icon-folder.gif   Conference Reports for NATAP  
  AIDS 2002 Barcelona
Barcelona, Spain July 7-12 2002
Back grey_arrow_rt.gif
A Lipodystrophy Reformation?
David Alain Wohl, MD University of North Carolina AIDS Research and Treatment Center
  For a time it was thought that most every body shape change accompanying HIV therapy was due to protease inhibitors - and to say otherwise was almost heresy. However, with reports from observational, comparative and treatment switch studies it has become abundantly clear that other causes of these changes exist and that some shape changes, such as lipoatrophy (fat loss), seem to have less to do with protease inhibitors than other drugs and patient-specific factors. One of the largest and most ambitious efforts to examine the prevalence and associated factors of HIV-associated metabolic complications is the acronym challenged Fat Redistribution and Metabolic Change in HIV Infection (FRAM) Study.
The preliminary and controversial results of this large cross sectional study supported by the U.S. National Institute of Diabetes and Digestive and Kidney Diseases were presented in Barcelona at the World AIDS Conference by Dr. Carl Grunfeld of the University of California at San Francisco who, although originally slated to provide a basic science overview of HIV-associated metabolic disorders during one of the 'bridging sessions', chose instead to use the time he was provided to present the preliminary FRAM results - results that challenge most of our assumptions about body shape alterations during HIV infection. This meant that the data in the presentation were not subject to the conference peer review process and are not included in the published abstract book (at least Luther nailed his theses to the church door for all to review).
In rapid fire powerpoint succession, Dr. Grunfeld, sped through a series of slides that detailed the study design and initial results. The study enrolls a random sample of over 1000 HIV-infected patients at centers throughout the country and several hundred presumably un-infected age matched control subjects from the CARDIA study - an ongoing long-term study examining the evolution of cardiovascular (CVD) risk factors in young adults. All subjects underwent one time surveys of body shape, body shape measurements, total body DEXA scan, head to toe MRI scans as well as blood and urine testing for metabolic and renal abnormalities.
The analysis presented included data collected from only 350 of the 800 HIV-infected men enrolled and from the age-matched HIV un-infected controls. Data from women were not presented. Of the HIV+ subjects included median age was 40 years, half were non-white, 77% were men who have sex with men, and median CD4 and viral load were 394/cu mm and 35,700 copies/mL, respectively. Over 80% were on an NRTI and 14% on no antiretrovirals. Lipoatrophy was classified as peripheral if involving the cheeks, face, buttocks, legs or arms or as being central if of the waist, abdomen, neck, back, and chest. Lipohypertrophy was used to describe fat accumulation at these areas. All body shape abnormalities were graded as mild, moderate or severe.
The bombshell of the presentation was the comparison of body shape between the randomly selected HIV+ men and the controls. By self report, peripheral lipoatrophy was more prevalent in HIV+ men but there was no difference in fat accumulation of the belly reported among HIV-infected and uninfected subjects. In fact, fat loss of the abdomen was more commonly reported in HIV+ men compared to those without HIV. Likewise buffalo humps were described by about 8% of the HIV-infected subjects but surprisingly by 12% of the HIV negatives, therefore this problem seemed, if anything, less common in HIV-infected men. About 40% of the included subjects had report of lipoatrophy of some area of the body.
Further, the concept of mixed peripheral lipoatrophy and central (abdominal) fat accumulation was found to be, basically, non-existent. Patients with peripheral fat loss not only did not tend to have associated fat accumulation of the belly, they actually reported more loss of central fat compared again to controls. Amazingly, patients with self-described central fat accumulation were less likely to have peripheral fat wasting, further standing on its head the observation of a mixed fat gain/loss picture. Dr. Grunfeld reported that there was strong concordance between self-report and physical exam.
Objective data from DEXA and MRI scans seemed to support the clinical findings (the slides were whizzing by). Peripheral lipoatrophy distinguished between HIV positive and negative subjects (more fat wasting of the peripheral areas in HIV+) and was most prevalent in those with clinical peripheral lipoatrophy, although even HIV-infected subjects without self-reported peripheral fat loss had less limb fat than controls. Central fat volume (visceral fat tissue) as measured by MRI was less in HIV positive subjects overall than controls and was lowest in those HIV+ men with peripheral lipoatrophy.
How to explain these results which fly in the face of the clinical observations of most clinicians and patients? After the presentation to a room full of the world's foremost experts on metabolic complications of HIV infection, there was (stunned?) silence. Not one question was asked and probably more than just this author wished Dr. Grunfeld could just repeat it all again slowly to make sure we got it right. Anticipating our skepticism, Dr. Grunfeld suggested that what patients and clinicians appreciate as fat accumulation of the abdomen or posterior neck may actually be relatively less fat loss of these areas (i.e. you are losing fat at the arms, legs and chest faster so everything else looks bigger). This explanation falls short of explaining the peculiar changes we have seen in our clinics, often times accompanied by fat loss of the face and limbs - the motion limiting buffalo humps requiring surgery, and the expanding bellies that result in increases in pants size and even umbilical hernias, not to mention all those CT scans of the abdomen demonstrating dense white fat accumulation in the visceral abdomen in patients with HIV.
Before we declare that central fat accumulation or buffalo humps are not a feature of HIV-associated lipodystrophy it is important to recognize the clear limitations of this study. Foremost, it was cross sectional. This means that only the body shape changes that were actually present when the researchers examined the patients and not the changes that may have occurred (or will occur) over time could be evaluated. Longitudinal follow-up would allow for individual changes to be much better appreciated and compared. Additionally, there has yet to be an analysis looking at subsets of patients who were on HIV treatment. These results may look different if, say, subjects on indinavir and d4T were compared to HIV-infected persons not on therapy and to HIV-uninfected controls. The control subjects also were fatter than the HIV-infected subjects. This could make comparison of certain body shape changes less appropriate (is a 'buffalo hump in an obese HIV-negative man the same kind of buffalo hump as in an HIV-infected skinnier man?) Lastly, these results were from a relatively small number of persons and only a portion of the overall group. It will be interesting to see how the findings evolve with the analysis of the entire cohort, including women, and what sub-analyses looking at drug exposures, clinical history and demographics will yield.
This important study will undoubtedly be a wellspring of information about metabolic issues in HIV including lipid abnormalities, bone density, kidney problems, etc. But, whether this study can best answer the most pressing questions about body shape is questionable as the one thing this one-time, slice of reality study can not detect, measure or describe about the bodies of HIV infected persons is what we are most interested in, how they change.