One of the greatest drags on the success of potent antiretroviral (ARV) therapy has been the fear of metabolic complications associated with these medications. Disfiguring body shape changes including the loss of fat in the face, as well as unhealthy cholesterol and triglyceride levels and pre-diabetes are troublesome counter-balances to the euphoria that arose when these drugs arrived and people stopped dying. Even as ARVs have become more user-friendly -- less pills, less frequent dosing, less diarrhea and nausea -- the specter of metabolic problems can still overshadow these advances, leading those in need of therapy to hesitate when ARVs are recommended. For those already on treatment, metabolic disorders may prompt a change in therapy or lead to the prescription of even more medication and can raise the volume of the little voice that says it is okay to skip doses.
A major frustration for people living with HIV and their health care providers has been a lack of information regarding the cause of metabolic problems during HIV infection and ways to prevent and treat them. The field of metabolic complications of HIV and its therapies is relatively young and much has been learned during a short period of time but some conclusions have been reached with little supportive data. Below is a list of some of the most common of these metabolic complications myths. Myths that emerged in a data vacuum and that even people in the "HIV-know" often still accept. Fortunately, over the past few years a slew of studies has painted a clearer picture of these changes and together suggest that some of our closely-held beliefs about the risks for metabolic complications have been wrong. Understanding that these assumptions are no longer valid, and why, is essential if people living with this infection and their health care providers are to make informed decisions about their care.
Like many myths, this one is based on a truth that has been stretched to extremes. People taking protease inhibitors can see an increase in their belly fat, both the deep down fat that surrounds our internal organs and the surface, pinch-an-inch fat so abundant in our land of amber waves of grain. But protease inhibitors hold no monopoly on an ability to expand trunk fat. Studies of efavirenz (Sustiva) have shown that people taking this non-nucleoside also tend to have increases in belly fat. In fact, increases in waist size have been seen in studies of every HIV regimen in which body shape has been objectively measured. For example, in a federally funded AIDS Clinical Trials Group (ACTG) clinical trial called study A5142 comparing the popular HIV medications lopinavir/ritonavir (Kaletra) and efavirenz, trunk fat was seen to increase in participants regardless of which drug they were assigned. Similarly, a Bristol-Myers Squibb sponsored head-to-head study of efavirenz and another protease inhibitor, atazanavir (Reyataz), in patients who were starting HIV therapy also found that both drugs when combined with zidovudine and lamivudine (Retrovir and Epivir, also Combivir) tended to increase abdominal fat over time. Interestingly, a recent Abbott Laboratories study that looked at using lopinavir/ritonavir by itself (i.e. monotherapy) in patients started on this protease inhibitor and zidovudine/lamivudine found that these patients experienced increases in belly fat to the same extent as a control group of patients who were maintained on zidovudine/lamivuine and efavirenz. Therefore, it looks like both protease inhibitors and, at least, the non-nucleoside efavirenz can lead to gains in belly fat.
A problem for most all of these studies is that they rely on a special type of scan called a DEXA to measure abdominal fat. This scan, commonly used to also measure bone density, cannot tell the difference between the deep and surface fat. So, one therapy could be causing accumulation of the deeper fat while another could be associated with surface fat. CT and MRI scans, however, can differentiate deep and surface fat. Unfortunately, we do not have much data regarding the relative changes in fat in deep and surface fat for most HIV regimens. Clearly, more studies need to be done on other regimens, including those that contain newer drugs, and should use CT scans when possible but one thing is clear: when it comes to increasing belly fat, protease inhibitors are not unique.
As if a big spare tire was not bad enough, some people taking HIV medications also experience loss of fat of the arms, legs, and face. The image of an apple-shaped body with skinny limbs is a frightening one that further turns many people off to HIV therapy. However, it has become clear that most people on HIV medications do not develop this body shape. In fact, a couple of studies of people starting a variety of HIV regimens have found that for most people limb and belly fat tend to increase or decrease together. That is, if someone experiences a gain in belly fat then they are more likely to also experience a gain rather than a loss in limb fat. In one study, only a quarter of people experienced a loss of arm and leg fat while gaining abdominal fat.
Most studies suggest that overall fat gain is a major problem for HIV-positive people. As in the general population, being overweight and obese is common. In a study of HIV-infected patients receiving care in Philadelphia, rates of being overweight and obesity were more of a problem than weight loss. As people with HIV infection look to decades of living with their infection, the problem of obesity is likely to take its toll since obesity increases the risk of diabetes, heart disease and death.
The profound loss of fat within the arms, legs and especially the face among people on HIV medication cocktails that was seen in the mid-1990s was quickly associated with one drug, stavudine (Zerit). The link between such disfigurement and this drug was so obvious that use of stavudine in the U.S. and Europe quickly fell and is now rarely prescribed (unfortunately, stavudine is still commonly used in developing nations as it is easy to make and, thus, cheap).
The drop in stavudine use was followed by a dramatic reduction in new cases of severe fat loss of the face and limbs. However, over time some doctors and their patients noticed a slower but undeniable depletion of fat in these same areas of the body. But, as these changes were slow to develop and DEXA, CT, and MRI scans are not routinely performed in clinics to measure and follow body fat changes, it was unclear whether these changes were real and, if so, what caused them. What was clear was that these people seemed to be losing limb and face fat but had never taken stavudine.
Some answers came from clinical trials that incorporated DEXA scans into their design. One study done several years ago by the ACTG found that people starting HIV therapy who took the protease inhibitor nelfinavir (Viracept) were more likely to lose limb fat -- as measured by DEXA scans -- than those taking efavirenz, even when the other medication taken was limited to zidovudine/lamivudine (Combivir). This meant that people on zidovudine/lamivudine were experiencing fat loss and that this was accelerated with nelfinavir use. Another study comparing zidovudine/lamivudine with tenofovir/emtricitabine (Truvada) when both were taken with efavirenz found that there was a progressive loss of fat among those assigned to zidovudine/lamivudine while those taking tenofovir/emtricitabine gained limb fat over time.
The ACTG study A5142 looking at people new to HIV medications also performed DEXA scans before HIV medications were initiated and then at regular intervals after starting the drugs. This was a large study of almost 750 people who were assigned to one of three different study treatments: a.) lopinavir/ritonavir plus two nucleosides, or b.) efavirenz plus two nucleosides, or c.) lopinavir/ritonavir plus efavirenz alone without nucleosides. Those taking nucleosides could use only lamivudine plus either stavudine, zidovudine or tenofovir (Viread). The study is very important as efavirenz and lopinavir/ritonavir are two of the most popular medications used to treat HIV infection yet, had never been compared before. The results of this trial have shaken the field of body shape changes during HIV treatment. Those taking stavudine had, as expected, the greatest loss of limb fat and those taking tenofovir had the least. But, zidovudine fell in between. This alone indicated that some people experienced limb fat loss even when not receiving stavudine and that zidovudine was capable of doing this to a greater extent than many had thought. In addition, the study found that no matter what nucleoside was used, efavirenz was more likely to cause significant fat loss compared to lopinavir/ritonavir. That is, efavirenz seemed to add to the fat loss that was associated with the nucleosides. The good news is that few of those on tenofovir lost significant amounts of limb fat at 96 weeks of study, even when on efavirenz, so fear of fat loss should not be a major concern for those who are taking or considering use of tenofovir plus efavirenz (two of the three medications in Atripla).
Taken together, these data indicate that fat loss of the arms and legs is not limited to stavudine and that other drugs can also produce these changes. Zidovudine appears to be worse than tenofovir (or abacavir [Ziagen]), albeit it is not as bad as stavudine. Additionally, efavirenz seems to be able to dial-up the fat loss effect of nucleosides to a greater extent than lopinavir/ritonavir. Unfortunately, there is not much information regarding face fat from any of these studies.
This myth falls into the same category as the belief that going out with wet hair will increase your risk for a death of a cold and that too much time spent self-pleasuring can wreak havoc on your visual acuity. A remarkable number of intelligent men and women arrive at their clinic visits complaining of increases in belly fat, and are frustrated that endless sit-ups have done nothing to reduce their mid-body girth.
Sit-ups, when done properly, can increase strength in the abdominal muscles. This leads to firmer muscles and an increase in core strength but will not melt away fat in that one area. Fat is lost when more energy is expended than taken in. While sit-ups require energy, they do not preferentially draw that energy from the deposit of fat cells found in those love handles. A better approach is to combine sit-ups with aerobic exercises that require heavy breathing and sweating for prolonged periods of time like running, cycling, stair climbing, rope jumping, etc. Small studies have shown decreases in abdominal fat when HIV-positive people followed a program of aerobic exercise and weight lifting several times a week.
Diet can also play a role here and a smart approach would be to limit simple sugars and the highly caloric fats that make up most of the so-called comfort foods of our society. For most people dietary modification need not be very complicated and can be summed up with a recommendation to greatly increase daily intake of fruits and vegetables, the latter preferably raw or lightly steamed. These are foods that are not packed with excess calories, contain cholesterol-lowering fiber and are filling -- leaving less room for the fatty, super-size-me foods at the root of many of our health problems.
In addition to eating like a Buddhist monk and joining a gym there are other interventions that have been studied to reduce excess fat. Unfortunately, few have panned out. Growth hormone is an injectable agent that has been found to reduce fat in the belly and buffalo hump and some people have benefited from this therapy. However, this is an expensive drug that is not usually covered by insurance carriers for the treatment of excess fat. Also, at the doses studied for the treatment of excess fat, growth hormone has been plagued by a number of troublesome side effects including worsening glucose levels, muscle and joint aches, and feet swelling. Interestingly, exercise is known to increase the body's own production of growth hormone.
Testosterone and other androgens ("male hormones") have also been studied as treatments for fat accumulation in people with HIV infection. These hormones, like growth hormone, can pop fat cells but in another ACTG study were found to preferentially reduce the surface fat and not the deep fat that made for most of the enlargement of the belly. Androgens can also worsen limb and face fat loss. Therefore, although beloved by many, the data suggest that androgens may do little to reduce abdominal girth and can aggravate loss of fat beyond the trunk.
A few drugs used for the treatment of diabetes have also been studied for fat accumulation, including metformin, rosiglitazone, and pioglitazone. Most of the data informing the use of these drugs in people with HIV come from small studies. Suffice to say that their effects, if present at all, seem to be mostly limited to those with diabetes or a pre-diabetes condition. The underwhelming study results and the toxicities of these medications have diminished any enthusiasm for dedicated use of these drugs to treat fat changes in people with HIV infection.
Take a survey of people living with HIV or even their docs and ask whether HIVers have higher cholesterol levels than those without HIV. Chances are most would respond that those who are HIV-positive would, on average, have higher levels than those who are uninfected. Actually, at least a couple of studies have found that people with HIV infection tend to have lower levels of LDL cholesterol, the "bad" cholesterol that has been strongly linked to heart disease, than people in general; this finding holds even when including those who are on HIV medications.
This does not mean that those with HIV infection have a better lipid profile than uninfected folks. A major problem is that levels of the "good" cholesterol, HDL cholesterol, are also lower in HIV-positive people. HDL cholesterol has been found to offer protection from heart disease and a low level is an independent risk factor for cardiovascular problems. Exercise and modest alcohol (not just red wine) intake can safely raise HDL cholesterol in some people. In addition, a little appreciated fact is that certain HIV medications also raise HDL cholesterol levels. The non-nucleosides efavirenz and nevirapine [Viramune] and the protease inhibitor atazanavir alone or in combination with ritonavir [Norvir] and most all other protease inhibitors that are boosted with ritonavir have all been found to raise HDL cholesterol levels.
Triglyceride levels, though, are a different story. Triglycerides are broken down in the body from fat and can be found floating free in the blood or in a complex with other lipids and proteins in the form of cholesterol. The more triglycerides in the cholesterol complex, the more dangerous it is in terms of cardiovascular risk with LDL cholesterol having more triglycerides than HDL cholesterol. Fasting triglyceride levels are, on average, higher in people with HIV infection and increases further with HIV therapy. While in some people the level of triglycerides can skyrocket to very concerning levels (greater than 500 mg/dL) most people with HIV infection have levels that are high but not alarming. In addition, by itself the level of triglycerides measured in the blood is not considered as nearly big a risk for cardiovascular disease as high LDL or low HDL cholesterol. Most all HIV regimens can raise triglyceride levels. The ritonavir-boosted protease inhibitors are a bit worse in this regard than efavirenz, and most studies suggest that lopinavir/ritonavir and fos-amprenavir/ritonavir (Lexiva/Norvir) may raise triglyceride levels a bit more than other commonly-used boosted protease inhibitors, but the clinical significance of these modest differences is not clear.
Overall, the data suggest that people with HIV may be at greater risk of cardiovascular problems like heart attacks due to their low HDL cholesterol levels and possibly increases in LDL cholesterol and triglyceride levels during HIV therapy. Additionally, there may be other factors such as inflammation caused by the virus that can lead to chemical changes in the body that can prompt clogging of the arteries. However, it is almost certain that smoking adds much more to the risk of cardiovascular disease than these other HIV-related factors and that of all the things a person with HIV infection could do to survive and thrive, beyond taking HIV medications when necessary, the most significant is to stop smoking.
Clinicians and their patients do not tolerate ambiguity well. Gaps in knowledge of a disease demand to be filled and when the research data come up short it is difficult not to extrapolate. In the 25 years since the AIDS pandemic ignited, much has been learned about HIV and the crowning achievement of the scientists, clinicians, and advocates dedicated to this disease has been the dramatic reversal of the lethality of this disease. However, in HIV, as in medicine in general, it has been difficult to not jump to conclusions when data are conflicting or just plain not in existence.
In the case of metabolic complications of HIV and its treatments, we have learned to learn. New investigations have revealed the accuracy and inaccuracy of previous assumptions and allow us opportunities to better choose among our options. The trick is we have to be willing to let go of our old beliefs and embrace findings that rigorously challenge these concepts. The old mantra that knowledge=power still holds, but we have to accept that better knowledge=even more power.
Dr. Wohl is an Associate Professor of Infectious Diseases and Co-Director of the AIDS Clinical Trials Unit at the University of North Carolina. Metabolic complications associated with HIV infection and the nexus between HIV and incarceration are his major areas of research interest. He can be reached via e-mail at firstname.lastname@example.org.