November 19, 2009
Until the mid-1990s in the U.S., packing on a few extra pounds seemed like a good thing: It helped protect a person against the dangers of wasting and HIV's destructive effects on the immune system. But today, for people on potent HIV treatment, those extra pounds may no longer help -- in fact, they may reduce the immune benefits of HIV meds, a new study suggests. In this interview, Nancy Crum-Cianflone, M.D., M.P.H., the study's lead author, explains the intriguing findings.
I'm Nancy Crum-Cianflone from the Infectious Disease Clinical Research Program, which is centered in Bethesda, Md. I am located, though, at the Navy Medical Center in San Diego, Calif. We performed a study entitled, "Obesity Among HIV-Infected Persons: Impact of Weight on CD4 Cell Counts," which was recently presented at the IDSA meeting [47th Annual Meeting of the Infectious Diseases Society of America] in Philadelphia, Pa.1
Nancy Crum-Cianflone, M.D., M.P.H.
Since we already know that obesity leads to negative health consequences in the general population -- such as heart disease, diabetes and certain cancers -- we wondered if obesity could have additional health consequences among HIV-positive persons. Specifically, we set out to test the hypothesis of whether obesity could affect their immune systems as measured by their CD4 cell counts.
We evaluated over a thousand HIV-infected persons as part of the U.S. Military HIV Natural History Study that's been conducted at seven sites across the U.S. Our participants have known dates of HIV seroconversion and have sequential information regarding their weights and CD4 counts over time. We evaluated the relationship between weight changes and CD4 counts over time by a variety of different statistical methods.
We found that since the availability of potent HIV medications, HIV-infected patients who are obese were more likely to have lower mean CD4 counts than those whose weight was appropriate. We also found that after beginning HIV medications for the first time, obese persons gained fewer CD4 cells than overweight persons did.
In summary, our study found that those who are obese had lower CD4 counts over time than those who had normal weight. We feel that this data suggests that lower CD4 counts may be another adverse consequence of being obese.
We hope that our study encourages HIV-infected patients to achieve and maintain normal weights. In addition, we hope our study encourages HIV care providers to become cognizant of the potential negative effect that obesity may have on their patients' immune systems.
How did you define "appropriate weight" in this study?
We measured each participant's height and weight and calculated a body mass index (BMI). This calculation can be found on the Internet: If you simply put in "BMI calculator," one will pop up, and a person can calculate, based on their weight and height, what their body mass index should be. It's based on the body mass index that the CDC, the U.S. Centers for Disease Control and Prevention, has defined as what a person's appropriate weight should be. If your body mass index is between 18.5 and 25, they would consider that normal. Anything above 25 would be overweight or obese.
Tell me a little bit more about the types of people who enrolled in this study. How old were they? Were they men or women?
We studied over 1,000 HIV-positive persons who participated in our U.S. Military HIV Natural History Study. This is a study that's been going on for over 20 years, in which participants who have military benefits voluntarily enroll and get followed as part of an observational natural history study. They are followed over time, and we collect various information including weight and CD4 counts. It was from this population that this study was conducted.
The study population's mean age was 29 years. Because it was drawn from a military consortium, 96% were male; as such, these data can be best generalized to men. Regarding race, 44% were Caucasian, 41% African American and 15% other races. One interesting characteristic of our population is that we have a fairly nice distribution of the races, particularly the 50-50 split between white and black persons. Many studies only focus on Caucasians, but we have a fair number of African-American participants in this particular study.
Did you notice any difference in the findings between people who were black or white?
We did not split people based on whether they were white or black to examine if there was an association between obesity and CD4 count based on race. We certainly included race in our adjusted models. When we looked at the relationship between weight changes over time and CD4 changes over time, there were many things that could potentially influence that. Of course, age and race could influence that, so we accounted for all these other variables in that model. But we didn't then divide our model up [by race], partly because we didn't want to divide it too finely since our study size was limited to 1,000 people.
Were the people involved in this study more overweight than the general population?
I don't believe they're more overweight or obese than the normal population, but I think the changes we're seeing in our HIV-infected patients over time are reflective of what we're seeing in the general population. There have been many publications and announcements noting that Americans are increasingly overweight or obese. I think that among the HIV-infected patients, similar trends are occurring. We didn't do a direct comparison, but I don't think our numbers suggest that they're heavier than the general population; what we have seen is they're definitely heavier over time, just like the general population is.
To my understanding, the military has a very well developed system of testing and ultimately treating people for HIV, so I'm going to guess that these people were diagnosed pretty early on?
That is correct. One of the things that our study was able to do that most previous studies have not -- in terms of trying to look at how weight influences CD4 count -- is that we had, as part of our study, only people who were documented seroconverters, meaning patients who had a documented last negative HIV test and a first positive HIV test. We actually restricted our analysis of these data to just those who were documented seroconverters because we felt that including people at various stages of HIV might bias some of the conclusions.
Our patients had a median time from last negative HIV test to first positive test of just over a year. So these were, in fact, very early diagnosed people. We thought it was very important to adjust for duration of HIV infection in the models used to examine the relationship between obesity and CD4 counts.
What were the average CD4 counts of these people when they were first diagnosed?
The average was around 520 cells. So you can see that it was a very early diagnosed cohort.
We used the DHHS [U.S. Department of Health and Human Services] guidelines to direct therapy in our patients. We ran our analysis accounting for when they were off therapy and when they went on therapy, since this can influence both your weight and your CD4 count. So we accounted for antiretroviral use in our models as well.
The study also compared, did it not, the impact of weight on CD4 count in the pre-HAART [highly active antiretroviral therapy] era to the HAART era? You found very different results from the two, right?
That's very correct. Thus far in our conversation, I've really focused on the HAART era, since we have these drugs available in the developed world. I've also focused on these results because there are very little data about the impact of weight on your CD4 count during the HAART era, or since the availability of these new medications.
As you mentioned, we found very different results for the pre- versus post-HAART eras. We found in the pre-HAART era, like many other studies did, that as you increase your weight, you had less reduction over time in your CD4 count. It appeared that as a patient had higher weight, it was probably beneficial. We thought that partly had to do with: If you had additional nutritional stores and then you developed an opportunistic infection or an AIDS-related crisis, there was some protection in the extra weight you had. As many studies have shown, before the introduction of highly potent HIV drugs, having extra weight on may have been a good idea -- it actually preserved your CD4 count.
I think many patients, and perhaps some providers, today still live in that paradigm where patients feel that it's best to achieve high weights to protect themselves from wasting and adverse health consequences. But one of the things that our study wanted to address was whether or not such an idea was still in play in the post-HAART era, and whether or not being overweight or obese continued to be advantageous.
As we've already mentioned, patients who were obese actually had poorer CD4 counts, suggesting that with the availability of these new HIV medications, gaining excessive weight is no longer needed and no longer a good idea. It potentially has negative consequences on your CD4 count.
We, of course, also know that being obese has negative consequences in terms of high blood pressure, high cholesterol and heart disease. We know that from the general population. Our study was really trying to change the mindset of HIV-infected patients and their providers to strive for a normal weight and not try to gain excessive amounts of weight, because the incidence of wasting has fallen and we don't see a protective effect of having higher weights anymore.
Do we have any idea what the reasons might be as to why people who are significantly overweight are at a greater risk of not having their CD4 count go up as much?
Unfortunately, our study did not examine the reason that this could have occurred. However, that being said, one could consider a couple of hypotheses.
I think the first hypothesis -- again, which we did not examine in our study, but may be an area of future research -- is that obese patients may have lower drug levels of their antiretrovirals, or their HIV medications, in their system. Most studies that are done on the proper dose of a medication are done among normal-weight persons.
However, we know that people who are obese have different distributions of those medications and different times to breakdown of those medications than those who are normal weight. So, possibly, some of the effects that we're seeing are that obese patients just do not have high enough drug concentrations of the HIV medications in their system, and that that may lead to poorer CD4 count responses. Again, that hypothesis would require further study to confirm that as the reason.
I think the second possible reason is that, in the general population, there's a lot of data that suggest obese patients have high inflammatory markers, which has become a very interesting area in HIV as well. These measures -- like CRP, or C-reactive protein -- seem to be elevated among obese versus normal-weight persons. There's some evidence coming out that high inflammatory marker levels may have adverse consequences on the immune system, including the CD4 counts. Again, this study did not examine the inflammatory marker levels, so this would be a study in the future to show if that could potentially be the reason for this relationship.
But those are two potential hypotheses to be considered. Certainly, we're interested in doing further study to try to see more precisely why obese patients have lower CD4 counts.
People who are in the military have the benefit of regular HIV testing so they can find out whether they're HIV positive before their CD4 count drops. However, research in the U.S. has indicated that many people do not actually get diagnosed until their CD4 count is much, much lower. Do we have any idea how much your findings might apply to people who were diagnosed at low CD4 counts instead of high CD4 counts?
I think that's a great point to make, that this is an early diagnosed cohort who had free and open access to medical care. I think our study results show that, in such a population, maintaining a normal weight versus becoming overweight or obese is preferable.
How does this apply to other populations? I think our study really cannot draw any conclusions. Certainly, in populations that don't have antiretroviral medications available, they may be more in line with what we saw in the pre-HAART era, in which weight may be actually a little beneficial.
But I think it would be interesting to repeat a study like ours in a resource-poor setting to then re-examine the impact of weight on CD4 counts, as these patients may not have antiretrovirals. Then, as antiretrovirals are introduced in those populations, one could see if there's a changing effect of how weight impacts CD4 counts.
Alright, let's try to break your study findings down in practical terms. Let's say you were diagnosed with a CD4 count of 250. You've been on treatment for a while. Your CD4 count is up to 500 or so, but you are about 40 pounds overweight. What do these study results mean to you?
If you're on treatment and you have access to these medications, and if you are 40 pounds overweight or you currently are obese, these study findings should encourage you to join a weight reduction program -- obviously through the guidance of your physician -- to try to get to a normal weight.
I think one thing that's important to note about our study findings is that we did not simply look at their weights at the beginning of diagnosis and how that influenced their CD4 count. We actually looked at their weights and how they changed over time, to look at the relationship with the CD4 count.
The data suggest that if you have the availability of HIV drugs, the most beneficial thing that you can do is to achieve a normal weight, and you might have a positive impact on your immune system.
Again, our study findings are preliminary. We would like other study cohorts to examine this question to see if our data can be replicated. I also want to emphasize that there are many other reasons for that patient to lose 40 pounds. If you're 40 pounds overweight, you're likely at a higher risk of diabetes, heart disease and many other medical consequences that we're seeing rising rates of in our HIV-infected patients.
There is an increasing focus on the reasons why our HIV-infected patients are having these comorbidities, but perhaps one of these reasons is that our patients are increasingly overweight or obese. Perhaps by maintaining normal weight, some of these other comorbidities could be reduced.
Should a doctor who's prescribing first-line therapy consider prescribing a regimen that research has shown may boost CD4 count a little bit more than others? I know that with Selzentry [maraviroc, Celsentri], research has suggested that it can increase CD4 count a little bit more than some other HIV meds. Would that be worth considering as an option if you're going into treatment significantly overweight?
I think that's a really interesting question. Again, our results are preliminary, and I don't think we can draw any conclusions linking our results to the management of patients. I know that there are different regimens that can potentially boost CD4 counts higher than other regimens. I don't think we're to a point yet that one can say, "If you're treating an obese patient, choose this regimen over another," based on CD4 counts. I'd like to see further research done in this area; then, future studies to look at potentially treating various weight patients with various drugs might be a consideration.
I would also like to see how these drugs' pharmacokinetics work in our obese patients a bit more. Perhaps, by adjusting dosages in obese patients, we could correct the issue with CD4 counts. But again, I think further research has to be done before we can make any clinical judgments in terms of impact on decision making.
We've focused pretty heavily on, at least in the post-HAART era, the impact of obesity on CD4 count. Is the impact of being underweight non-existent in the post-HAART era, or is it just a lesser one compared to being obese?
We found in our study that being underweight continued to have a negative impact in the HAART era. It's been shown in the pre-HAART era, by other studies, that if you are underweight, your HIV disease progresses faster and you have lower CD4 counts. But we also found similar findings in the HAART era, that those people also have lower CD4 counts [if they are underweight].
I think it's a very important point to go back to: We really emphasize achieving and maintaining normal weights as based on the body mass index. Underweight is not optimal, but we also feel, from our research, that obese is not optimal either. Just like in the general population, we really are encouraging our HIV-infected patients to achieve and maintain what would be considered a normal weight.
How did studies like this -- which are ultimately so important, so common-sense -- not get done five or more years ago?
We do a lot of clinical-based research where I work, and we've noticed that patients who are HIV positive are increasingly overweight and obese. It struck us that we're no longer thinking about wasting; now we're dealing with blood pressure issues and diabetic issues, and a lot of this is driven by their weight. Our patients were telling us that, at least early in the epidemic, HIV-infected patients were told to gain excessive weight because that might help protect them. Plus, some HIV-infected patients think that if they look too thin, their diagnosis may be revealed, whereas if they look heavy, people wouldn't expect them to be HIV positive.
We heard these comments from our patients and we began to realize that perhaps at the beginning of the epidemic, HIV-infected patients were becoming overweight or obese because of these issues. We wondered how this was impacting people today, because we saw blood pressure issues and diabetic issues. We began to wonder, "Is it really good for their HIV if they're overweight?" We knew it was probably OK in the pre-HAART era, but how about now, when we're seeing these other complications?
That's what led us to do this study. I'm fortunate to work in a large consortium of military beneficiaries who have collected this information on early diagnosed patients, which allowed us to do this research and try to answer this specific question.
You've laid out some pretty clear further directions for research that can help us better understand the link between weight and CD4 count. Are you continuing with this particular study, or planning any additional ones to further flesh out these results?
I really appreciate that question. We're in the process of disseminating these results a bit wider, and then we are looking into possible areas where we could explore, going back to the two hypotheses I brought up earlier to try to explain why obese patients are having lower CD4 counts. Is it an issue of drug levels? Is it an issue of inflammatory markers? Or is it something completely different?
Our group is interested in continuing this research in terms of trying to explain our findings. I would highly encourage other groups to look at this question in their cohorts, both in the developing and in the developed world, to see if this finding is present in other cohorts.
As you mentioned, in cohorts that maybe do not get diagnosed so early and maybe who don't have access to medications, how do these findings apply? Our study did not examine those populations, but I think it would be important to look at the same question in some alternate populations.
In the meantime, it seems like the moral of this story is like the moral of so many other stories: For the time being, at least, it appears to be yet another reason why it's good to keep your weight within relatively normal bounds, if you can.
I think that's a great concluding statement.
Dr. Crum-Cianflone, thank you so much for taking the time to explain your study.
This transcript has been lightly edited for clarity.