Objective: We evaluated regional brain volumes and cerebral metabolite levels as correlates of HIV DNA in peripheral blood mononuclear cells (PBMCs).
Methods:In this cross-sectional study, 35 HIV1 subjects aged $40 years (25 with detectable PBMC HIV DNA; 10 with HIV DNA ,10 copies/10 6 cells, the threshold of detection) and 12 seronegative controls underwent structural brain MRI and magnetic resonance spectroscopy at 3 T. HIV1 subjects were on combination antiretroviral therapy $1 year; all but 1 had plasma HIV RNA ,50 copies/mL. We used logistic regression to evaluate relationships of likely predictor variables to the outcome of PBMC HIV DNA detectability in the HIV1 subjects. Effects of serostatus and HIV DNA on regional brain volumes (normalized to intracranial volume) and on metabolite ratios over creatine were evaluated by analyses of covariance, controlling for age.Results: Relative to the HIV1 group with undetectable HIV DNA, subjects with detectable HIV DNA demonstrated decreased volumes of cerebellar (214%, p 5 0.020) and total subcortical (210%, p 5 0.024) gray matter. Compared to healthy controls, only the detectable HIV DNA group showed significant (p , 0.05) enlargement of lateral ventricles and volumetric reductions of caudate, putamen, thalamus, hippocampus, nucleus accumbens, brainstem, total cortical gray matter, and cerebral white matter. Detectable HIV DNA was not associated with significantly altered cerebral metabolite levels. Glu 5 glutamate; HAND 5 HIV-associated neurocognitive disorders; ICV 5 intracranial volume; MI 5 myo-inositol; MRS 5 magnetic resonance spectroscopy; NAA 5 N-acetylaspartate; PBMC 5 peripheral blood mononuclear cell; PLSD 5 protected least significant difference; SN 5 seronegative; U-HIV 5 HIV1 infected subjects with undetectable peripheral blood mononuclear cells HIV DNA; VRS 5 Virchow-Robin spaces.
Conclusion: