Effects of age, HIV, and HIV-associated clinical factors on neuropsychological functioning and brain regional volume in HIV+ patients on effective treatment

Due to combination antiretroviral therapy (cART), human immunodeficiency virus (HIV) infection has become a manageable, chronic illness and many HIV+ patients reach advanced age. In Europe, the proportion of individuals presenting several years after being infected is high among people older than age 50 (ECDC/WHO 2017). Also, a significant proportion of new diagnoses are being made in this age group (Tavoschi et al. 2017). Studies suggest that HIV may continue to affect the brain even in the presence of cART. The impact of the infection on the aging brain structure and cognitive functioning is not yet fully understood. In this article, we depict the research evidence showing the influence of HIV infection, age, and HIV-associated clinical factors on neuropsychological performance and brain volume in HIV-positive (HIV+) Polish men receiving cART.

The introduction of cART has decreased the incidence of HIV-associated neurological complications and dementia. Neuropsychological abnormalities, however, remain common despite improved viral suppression (Heaton et al. 2010). Even HIV+ patients with undetectable HIV-1 RNA concentrations in the plasma show declines in attention, memory, psychomotor ability, or executive functions (Cysique et al. 2014; Muñoz-Moreno et al. 2008; Saktor et al. 2010). As individuals with HIV have longer life expectancies, the prevalence of cognitive impairment is likely to increase in the elderly (Sheppard et al. 2015). It is unclear, however, if HIV+ patients decline with age at the same or greater rate than matched HIV− comparators, i.e., if the HIV-age effects on cognition are independent or interactive. Many studies have demonstrated that the degree to which cognitive outcomes decline due to HIV is not increased by older age (Seider et al. 2016; Valcour et al. 2011). Yet, significant interactive effects of HIV and age have been reported in one longitudinal study in which older HIV+ individuals declined in verbal memory over one year at greater rate than demographically similar HIV− controls (Seider et al. 2014).

There also is substantial evidence that even in effectively treated HIV+ individuals, the infection contributes to decreases in regional brain volume. Cortical reductions due to HIV have been found in frontal, orbitofrontal, temporal, cingulate, primary motor, and sensory areas (Küper et al. 2011; Sandford et al. 2017; Towgood et al. 2012). Subcortical reductions due to HIV have been revealed within the amygdala, caudate, corpus callosum (Ances et al. 2012; Guha et al. 2016), basal ganglia (Küper et al. 2011), brainstem and thalamus (Sanford et al. 2017), or white matter (Hua et al. 2013). It is not well understood if HIV infection accelerates brain changes associated with aging. To date, Pfefferbaum and collaborators (Pfefferbaum et al. 2014) reported cortical volume decreases intensifying with age in older HIV+ individuals. Cardenas and collaborators (Cardenas et al. 2009) noted faster white matter volume loss in HIV+ patients with viral suppression than in matched HIV− controls, while Seider and collaborators (Seider et al. 2016) elucidated more pronounced increases in white matter hyperintensities in older HIV+ patients than in controls.

The development of brain abnormalities and cognitive deficits in HIV+ patients may be related to parameters such as the lowest CD4 lymphocyte cell count in blood (nadir CD4), detectable plasma HIV RNA level, or duration of cART (Ellis et al. 2011; van den Dries et al. 2017). CD4 cells are a type of lymphocyte cells. They stimulate other immune cells to fight infection. Nadir CD4 cell count is the lowest point to which CD4 count has dropped due to HIV infection, i.e., the greatest immunosuppression, and is a particularly significant predictor of neurological complications and cognitive impairment (Valcour et al. 2006). Strong evidence demonstrates that the incidence of neuropsychological deficits is highest in HIV+ patients who experienced CD4 cell counts below 200 cells/mm³ (Ellis et al. 2011; Muñoz-Moreno et al. 2008; Seider et al. 2014). CD4 recovery with continuous viral suppression also takes longer in patients who initiate cART at low CD4 cell counts (Costagliola et al. 2014).

Neuroimaging studies of HIV+ populations have reported significant associations between low nadir CD4 cell count and brain regional volume. Nadir CD4 levels have been correlated with volume reductions in gray matter (Küper et al. 2011; McCombe et al. 2013), white matter, and subcortical gray matter (Hua et al. 2013), as well as increased amounts of the cerebrospinal fluid (Jerningan et al. 2011; Su et al. 2016). Reductions in brainstem white matter, internal capsule, globus pallidus, and caudate also have been reported (Sanford et al. 2017). Low Current CD4 cell count and detectable plasma HIV RNA levels have been reported as correlates of ventricular enlargement and reductions in the basal ganglia (Hua et al. 2013).

It is not clear if the extent of cognitive decline and neural injury may be determined by interactions between clinical factors and older age. It has been hypothesized that older HIV+ patients are at risk of accelerated aging of the immunological system (Hong Banks 2015). Older age also may constraint the capacity of treated patients to reconstitute their CD4 cell resources (Appay et al. 2011).

Despite these findings, our knowledge on aging in HIV+ populations remains limited. The evaluation of HIV-age interactions within a sizable group of relatively healthy aviremic HIV+ subjects on cART would add to our scientific understanding of this disease. The current cross-sectional study investigated independent and interactive effects of HIV infection and aging on neuropsychological functioning and gray and white matter as measured with brain morphometry in HIV-positive Polish men with viral suppression below the level of 60 copies/mL. Our second objective was to explore the impact of HIV-associated clinical parameters and duration of infection while controlling for age. In accordance with the existing evidence, we hypothesized that both HIV infection and age would have adverse effects on cognitive status and brain regional volume despite stable antiretroviral regimen and undetectable viral loads in plasma.

The results of this study show that for well-educated HIV+ men of various ages, with viral suppression and no active comorbidities, (1) HIV and age led to neuropsychological declines; (2) older age but not HIV was associated with less volume in several cortical and subcortical brain regions; (3) essentially, the adverse effects of HIV infection and age were independent; and (4) in the oldest HIV+ participants, older age decreased speed of visuomotor performance along with every year on cART.

Overall, such findings are consistent with previous evidence showing that HIV infection is associated with cognitive declines despite successful and, in some cases, long-term viral suppression due to cART (Coban et al. 2017; Crum Cianflone et al. 2013; Seider et al. 2014; Sheppard et al. 2015; Simioni et al. 2010). Our results also support prior observations that HIV and age exert independent but not interactive effects on cognitive performance (Ances et al. 2012; Cysique et al. 2011; Valcour et al. 2011). Such results are relevant to the debate on the relationship between HIV and aging because they seem to contradict the hypothesis that age is an additive risk for neuropsychological dysfunction among HIV+ individuals. On the other hand, according to the literature, it is possible that greater cognitive declines among older HIV+ adults compared to demographically similar controls would emerge in follow-up assessments rather than cross-sectionally (Saktor et al. 2010; Seider et al. 2014).

Another important finding is that we have not found volume decreases of gray matter related to HIV infection and the observed HIV-associated loss of cortical white matter as well as the enlargement of the fourth ventricle was not significant after correcting for multiple comparisons. Such results are in line with evidence suggesting that in virally suppressed patients, brain destruction due to HIV replication and inflammation may develop slowly and subtly, starting in subcortical regions (Ances et al. 2012; Corrêa et al. 2016; Cysique et al. 2017; Gelman 2015). The impact of HIV and aging on individual cortical regions in our sample was subject of a separate study conducted with machine-learning technique that has revealed that the best between-group classification accuracy was obtained based on volumetric measures of subcortical regions such as white matter; 3rd, 4th, and lateral ventricles; amygdala; caudate; and putamen (Pluta et al. in press).

This study has also failed to demonstrate eminent HIV-age interactive effects on brain volume, suggesting that advanced age does not accelerate changes associated with HIV. This finding matches the results of a similar cross-sectional study addressing HIV and age effects on the brain in younger and older aviremic patients (Towgood et al. 2012), albeit in that sample, the HIV+ group had less gray matter in several regions when compared to HIV− controls. Interestingly, the present study has proven interactive effects of age and duration of cART, showing that in older HIV+ patients, longer exposure to cART increased the risk of motor slowing. Such effects were limited to performance in GPT and should be confirmed in other motor tasks for more solid conclusions. The possible neuroanatomical changes underlying motor decline in HIV+ subjects may be disrupted cortico-striatal networks and decrease of functional connectivity (Ortega et al. 2015). Our results indicate detrimental effects of cART in the oldest HIV+ participants. However, declines in fine motor performance do not necessarily have to be related only to cART, which in our sample was administered in standard schedules (2 NRTI + PI/r 55.7% of all HIV+ participants, 2NRTI + NNRTI 22.7%, other 20.6%). Literature has emphasized that the etiology of cognitive decline or brain abnormalities in older patients with HIV is multifactorial. In addition to side effects of cART, the involved processes may include chronic inflammation, neurovascular abnormalities, and metabolic and age-related changes (Clark Cohen 2010; Gelman 2015). The literature on the outcomes of cART has been inconclusive. Many prior studies that have examined aviremic individuals with normal levels of CD4 counts reported negative influence of long-term cART on cognitive status or brain (Ellis et al. 2011; Muñoz-Moreno et al. 2008; van den Dries et al. 2017; Walker Brown 2017), while others have found beneficial effects such as higher brain functional connections (Ortega et al. 2015). Future research on the effects of cART in HIV+ population should comprise larger groups of older individuals.

Despite that, on the whole, our results seem to reflect the efficacy of cART in preventing neuroinflammation due to HIV and subsequent extensive neural and glial dropout (Gelman 2015; Hong Banks 2015) as well as notable cognitive impairment. Such belief is supported by the fact that in our sample most patients had cART introduced immediately after HIV diagnosis (51.6%) or within 1 year (19%) or 2 years (11%), and all were successfully suppressed at the moment of the assessment. Although promising, our results should be treated with caution. The sample was relatively young (mean age in HIV+ group 41.2 ± 11.9 vs HIV− group 44.3 ± 12.5; 19 HIV+ subjects age > 50 years vs. 35 controls) and free of active opportunistic infections or comorbidities that may elicit structural brain injury such as cardiovascular conditions or diabetes.

The lack of association between other studied clinical variables (e.g., nadir CD4 cell count, current cell count, duration of HIV infection) and cognition or brain volume also may reflect the small degree of infection severity in the studied sample (28.6% HIV+ participants experienced nadir CD4 > 200 cells/μL) and/or the successful cessation of brain injury due to quick suppression. This result differs from prior evidence revealing low nadir CD4 as an important risk factor for cognitive changes (Ellis et al. 2011) and volume decreases in the putamen (Wright et al. 2016), thalamus (Cohen et al. 2010), white matter (Hua et al. 2013), and other subcortical regions (Jerningan et al. 2011; Sanford et al. 2017). Still, the role of T cells in the development of brain abnormalities and cognitive deficits in HIV patients is not completely understood. Contradictory hypotheses have attributed CD4 cells a protective role against neuroinflammation or a contributory role to it in virally suppressed patients. Other factors such as aging or past comorbid conditions also may have an effect (Hong Banks 2015).

The differences between our results and other studies may be due to dissimilarities in demographic, clinical, treatment, or other characteristics. To minimize confounding factors, we enrolled only men with secondary or university education and who were employed, without substance abuse, psychiatric diseases, or dementia. HIV+ participants were homogeneously neurologically asymptomatic, aviremic, and without HCV co-infection or opportunistic or sexually transmitted diseases at the time (although a considerable proportion of the HIV+ group had been diagnosed with syphilis, neurosyphilis, or HCV in the past). Most of the HIV+ participants were cognitively normal according to HAND criteria (81.5% were classified as no HAND, Egbert et al. 2018). This differentiates our sample from other recent research that pooled participants of both sexes, multiple races/ethnicities, patients with different levels of viremia, HCV co-infection, neurocognitive impairment, depression, heterogeneous socioeconomic background, substance use disorders, etc. (e.g., Coban et al. 2017; Cohen et al. 2010; Sanford et al. 2017; Sheppard et al. 2015).

Such characteristics of our HIV+ group may reflect the existent clinical observations of patients undergoing treatment and provide important information about the Polish male HIV+ sample on cART. They constrain, however, generalization of our findings to healthy, educated, wealthy, and demographically/clinically homogeneous cohorts (e.g., Crum Cianflone et al. 2013; Cysique et al. 2011; McDonnell et al. 2014; Seider et al. 2014; Simioni et al. 2010). Further studies on larger samples are needed to determine if brain reductions occur in such samples despite successful implementation of cART. If CD4 drop mechanisms or duration of exposure to cART contribute to such a process remains an open question.

This study has some limitations. First, it was cross-sectional, and we collected the data only once. Follow-up assessments are necessary to directly estimate if cognition or brain volume differentially decline as a function of HIV status and age in aviremic HIV+ cohorts. This is particularly important in older patients who get the infection at older ages and are supposed to be more vulnerable. Another limitation is that we did not analyze the impact of past comorbid conditions on cognition and brain structure—conditions such as AIDS-associated opportunistic infections and sexually transmitted diseases (STDs) such as syphilis, neurosyphilis, gonorrhea, and chlamydia. Comorbid conditions may increase the risk of neurocognitive decline in HIV+ patients (Heaton et al. 2015; van den Dries et al. 2017). Other co-infections, such as HCV, also can lead to brain inflammation and injury, which might exacerbate further the aging process in the HIV-infected brain (Holt et al. 2012). Less is known about AIDS-associated opportunistic infections, including the herpes virus, cytomegalovirus, and STDs such as syphilis. These diseases can lead to severe neurological complications but few studies have evaluated their impact on the HIV-infected brain, especially the aging brain (Hong Banks 2015). The relationships among history of comorbid conditions, advanced age, and cognitive or brain declines in successfully treated HIV+ adults remain uncertain.

Despite the limitations, the findings of this study provide a unique contribution to the existing literature on aging with HIV infection and have important potential clinical implications for neuropsychological outcomes in contemporary HIV+ cohorts where a substantial number of people are older than 50. The results underscore the importance of monitoring neuropsychological functioning and brain structure in chronically infected patients. This is extremely important in older patients with long adherence to cART therapies who may be at greatest risk of developing cognitive and brain abnormalities. The results of the present study also support recommendations that early antiretroviral intervention may offer an effective approach to prevent or lessen brain atrophy and cognitive decline over time, although further research is needed to confirm this.