Cardiovascular disease, impaired renal function and diabetes all predicted cognitive impairment or performance in HIV-positive men with an undetectable viral load in Amsterdam's AGEhIV cohort. This comparison with similar HIV-negative men also linked marijuana use, high waist-to-hip ratio, depressive symptoms and lower nadir CD4 count to worse cognitive performance.
Cognitive impairment persists in the current antiretroviral era, although prevalence of impaired cognition remains difficult to pin down because of differences in cognitive testing and populations studied. AGEhIV investigators believe the often-used Frascati criteria yield many false-positive results and thus overestimate prevalence of cognitive impairment. They recently assessed multivariate normative comparison (MNC) in a subset of the AGEhIV cohort because they believe it limits false-positives while retaining sensitivity in identifying impairment. The AGEhIV team conducted a new study to identify factors contributing to decreased cognitive performance and cognitive impairment in the AGEhIV cohort. The results were published in the April 24 issue of AIDS.
AGEhIV is an ongoing prospective comparison of HIV-positive people at least 45 years old and an HIV-negative group recruited from a sexually transmitted disease clinic and selected to match the HIV group in age, sex and race. The primary aim is to assess age-associated morbidity in the two groups. The cognition substudy included only HIV-positive antiretroviral-treated men with a viral load below 40 copies/mL for at least 12 months and matched HIV-negative controls. The analysis excluded people with a history of severe neurologic disorder, HIV-associated dementia, central nervous system infection or tumor, current severe psychiatric disorder and current excessive alcohol consumption or intravenous or illicit drug use, except for marijuana.
Participants underwent neuropsychological assessment covering six cognitive domains, and researchers used MNC to diagnose cognitive impairment. MNC provides both a dichotomous result (cognitive impairment or not) and a continuous measure of cognitive performance (better or worse cognitive performance in the HIV group than in the HIV-negative group).
The analysis focused on 103 men with HIV and 74 HIV-negative men. Both groups had a median age of 54, more than 90% were men who have sex with men (MSM) and more than 86% were Dutch. Men with HIV had been diagnosed for a median of 13.5 years, and they had a median CD4 count of 625 cells/mm3. Only 2% of HIV-positive men were also positive for hepatitis B or C, and no HIV-negative men had hepatitis virus infection. MNC-determined cognitive impairment affected 17% of men with HIV and 5% of men without HIV (P = .02).
Linear regression analysis identified seven factors associated with worse cognitive performance in men with than without HIV (the continuous measure): (1) daily to monthly marijuana use (β coefficient -0.77, P = .002), (2) past cardiovascular disease (β -0.64, P = .066), (3) impaired renal function (β -0.36, P = .096), (4) diabetes mellitus (β -0.73, P = .036), (5) above-normal waist-to-hip ratio (β -0.46, P = .055), (6) depressive symptoms (β -0.69, P = .061) and (7) ever 50-cell/mm3 lower nadir CD4 count (β -0.09, P = .008).
Logistic regression analysis found that the first four of those variables were associated with cognitive impairment (the dichotomous measure): (1) daily to monthly marijuana use (odds ratio [OR] 27.76, P < .001), (2) past cardiovascular disease (OR 17.71, P = .014), (3) impaired renal function (OR 8.76, P = .017) and (4) diabetes mellitus (OR 5.71, P = .097).
The authors note that their study found associations with cognitive performance or impairment, not causes of cognitive change. Because the study group consisted largely of white MSM without hepatitis virus coinfection, the researchers caution that results may not apply to other HIV groups.
The AGEhIV team observes that marijuana use is highly prevalent in some HIV populations, who use it recreationally or medicinally. While marijuana directly affects cognition, its medicinal use may indicate conditions linked to cognition, such as pain and mood disturbances. The authors suggest that cardiovascular, metabolic and renal factors may affect cognition through microvascular or organ damage that may share pathophysiological mechanisms with cerebral damage.