icon-folder.gif   Conference Reports for NATAP  
Noninfectious Conditions May Drive Neurocognitive Problems With HIV
  6th International Workshop on HIV and Aging
October 5-6, 2015, Washington, DC
Mark Mascolini
Noninfectious comorbidities including vascular and metabolic variables accounted for neurocognitive impairment more than infectious comorbidities in an analysis of 68 veterans with HIV [1]. The study also found a link between hemoglobin A1c (HbA1c), a marker of blood glucose over time, and worse neuropsychological test performance.
Research has not consistently identified associations between infectious disease markers and neurocognitive performance, noted Veterans Administration (VA) workers who conducted this study. Therefore they assessed both infectious and noninfectious disease markers in a small group of veterans referred for neuropsychological assessment because of cognitive concerns.
The VA team extracted data from medical records to tally composite scores for infectious diseases and noninfectious diseases. Infectious disease burden included 6 factors: number of infectious diseases, time since diagnosis, CD4 count below 200, detectable HIV load, number of opportunistic infections, and HCV infection. Noninfectious disease burden included 7 factors: vascular disease, cancer, neurologic disease, kidney dysfunction, immune disorders, vitamin deficiency, and metabolic markers. The investigators used hierarchical multiple logistic regression to gauge the relative contribution of infectious disease markers and noninfectious markers to neurocognitive impairment. The analysis first controlled for the influence of infectious disease factors to see if the noninfectious variables accounted for significant variance in cognitive scores independently of infectious factors.
Of the 68 participants, 67 were men, 57 (84%) were black, and half had HCV infection. Age averaged 55.9 years and education averaged 12.1 years. After controlling for infectious disease factors, logistic regression determined that noninfectious factors accounted for significant variance in scores for learning/memory (r2 delta = 0.17), attention/working memory (r2 delta = 0.22), and processing speed (r2 delta = 0.21) (all P < 0.05). These results mean noninfectious factors accounted for about 20% of variance in these three neurocognitive domains independently of infectious factors. When the VA team reversed the statistical analysis and controlled for noninfectious variables, infectious disease factors did not account for significant variance in neurocognitive scores.
Next the researchers conducted exploratory zero-order correlations between neurocognitive scores and markers of metabolic function such as HbA1c and blood glucose levels. This analysis determined that elevated dispersion of HbA1c values was associated with worse performance in the domains of learning (r = -0.34, P < 0.05) and attention/working memory (r = -0.38, P < 0.05).
The VA investigators proposed that neurocognitive impairment cannot be attributed exclusively to HIV and other infectious disease-related factors and is in fact more strongly linked to noninfectious comorbidities. They observed that some of these noninfectious contributors to neurocognitive impairment are modifiable. The researchers added that their composite score approach is "a rudimentary way of understanding the impact of comorbidities on neurocognitive impairment" and suggested that more sophisticated techniques may enrich this line of enquiry.
1. Perra A, Dux M, Lee-Wilk T. The association between comorbidities and neurocognitive impairment in aging veterans with HIV. 6th International Workshop on HIV and Aging. October 5-6, 2015, Washington, DC. Abstract 8.