Imaging NeuroAIDS
=================

* R. Gilberto González

The neurologic complications of HIV infection remain a clinical challenge. In the early days of the epidemic, it became clear that the virus has a predilection for the CNS. Before the advent of antiretroviral therapy, a large proportion of HIV-infected individuals developed neurocognitive disorders of varying severity, including a profound dementia. With the introduction of antiretroviral therapy, and more recently with the clinical implementation of highly active antiretroviral therapy (HAART), the incidence of neurocognitive disorders has decreased dramatically. Although HAART has prolonged the life expectancy of HIV-infected individuals, it has raised the fear of an increasing prevalence of neurologic disorders in this population.

Neuroimaging has played a central role in the management of HIV/AIDS patients, particularly in the diagnosis of opportunistic infections and neoplasms that are seen in this population. Structural neuroimaging has had a limited role in the study of what has come to be known as “neuroAIDS.” The neurocognitive abnormalities caused by HIV are thought to arise from injury and death of neurons; however, because the virus does not infect neurons themselves, neuronal injury is thought to arise by indirect mechanisms. The virus infects perivascular macrophages and microglia within the CNS, and it is believed that substances including cytokines and viral products generated by these cells ultimately result in neuronal injury. Early in the epidemic, severe atrophy and white matter abnormalities could be detected by use of imaging in patients with advanced neuroAIDS. In the era of HAART, by contrast, these findings are uncommon, except in populations with little or no access to health care.

Because few or no abnormalities are detectable by use of CT or MR imaging in the early stages of neuroAIDS, clinicians and investigators have used functional neuroimaging methods to measure objectively changes in the brain that are caused by the virus. Many methods have been employed successfully, including positron-emission tomography, single-photon emission tomography, blood oxygen level–dependent functional MR (fMR) imaging, dynamic contrast fMR imaging, MR spectroscopy, and standard diffusion MR imaging. In this issue of the *AJNR,* Ragin et al report that diffusion tensor (DT) imaging may be useful quantifying cumulative brain injury in neurocognitively impaired HIV-infected patients.

There would be many advantages to having a sensitive and reliable neuroimaging method suitable for study of early neuroAIDS. Despite the predilection of the virus for the brain, HIV encephalitis occurs in only one third of individuals who do not undergo therapy. This is thought to be due to both viral and host factors. Finding those who are susceptible to HIV-induced brain injury would be important to the administration of adjunctive therapies to prevent neuronal injury in this subset of patients. Such methods could also be important in elucidating the pathogenesis of neuroAIDS and developing suitable adjunctive therapy. The methods developed and used to date are not sufficiently sensitive to depict early changes in individual patients. The successful studies that have been performed to date have relied on cohorts of at least 10 patients. The great variability in progression of disease, not to mention the modification of its progression with HAART, makes studies of neuroAIDS exceedingly difficult to conduct.

Ragin et al demonstrate that DT imaging has joined the collection of functional neuroimaging techniques that may be useful in guiding the clinical management of neuroAIDS. They report that whole-brain fractional anisotropy (FA) was reduced in HIV-infected subjects and significantly associated with severity of dementia, as indicated by several widely used clinical and functional status measures. They also show that FA measures were more prognostic of dementia status than were apparent diffusion coefficient measures. They propose that FA measures the cumulative injury induced in the brain by HIV, that this methodology may provide insights into the biophysical basis of this injury, and that it may prove useful in understanding the pathogenesis of this disease. It is unclear whether DT imaging has the sensitivity for the early detection of this disease process. The difference in FA that distinguished HIV-infected individuals from control subjects was quite small in Ragin et al’s study. On the other hand, the dissemination of MR imaging systems with echo planar imaging capabilities has made DT imaging widely available. Also, there is little doubt that DT methodology will continue to evolve. It will be interesting to see how the evolution and application of DT imaging will affect our understanding of neuroAIDS and how it may help us in the management of patients with this disease.

*   Copyright © American Society of Neuroradiology