| |
Abnormal cognitive aging in people with HIV:
Evidence from Data Integration between two countries' cohort studies
|
| |
| |
AIDS: April 27, 2022 - Htein Linn AUNG 1,2, Thomas M. GATES 1,3, Limin MAO 4, Bruce J. BREW 1,2,5, Sean. B ROURKE6,7, Lucette A. CYSIQUE1,3,6
Download the PDF here
Based on the sample size calculation in our review [3], the current study sample size had at least 80% power to detect a small-medium effect size of age on NCI at a statistically significant level (p<0.05).
Older PLWH >50 Had Higher Chance for NCI, Older PLWH Had Higher Probability of Increases in NCI Over Follow-Up. African-Americans/Blacks have "much higher risk". More comorbidities & renal impairment are associated with NCI. HIV viral suppression reduces risk for NCI but does not prevent it, a the "abnormalities aging affects" of NCI were also detected for virally suppressed <200 c/ml. Our study advocates for HIV clinics to immediately implement routine cognitive screening in all PLHIV age 50+ years, especially those with comorbidities, subjective cognitive complaints, and those who fall into a vulnerable demographic group within their countries.
Aging among people living with HIV (PLHIV) adds additional risks for neurocognitive impairment (NCI) [1]. Previous research has suggested that cognitive aging can be accentuated and accelerated among PLHIV [1, 2]. However, based on a systematic review [3] our group has recently conducted, current available evidence on these abnormal patterns of neurocognitive aging among PLHIV have been inconclusive.
We therefore combined data from these two studies to:
1. demonstrate the feasibility of data integration in the NeuroHIV field. 2. To determine whether older age increases the risk for NCI among PLHIV both cross-sectionally and longitudinally while assessing for potential cohort effects and 3. To identify biomarkers and health conditions associated with cognitive impairment among PLHIV.
RESULTS:
Our study further confirms that older age is associated with increased risk for NCI both cross-sectionally and longitudinally even after considering for cohort, demographic factors, comorbidities, and HIV disease factors.
1. Our analyses reveal that participants older than 50 years of age had 83% higher chance of NCI compared to the younger participants.
2. - Older participants also had a higher probability of increases in NCI over the follow-up compared to the younger participants (OR=1.66).
3. More severe comorbidity ratings were associated with a significant higher risk of NCI. Specifically, having a contributing or confounding comorbidity increased the chance of NCI by 1.59 times and 11.9 times respectively compared to incidental or no comorbidity rating. The Australian cohort included 30% of people with contributing comorbidities and only 2% with confounding comorbidities. Therefore, the confounding effect essentially came from the US cohort. Since age-related comorbidities such as cardiovascular diseases and diabetes, and non-age-related comorbidities such as depression, substance and alcohol use disorders and hepatitis C infection (which can all have an impact on cognition) are more common among PLHIV than in the general population, risk for NCI may be much higher among PLHIV, especially in elders.
However, it also follows that minimization of NCI is possible via early detection and management of these comorbidities. With the aging of the HIV population, screening and management strategies for these comorbidities is probably best considered within the context of an HIV geriatric clinic [47, 48]
3. Despite normative correction of cognitive scores for racial/ethnic background, we still found much higher risk of NCI among Black/African American participants (1.60 times higher risk compared to White participants) and those from "Other" racial/ethnic backgrounds (2.41 times greater risk compared to White participants). …..We interpret these results not so much as the "imperfect" corrections of the norms, but as effects that the norms cannot correct because they represent the cumulative or synergistic impact of racial/ethnic disparities impacting many aspects of a person's life (e.g., lower quality of education, lower socio-economic status and intergenerational poverty, poorer housing, lower social opportunities, and psychosocial stress associated with racial discrimination) [41-44]. This interpretation is supported by the fact that African American and other racial/ethnic background participants also had higher contributing and confounding comorbidities.
4. Our study also showed that abnormal creatinine level is associated with increased risk of NCI. Having abnormal creatinine level was associated with 83% higher chance of NCI. Previous studies have reported an association between renal dysfunction and cognitive impairment in both PLHIV [49] and general population [50, 51]. Renal impairment is often associated with other risk factors for cognitive impairment such as hypertension, diabetes, cerebrovascular disease, and anemia; these conditions are likely to mediate the effect of renal impairment on cognition [50]. In our study, participants who had renal impairment were more likely to have higher comorbidity rating, diabetes, history of myocardial infarction and stroke and abnormal hemoglobin level.
5. Lower self-perceived cognitive functioning (higher PAOFI score) was associated with an increased risk for NCI. This finding supports results from previous studies indicating that objective cognitive impairment is associated with subjective cognitive symptoms in PLHIV [52-54]. The PAOFI is often associated with depressive symptoms in PLHIV and thus, cognitive deficit among those with higher PAOFI score may partly be mediated by depression [55, 56]. In the current study, PAOFI score was significantly higher among those who were depressed at baseline (P=<0.0001).
*** 6. Lastly, a plasma viral count <200 copies/ml reduced the risk for cognitive impairment by 37%. This is congruent with previous studies [57-59]. This finding reinforces the importance of access and adherence to ART to keep HIV viral replication suppressed [60]. Yet it is important to note that even in virally suppressed individuals, the abnormal aging effects were also detected.
These effects were still significant when the analysis was confined to participants who had already achieved viral suppression: 65% increased risk cross-sectionally and 68% increased risk longitudinally among older participants in relation to younger participants. While we did not have HIV-negative controls in this study, cognitive scores were corrected for demographic effects with robust regression-based norms derived from appropriate controls' sample, and we are therefore confident that the residual age effect detected was representative of pathological cognitive aging above and beyond that of normal aging.
AIDS: April 27, 2022
Htein Linn AUNG 1,2, Thomas M. GATES 1,3, Limin MAO 4, Bruce J. BREW 1,2,5, Sean. B ROURKE6,7, Lucette A. CYSIQUE1,3,6
1 Departments of Neurology and HIV Medicine, St Vincent's Hospital and Peter Duncan Neurosciences Unit, St Vincent's Centre for Applied Medical Research, Sydney, Australia
2 Faculty of Medicine, UNSW, Sydney, Australia
3 School of Psychology, University of New South Wales, Sydney, NSW, Australia
4 Centre for Social Research in Health, UNSW, Sydney, Australia
5 Faculty of Medicine, University of Notre Dame, Sydney, Australia
6 MAP Centre for Urban Health Solutions, Li Ka Shing Knowledge Institute, St. Michael's Hospital
7 Department of Psychiatry, University of Toronto
Abstract
Objectives:
Previous research has shown inconsistent results on whether cognitive aging is abnormal in people with HIV (PWH) because of low sample size, cross-sectional design, and nonstandard neuropsychological methods. To address these issues, we integrated data from two longitudinal studies: Australian HIV and Brain Ageing Research Program (N = 102) and CNS HIV Antiretroviral Therapy Effects Research (CHARTER) study (N = 924) and determined the effect of abnormal aging on neurocognitive impairment (NCI) among PWH.
Methods:
Both studies used the same neuropsychological test battery. NCI was defined based on demographically corrected global deficit score (≥0.5 = impaired). Both studies also assessed comorbidities, neuropsychiatric conditions and functional status using similar tools. To determine the cross-sectional and longitudinal effects of age on the risk of NCI, a generalized linear mixed-effect model tested main and interaction effects of age group (young, <50 vs. old, ≥50) and time on NCI adjusting the effects of covariates.
Results:
Older PWH had 83% higher chance of NCI compared with younger PWH [odds ratio (OR) = 1.83 (1.15-2.90), P < 0.05].
Older participants also had a greater risk of increases in NCI over the follow-up [OR = 1.66 (1.05-2.64), P < 0.05] than younger participants.
Nonwhite ethnicity (P < 0.05), having a contributing (P < 0.05) or confounding (P < 0.001) comorbidity, greater cognitive symptoms (P < 0.001), and abnormal creatinine level (P < 0.05), plasma viral load greater than 200 copies/ml (P < 0.05), being from the Australian cohort (P < 0.05) were also associated with a higher risk of NCI.
Conclusion:
Data integration may serve as a strategy to increase sample size and study power to better assess abnormal cognitive aging effect in PWH, which was significant in the current study.
|
|
| |
| |
|
|
|
