Neurocognitive dysfunction in HIV-infected youth: investigating the relationship with immune activation

Abstract

Background: HIV-infected individuals are at increased risk of neurocognitive impairment compared to the general population. Studies suggest that, despite combination antiretroviral therapy (cART), HIV infection causes immune activation which results in neural damage; however, few data exist in HIV-infected youth.

Methods: HIV-infected youth 8-26-years-old on cART with virological suppression were prospectively enrolled along with healthy controls. Neurocognitive performance was assessed by age-appropriate Wechsler Intelligence Scales. Soluble and cellular markers of T-lymphocyte and monocyte activation were measured by ELISA and flow cytometry, respectively.

Results: 45 HIV-infected subjects and 21 controls were enrolled. Markers of T-cell and monocyte activation were higher in the HIV-infected subjects compared to controls, but proportions of inflammatory and patrolling monocytes were similar. Although there were no significant differences in neurocognitive scores between the HIV-infected and control groups, scores were low-average for four of five testing domains for the HIV-infected subjects and average for all five in the controls, and % of HIV-infected subjects with scores classified as 'low average' or below was higher than in the controls. Variables most associated with neurocognitive performance among HIV-infected subjects included activated CD4+ T-cells (% CD4+CD38+HLA-DR), monocyte activation (soluble CD14), HIV duration, age and sex.

Conclusions: HIV-infected youth on cART with virological suppression show subtle evidence of neurocognitive impairment compared to healthy controls, and increased immune activation appears to play a role. Additional studies are needed to develop strategic interventions beyond cART to potentially improve neurocognitive performance and/or minimize further impairment in this vulnerable population. ClinicalTrials.gov Identifier: NCT01523496.

Conflict of interest statement

Conflicts of Interest: ARE has received research funding from Bristol-Myers Squibb, Cubist Pharmaceuticals, and GlaxoSmithKline and has served as an advisor and speaker for Gilead. GAM serves as a consultant for Bristol-Myers Squibb, ViiV/GlaxoSmithKline, Gilead, Pfizer, and ICON, and has received grant funding from Bristol-Myers Squibb, ViiV/GlaxoSmithKline, and Gilead. All others declare no conflicts of interest.

Figures

Figure 1. Neurocognitive performance by study group

These scatter plots show the distribution of neurocognitive testing scores for HIV-infected subjects compared to healthy controls for (A) verbal comprehension, (B) perceptual reasoning, (C) working memory, (D) processing speed and (E) full-scale intelligence quotient (FSIQ). In the HIV-infected group, median scores were in the low-average range for 4 of 5 composite scores [median (Q1, Q3) verbal comprehension: 89 (77, 96); perceptual reasoning: 88 (79, 100); working memory: 89 (77, 99); FSIQ: 87 (77, 95)], and in the average range for processing speed [97 (84, 102)]. For the control subjects, median scores were in the average range for all 5 composite scores [median (Q1, Q3) verbal comprehension: 91 (76, 106); perceptual reasoning: 92 (81, 104); working memory: 97 (86, 113); processing speed: 94 (89, 108); FSIQ: 94 (80, 110)]. However, there were no statistically significant differences in the scores for any of the categories between the HIV-infected subjects and healthy controls. Horizontal lines denote medians.

Figure 2. Correlations between full-scale IQ and significant variables

Scatter plots depict the bivariate relationship between full-scale IQ and A. sCD14 and B. HIV duration for the HIV-infected subjects. Full-scale IQ was significantly negatively correlated with both sCD14 and HIV duration. *Spearman correlation coefficient. IQ, intelligence quotient; sCD14, soluble CD14.

Figure 3. Comparison of the effect of selected variables on the neurocognitive scales

These plots show the beta-coefficients +/− standard errors relative to a null effect for selected variables from the multivariable models for ease of comparison across the neurocognitive testing categories (A. HIV duration, B. % CD4+CD38+HLA-DR+ and C. sCD14). sCD14, soluble CD14; FSIQ, full-scale intelligence quotient.