Fat Redistribution and Diabetes (by Larry Cohen)

People with HIV taking antiretroviral therapy have reported changes in their body composition including a redistribution of body fat. Researchers have begun to study these abnormal fat accumulations, but we do not yet understand this phenomenon. Generally, it is agreed that you should not discontinue, stop, or not take therapy because of these bodily changes, except under extreme conditions. There is no exact definition of what is an abnormal fat redistribution, and not all of the bodily changes are seen in all individuals reporting a redistribution of fat.

Some of the bodily changes reported are an increase in the size of the abdomen, sometimes referred to as "protease paunch", and a less frequently reported change, "buffalo hump," an enlarging of tissues at the back of the neck. Others report thinning of the legs, buttocks, and face. Women have reported an increase of the size of their breasts and hips. Usually there is not an increase in weight, however one study has reported an overall weight gain. The fat changes are not associated with a particular protease inhibitor and there were cases of fat redistribution prior to the advent of antiretrovirals. In addititon, individuals only taking nucleosides have developed buffalo hump.

The fat changes could be due to HIV disease, recovery from HIV, antiretroviral therapy, or a combination of all three factors. Dr. Donald Kotler of St Luke's Roosevelt Hospital, suggests fat redistribution may be more noticeable due to the advent of antiretroviral therapy, but it is not a new phenomenon. HIV was associated with weight loss prior to the existence of antiretroviral therapy and there were incidences of fat redistribution. As individuals stopped loosing weight while taking antiretroviral therapy, fat redistribution would be more noticeable.

Kotler suggests that fat redistribution may be a phenomena of recovery. AIDS could be the first time people suffering from such a life threatening disease recover from it. An HIV positive person is under incredible assault by HIV. Lowering viral load decreases the amount of HIV in the body, but there remains a memory and stress from the HIV assault. Kotler looked at 9 individuals who were taking antiretroviral therapy, seven of them had high urinary cortisal levels. Since stress is associated with increased cortisal levels, high cortisal levels may indicate that the body is still under stress— even after viral load is reduced. In addition, increased cortisal levels are associated with changes in body fat. Possibly, the more dramatic the change in viral load, the more dramatic the recovery and the more dramatic the fat redistribution. Furthermore, stress lowers testosterone levels and lower testosterone is a symptom of some people with fat redistribution.

Fat redistribution may be associated with stress and recovery, but a HAART regimen containing a protease inhibitor may exacerbate the problem. However, when the benefit of a low viral load and lessened disease progression are weighed against this phenomena, fat redistribution and metabolic changes such as triglyceride, cholesterol and glucose levels should be monitored. Experts recommend antiretroviral therapy should not be stopped unless circumstances are extreme. Some of the symptoms associated with fat redistribution may be associated with risk for heart disease and stroke. Dr Keith Henry briefly discussed these concerns in the May 2 issue of The Lancet. His group obtains a fasting lipid profile prior to antiretroviral therapy and then another profile 3-6 months after therapy. Based on these test results, they use the National Cholesterol Education Program Guidelines to treat abnormalities. The program emphasizes diet and exercise and can be found at the American Heart Association's web site at <www.amhrt.org>.

Some of Henry’s patients who have higher lipid or trygliceride levels were referred for treatment with drugs. 6 patients were treated with gemifbrozil for 3 months and then with atorvastatin. See the ritonavir/saquinavir section on page 28 where there is a discussion of drugs used for treatment of elevated cholesterol and triglycerides. Drugs used to treat elevated cholesterol may interact with protease inhibitors. Research links elevated cholesterol with risk of heart disease, but it may be unclear whether or not elevated triglycerides are associated with heart disease.

Frequency of Fat Redistribution. There were several reports on fat redistribution presented at the Retrovirus Conference. A problem in reporting frequency is that the studies examining incidence rates do not use a standard measure of abnormal fat redistribution, studies have not been conducted in a large number of patients, and to date the studies conducted have been retrospective, looking back at patients. The ACTG will be conducting prospective studies, looking at patients prior to their beginning protease inhibitors. A variety of studies will compare protease inhibitor regimens to non-protease regimens looking at potential changes in body composition, metabolism, weight and fat redistribution. They will also examine cardiovascular related changes and the effects of diet and exercise. They will examine factors that may effect changes including age, family history, and additional baseline measures. In the meantime individuals taking antiretroviral therapy should follow glucose, tryglicerides and cholesterol levels closely.

A group at Cornell University evaluated 72 patients receiving care and taking protease inhibitors for "protease paunch." They reported 11 patients (11%) had protease paunch at a mean of 18 months into therapy. A study from Australia looked at 116 patients taking at least one protease inhibitor and examined 32 patients who were protease inhibitor naive. They reported 72 patients (64%) developed an "abnormal accumulation of fat." All but one person was taking protease inhibitors. The mean time on therapy was 10 months.

The Australian group compared insulin, triglyceride, and cholesterol levels of patients with no abnormal fat redistribution with those who had a redistribution. The fat changes were associated with a mean weight loss of .51 kg relative to protease treated patients without fat redistribution and a mean fat loss of 5.5 kg in all regions except the abdomen. Patients with abnormal fat accumulation had significantly higher triglycerides, cholesterol, insulin, and c-peptide levels and insulin resistance scores. The authors suggest that protease inhibitors may play a direct role in fat accumulation. They report that the site of HIV protease enzyme has 12 amino acids that are 70% similar in their structure with a certain protein associated with lipids, a low-density lipoprotein receptor-like protein. The authors suggest that fat redistribution may be the result of an interaction between protease inhibitors and lipoprotein receptor-like proteins. However, they expressed uncertainty regarding this potential association.

A group from Atlanta reported 4 out of 4,000 studied patients developed buffalo hump. One female patient had an increase in breast size. All were taking indinavir. Two patients stopped medication with resolution of symptoms and one patient continued medication and the accumulation of fat resolved spontaneously. A group from Canada looked at buffalo hump in over 800 patients, two-thirds of whom were taking protease inhibitors. Seven out of the 800+ had buffalo hump. All were on protease inhibitor therapy and were NRTI experienced. The abnormal fat accumulation occurred 4 to 61 weeks after initiation of therapy. There was a weight gain of 3.5 kg. In contrast, a study from San Francisco found that three out of seven patients who developed buffalo hump were taking protease inhibitors. The other 4 were taking antiretroviral therapy without protease inhibitors. They reported triglycerides were elevated, but cholesterol levels and cortisal levels were normal.

The National Institutes of Health conducted a study of 37 HIV infected men and the protease indinavir. Investigators said they chose indinavir because most of their patients were taking it, but they also stated that they do not believe fat redistribution is unique to indinavir. 13 of the patients were not taking indinavir and did not have abdominal complaints. 14 were taking indinavir for more than 6 months and had no abdominal symptoms, and 10 had been taking indinavir for more than 6 months and had abdominal complaints. Fat located in the midsection can accumulate in two different regions. In the intra-abdomen called visceral fat or near the surface of the skin, the subcutaneous layer. Men have more visceral fat in the intra-abdomen area than women.

The researchers examined visercal and abdominal fat by using a computerized tomography scan. The ratios of visceral to abdominal fat for the three groups were .44, .60, and .73, respectively. Individuals who were not taking indinavir had the least amount of visceral fat and patients taking indinavir with symptoms of fat redistribution had the highest percentage of visceral fat. There was no difference in body weight between the three groups. Obese HIV negative patients with visceral fat accumulations frequently have more of an impairment of glucose and lipid metabolism and a higher incidence of heart attack and stroke than patients with subcutaneous fat accumulations.

Diabetes. There were two reports at the Retrovirus Conference on diabetes. A report from Johns Hopkins found 8 out of 417 (1.9%) individuals followed developed diabetes in a median time of 53 days. There were 5 moderate to severe cases and 3 mild cases. Two of the three mild cases remained on protease therapy with diet control and had a resolution of diabetes. All 5 moderate to severe cases were treated with insulin or an oral agent. Two of the eight developed ketoacidosis and required hospitalization. The National HIV Telephone Consultation Service reported that out of 1300 calls, 15 cases of diabetes were found. Cases were associated with all 4 currently approved protease inhibitors. The mean duration of therapy was 8.8 months and protease inhibitor therapy was continued for 10 out of the 15 individuals. People with HIV should monitor their glucose, especially if they have a family history of diabetes.