Studies look at brain and cognitive changes in people with HIV as they age

This article is more than 7 years old. Click here for more recent articles on this topic

People with HIV often show persistent signs of cognitive impairment and abnormalities in brain structure despite suppressive antiretroviral therapy (ART), but they do not appear to experience accelerated decline compared to HIV-negative people as they age, according to research presented at the 2017 Conference on Retroviruses and Opportunistic Infections (CROI) last month in Seattle.

HIV-associated neurocognitive disorder remains a poorly-understood co-morbidity in HIV-positive people. While frank AIDS dementia is now rare among people who receive effective antiretroviral treatment, more subtle cognitive problems – some of which may only be revealed by specialised testing – remain common. Cognitive decline is a concern as the HIV population ages; currently more than half of people living with HIV in the US are over age 50.

Ageing and cognitive impairment

Hamza Coban of the University of San Diego and colleagues looked at the relationship between ageing and neurocognitive performance over time, comparing changes in older and younger people who had been on ART for at least two years – at which time problems related to active, uncontrolled HIV replication should have been resolved.

The researchers analysed people in the AIDS Clinical Trials Group (ACTG) Longitudinal Linked Randomized Trials (ALLRT) cohort, which includes participants from seven ACTG randomised trials of first-line ART.



Lowest of a series of measurements. For example, an individual’s CD4 nadir is their lowest ever measured CD4 count.

virological suppression

Halting of the function or replication of a virus. In HIV, optimal viral suppression is measured as the reduction of viral load (HIV RNA) to undetectable levels and is the goal of antiretroviral therapy.

cognitive impairment

Loss of the ability to process, learn, and remember information. Potential causes include alcohol or drug abuse, depression, anxiety, vascular cognitive impairment, Alzheimer’s disease and HIV-associated neurocognitive disorder (HAND). 

longitudinal study

A study in which information is collected on people over several weeks, months or years. People may be followed forward in time (a prospective study), or information may be collected on past events (a retrospective study).

magnetic resonance imaging (MRI)

A non-invasive, non-x-ray diagnostic technique that provides computer-generated images of the body's internal tissues and organs.

Participants underwent annual neurocognitive assessment using tests measuring different aspects of cognitive performance and memory. Overall performance was summarised using Z-scores across the four tests (NPZ-4) standardised against normative results from the general population of the same age.

The cohort included 3313 people, mostly men, starting antiretrovirals for the first time. A quarter were aged 30 or younger, about a third each were in the 31-40 and 41-50 age groups, 10% were aged 50-60 and 2% were over 60. Two-thirds had some education beyond high-school level (grade 12); 8% had co-infection with hepatitis C virus (HCV), which can also affect neurocognitive function.

At the time of the first neurocognitive test, more than 90% had suppressed viral load (< 200 copies/ml). Just over half had a current CD4 count above 500 cells/mm3, but the median nadir (lowest-ever) count was 221, indicating that many had advanced immune suppression before starting ART.

Initial testing found that 42% showed impairment on the Hopkins Verbal Learning Test and 39% showed overall NPZ-4 impairment (average score -0.23). In the cohort as a whole neurocognitive impairment declined over time after starting ART, from 23% meeting the definition at the first evaluation to 13% at the last evaluation.

But the odds of neurocognitive impairment increased with age, by 18% for each decade older. Being older than 31-40 years at ART initiation and HCV co-infection were significantly associated with poorer neurocognitive test results, but there was no significant link with first-line antiretroviral drug class.

At two or more years after ART initiation, older age remained a significant risk factor for neurocognitive impairment, and age-related impairment was seen despite viral suppression and despite improvement in the cohort as a whole, the researchers concluded.

They suggested that causes of age-related neurocognitive impairment might include ART worsening common age-related conditions such as diabetes, hypertension and abnormal blood lipids, and possibly more central nervous system toxicities in older people. Future studies, they recommended, should look at the effects of inflammation, other co-infections such as syphilis and cytomegalovirus, obesity and cardiovascular risk factors.

Changes in HIV-positive vs HIV-negative people

Coban's study showed that cognitive decline was associated with ageing in the ALLRT cohort, but this typically occurs as people age in the general population. How does progression of cognitive decline in people with HIV compare with that of HIV-negative people? James Cole of Imperial College London and colleagues aimed to answer this question.

This study analysed longitudinal neuroimaging and neuropsychological data from HIV-positive participants in the European COBRA (Co-morbidity in Relation to AIDS) collaboration and a group of demographically similar HIV-negative people, looking at whether successfully treated HIV is associated with accelerated age-related changes to brain structure and function.

The analysis included 134 HIV-positive people at centres in Amsterdam and London who were on ART with HIV RNA < 50 copies/ml for more than a year. More than 90% were men and the mean age was 57 years. Again, the current CD4 count of 646 cells/mm3 was much higher than the nadir of 185. The control group included 79 HIV-negative people of a similar age and gender distribution.

Magnetic resonance imaging (MRI) and a neuropsychological evaluation including tests of language, memory, executive function, motor function and processing speed were performed at baseline and two years later.

Baseline MRIs showed that people with HIV had less brain grey matter volume and abnormal white matter microstructure compared to HIV-negative people, as well as poorer cognitive function.

But there were no notable differences in age-related changes in the HIV-positive and HIV-negative groups. Both groups showed declines in neuroimaging measures. HIV-positive people lost 0.82% of their brain volume per year while HIV-negative people lost 0.77%, not a significant difference.

Measures of cognitive function did not change much overall. A measure of global cognition (T-score) increased a bit in both groups, by +0.8 in the HIV-positive group and +0.5 in the HIV-negative group. Attention was the only measure to show a significant difference in the rate of change between the two groups, rising in the HIV-positive group and falling in the HIV-negative group to make their scores more similar.

HIV-positive people with suppressed virus on combination ART "had abnormalities in measures of brain structure and function at baseline," but "there was no difference in the dynamics of these measures over time between [HIV-positive] and HIV-negative controls," the researchers concluded. "Cognitive performance did not decline over two years."

"Although we previously found evidence for increased brain age in this cohort of people living with HIV," they continued, "this analysis does not find evidence for brain aging to be accelerated over time" in people with continued viral suppression on ART.

Cognitive changes in women

Brain function and structure changes in women with HIV may not be the same as those seen in HIV-positive men, who historically have had a higher socioeconomic status and higher education levels.

Leah Rubin of the University of Illinois in Chicago and colleagues compared cognitive trajectories of HIV-positive women with viral suppression on ART, HIV-positive women with poorly controlled HIV and HIV-negative women in the Women's Interagency HIV Study (WIHS). They hypothesised that virally suppressed women would do better than women with poorly controlled virus, but worse than uninfected women.

Between 2009 and 2015 a total of 932 WIHS participants underwent neurocognitive testing, including measures of learning, memory and attention, at baseline and then every other year.

Of these, 239 were HIV-positive with viral suppression, 392 were HIV-positive without viral suppression and 301 were HIV-negative. The groups were demographically similar. The average age was approximately 45 years, about two-thirds were black and half had a high school education or less.

The virally suppressed women reported continued ART use over four years with better than 90% adherence, while in the non-suppressed group 44% used it intermittently with 74% adherence. The median viral load was 48 copies/ml in the suppressed group and 360 copies/ml in the non-suppressed group. Current CD4 counts were 657 and 437 cells/mm3 in these two groups, while nadir counts were 244 and 170 cells/mm3, respectively.

Overall, HIV-positive women had significantly lower baseline T-scores for global neuropsychological performance, as well as for memory, attention and learning, compared to HIV-negative women.

Among women with HIV, those with viral suppression had higher learning, memory and motor skill scores than non-suppressed women, but global performance was essentially the same, and non-suppressed women actually did better on a couple of measures such as attention and fluency.

Having an AIDS diagnosis and lower CD4 nadir levels were associated with lower scores in some areas. Spending more follow-up time with a suppressed viral load was associated with higher scores in most domains.

Over time, women in all three groups experienced declines in global performance, memory, attention and learning. Women with HIV saw a decrease in processing speed while HIV-negative women improved – the only area of improvement seen in any group. There was no consistent pattern in the steepness of declines in different domains across the groups.

"Longitudinal findings confirm persistent cognitive impairment despite continued viral suppression," the researchers concluded. "Patterns of group differences indicates persistent vulnerability in attention, learning, memory and fluency, and increased vulnerability in motor skills over time despite optimal suppression among HIV-positive women."

Brain volume and cerebral vessel disease

The anatomical changes in the brain that may contribute to impaired cognitive function in people with HIV are not well understood.

Ryan Sanford of the Montreal Neurological Institute at McGill University and colleagues longitudinally assessed brain volume in 46 HIV-positive people with well-controlled virus and 31 demographically matched HIV-negative individuals. About half were men, the average age was approximately 50 years and they had 14 years of education on average. In the HIV-positive group the current and nadir CD4 counts were 641 and 200 cells/mm3, respectively.

Participants completed two sessions of neuroimaging and neuropsychological testing done approximately two years apart. The assessment covered six cognitive domains with eight standard tests. Standardised Z-scores were calculated for each test, as well as an overall neuropsychological summary score (NPZ-8).

At both visits HIV-positive participants had significantly worse cognitive performance than HIV-negative people, with lower scores for NPZ-8 and the executive function, attention, working memory and processing speed components.

However, there was no significant changes in NPZ-8 score over time in either group, nor a difference in rates of change between the two groups. The HIV-positive group saw a greater improvement in memory and a trend toward a larger decline in executive function.

Likewise, tensor-based morphometry revealed significantly reduced subcortical volume in the thalamus, caudate, putamen, globus pallidus and mid-brain in HIV-positive people at both visits. But no major changes in brain volume were observed over time in any region in either group, nor different rates of change between the groups. Lower nadir CD4 count had a weak correlation with smaller brain volumes.

"No evidence of on-going brain injury or overall cognitive decline were detected," the researchers concluded. "These findings support the hypothesis that cognitive and structural brain differences in HIV-positive patients most likely occur during the period of untreated infection, suggesting a possible neurocognitive benefit from early combination ART initiation."

Finally, Dominique Costagliola of the Sorbonne and INSERM in Paris and colleagues looked at the prevalence of cerebral small vessel disease (CSVD) in people living with HIV compared to HIV-negative people age 50 and older.

CSVD – defined by abnormalities in white matter, silent brain infarction (obstructed blood supply) or micro-bleeds – is a major cause of future vascular events such as stroke, cognitive impairment, frailty and poor survival, the researchers noted as background.

This analysis included 456 HIV-positive people in the French ANRS EP51 MICROBREAK cohort and 154 HIV-negative people. About 80% were men and the median age was about 56 years. Participants in the HIV-positive group were on ART with suppressed viral load for at least a year. The median and nadir CD4 counts were 655 and 195 cells/mm3; they did not have HCV co-infection. Most cardiovascular risk factors were seen more often among people with HIV, including hypertension and abnormal blood lipid levels, as was regular alcohol use; more than 40% in both groups smoked.

In this cross-sectional study each participant received a single MRI, and results were analysed by two experienced neuroradiologists who were blinded to their HIV status.

CSVD was detected in 52% of HIV-positive and 36% of HIV-negative participants, a significant difference (adjusted odds ratio [OR] 2.3). Severe CSVD was observed in 19% and 14%, respectively (adjusted OR 1.6).

But the impact of HIV differed according to age. HIV-positive people under age 54 were about five times more likely, and those in the 54-60 age group were nearly four times more likely, to develop CSVD compared to HIV-negative people; however, for people over age 60 there was essentially no difference (adjusted OR 5.3, 3.7 and 1.2, respectively).

In addition to HIV, being aged over 60, hypertension and nadir CD4 count below 200 cells/mm3 were also associated with greater risk of CSVD, the researchers reported.


Coban H et al. Impact of advancing age on cognition in HIV+ persons on a first suppressive regimen. Conference on Retroviruses and Opportunistic Infections (CROI 2017), Seattle, abstract 343, 2017.

View the abstract on the conference on website.

Download the poster from the conference website.

Cole JH et al (van Zoest R presenting). Longitudinal analysis shows no evidence for accelerated brain ageing in treated HIV. Conference on Retroviruses and Opportunistic Infections (CROI 2017), Seattle, abstract 32LB, 2017.

View the abstract on the conference website.

Download the poster from the conference website.

Rubin LH et al. Cognitive trajectories over 4 years among HIV+ women with optimal viral suppression. Conference on Retroviruses and Opportunistic Infections (CROI 2017), Seattle, abstract 350, 2017.

View the abstract on the conference website.

Watch the webcast of this presentation on the conference website.

Sanford R et al. Longitudinal assessment of regionally specific brain volumes in treated HIV+ patients. Conference on Retroviruses and Opportunistic Infections (CROI 2017), Seattle, abstract 397, 2017.

View the abstract on the conference website.

Download the poster from the conference website.

Moulignier A et al (Costagliola D presenting). Cerebral small-vessel disease in HIV-infected patients well controlled on cART. Conference on Retroviruses and Opportunistic Infections (CROI 2017), Seattle, abstract 75, 2017.

View the abstract on the conference website.

Watch the webcast of this presentation on the conference website.