlargement of the breasts; and loss of fat in the buttocks, limbs (peripheral lipodystrophy) and in the cheeks (facial fat atrophy), causing the hollowing of the cheeks and grooving of the nasolabial fold. It may or may not be accompanied by elevations in blood lipids (cholesterol and triglycerides) and/or blood glucose levels, which in many cases can accelerate the progression of coronary artery and cerebral artery disease leading to heart attacks and strokes.

However, some researchers, such as Kathleen Mulligan, Ph.D., from UCSF, think that these characteristics may not all part of one syndrome (abstracts S20, 202). Rather, she posits that there may be more than one syndrome with distinct etiologies, some of which may actually be caused by nucleoside analogs not protease inhibitors (see box on facing page). This could at least partly explain why the basic science and animal studies evaluating possible causes of the fat metabolism do not seem to consistently explain all the various facets of the disorder and sometimes are contradictory. Some conflicting results may depend on which company was conducting the study. For example, one showed that amprenavir and saquinavir reduce triglycerides in mice fed a low fat diet, whereas another showed that indinavir, nelfinavir and saquinavir all lead to rises in triglyceride levels (abstract 37).

Studies by Lenhard and others from Glaxo Wellcome (abstract 38) continued to show the effect of indinavir on retinoid receptors, which are involved not only in fat metabolism but also in the generation of skin and its appendages. This may explain the disorders of dry skin (xerosis), dry mouth (xerostomia), ingrown toe nails (paronychia), and alopecia (hair loss) seen sometimes with indinavir use. Another in vitro study showed that nelfinavir and ritonavir increased the production of triglycerides in the liver (abstract 39).

Despite these suggestive studies, none can establish cause and effect, and many postulate that the disorder may have nothing to do with the therapy but may be a long-term consequence of HIV infection, since it has been seen in patients who have never been on protease inhibitors, those who have been on nucleoside therapy alone, and even a few that have never received treatment.


Several studies attempted to evaluate the overall prevalence of the fat redistribution disorder in the HIV-infected population in their community. A very large Australian study (1350 patients) found a prevalence of 53% of body habitus changes by physician assessment, more commonly with those who used protease inhibitors (81%) than those who had not (33%) (abstract 201).

One provocative study by Wanke and others from Boston found a significantly lower rate of fat deposition and atrophy among African-Americans and women than others. However, this study based its definition of fat atrophy and deposition on strict anthropometric measurements such as waist/hip ratios and triceps skinfold thickness, which may not truly measure the severity of the fat redistribution (abstract 24). In addition it found no association between HAART and the development of the syndrome, or an association with drug use, use of anabolics or appetite stimulants.

The largest study (HIV Outpatient Study), conducted by the CDC, found that the risk factors for the development of the syndrome included age, years of HIV infection or years since an AIDS diagnosis, use of indinavir and/or d4T, and months since the lowest CD4 cell count (abstract 23). Another large study from Barcelona, Spain, reported that 13% of their cohort of 506 patients developed lipodystrophy after an average of 18 months. It found an association with older age, sexual transmission of HIV and longer duration of d4T therapy (abstract 15).

Another European study by Rozenbaum and others found significantly higher incidence of the syndrome among the 203 patients: 60% had at least one sign of the condition, such as atrophy in 16%, hypertrophy in 23%, and both in 20% (abstract 14). They evaluated the metabolic abnormalities and found that elevated cholesterol and triglyceride levels were found in 34% and 30% respectively, and diabetes mellitus in 3%.

One study (abstract 22) suggested that single slice computed tomography (CT) scanning through the umbilicus was very effective in detecting fat deposition in the subcutaneous and the intra-abdominal cavity, using the VAT:TAT ratio (visceral to total adipose tissue), which was highest in patients with clinical HIV disease on HAART and not significantly elevated in non-HIV-infected controls or in naive patients.

Other studies found a correlation between lipodystrophy and elevated cortisol/DHEA ratios and in serum alpha interferon levels, both of which are involved in fat metabolism (abstract 203).

Two studies also found an association with accelerated bone mineral loss (osteopenia and osteoporosis) using dual energy absorptiometry (abstracts 207 and 208). In one the rate among patients receiving HAART was 21% compared to 6% among those not receiving HAART; in the other study of patients with lipodystrophy, 28% had osteopenia and 9.5% had osteoporosis.

Back to the GMHC Treatment Issues Spring 2000 contents page.