Bulletin of Experimental Treatments for AIDS, January, 1999
Over the past two years, a wide range of body fat abnormalities has been reported in people with HIV/AIDS. Often, these reports are made by people taking antiretroviral therapy, many of whom are experiencing therapeutic success as indicated by viral load decreases, CD4 cell increases, and a general return to good health. Ironically, the first reports of abnormal body fat changes began emerging amid the euphoria that followed the XI International Conference on AIDS in Vancouver in 1996, with its reports of the promise of protease inhibitors and highly active antiretroviral therapy (HAART). The number of case reports of body fat changes grew rapidly in 1998, but the causes remain unclear and the pathogenesis poorly understood.
This article will discuss body fat changes, a topic of great current interest and controversy in the HIV community. Terms used to refer to these symptoms include body fat redistribution and fat derangement. "Lipodystrophy" has been frequently but incorrectly used as a general term for diverse body shape changes; the term properly describes fat loss only.
A discussion of body fat redistribution requires mention of the metabolic abnormalities that also have been widely observed. The connection between body fat redistribution and metabolic changes--primarily elevations of plasma sugar and fats--is presently uncertain. Metabolic symptoms may be related to body fat changes and may be associated with HAART.
Body fat changes may cause secondary problems, ranging from restricted mobility to sleep apnea. In addition, wide-scale public health problems related to body fat changes are also a concern, should, for example, some people with HIV prematurely discontinue or change medications they suspect are causing body shape changes, or refuse potent antiretroviral treatment from which they might benefit.
An exact definition of body fat redistribution syndrome does not yet exist. At least four syndromes have been described that are characterized by the accumulation of fat, and one by the loss of fat; combinations of these may occur in an individual. Nor are there reliable data on prevalence. At the 12th World AIDS Conference in Geneva in June 1998, the reported prevalence of body fat changes in people taking protease inhibitors ranged widely, from 2% to 84%.
At one end of the spectrum, Boix reported that a chart review of 272 patients revealed a prevalence of 2%. At the other end, Andrew Carr, MD, of St. Vincent's Hospital in Sydney, Australia, reported at a late-breaker session a prevalence of 84% in his cohort of 116 people (by participant self-report, exam, and dual energy X-ray absorptiometry [DEXA] scans, or imaging studies). Clinical studies of drugs have not yet begun to routinely include lipid and other metabolic measurements. In the absence of a standardized definition, it has not been possible to collect consistent data or produce coherent reports of fat abnormalities.
At the moment, considerable effort is going into clearly defining the features of a presumed syndrome. Various groups within the HIV community are working to develop a standardized, interim definition, as well as research protocols to study these phenomena.
At present, reports of people with body fat abnormalities constitute something of a mixed bag of potential signs and symptoms. Although body fat changes have been associated with HAART, it is not yet clear what role antiretroviral therapy plays. Protease inhibitors, in particular, have been strongly implicated, since they are known to alter levels of blood lipids and to disrupt glucose and insulin metabolism, and since the complaints first surfaced in the era of protease inhibitors. HIV itself probably plays a role as well. Early in HIV infection, metabolism is somewhat altered. These abnormal changes may be related to HIV disease progression, to successful antiviral suppression, and to immune system restoration. In addition, being in a chronic disease state, coupled with chronic drug treatment, which continuously affects enzymes in the liver, may contribute to the development of this syndrome.
In conclusion, the causes of body fat redistribution are likely to be multifactorial. Most experts agree that research cannot move forward until there is a clear definition of the syndrome and a better understanding of its prevalence.
Both men and women with HIV have reported body fat redistribution. At the 38th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC), F. E. Babl of Boston Medical Center reported body fat changes in school-aged and adolescent children with HIV, as well. Children taking protease inhibitors were four times as likely to have developed body fat redistribution as those not taking these drugs.
The most common manifestations of body fat redistribution are accumulations of fat in the central body, along with the loss of subcutaneous (under the skin) fat in the face, limbs, and gluteal region, or buttocks. Fat accumulation includes the development of a fat pad on the back of the neck ("buffalo hump") and an accumulation of visceral fat in the abdomen or belly ("protease paunch"). Women may experience breast enlargement; more rarely, men also have reported breast growth. Overall, the most common finding is increased abdominal girth, or truncal obesity, particularly in the visceral cavity. With fat loss (lipodystrophy), thinning regions such as the arms and legs display subcutaneous veins more prominently, and gaunt faces show wrinkling.
Body fat changes may be so pronounced as to be disfiguring. Most people have concerns that go far beyond the cosmetic; body fat changes can also interfere with one's ability to function normally. For instance, people with extremely large buffalo humps have reported severe migraine headaches, inability to sleep, and loss of the normal range of motion of the head, neck, and shoulders. Others with gluteal wasting may be unable to remain in a seated position. People with vastly increased visceral (internal) fat may suffer problems related to pressure and adverse effects upon the organs, with symptoms of bloating and pain. Any of these changes may also lead to anxiety and depression.
At the 5th Conference on Retroviruses and Opportunistic Infections (CROI) in February 1998, many scientific reports were presented on body fat changes. More such reports were presented at the Geneva AIDS conference. In addition to other studies which will be discussed throughout this article, Donald Kotler, MD, Chief of Gastrointestinal Immunology at St. Luke's-Roosevelt Hospital in New York, presented results of a study of the effects of HAART on body fat redistribution. Kotler's group compared data from 96 HIV positive persons evaluated since the beginning of 1996, 96 HIV positive persons evaluated before 1996, and 96 HIV negative controls. The HIV positive persons seen most recently had higher body weights, higher amounts of lean body mass, and lower proportions of fat than both the HIV positive persons seen earlier and the HIV negative controls. Both groups of HIV positive persons had less subcutaneous fat than the HIV negative persons, indicating that people with HIV who are not taking protease inhibitors still experience changes in fat distribution. Being on antiviral therapy in general was found to be a predictor of body fat changes, but it was not significant that therapy include a protease inhibitor. Older men were more likely to have increased abdominal girth. Those with greater visceral fat -- a greater degree of abnormal fat redistribution -- had lower viral load levels.
Another study of body composition presented by Kotler in Geneva showed that HIV positive people with increased waist size had significant increases in visceral (rather than subcutaneous) abdominal fat compared to HIV positive people without waist size increases. The study involved 15 men and 11 women. Those with waist size increases also weighed more and had higher CD4 cell counts and lower viral loads. Based on these studies, Kotler concluded that body fat changes may not be directly related to protease inhibitors, but indirectly related to antiretroviral therapy and the extent of viral suppression. That is, disease processes alone may be associated with body fat changes.
A buffalo hump, as it has been commonly called, is an accumulation of fat underneath the skin at the back of the neck, or dorsocervical region. Examination has revealed that the fat has a fibrous nature. Buffalo hump has been classically described in Cushing's syndrome, where it is attributed to elevations of the adrenal gland hormone cortisol and is usually associated with steroid use. When an individual with Cushing's syndrome stops taking exogenous steroids, the buffalo hump disappears. However, studies in people with HIV generally reveal normal levels of cortisol. Stopping steroids does not cause buffalo hump to disappear in people with HIV.
In a poster session at the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) in Toronto in September 1997, Peter Ruane, MD, of Tower Infectious Diseases Medical Associates in Los Angeles reported on three patients in his HIV practice who had presented with unusual deposits of fat at the base of the neck. All three men were taking triple combination antiretroviral therapy. Their complaints included discomfort and trouble sleeping, as well as cosmetic concerns. All three had extensive past use of antiretroviral drugs, and all had begun triple combination therapy that included indinavir (Crixivan) in 1996. Two reported that their swelling began about six or seven months after beginning therapy; one man reported that he had noticed a swelling sometime earlier, before he began indinavir, but that it grew rapidly about two months afterward. All of the men had normal cortisol levels and an average weight gain of four pounds at the time of diagnosis. One of the men had a magnetic resonance imaging (MRI) test, which revealed a 15 x 7 cm mass. At the time, Ruane stated that the relationship, if any, between the fat pads and triple combination therapy with a protease inhibitor was unclear.
Joan Lo, MD, and colleagues from San Francisco General Hospital (SFGH) first described buffalo hump in eight persons at the 1998 CROI. The data were published in the March 21, 1998 issue of The Lancet. All eight men studied had developed buffalo hump while on a stable antiretroviral regimen, but only four of the eight were taking protease inhibitors, which at that time were available only through clinical trials. Investigators performed tests for cortisol levels, which allowed them to rule out Cushing's syndrome as a possible cause. When compared with HIV positive controls, the eight men with buffalo humps also had a significantly higher proportion of abdominal fat. Triglyceride levels were somewhat higher in the men with buffalo humps than in the controls, but not significantly. Cholesterol and fasting glucose values were similar in the two groups. This small but important study established that buffalo humps in these HIV positive men were not related to hypercortisolism (elevated cortisol levels), nor were they unique to people taking protease inhibitor therapy.
Computed tomography (CT) scans of the abdominal region in people with increased abdominalgirth ("protease paunch") show that the fat accumulation is deep within the visceral cavity. Increased abdominal girth has been reported in people whose drug regimens do not include a protease inhibitor. This intra-abdominal fat accumulation is also referred to as truncal obesity. Often, increases in visceral fat are observed along with decreases in subcutaneous fat elsewhere, e.g., in the extremities (arms and legs). People with increased abdominal girth may also experience heartburn and discomfort associated with a sense of bloating.
A report by Kirk Miller in the March 21, 1998 issue of The Lancet described abdominal fat composition in 30 HIV positive men. Of 20 men who were taking indinavir, ten had experienced increased abdominal girth and gastrointestinal discomfort about three months after starting the drug, and ten had not experienced any fat accumulation or discomfort; the other ten were not using indinavir and served as controls. All 30 men had stable body weight. Miller used CT scans to evaluate the type of fat accumulation and distribution, and compared the ratio of visceral to total fat. Those on indinavir with increased abdominal girth had the highest ratios of visceral to total fat, compared to those not taking indinavir. Asymptomatic men taking indinavir also had a higher ratio of visceral to total fat than those not taking the drug, but not as high as symptomatic men taking indinavir. Researchers concluded that the higher ratios of visceral to abdominal fat were related to longer duration of indinavir use and were associated with hyperlipidemia (high blood plasma fat levels).
Another distinct fat accumulation syndrome that has been reported in people with HIV is called multiple symmetric lipomatosis or benign symmetric lipomatosis (BSL). BSL is distinguished by a symmetric accumulation of fat (i.e., similar on both sides of the body), usually in the head (mid to lower face), neck, and shoulders (often described as a "horse collar" distribution), but sometimes occurring in the chest, abdomen, upper thighs, or groin. An uncommon condition in the general population, BSL occurs most frequently in male alcoholics. BSL is also associated with glucose intolerance and hyperlipidemia.
The syndrome in the context of HIV was first reported in 1997, when R. L. Hengel and colleagues published a case report in The Lancet concerning a 34-year-old HIV positive man with BSL. The man developed BSL in the supraclavicular area (above the collar bones) and a buffalo hump while taking indinavir, 3TC, and AZT. All other causes were ruled out, including alcohol use. Other tests revealed slightly elevated blood glucose, but normal cholesterol and cortisol levels. At the 1998 CROI, Hengel reported four cases of BSL. All four people (three men, one woman) had been taking indinavir for a median of about six months before developing body changes. One had elevated glucose levels and two had slightly elevated triglyceride levels. Morris Schambelan, MD, of SFGH says that physicians are beginning to see an increasing number of cases of BSL at this hospital.
Breast hypertrophy (enlargement) has been seen in HIV positive women, who usually also report an increase in abdominal girth and loss of fat in the buttocks, or development of a buffalo hump. Breast hypertrophy also has been reported in a few men.
At the Geneva AIDS conference, Krista Dong of Brown University reported on body changes in 118 women who began protease inhibitor therapy. Of the 118 women, 19 self-reported changes in body shape, as follows: breast hypertrophy (71%), increased abdominal girth (71%), peripheral wasting (47%), weight gain (47%), gluteal wasting (29%), and buffalo hump (23%). Lipid abnormalities included low high-density lipoprotein (HDL) or "good" cholesterol (6%), elevated low-density lipoprotein (LDL) or "bad" cholesterol (47%), and elevated triglycerides (16%). (See the sidebar "Lipids in Brief" for a description of types of fats.) After one year on therapy, the prevalence of body fat or blood lipid changes was 16%.
At the 1998 ICAAC, a team of researchers from Italy described a pattern of fat redistribution in a cohort of 311 women being treated with antiretroviral drugs. About 10% of the women had significant breast enlargement (greater than two bra sizes) and increased abdominal girth, along with a reduction of fat in the legs and gluteal region. None showed changes in total body weight. DEXA scans showed that the changes in body mass distribution were related to changes in fat, rather than bone or muscle. All women were on stable antiretroviral treatment to which they had reportedly responded well.
Women taking a three-drug combination that included a protease inhibitor were significantly more likely to have developed body fat changes, compared to women taking two nucleoside analogs (20 of 137 vs 12 of 174). Blood endocrine tests (plasma cortisol, ACTH, GH, C-peptide, testosterone) were generally normal, as were prolactin, glucose, cholesterol, and triglyceride levels. The researchers concluded that "the high incidence of these side effects may cause serious psychological problems for HIV positive women in the near future and significantly reduce their adherence to antiretroviral therapies."
Lipodystrophy denotes lack or loss of subcutaneous fat. Outside the context of HIV, lipodystrophy syndrome is rare, and may be inherited or acquired. Inherited, or congenital, lipodystrophy is characterized by a near total lack of subcutaneous fat and involves the internal organs. Acquired lipodystrophy involves a progressive loss of subcutaneous fat in the upper body. The etiology of non-HIV-related lipodystrophy is not known, but involves an inability of certain adipocytes (fat cells) to store fat. Associated features include a tendency for insulin resistance, diabetes, and elevated triglycerides, plus increased fat accumulation in other areas (e.g., the pelvic region) and muscular hypertrophy (enlarged muscle cells). Apoptosis (programmed cell death) may be a factor, by causing the programmed death of fat cells in the periphery of the body.
Lipodystrophy in HIV is commonly characterized by the loss of fat in the face, especially subzygomatic (below the cheekbones) fat, and in the arms, legs, buttocks, and sometimes even in the subcutaneous abdominal area. When severe, facial wasting occurs at the temples and in the eye sockets. Increased facial wrinkling and prominence of veins, particularly in the extremities, have been observed with lipodystrophy.
The reported prevalence of lipodystrophy in people taking protease inhibitors ranges widely. Some of the best descriptions of HIV-related lipodystrophy made so far are the reports by Carr, Cooper, and colleagues. At the 1998 CROI, they reported seeing in a one-month period 160 people with HIV, 116 of whom were taking protease inhibitors. By patient self-report and subsequent confirmation by physical exam, 64% of the 116 on protease inhibitors had fat loss (the higher figure mentioned earlier, 84%, was a revised and updated finding the team reported later at the Geneva AIDS conference). A relative preservation of abdominal fat, as well as muscle in the extremities, was reported. Changes were associated with elevated fat and glucose levels, and insulin resistance. The findings were most pronounced in those taking ritonavir (Norvir) plus saquinavir (Invirase), followed by nelfinavir (Viracept), then indinavir.
Changes in blood plasma lipid levels have been widely reported, including hypercholesteremia (high blood cholesterol levels) and hypertriglyceridemia (high blood levels of triglycerides, a type of fat), collectively termed hyperlipidemia (high fat levels in the blood). Insulin resistance and hyperglycemia (high levels of glucose, or blood sugar) have also been reported. Some people have reported hyperuricemia (increased blood levels of uric acid). The relationship between body fat changes and changes in lipid levels, glucose levels, and insulin is not yet understood. There also have been reports of cardiovascular disease and heart attacks, as well as diabetes requiring treatment, and gout associated with hyperuricemia.
Metabolic changes in lipids, glucose, and insulin will be the focus of a future article. Current data on the relationship between body fat changes and metabolic changes are somewhat contradictory. In Carr's Geneva report, hyperlipidemia was common but did not clearly correlate with body fat changes. In one report, body fat changes appeared to follow the development of changes in blood fat levels; in another, there is no correlation. In Carr's cohort, amounts of insulin were elevated, yet glucose levels remained relatively close to normal.
A general principle concerns the fact that protease inhibitors are metabolized in the liver by the cytochrome P450 system. Interactions between drugs that are metabolized by these isoenzymes may cause side effects. Studies to date indicate that protease inhibitors are associated with abnormalities in both lipid metabolism and glucose tolerance. Yet high levels of triglycerides were known to be a common feature of HIV infection before the protease inhibitor era. In studies published in 1989 and 1991 by Kotler and Carl Grunfeld, MD, PhD, of the UCSF Veterans Affairs Medical Center, one-half of HIV positive participants had elevated blood triglyceride levels in fasting samples. Finally, in several of the rare, non-HIV-related fat redistribution syndromes, similar metabolic changes are common.
Little is known about the causes of body fat redistribution syndrome, nor its pathogenesis. Researchers like Kotler contend that HIV infection and disease progression themselves probably play roles in fat redistribution, pointing out that fat abnormalities are seen in other chronic disease states. Kotler also feels that the process of recovery may be associated with the development of fat abnormalities -- that it may be the degree of viral load reduction achieved with HAART and immune restoration, rather than protease inhibitor therapy per se, that are associated with these syndromes. This remains to be proven.
Grunfeld, as well as Schambelan and others at SFGH, also are evaluating HIV disease as a major factor in the development of fat abnormalities. On the other hand, Cooper and Carr continue to elaborate upon their theory that that protease inhibitors are responsible for these changes.
According to Grunfeld, HIV itself is a likely culprit. Lipid abnormalities in HIV disease were known to exist long before protease inhibitors were developed. In a study of wasting conducted in 1989, investigators compared triglyceride levels (based on fasting samples) in people with AIDS, people with HIV, and HIV negative controls. Compared to HIV negative participants, all people with HIV had high triglyceride levels. Half of the people with AIDS had hypertriglyceridemia; these same people met the definition of wasting (the study was done before the 200 cells/mm3 CD4 count criteria was added to the CDC definition of AIDS).
Much of Grunfeld's current effort to understand fat derangement have to do with metabolism and metabolic changes related to HIV infection. These topics will be discussed in depth in a later article. Following are some of the high points.
Studies of people with wasting and hypertriglyceridemia led to the elucidation of the roles of cytokines (chemical messengers) in AIDS-related wasting. A cytokine called interferon-alpha was found to be linked to abnormal triglyceride levels. As HIV disease progresses, levels of interferon-alpha increase and triglycerides cannot be cleared from the blood. Interferon-alpha was correlated with HIV disease progression and the development of AIDS, with elevated triglycerides, and with the generation of new lipids (lipogenesis).
Another feature of HIV disease observed before the era of protease inhibitors was a decrease in blood HDL cholesterol. People with AIDS have lower HDL levels, which puts them at risk for cardiovascular disease. Plasma LDL is also decreased.
Grunfeld explains that "people with AIDS have a form of LDL cholesteremia that puts them at increased risk of atherosclerosis -- if they live long enough for it to emerge." Atherosclerosis is a type of hardening of the arteries. It is the main form of cardiovascular disease and the primary cause of heart attacks and death in the U.S. Characterized by lipid deposits in large and medium-sized arteries, it is normally associated with aging, along with cofactors such as smoking, diabetes, and high blood pressure. Atherosclerosis is also associated with increased risk of stroke, another leading cause of death in the U.S.
Essentially, protease inhibitors appear to be allowing people with a proatherogenic profile (an elevated risk for heart disease) to live longer, which provides the time it takes for heart disease to develop.
The Australian researchers Carr and Cooper have offered an explanation of the pathogenesis of "protease inhibitor-associated peripheral lipodystrophy, hyperlipidemia, and insulin resistance." In the June 20, 1998 issue of The Lancet, Cooper and Carr published their hypothesis, and their belief that most people taking protease inhibitors will at some point experience metabolic and fat abnormalities. (Cooper also presented the hypothesis at the Geneva AIDS conference.)
The researchers suggest that the HIV protease enzyme, to which protease inhibitors bind, is structurally very similar to two human proteins that are involved in the metabolism of lipids (that is, they have a considerable homology). One of the human proteins is called low-density lipoprotein-receptor-related protein (LPR) and the other is called cytoplasmic retinoic-acid binding protein type 1 (CRABP-1). Carr and Cooper reason that protease inhibitors may also inhibit these two human proteins, in addition to inhibiting the HIV protease. This may lead to hyperlipidemia, which contributes to body fat redistribution and lipodystrophy, as well as to insulin resistance and diabetes.
The homology or structural similarity between the human proteins and HIV protease is large, about 65%. However, neither LPR nor CRABP-1 have yet been well studied. Skeptical about Carr's hypothesis, Schambelan refers to a study by Rohlmann and others in the February 1998 Journal of Clinical Investigation in which inactivating the LPR gene in mice had no effect on the general health of the animals. While inactivating LPR resulted in the accumulation of cholesterol-rich lipoproteins (fat/protein compounds), an accompanying upregulation of LDL occurred in the liver, which achieved a balance of sorts and prevented harm. "Despite the large homology, this study casts some doubt on Carr's theories," says Schambelan, "at least regarding LPR."
The major uncertainty cast upon the homology hypothesis relates to the fact that there are reports of body fat changes in people who are not taking protease inhibitors. The 1997 study by Lo and colleagues, in which half the body fat changes observed were seen in people not taking protease inhibitors, established early that protease inhibitors were likely not the only factor involved in the development of this syndrome.
Unfortunately, studies of people with body shape changes taking protease inhibitors have not been able to determine to what extent protease inhibitors influence the changes. So the question remains, could protease inhibitors be having an indirect effect? It is possible that the degree of viral suppression achieved with protease inhibitors may be operative. For instance, in studies that considered the role of protease inhibitors, control groups that took non-protease inhibitor-containing regimens were not achieving the same degree of antiviral suppression. This confounding observation raises a question about the role of the strength of antiviral effect, and possibly the role of immune reconstitution. In other words, the magnitude of the antiviral effect may be a cofactor in the development of body fat redistribution syndrome.
Other aspects of protease inhibitor therapy that may contribute to this indirect effect remain to be elucidated.
In the HAART pool of options, protease inhibitors are known to cause elevations in blood lipids. In contrast, the earliest approved antiretroviral drugs, the nucleoside analogs (AZT, ddI, ddC, d4T, 3TC) did not provoke hyperlipidemia, nor did they suppress the virus as effectively. A presentation at the 4th International Congress on Drug Therapy in HIV Infection held in November 1998 in Glasgow added new information to the discussion of antiviral suppression and fat abnormalities. Schlomo Staszewski, MD, presented further results of a Phase III multicenter study that compared the antiretroviral activity of three-drug combinations in 450 people (results to 24 weeks were presented in Geneva).
Participants, 86% of whom were male and 85% of whom were treatment-naive, were randomized to take either efavirenz (Sustiva)/AZT/3TC, or indinavir/AZT/3TC, or efavirenz/indinavir. At entry, the mean viral load was 4.77 log copies/mL, and the mean CD4 count was 345 cells/mm3.
The study was amended to include the collection of lipid level data. In Glasgow, researchers presented information on the effects of efavirenz on lipid levels. Participants taking efavirenz experienced 10-20% increases in total and HDL cholesterol, and in triglyceride levels. Since blood was not drawn when participants were in a fasting state, the results are valid for total cholesterol and HDL, but not for triglycerides, or for LDL or VLDL (another form of cholesterol), since these are calculated based on triglycerides.
Increases in total cholesterol leveled off after four to eight weeks, and were greatest in those taking efavirenz/indinavir.
Increases in HDL cholesterol were seen only in people taking efavirenz-containing combinations, and increases continued through week 24. In terms of viral suppression, intent-to-treat analysis results to 36 weeks indicate that efavirenz/AZT/3TC was superior to indinavir/AZT/3TC, which was comparable to efavirenz/indinavir.
The observation that HDL cholesterol appears to increase with efavirenz, a potent NNRTI, lends support to the theory that viral suppression, rather than protease inhibitors per se, may influence fat metabolism, altering lipid levels. (HDL may be protective against cardiac risk, so the particular fat changes associated with the use of efavirenz may be beneficial.)
Since the syndrome(s) involving body fat redistribution are poorly understood, little is known about treating or preventing them. This creates obvious problems for people currently experiencing one or more of these body changes. Such changes -- from severe discomfort to disfigurement -- are significant enough to demand some type of response. Cardiovascular disease is a concern for those with abnormal lipid levels. Matt Sharp of the Asian/Pacific Islander Wellness Center sums up the feelings of many in the treatment community regarding body fat redistribution when he says, "There's a real division: the researchers keep talking about how to define it, while the community keeps asking how to treat it." While studies are being planned to answer key questions about body fat redistribution, the results will not be available for months or years, and people would like to know what they can do to treat or manage the fat abnormalities they are experiencing now. "We know it's happening," said one woman. "I can see it happening to me. What I'm interested in is what can be done about it."
The following section briefly outlines some of the experimental treatment leads being discussed.
The drugs most commonly prescribed in the general population to reduce lipid levels are the "statins." There is relatively little information about the safety and utility of statins in people with HIV who have hyperlipidemia. Since the statins, protease inhibitors, and NNRTIs are metabolized by the same CYP3A isoenzyme in the liver, there is potential concern about how they may interact with HAART and the toll that would be exacted upon the liver. Two of the most commonly used drugs, lovastatin (Mevacor) and simvastatin (Zocor), are known to be highly dependent on the CYP3A isoenzyme, and thus would be among the most likely to interact with protease inhibitors, causing toxicity and other adverse effects.
At the 1998 ICAAC, R.G. Hewitt and colleagues from the State University of New York at Buffalo concluded that gemfibrozil (Lopid) was an effective treatment for hypertriglyceridemia. Eight HIV positive men taking HAART added gemfibrozil, and their triglyceride levels on average fell from a high of 1,803 to 300 mg/dl, which was nearly the median pre-protease inhibitor level (298 mg/dl). Hewitt noted that men who switched from indinavir to nelfinavir experienced rebounds in triglyceride levels, even though they continued to use gemfibrozil.
Keith Henry's group found little benefit from the lipid-lowering drugs gemfibrozol and atorvastatin (Lipitor), although they did consider the drugs relatively safe. Atorvastatin is thought to lower triglyceride levels more than the other statins. According to Grunfeld, gemfibrozol and niacin might be tried. However, Grunfeld does not consider gemfibrozol very effective. He also cautions that "the problem with niacin is that it's not very easy to take -- compliance is usually low, around 50%."
Since the statin drugs are considered most effective, Grunfeld suggests trying fluvastatin (Lescol) and pravastatin (Pravachol). Fluvastatin and pravastatin are not metabolized by the CYP3A isoenzyme, so they are less likely to interact with protease inhibitors and are therefore the best candidates for use. "So far," summarizes Grunfeld, "reported responses are variable."
Human growth hormone (Serostim) is currently being evaluated in research studies as a possible treatment for fat redistribution. Growth hormone is known to restore lean body mass in people with AIDS-related wasting; it promotes increases in lean tissue and destruction of fat, both subcutaneous and visceral. However, growth hormone does not appear to reduce lipid levels.
Gabriel Torres, MD, reported in Geneva that five patients in his practice succeeded in reducing buffalo humps and abdominal girth (from 25-100% regression) with growth hormone. Reported adverse effects included hyperglycemia, which may require treatment, and elevated pancreatic enzymes. No optimal dose or duration was established. This report has inspired many to consider growth hormone an attractive possibility.
Similar anecdotal reports of people who have taken growth hormone and experienced reductions in buffalo hump and protease paunch continue to emerge. Sharp credits low-dose growth hormone (2.5 mg every other day) for contributing to a reduction in his abdominal girth, which had increased after beginning protease inhibitors in 1996. He says that after daily injections of 5 mg of growth hormone for one week, his paunch was significantly reduced. After the first week, he switched to a low-dose, every other day schedule.
On a cautionary note, growth hormone can cause hyperglycemia in a small number of people. Since people with fat redistribution taking antiretroviral therapy may already be at risk for developing diabetes, growth hormone should be used with caution. Growth hormone is likely not to be appropriate for all people, particularly for those experiencing severe metabolic disturbances. Finally, its considerable cost will probably prevent it from being a viable option for many people, particularly since many healthcare plans will not pay for growth hormone.
Some people with fat redistribution, particularly buffalo hump, have tried liposuction, the surgical removal of subcutaneous fat using high vacuum pressure. Liposuction is inappropriate for increased abdominal girth because the fat accumulation lies within the abdominal cavity, surrounding the organs, rather than under the skin. The benefit of liposuction appears limited; fat deposits may reappear. Anecdotally, a San Francisco man whose collar size went from 16 to 24 underwent liposuction with the support of his doctor. The procedure initially succeeded in alleviating his headaches and other discomfort, as well as problems with motion. However, within two to three months, he experienced a significant fat reaccumulation which required him to repeat the procedure. Payment is another issue; insurance companies do not routinely reimburse the cost of the procedure, which is generally considered cosmetic.
Finally, no controlled research studies have been conducted to evaluate the safety and utility of liposuction as a treatment for fat abnormalities.
Cardiovascular disease is the leading cause of death in Americans. Prevention, primarily through the reduction of risk factors like high cholesterol levels, has proved to be an important and widely respected strategy. Reducing dietary cholesterol is known to reduce risk.
When Henry reported cardiovascular disease in three young (26- to 40-year-old) men taking protease inhibitors, he emphasized that the patients had a history of smoking, a family history of heart disease, or diabetes. Traditional risk factors for cardiovascular disease include a high-fat diet, smoking, testosterone use, hypertension (high blood pressure), and diabetes. Traditional risk reduction strategies include smoking cessation, cardiovascular (aerobic) exercise, and a low-fat diet. These strategies are considered especially important for people with a family history of heart disease.
Henry looked at the effects of diet and exercise on a group of 22 people with HIV and hyperlipidemia, to see how their lipid levels would respond, compared to a similar group of 22 people treated with lipid-lowering drugs (gemfibrozil and/or atorvastatin). No significant effects were seen in the diet/exercise group, and only modest improvements were seen in the drug group. A Swiss study presented in Geneva on the prevalence of metabolic abnormalities reported no reduction in lipid levels following diet modification in a group of people with hyperlipidemia taking protease inhibitors.
Sharp considers diet and exercise an important component of his fat management regimen. He has cut back on dietary fat, using low-fat dairy products and reducing his intake of red meat and increasing his intake of chicken. Sharp says he was "eating a big greasy cheeseburger with fries" and commiserating with a friend about his paunch when a lightbulb went off. "We've been brainwashed into eating all that we can to gain weight. For years we've been eating lots of fat that we don't burn off with any kind of cardiovascular exercise. But now I've found that a regular cardio workout with weight-lifting, plus cutting back on high-fat and saturated-fat foods, has helped a great deal with the extra fat accumulating in weird places." Sharp's regular exercise regimen now includes a cardiovascular workout three to five times weekly, and he continues to lift weights three to four times a week. (Note: always consult with a physician before beginning a major exercise program.)
Overall, the possible benefits of interventions like diet and exercise are unknown. Nonetheless, Grunfeld also considers stopping smoking, beginning cardiovascular exercise, and eating a low-fat diet worthwhile interventions for a person with fat abnormalities.
Anabolic steroids have been discussed as possible interventions to reduce fat abnormalities. SFGH is currently studying nandrolone for the treatment of body fat redistribution syndrome (see "Current and Planned Research Studies," below); people in the treatment community continue to discuss using testosterone in various formulations (patch, gel, cream, and injection forms).
At the 36th Annual Meeting of the Infectious Diseases Society of America (IDSA) in Denver in November 1998, Douglas Dieterich gave a poster presentation of a chart review conducted by himself and colleagues at New York University. The review looked at 700 HIV positive persons, 91% of whom were male and 73% of whom were Caucasian, with a median age of 40 years. Of the 560 people on antiretroviral therapy, 96% were taking one or more protease inhibitors. More than half of the participants (62.4%) had taken anabolic steroids: 243 took testosterone; 101, oxandrolone; and 89, nandrolone.
In the entire cohort, only 31 had "physically apparent changes," of whom 14 were taking nandrolone; 12, testosterone; and 4, oxandrolone. In other words, body changes were seen in only 7% of those who had received anabolic or androgenic therapy. Dieterich decided to evaluate this cohort after being struck by the high prevalence of fat abnormalities reported in Geneva by Carr and Cooper, after learning that none of the Australian cohort were taking any sort of steroid therapy to promote weight gain (unlike Dieterich's own patients and unlike many people with HIV in the U.S.).
Finally, testosterone deficiency is known to occur in HIV positive men and even women. In men, testosterone deficiency is known to contribute to insulin resistance. However, there are insufficient data with which to evaluate adequately the utility of steroids to treat fat abnormalities. Moreover, testosterone and anabolic steroids may actually enhance hyperlipidemia. More data is needed on endocrine profiles and effects in people taking HAART, as well as on the use of exogenous steroids to treat fat abnormalities.
Some people with lipodystrophy have tried fat injections or silicon implants in the cheeks. The results do not appear to be particularly good; allegedly, fat implants have disappeared and silicon implants are said to have eroded, according to Grunfeld. There is no reason to think that these interventions will provide real benefit, but they may invoke a real risk of secondary problems such as infection.
Many issues remain to be resolved. The following are needed:
Newly emerging adverse reactions to drugs continue to be reported (e.g., cracked lips, dry skin, ingrown toenails), adding to the mass of anecdotal reports without clarifying associations. As a relatively new and poorly understood syndrome, body fat abnormalities are a reminder of how much remains unclear about HIV disease.
Although protease inhibitors are under suspicion as a factor in the development of body fat redistribution, more data are needed. For the time being, physicians are not recommending that people who are succeeding on a protease inhibitor-containing regimen discontinue or switch treatment, even when body fat changes emerge. Insufficient data exist to suggest that switching is appropriate, although research continues to pursue this possibility. Protease inhibitors continue to appear significantly beneficial for many people with HIV.
In 1996, after six months on a protease inhibitor, I noticed changes in my body. My legs, arms, hips, and buttocks all experienced a decrease in fat. I also started noticing fat accumulating in a buffalo hump and in what was then called a "crix belly." At the same time, my face was thinning, and in particular, I developed hollow cheeks.
Colleagues of mine, both males and females, began showing and reporting the same symptoms, or a similar syndrome.
At first, when I started complaining, my doctors considered it a cosmetic problem. The first question was always, "Are you in menopause?" (I am not.) Then, lab results showed an increase in blood values of cholesterol and triglycerides. I also had a biopsy of the fat on my neck, but it provided no insight. I documented the body changes in limbs and belly through photography. I felt a sense of helplessness even working through my doctors. Not only in our outpatient ward, but all over Hamburg, these doctor/patient interactions resulted in similar complaints and helplessness.
Then came the 12th World AIDS Conference in Geneva. All of a sudden, posters and oral presentations on the lipodystrophy syndrome were displayed prominently -- and then the doctors considered it more seriously.
Still, no solutions have been found yet. Doctors here are not enthusiastic about liposuction or about trying growth hormone, which is also frightfully expensive and not covered by health insurance. So far no one really has any documented experience with any treatments. A friend of mine in Munich takes growth hormone injections daily and has seen some results by regaining fat in his limbs. I've not tried anything yet myself.
From what I can tell, men and women are equally affected by this syndrome. I feel that the only thing that will help those of us affected is research, research, and more research. And international exchanges of our experiences can help. For example, ACT UP's LIPIDLIST helps a lot -- more self-empowerment by people living with AIDS, which is especially important for minorities.
''Positive" regards... --Petra
Hamburg, Germany
The LIPIDLIST is an Internet e-mail discussion list that allows people to discuss rumors, research, treatments, and shared concerns about fat abnormalities. Topics discussed on the LIPIDLIST range from laboratory tests and reports, to research results, to lipid-lowering drugs and other subjects mentioned elsewhere in this article. To subscribe, send e-mail to: listproc@critpath.org. In the body of the message, type: subscribe lipidlist <first-name> <last-name> (enter the actual names without brackets).
The CRIX-LIST is another e-mail discussion list, for people who are taking or considering taking indinavir (Crixivan) and other protease inhibitors. A variety of topics are discussed. For more information, refer to the website at crix.pinkpage.com.
Lipoproteins with a higher percentage of lipids, or those fatty ones that rise, have a lower density: low-density lipoproteins (LDL), commonly referred to as "bad" cholesterol because high LDL levels are associated with an increased risk of heart attack.
Lipoproteins with a higher percentage of proteins have a higher density, and fall out to the bottom in the spinning test tube: high-density lipoproteins (HDL), commonly referred to as "good" cholesterol because normal or high HDL levels are associated with a reduced risk of heart attack.
During the past few months, people have been gathering at meetings large and small, formal and informal, to discuss fat abnormalities. One of the most significant so far was held on October 26, 1998 when the Forum for Collaborative HIV Research (FCHR) sponsored a meeting specifically to discuss HIV-related body fat changes and metabolic abnormalities. (The FCHR is an organization dedicated to bringing together experts from industry, government, and academia to foster innovative HIV research.)
Participants included clinical researchers, physicians, treatment advocates, and representatives from the National Institutes of Health (NIH), the Centers for Disease Control and Prevention (CDC), the Food and Drug Administration (FDA), and the pharmaceutical industry. One of the goals of the meeting was to develop an interim case definition of metabolically associated syndromes for widespread use by those in the research and clinical arenas; another goal was to discuss designing a cross-sectional study of the prevalence of fat changes.
Although the goal of developing an interim definition was not met, participants assembled a list of abnormalities to be considered for inclusion in the definition. A subcommittee called the Design Working Group will use this list to develop a specific definition, which is intended to be an aid in the design of a prevalence study, as well as to assist other research efforts.
FCHR envisions a prevalence study that will look at people from different populations, at varying stages of HIV disease and with varied antiretroviral histories. The goal is to gain information about how the signs and symptoms of these syndromes may be related to one another, and to determine the causes. A separate working group will continue developing the protocol.
Other research protocols are at other stages of development. Both the AIDS Clinical Trials Group (ACTG) and the Terry Beirn Community Programs for Clinical Research in AIDS (CPCRA) will conduct large, multicenter studies of antiretroviral-naive people who begin HAART with a protease inhibitor, a NNRTI, or both. ACTG 384 will look prospectively at what happens with people who initiate efavirenz plus indinavir. The CPCRA FIRST study includes a metabolic substudy, which will evaluate changes in total cholesterol, triglycerides, waist circumference, and waist/hip ratio.
Both studies will evaluate lipid and glucose abnormalities, and body composition, collecting data on the incidence of both metabolic and body composition abnormalities. The studies will provide information on the features of the syndrome(s), as well as insight into how they may or may not be related to protease inhibitors. Since most participants will be antiretroviral-naive, these studies will not address the issue of advanced HIV disease as a potential cofactor.
Some studies are presently enrolling or underway. The SeronAIDS Lipodystrophy Syndrome Assessment (SALSA) questionnaire is a Serono-sponsored project intended to help monitor metabolic and body shape changes. At 40 sites, HIV positive persons and their healthcare providers will complete questionnaires. The patient will provide data on demographics, body shape, health habits, and personal and family medical history. The healthcare provider will provide more detailed information on medical history, clinical status (HIV data, laboratory values, etc.), and medications. To ensure anonymity, study participants will be assigned a patient identification code and will not attach their names to the surveys.
Two controls will be compared to each HIV positive person with metabolic or body shape changes (the "case" participant). One control will be HIV positive, matched for CD4 cell count and HIV viral load. The other control will be HIV negative, without body shape abnormalities, but matched for gender, ethnicity, and age. Data will be centralized and ultimately analyzed to assess differences between cases and controls, and other factors associated with body shape changes. Serono hopes that eventually the questionnaire will be used as a diagnostic and monitoring tool, as well as to help gather data on incidence and prevalence.
New York's Community Research Initiative on AIDS (CRIA) has begun two studies of metabolic complications. One study will follow 30 people just beginning protease inhibitor treatment over a three-month period, monitoring their glucose levels and ability to process sugar, in order to detect pre-diabetic conditions. Preliminary results are expected to be released soon. The second, 24-week study is evaluating the safety and efficacy of human growth hormone (HGH) in 25 participants as a treatment for body fat redistribution. This study is being conducted through St. Luke's-Roosevelt Hospital Center in Manhattan, in cooperation with Donald Kotler. Preliminary results may be available by the spring. For more information, call CRIA at 212-924-3934.
At SFGH, two studies are enrolling participants. One is a prospective study of the effects of the initiation of protease inhibitor therapy on glucose and lipid metabolism, and on body composition. Researchers will also evaluate the roles of viral, hormonal, and immunologic factors. Participants will be followed for approximately 18 months. The second, 24-week study will evaluate the utility of growth hormone and nandrolone for people with body fat redistribution (buffalo hump, benign symmetric lipomatosis, and/or truncal obesity). Data also will be collected on glucose and lipid metabolism and body composition. For more information about either study, call Kathy Mulligan, PhD, at 415-206-5882 or Joan Lo, MD, at 415-206-4430. (Information about other related studies at SFGH will be made available as the studies open; please call Mulligan or Lo.) Leslie Hanna is Acting Editor of BETA.
Babl, F.E. Metabolic effects of protease inhibitors explored. 38th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Diego, California. September 24-27, 1998. Abstract I-86.
Carr, A. and others. A syndrome of peripheral lipodystrophy (LD), hyperlipidemia and insulin resistance due to HIV protease inhibitors (PIs). 5th Conference on Retrovirus and Opportunistic Infections. Chicago, Illinois. February 1-5, 1998. Abstract 410.
Carr, A. and others. Pathogenesis of HIV-1-protease inhibitor-associated peripheral lipodystrophy, hyperlipidaemia, and insulin resistance. Lancet 1998 Jun 20;351(9119):1881-3.
Dong, K. Changes in body habitus in HIV (+) women after initiation of protease inhibitor therapy. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 177/12373.
Gervasoni, C. and others. Redistribution of body fat in HIV-infected women undergoing combined antiretroviral therapy: clinical, immunological, and metabolic analyses. 38th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Diego, California. September 24-27, 1998. Abstract I-93.
Hengel, R.L and others. Multiple symmetrical lipomatosis associated with protease inhibitor therapy. 5th Conference on Retrovirus and Opportunistic Infections. Chicago, Illinois. February 1-5, 1998. Abstract 407.
Hengel, R.L. and others. Benign symmetric lipomatosis associated with protease inhibitors. Lancet 1997 Nov 29;350(9091):1596.
Henry, K. Lipid abnormalities associated with use of protease inhibitors: prevalence, clinical sequelae and treatment. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 12319.
Hewitt, R.G. and others. Gemfibrozil effectively lowers protease inhibitor-associated hypertriglyceridemia in HIV-1+ patients. 38th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Diego, California. September 24-27, 1998. Abstract I-88.
Kotler, D. Alterations in body fat distribution in HIV-infected men and women. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 336/32173.
Lo, J.C. and others. Body shape changes in HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes and Retrovirology 19(3): 307-308. November 1, 1998.
Lo, J.C. and others. "Buffalo hump" in men with HIV-1 infection. Lancet 1998 Mar 21;351(9106):867-70.
Miller, K.D. and others. Visceral abdominal-fat accumulation associated with use of indinavir. Lancet 1998 Mar 21;351(9106):871-5.
Rohlmann, A. and others. Inducible inactivation of hepatic LRP gene by mediated recombination confirms role of LRP in clearance of chylomicron remnants.J Clin Invest 1998 Feb 1;101(3):689-95.
Ruane, P.J. Atypical accumulations of fatty tissue. 37th Interscience Conference on Antimicrobial Agents and Chemotherapy. Toronto, Ontario, Canada. September 28-October 1, 1997. Abstract I-185.
Staszewski, S. and others. A phase III, multicenter, randomized, open-label study to compare the antiretroviral activity and tolerability of efavirenz (EFV) + zidovudine (ZDV) + lamivudine (3TC), versus indinavir (IDV) + ZDV + 3TC, versus EFV + IDV (DMP 266-006). 4th International Congress on Drug Therapy in HIV Infection. Glasgow, Scotland. November 1998.
Torres, G. Treatment of dorsocervical fat pads and truncal adiposity with Serostim [recombinant human growth hormone] in patients with AIDS maintained on HAART. 12th World AIDS Conference. Geneva, Switzerland, June 28-July 3, 1998. Abstract 32164
DT 19990110
DOCN BE990108
Copyright © 1999 - San Francisco AIDS Foundation. Reproduced by permission. Reproduction of this article (other than one copy for personal reference) must be cleared through BETA: PO Box 426182, San Francisco, CA 94142-6182. Tel: 415 487 8060 Fax: 415 487 8069 URL: http://www.sfaf.org/beta E-mail: beta@sfaf.org
This information is designed to support, not replace, the relationship that exists between you and your doctor.