The Lipodystrophy Syndrome: More Questions than Answers

By Bill Valenti, M.D.

One of the major topics at the 12th World Conference on AIDS in Geneva, Switzerland was the discussion of the metabolic changes and changes in body configuration seen in the so-called Liposdystrophy Syndrome (LDS) that has been associated with HAART or highly active antiretroviral therapy. While these changes have been attributed to the protease inhibitor component of HAART, the cause of the syndrome and the role of protease inhibitors were the subject of considerable debate and speculation.

The discussion that follows reviews information from the 12th World AIDS Conference in Geneva and the scant, available literature on the subject. While the cause and management of the syndrome are still unclear, it is still important to use our clinical judgment to help patients make the best treatment choices.

Definition and Components
An overview of the Metabolic Lipodystrophy Syndrome or Liposdystrophy Syndrome (LDS) was provided by Dr. David Cooper of the HIV Medicine Unit, St Vincent's Hospital, Sydney, Australia, who discussed 5 issues: definition of the syndrome and its components, the metabolic and body composition changes, a proposed mechanism for the syndrome, work to be done, and what to do "in the meantime." Later in the week, others, including Dr. Donald Kotler from the US, and several poster sessions discussed various aspects of the syndrome and some alternative hypotheses. There is no official case definition for HIV lipodystrophy yet. This can create confusion since some investigators lump together the serum lipid and glucose abnormalities often seen with HAART with changes in body habitus; others separate the physical findings from the laboratory or metabolic abnormalities. While it is true that most patients with abnormal fat redistribution also have changes in their lipid profile, a causal relationship between the two has not been established. At this point, the lipodystrophy syndrome has several components not all of which are present in any one patient.

According to Cooper, the principle features of LDS are:
1. Peripheral fat wasting or lipodystrophy. This occurs in the face, neck, arms, legs or buttocks. An associated feature is a prominent superficial venous pattern of the arms and/ or legs due to the loss of subcutaneous fat. The facial appearance can be quite striking, "as though the patient had AIDS, " according to Dr. Cooper.

2. Central adiposity. This has also been called central truncal obesity. This feature consists of abdominal visceral fat accumulation, not subcutaneous fat. Patients with this variant will have an increased ratio of visceral abdominal tissue (VAT) to total abdominal tissue (TAT). The table below demonstrates these findings in protease inhibitor-naive patients, patients receiving protease inhibitors (PI) and patients receiving protease inhibitors with truncal obesity (PI + belly).

The basis for comparison in this table is the protease-naive group of people with HIV. Compared to this group, patients on protease inhibitors have an increased VAT/TAT ratio; with the highest ratio being in the protease experienced patients with body composition changes. The VAT increases in each of the groups from 106 cm2 in the PI naive group to 202 in the PI + belly group.

3. Breast hypertrophy. This is another manifestation of central fat deposition and is seen in women. It may be significant enough to warrant a change of dress and bra size.

4. Dorsal fat pad. "Buffalo hump", and lipomatosis: Presumably these are other signs of abnormal fat redistribution

5. Ectodermal dysplasia. This is a related problem which includes dry skin, cracked lips, loss of body hair, premature coronary artery disease, and gynecomastia in men.

6. Metabolic abnormalities. Diabetes mellitus is due to insulin resistance, not insulin deficiency. Metabolically, insulin resistance or frank diabetes mellitus may also be present. The hyperlipidemias are a significant part of this syndrome and have been particularly noticeable and troublesome for physicians and their patients. Lipid lowering drugs may be effective in many patients, but it is not known if their role extends to prevention or reversal of the syndrome.

Dr. Cooper reported ongoing body composition studies comparing patients on protease inhibitors to healthy HIV positive, protease inhibitor-naive patients. Looking at a variety of parameters, patients receiving protease inhibitors have decreases in fat-free mass, total fat and arm fat, leg fat, and trunk fat when compared to protease inhibitor-naive patients.

Cooper emphasized a number of times during his review that this is a progressive process with subtle changes present early in PI use. Interestingly, patients may not be aware of these body composition changes early on. Others patients have reported that "my body is changing" as early as 4 weeks after the start of HAART. The most precise way to assess these changes is to do body composition studies. Bioelectrical impedance (BIA) is a helpful assessment tool for ambulatory settings. A fairly typical clinical picture is increased total body fat and decreased lean body mass. More formal studies using dual-energy x-ray absorptiometry (DEXA) have also been done, but this is primarily a research tool.³ However, experienced clinicians can also recognize the syndrome in a clinical setting with a good patient history and physical examination. BIA studies early in the course of a patient's medical care may help establish baseline measurements that can be followed over time.

Dr. Cooper noted that there have been no reports of pancreatitis with this syndrome, although other investigators reported several cases in patients with elevated triglyceride levels.

Hypotheses
The hypotheses for this syndrome are only tentative at this point, but Cooper pointed out that understanding the syndrome's pathogenesis should lead to treatment strategies and to the design of protease inhibitors that do not cause the syndrome. Cooper's group presented an intriguing hypothesis in which his group compared the genetic sequence of the HIV-1 protease to a number of gene banks to search for matches. There were two matches, each with about 60% homology to two proteins involved in human fat transport and metabolism. One is cytoplasmic retinoic acid binding protein type 1 (CRAPB-1) and the other is low density lipoprotein-receptor-related protein (LRP). They speculated further that regulation of lipid metabolism is altered in several ways. This competitive inhibition of CRAPB-1 is thought to be responsible for the peripheral fat wasting and the hyperlipidemia seen with the syndrome. The inhibition of LRP's lipid binding capability exacerbates the hyperlipidemia, which contributes to the central fat deposition, insulin resistance, and, in susceptible individuals, type 2 diabetes. These pathways are discussed in detail by Cooper's group.¹

Dr. Donald Kotler provided a different view on LDS in his presentation in Geneva [Abstract 32173]. The question at hand is whether this is a new syndrome or an old one, he said. Kotler believes that lipodystrophy is not a new phenomenon. His experience, similar to many clinicians', is that some cases of LDS have occurred in patients who have never taken protease inhibitors but were treated with nucleoside RTIs or NNRTIs. He also pointed out that LDS is not unique to HIV; the kind of fat redistribution associated with HAART is also seen in HIV-negative patients with a variety of other genetic lipodystrophy disorders, including the pediatric syndrome called "Syndrome X." His line of reasoning suggests that LDS is a secondary consequence of suppressing chronic HIV infection, rather than as a primary side effect of drug therapy.

More recently, Spanish researchers Martinez and Gatell have stated that protease inhibitors are non-specific and, therefore, may alter proteins including insulin and its substrates.⁵ The drugs may cause a reduction in hepatic insulin catabolism which would lead to primary hyperinsulinemia and a resulting increase in fat. Hyperinsulinemia would eventually lead to insulin resistance, causing the inhibition of lipogenesis and stimulation of lypolysis. This, in turn, would cause an increase in the plasma-free fatty acids and a following increase in hepatic triglyceride synthesis. Hepatic glucose output would also follow, inducing a compensatory insulin hypersecretion. The overstimulation of beta-cells could result in the emergence of type two diabetes mellitus. The authors note that the progressive loss of subcutaneous fatresulting in the development of lipidostrophycould be due to the inhibition of the antilipolytic effect of insulin once insulin resistance has developed.

What Next Clinically?
Cooper made the important point that first we need to overcome our denial that this syndrome even exists. In the meantime, we should monitor the situation in our patients, he said. However, this syndrome brings up a whole new series of questions to be addressed in clinical settings beyond merely monitoring it. Some of the clinical unknowns are the potential for reversibility of the syndrome, the advisability of changing drugs in the setting of lipodystrophy, and the treatment of some of the lipid disorders associated with the syndrome.

In symptomatic or advanced disease, the benefits of therapy outweigh the liabilities. The risk-benefit analysis of protease inhibitors in asymptomatic patients, on the other hand, is not known, according to Dr. Cooper. However, many clinicians and patients are opting for so-called protease sparing regimens by using non-nucleoside drugs such as nevirapine, delavirdine, and efavirenz in place of protease inhibitors.

Some clinicians have begun to study the possibility of reversal of the syndrome if and when PI treatment is stopped, but there is no clear answer on this yet. Since the syndrome may consist of more than one metabolic and clinical entity, it needs to be further defined before this question can be answered clearly. While changing drugs may reverse some of the lipid abnormalities, there is no convincing evidence that changing drugs will reverse the body shape changes associated with lipodystrophy.

Some of the choices for management of LDS include changing protease inhibitors (PI), avoiding PIs altogether, treatment to reverse changes in body shape, and treatment of metabolic abnormalities, including increased triglycerides and cholesterol. Liposuction is generally not an alternative because the truncal fat distribution is not subcutaneous, but visceral. The exception is the dorsal fat accumulation ("buffalo hump"), which is more superficial. However, limited experience with liposuction suggests that it reaccumulates after liposuction.

It is not clear if the syndrome is more likely to be associated with one protease inhibitor or another. One small study from Australia² suggested a change of PIs may be helpful, but the debate is far from over nor are the results conclusive. The investigators reported that switching LDS patients from indinavir or ritonavir/saquinavir to nelfinavir led to improved triglycerides but no change in cholesterol. In addition, 11 of 21 subjects reported some improvement in body habitus. However, this observational report was potentially biased because it was small, unblinded, and uncontrolled. On the other hand, it does suggest that clinical judgment is still important here. While changing drugs may not change body habitus significantly, changing drugs in patients with abnormally high cholesterol and triglyceride levels may be indicated. The other choice involves the use of lipid lowering drugs because of the potential for development of coronary artery disease.

Another clinical choice involves the use of protease-sparing regimens, which received a lot of attention at this conference. While Cooper reemphasized that he would not recommend changing a regimen that was showing immunological or virologic benefit, changing PIs to non-nucleoside reverse transcriptase inhibitors (NNRTIs) is still an option for some patients. Regimens containing non-nucleoside drugs instead of PIs were shown to be equally efficacious to PI containing regimens in treatment-naive patients at this conference. Protease-sparing regimens offers another choice as a starting regimen for patients and physicians.

Dr. Cooper indicated that the syndrome has not been associated with non-nucleoside drugs, but there is not universal agreement on this point either. Some clinicians have suggested that, if Kotler's hypothesis is true, then the increased use of non-nucleoside drugs might unmask the syndrome as well.

There was considerable speculation regarding specific treatments for this syndrome. On poster by Torres et al reported a limited experience with use of growth hormone of LDS in 3 patients.⁴ All 3 patients had complete or partial resolution of buffalo hump with subjective improvement appearance. However, there were no measurable changes in triglycerides or cholesterol and inconsistent changes in BIA measurements.

Summary and Recommendations
The lipodystrophy syndrome appears to be, at least in part, a new version of an old disease. Many of the features of LDS such as wasting (especially facial and peripheral fat wasting) and increased triglyceride levels, have been seen before the PI era in patients with AIDS. The widespread use of PI drugs may unmask or exacerbate some of these features. The lack of a uniform case definition makes it difficult to truly characterize this disease. Treatment up to now has also been problematic because it is directed toward reversal of one or more of the signs and symptoms of LDS.

We do not know the best choice for management of LDS, but for now, treatment will need to be individualized depending on the needs of the individual patient. While we are waiting for these data and standards of care, I suggest we start by following Cooper's advice and overcoming our denial that the syndrome exists and try to recognize and characterize the syndrome in our patients. Beyond that, we need to use our clinical judgment to address the metabolic abnormalities such as diabetes and increased cholesterol and triglyceride levels. The choice to change drugs vs. treating the metabolic abnormalities will depend on how well the drugs are working. My experience with drug-experienced patients is that most would rather continue on a regimen that results in undetectable viral load and stable or elevated CD4 counts. In those patients, continued follow-up and treatment of metabolic abnormalities with lipid lowering drugs is a choice. In other patients, especially in patients with less antiretroviral experience, change of antiretrovirals may be a way of correcting hyperlipidemias.

The decision to treat with HAART will also need to be addressed with our patients. As Dr. Cooper said, the risk-benefit of HAART in symptomatic patients with AIDS comes out in favor of treatment. For asymptomatic patients, I believe we have sufficient evidence to support continued treatment with HAART for asymptomatic patients, whether we use protease inhibitors or other drugs. Also, concerns over the possible development of LDS are not necessarily the reason to delay therapy. A better choice is HAART with careful monitoring of key metabolic parameters along with baseline and ongoing BIA measurements. Whatever the choice, it is important to remember the major advances in treatment over the past three years have improved outcome for patients and made HIV an ambulatory illness and decreased death rates for many patients . It is important to remember that in this new treatment era, HIV infection can no longer be considered untreatable.

1. Carr A, Samaras K, Chisholm DJ, Cooper DA. - Pathogenesis of HIV-1-protease inhibitor-associated peripheral lipodystrophy, hyperlipidaemia, and insulin resistance. Lancet 1998 Jun 20;351(9119):1881-3.

2. Duncombe, C, Bloch, M, and Austin, D. - Reversal of hyperlipidemia and lipodystrophy in patients switching therapy to nelfinavir. Abstract 12287. 12th World Conference on AIDS, 1998.

3. Engelson , E, Kotler, DP, Tan, SY et al ] - Altered body fat distribution in HIV infection: Regional body composition measurements by whole body MRI and DXA scans. Abstract 32181. 12th World Conference on AIDS, 1998.

4. Torres, R. - Treatment of dorsocervical fat pads and truncal adiposity with serostim (recombinant human growth hormone) in patients with AIDS maintained on HAART. Abstract 32164. 12th World Conference on AIDS, 1998.

5. Martinez, E and Gatell, J, - Metabolic Abnormalities and Use of HIV-1 Protease Inhibitors" Lancet. 1998; 352: 821.