February 11, 2009
There's nothing like hearing the results of studies directly from those who actually conducted the research. In this interview, you'll meet one of these impressive HIV researchers and read her explanation of a study she presented at CROI 2009.
My name is Kathleen Mulligan. I'm at the University of California - San Francisco. This poster reports work that was done on behalf of the Adolescent Trials Network.1 It's a a survey of metabolic and morphologic abnormalities in behaviorally infected young adolescent women.
We wanted to see if there were disorders of glucose metabolism, lipids, fat distribution and inflammation that might be targets of subsequent intervention studies.
Kathleen Mulligan, Ph.D.
This is a cross-sectional study. We recruited HIV-positive young women into four different groups:
Clearly, most of these people were on triple nucleoside regimens, including abacavir [Ziagen].
The majority of the data were collected around 2005. So measurements, including fasting lipids, a two-hour oral glucose tolerance test, high sensitivity CRP as an index of inflammation. We did conventional anthropometry, whole body DEXA scanning. We also had extensive medical and antiretroviral histories, through chart review and interview, and questionnaires to get at lifestyle issues.
In addition to the HIV-positive groups, we recruited an HIV-negative group from the same age, racial and ethnic distribution, as the HIV positives. We wound up with a total of 61 HIV-negative subjects, and 173 HIV-positive participants.
Since these were behaviorally infected young women, the average duration of known HIV infection was relatively short, and ranged from under one year in the antiretroviral-naive group, up to a little under four years in the group on protease inhibitors.
What was the mean age?
The median age in both groups was 20. The upper limit was 24 and the lower limit was around 14 or 15. The group was predominantly minority, with 77 percent of the HIV-positive group being black or African-American.
Just to give you a snapshot of what this group looks like: over half of the HIV-positive young women had had at least one pregnancy. Close to two-thirds of them used drugs, and it looked like cannabinoids were the primary drugs of choice.
About 60 percent consumed alcohol. More than one third smoked cigarettes. Fewer than one third exercised regularly. A high proportion of people in both the HIV-positive and HIV-negative groups had family histories of type 2 diabetes and heart disease. So there were a lot of cardiovascular risk factors in this group, even before we started looking at biochemical and objective measurements.
In both the HIV-positive and the HIV-negative group?
In the positive and negative groups, yes.
So they were matched with these risk factors?
Correct. Although, as it turns out, a higher proportion of them HIV positive group had been pregnant, and I think a lower proportion exercised regularly, unfortunately.
As you might expect, the groups who were on antiretroviral therapy had evidence of more advanced disease. Even though current CD4 counts were similar, nadir CD4 count was lower in the groups on antiretroviral therapy, and peak HIV RNA levels were higher. A higher proportion of the groups on antiretroviral therapy had more advanced disease, by CDC staging.
In the NNRTI group, the use of efavirenz [Sustiva] and nevirapine [Viramune] was roughly equally divided. In the protease inhibitor group, almost 60% were using nelfinavir [Viracept]. Another 25% were on lopinavir [Kaletra], and then another 17% were on some other kind of ritonavir-containing regimen. The majority of the people in the no PI/no NNRTI group were using abacavir, as might be expected.
For me, the most striking finding in this group was the high prevalence of obesity. In both the HIV-positive and HIV-negative subjects, over 40 percent had a BMI (a body mass index) that put them in the category of being overweight or obese. This is consistent with what's seen in women of color in other surveys in the United States. So this was not surprising, but it's striking, and a bit frightening, to see it in this population.
When we looked at lipid profiles, not surprisingly, the HIV-positive groups who were on antiretroviral therapy had higher levels of triglycerides and cholesterol, lower HDL cholesterol, and higher non-HDL cholesterol.
Interestingly, when we compared indices of glucose metabolism, both fasting and two-hour glucose and insulin across the study groups, we saw no difference between the HIV positives and HIV negatives, and also no difference between the groups on antiretroviral therapy, compared with antiretroviral-naive HIV-positive subjects.
However, when we took those same measurements and sorted them according to BMI categories, the people who were overweight and obese tended to have higher -- significantly higher -- fasting and two-hour glucose levels, fasting and two-hour insulin levels, and then calculated HOMA score. Also, indices of lipid metabolism generally worsened with increasing body mass index. So this obesity played a real dominant role in our findings, of both glucose and lipid metabolism.
This would be true with HIV-negative women, as well, no?
Correct. That's exactly what you would expect to see in HIV-negative women. We also looked at high-sensitivity CRP as an index of inflammation. CRP was higher in the HIV-positive groups on antiretroviral therapy, and also higher in the women who were classified as being obese.
To summarize, we saw a high prevalence of obesity, of dyslipidemia and inflammation in this group of behaviorally infected adolescent women, despite a relatively short duration of HIV infection.
We concluded that dyslipidemia was associated with HIV infection, antiretroviral therapy, and increasing body mass index.
On the other hand, abnormalities in glucose metabolism were primarily associated with increasing body mass index, but not with HIV infection or antiretroviral therapy. Also, the major impact of body habitus on the metabolic outcomes seem to be consistent with that of garden-variety obesity, rather than alterations in fat distribution.
We saw a high prevalence of class risk factors, such as smoking, family history, inactivity. These findings point us toward a crying need, I think, for behavioral interventions in this group to reduce cardiovascular risk in this group of people who are probably going to be on antiretroviral therapy for their entire lifetimes.
Was any antiretroviral therapy associated with increased risk?
We looked at individual antiretroviral therapies, and our sample size was not large enough for us to say anything definitive about that.
In a similar study that we did in perinatally infected children in the PACTG [The Pediatric AIDS Clinical Trials Group], we found that ritonavir was associated with increased risk of both lipids and abnormalities of glucose metabolism.
The average time that these people were on therapy was less than four years?
Were you surprised by the data?
I was sobered by the data, not necessarily surprised. We know that obesity is a big problem in this population. I think it's a reminder, though, or a sneak preview, of what's to come -- that surveys in HIV-negative women show that they only get more obese with age, and that we have an opportunity here to intervene and at least stem the tide, and take measures to increase activity or make other lifestyle changes that might at least not put people at risk for the other comorbidities associated with a lifetime of obesity.
What about the surprising numbers in terms of alcohol? That's a pretty high number, 61%.
I was surprised by that, too. Remember that these were people who were infected through high-risk behavior, and so I guess we're seeing a cluster of behaviors that might contribute to the increased risk. Also, a third of them already smoke cigarettes.
But it's lower than the average, no?
Lower than the average, but they are still young. Perhaps we can intervene now and help them reverse some of those bad habits. I want to mention that even though we're seeing a high penetrance of obesity among HIV-infected women in the U.S., I don't think that it begins and ends there. I've seen studies coming out of Africa, in anti-natal clinics, for instance, that are showing an increasing prevalence of obesity. So I think we need to look at this group as a model for what might be seen in other groups, as well.
Thank you very much.