Up to Half of HIV-Infected Patients on Effective HAART Still Ultimately Develop Neurocognitive Impairment, Study Finds

A Discussion With Igor Grant, M.D.

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 his explanation of a study he presented at CROI 2009. After his explanation, he then answers several questions from the audience. This discussion was moderated by Judith Currier, M.D.

Igor Grant: The CHARTER study was designed to take a detailed look at the question of whether the brain effects of HIV are as well controlled as are the other medical complications of HIV in the context of modern treatments.1

Mortality and many other kinds of HIV disease-related variables have declined with modern treatments. But there have been some suggestions that this is not so true of the brain disease that's associated with HIV. The CHARTER study was designed to answer, or at least address, this question.

Igor Grant, M.D.
Igor Grant, M.D.

The CHARTER study consists of detailed examinations of almost 1,600 patients who were attending clinics at six different university medical centers [in the United States]. The data were collected in a systematized way and then analyzed.

The findings are as follows: First, there is still a fairly high rate of neurocognitive impairment in HIV-infected persons who are receiving antiviral treatment. What do I mean by "neurocognitive impairment"? This is a disturbance in functions such as learning, remembering, planning, making decisions, attentional abilities, concentration; those kinds of skills are under "neurocognitive."

In addition, a subset of these patients received brain-imaging studies, including studies of the structure of their brains and whether there might be the presence of abnormal signals in the brain. It was found that about a third of these patients -- most of whom, as I say, were receiving treatment -- had some evidence of brain disease, based on brain imagining.

A third variable that was looked at is peripheral neuropathy symptoms. Peripheral neuropathy symptoms include burning, tingling, pain and other kinds of discomfort -- primarily in the feet, but it also can affect the hands and arms. About half, or a little bit more, of the patients in CHARTER had these problems. In this case, they were related, to some degree, to the kinds of specific treatments they received, such as a type of drug called the "D" drugs [didanosine (ddI, Videx) and stavudine (d4T, Zerit)].

Coming back to the brain involvement, we also tried to look at [whether there are] some things about the disease or the person that particularly predict who is at risk for these neurocognitive complications. The findings were that one important predictor was something called nadir CD4 count. Nadir CD4 count refers to the lowest level of CD4 that a person has experienced at some point -- not what it is now, but the lowest at some point in time. That was associated with likelihood of cognitive impairment.

Put another way, people who never had severe immunosuppression, and whose treatment also was effective in eliminating virus from plasma, those people were least likely to have brain impairment.

The implications of the study could be that there may be events that perhaps occur early in the course of infection, such as severe immunosuppression, that may later rebound, but that still leave a legacy -- an effect, a reverberation on the brain. If that were true, that would raise the question of whether treatment guidelines might need to be adjusted in order to begin earlier, more aggressive treatment. Again, the study wasn't designed to answer that question. That's an implication that may be raised. Thank you.

Reporter #1: This study is a community study, essentially. A lot of the people in the study, if I read the abstract correctly, had viral loads greater than 50 copies/mL. Can you talk a little bit about how viral load, as opposed to the CD4 count, affects your results? Is it more likely, with a higher viral load, that people will have dementia?

Igor Grant: This is kind of a complicated issue. We've looked both at plasma viral load and viral load in the cerebrospinal fluid. And it looks to be the case that if there is persistent viral replication in the cerebrospinal fluid, that is more likely to be associated with the presence of these neurocognitive impairments.

What CHARTER found, in particular, was that there is a group of patients in whom plasma viral load is undetectable. But if you use a very sensitive assay, you can still detect virus in the cerebrospinal fluid. And in those people, neurocognitive impairment was somewhat more likely.

But the relationships are complex, to say the least. It's not possible just to kind of say, "Well, if this is high, therefore you're going to have more neurological problems."

I think the main finding from CHARTER, though, is that 40 to 50 percent of people who are receiving very good treatment still have these neurocognitive complications. So there's some kind of divergence between the process going on in the brain and what's going on in the rest of the body. That's a serious issue.

Judith Currier: Could you just say a little bit more about whether these would be the types of impairments that the patient, or their family, would notice? Or they are only picked up on sensitive tests? Just to put into perspective what kind of impairment you're talking about.

Igor Grant: Right. Well, as I say, something like 40 or 50 percent have some kind of impairment, based on extensive testing. Probably 25 percent have test impairments that the person themselves may not be aware of, or families may not routinely pick up. And then the rest are impairments that may not be profound, but they are enough that they affect day-to-day functioning in some manner. For example, they may affect ability to manage complex medication regimens or perform work at the same level as the person was used to. I would say, roughly speaking, of the impaired, maybe half or so are asymptomatically impaired, and others have more serious impairment.

By the way, the prevalence of very severe dementia, what we call HIV-associated dementia, is actually down very low, at least in the CHARTER cohort; it's like 2 percent. It's about 2 percent who have such cognitive impairments that their lives are shattered, in effect. That used to be a much higher proportion in the pre-treatment era, and in the early '80s, certainly.

Reporter #2: Could you talk a little bit about how well specific drugs penetrate the CNS, and how that might affect cognitive impairments? We know that some drugs in a regimen penetrate very well; others do not. But I haven't seen a good sense as to a listing of what goes in and what doesn't go in.

Igor Grant: Right. There's been quite a lot of interest in this topic. In fact, Dr. Scott Letendre, Ron Ellis and other investigators in our group have been looking at that in some detail. As a general rule, the NNRTI drugs fairly well penetrate into the CNS, whereas most protease inhibitors are poor penetrators. Drugs like zidovudine [AZT, Retrovir] are somewhere in between.

Dr. Letendre and his group have looked at the relationship of neurocognitive impairment and what are called CPE (central nervous system penetration-effectiveness) scores. And there is a modest association: That is, people on better-penetrating drugs first of all are more likely to have undetectable viral load in the cerebrospinal fluid, and also, they are somewhat less likely to have cognitive impairment. But I would say the effects are there, but not gigantic. It's definitely an effect, but. ...

Reporter #2: What about the penetration of the two newest classes, the entry inhibitors and the integrase inhibitors? Is there any sense as to how well those penetrate?

Igor Grant: I don't know how good those data are. I don't have an answer to that question. That's still under investigation.

Judith Currier: What does smoking do to cognitive impairment?

Igor Grant: We've actually looked at that as a covariate. Most of the people we see are still in their 40s and 50s. There are very few older people. So smoking was not a predictor of cognitive impairment. But that's not to say that as the population ages, and we start looking at those age-by-smoking, by-disease interactions, that that may not show up there.

Reporter #3: Are there any other things that might show up as being modifiable for people who are dealing with some sort of cognitive impairment, or possibly small dementia?

Igor Grant: Yes. That's a very important question. In addition to HIV disease variables themselves, there are a number of other factors that can contribute to the likelihood of cognitive impairment. One is substance abuse. Not within CHARTER so much -- we didn't have a huge number of [substance abusers] -- but in other studies it's been shown that methamphetamine abuse increases the likelihood of cognitive impairment in HIV-infected people. So that's one important factor.

Another one is hepatitis C coinfection. Hepatitis C, in some studies -- particularly a study we're involved with in China, where people were HCV positive but HIV negative; they had more cognitive impairment. In some of the American studies we see, again, this increasing risk of cognitive impairment in people who are both hepatitis C and HIV infected. So, those are two examples.

Other possibilities are that people who have had neurodevelopmental or other kinds of neurological injuries earlier -- unrelated to HIV -- may also have increased risk, although that hasn't been as well studied so far.

Reporter #4: You did some brain scan work, I think, in this population. Not everybody, but a proportion. Can you tell us what that investigation found? Is there serious atrophy in brain tissue? How is it correlated?

Igor Grant: What we have found is, first of all, if you do careful, morphometric measures -- that is, being able to measure the thickness of the gray matter of the brain, for example, or the volume of white matter, and so forth -- there seems to be a reduction in the volumes of these compartments in the brain as [HIV] disease progresses. Another thing one sees is that abnormal signal -- which probably reflects small areas of inflammation -- increases with HIV disease. These are also related to the degree of cognitive impairment.

The amount of brain shrinkage, and both corticol and white matter loss, and also loss of deep gray matter -- basal ganglia, for instance -- all of these effects are seen in HIV. And they do have a relationship to cognitive impairment.

Another kind of imaging data that CHARTER looked at is called spectroscopy, which is looking at certain kinds of metabolites in the brain that are thought to reflect certain processes in the brain. For example, there's a metabolite called N-acetylaspartate (NAA), which is thought to reflect the integrity of the neurons themselves. That signal goes down with HIV disease.

There's another marker called myo-inositol, which is thought to reflect inflammatory processes, at least in part. That signal tends to go up. There are relationships between likelihood of cognitive impairment and those kinds of brain imaging changes, as well.


  1. Heaton R, Franklin D, Clifford D, et al. HIV-associated neurocognitive impairment remains prevalent in the era of combination ART: the CHARTER study. In: Program and abstracts of the 16th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2009; Montréal, Canada. Abstract 154.

Download the audio file