Secondary Logo

Journal Logo

Basic and Translational Science

Neurovirological Correlation With HIV-Associated Neurocognitive Disorders and Encephalitis in a HAART-Era Cohort

Gelman, Benjamin B. MD, PhD*; Lisinicchia, Joshua G. BS*; Morgello, Susan MD; Masliah, Eliezer MD, PhD; Commins, Deborah MD, PhD§; Achim, Cristian L. MD, PhD; Fox, Howard S. MD, PhD; Kolson, Dennis L. MD, PhD; Grant, Igor MD; Singer, Elyse MD#; Yiannoutsos, Constantin T. PhD**; Sherman, Seth PhD††; Gensler, Gary MS††; Moore, David J. PhD; Chen, Tiansheng MD*; Soukup, Vicki M. PhD*

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: April 15, 2013 - Volume 62 - Issue 5 - p 487-495
doi: 10.1097/QAI.0b013e31827f1bdb



HIV-associated neurocognitive disorders (HAND) in the era of highly active antiretroviral therapy (HAART) occur in an estimated 50% of unselected cohorts.1,2 One concept regarding the pathophysiology of HAND posits that replication of HIV-1 in mononuclear phagocytes of the central nervous system (CNS) is the critical driving force. Neuropathologically that concept was suggested largely using studies conducted before the HAART era in subjects with HIV encephalitis (HIVE) and before the development of the most recently applied nosological schema used to diagnose HAND.3 It is well documented that HIVE is associated with high rates of replicating HIV-1 in the brain,4,5 which suggests that the pathophysiology of HAND might be driven substantially by a high rate of virus replication in the CNS. However, it still remains unclear whether mildly increased virus replication can influence HAND or whether a “no-effect” concentration of HIV in the brain can be achieved using HAART. The association of brain viral replication with cognitive impairment has not been broadly documented in humans, in part because of the limited brain specimen resources.

Mechanistic scenarios for HAND that have been suggested emphasize the production of neurotoxic HIV-1 proteins in infected macrophages, microglial cells, and astrocytes, which leads to the production of inflammatory mediators in many types of brain cells.6,7 In turn, virally driven inflammatory cascades in the CNS are believed to produce lasting neurodegenerative-type changes in “bystander” neurons and dysregulation of glial cells or neurovascular elements, which then lead to the neurocognitive dysfunction observed in HAND.8 Evidence obtained before the era of HAART suggested that the intensity of the inflammatory reaction in the brain based on cell staining was more strongly correlated with HAND than the load of HIV-1 in the brain.9

Findings that contradict the prevailing concepts established before HAART have been noted, including decedents with neuropsychological (NP) impairment but not HIVE and vice versa.10–12 A survey of brain specimens from the National NeuroAIDS Tissue Consortium (NNTC) suggested that the divergence between HAND and the neuropathological diagnosis of HIVE has increased in the HAART era to the point that most patients with HAND do not have HIV-related brain pathologies.13 Mainstream thinking regarding the pathophysiology of HAND has not successfully incorporated the conflicting data concerning the neurovirological correlation.14 It is important to clarify the relationship between CNS virus replication and HAND to justify the rationale to further reduce or eradicate virus replication in the CNS compartment. To that end, a comparison between brain HIV-1 load and HAND was conducted using clinically and neuropathologically characterized subjects.



HAART-era subjects with HIV/AIDS who underwent autopsy and neuropathological evaluation were examined. Decedents were chosen from the NNTC, which is a multisite consortium of clinical sites that has acquired clinically characterized tissue specimens.15 Two cohorts were assembled (Table 1). One was focused on subjects with a NP diagnosis that was matched to a neuropathological diagnosis (n = 148). A second larger cohort included all of the available neuropathological information regardless of whether matching NP data were available (n = 195). Inclusion in the NP focused cohort was based upon the following criteria: (1) HAND that was designated HIV-associated dementia, (2) HAND of any severity that was associated with a neuropathological diagnosis of HIVE or microglial nodule encephalitis (MGNE), or (3) normal NP test performance and no HAND. All qualified subjects were assigned an NP diagnosis. Four subjects who did not have HAND initially had HIVE at autopsy; those 4 subjects were excluded from group assignment but were not excluded from the correlation analyses. In the second cohort, inclusion was based on the following criteria: (1) neuropathologically diagnosed HIVE, with or without available NP testing and with or without a diagnosis of HAND (n = 38), (2) MGNE with or without available NP testing and with or without HAND (n = 13), or (3) other neuropathological diagnoses present in the subjects who already were selected for the NP-based cohort (see Table 1). There were no exclusion criteria.

TABLE 1-a:
Cohort Data
TABLE 1-b:
Cohort Data

Neuropsychological Testing

Subjects in the NP-focused cohort underwent testing with the NNTC neurocognitive test panel.15,16 Testing was given at 6-month intervals. Evaluations were scored according to age-, gender-, ethnicity-, and education-adjusted norms, and a composite normalized T score for NP performance was assigned to patients who completed at least 10 of 14 tests. Subjects who completed 9 or fewer tests were given a nosological diagnosis and domain T scores as appropriate but were not assigned a composite NP T score. A diagnosis of HAND was assigned guided by American Academy of Neurology criteria17 as modified by the Frascati Criteria.3


Neuropathological diagnoses were rendered by NNTC site neuropathologists. Criteria used for the nosological diagnosis of HIVE were according to Budka et al.18 The diagnosis of MGNE was made when the case material contained MGNE as the predominant finding, with the absence of multinucleated cells and other morphological criteria associated with HIVE.

Brain Dissection and Extraction of RNA and DNA

Dorsolateral prefrontal cortex was dissected from Brodmann areas 9 or 10. The Qiagen RNeasy Lipid Tissue Mini Kit (Cat. No. 74804; Valencia, CA) was used to prepare RNA. Briefly, approximately 100 mg of brain tissue was dissected on dry ice and homogenized in a minibead beater. After extraction with chloroform, the RNA was centrifuged in RNeasy mini spin columns, washed, and eluted. HIV cDNA was prepared using Bio-Rad iScript cDNA Synthesis Kit (Cat. No. 170-8891; Hercules, CA). One μg of brain RNA, 4 μL of 5× iScript reaction mix, 1 μL of iScript reverse transcriptase, and 1 μL of 20 μmol/L HIV antisense primer 84R were adjusted to 20 μL with nuclease-free water. After incubation, DNA was extracted (Stratagene, Cat. #300600; Agilent Technologies, Inc, Santa Clara, CA). Total DNA was extracted from brain tissue in the same way.

Assay of HIV-1 RNA, DNA, and Inflammatory mRNAs

Brain HIV RNA and DNA were quantified using cDNA and total genomic DNA using HIV gag/pol primer and probe sequences from Palmer et al19 by polymerase chain reaction (PCR). The reaction contained 4 μL of cDNA or 1 μg of total DNA, 12.5 μL of Sigma JumpStart Taq ReadyMix (Cat. No. D7440; Sigma, St. Louis, MO), 3.5 μL of 25 mmol/L MgCl2, 0.8 μL of 10 μmol/L HIV primer mix, and 0.5 μL of 10 μmol/L HIV probe adjusted to 25 μL. Conditions were 2 minutes. at 95°C, 40 cycles of 15 seconds. at 95°C, and 60 seconds at 60°C. Real time (RT)-PCR was run using an Eppendorf RealPlex (Hamburg, Germany). Copies per microgram of total RNA were calculated with a standard curve using a previously quantified HIV-positive RNA primary standard.20 The same procedure was used for HIV DNA concentration. The mRNAs for myxovirus resistance 1 (MX1), ISG15 ubiquitin-like modifier (ISG15), interferon regulatory factor 1 (IFR1), and type 2 dopamine receptor long isoform (DRD2L) were quantified in cDNA made from mRNA extracts using commercial reagents for RT-PCR. One microliter of 20× DRD2L primers and probe mix (Cat. Hs01024460_m1; Applied Biosystems, Foster City, CA) was combined with 1 μL of cDNA, 10 μL of 2× JumpStart Taq ReadyMix, and 2.5 μL of 25 mmol/L MgCl2 adjusted to 20 μL with water. GAPDH mRNA was used as the normalizing transcript in reactions analogous to the above using 1 μL of 10 μmol/L GAPDH primer mix and 0.5 μL of 10 μmol/L GAPDH probe. For IRF1 mRNA, IRF1 mix (Hs00971959_m1), GAPDH mix (Hs99999905_m1), and TaqMan Universal PCR Master Mix (Part No. 4304437) were used (Applied Biosystems) with conditions as above. For MX1 mRNA, MX1 mix (Hs00182073_mL) was used. For ISG15 mRNA, ISG15 mix (Hs00192713_mL) was used. RT-PCR was run in duplicate, and relative expression was calculated using the ΔΔCt method.

Demographic, Clinical, and Pathological Data

Demographic and medical data were obtained from the NNTC data archive15 as listed in Table 1. The concentration of HIV gag/pol RNA in blood plasma and cerebrospinal fluid (CSF) was quantified by NNTC sites using the Roche Amplicor HIV-1 Monitor test v1.1 through v1.5 (Basel, Switzerland). With few exceptions, the blood and CSF samples were obtained on the day that NP testing was done. Lifetime histories of substance abuse and dependence and of major depression were obtained using the Psychiatric Research Interview for Substance and Mental Disorders (PRISM) or the Composite International Diagnostic Interview (CIDI).21 HAART status was defined as being active if the subject was given at least 2 nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) or 1 nonnucleoside reverse transcriptase inhibitor (NNRTI) and 1 protease inhibitor within 1 year of death.

Statistic Analysis

HIV RNA and DNA levels were logarithm transformed using (log10x + 200) where x is copies of HIV RNA per gram, and 200 represents the observed threshold of HIV RNA detection of the assay. Effects between groups were evaluated using one-way analysis of variance with Tukey-Kramer tests. The normalized composite impairment T scores and 7 normalized component domain T scores were correlated with brain HIV RNA and DNA using Spearman test. The false discovery rate due to multiple comparisons for 7 domain T scores was controlled by the method of Benjamini and Hochberg.22 Correlations pertaining to plasma, CSF, and inflammatory markers were done using Spearman test. Fisher r-to-Z transformations were done to determine whether a correlation coefficient from 1 group was significantly different from another. The significance threshold was P < 0.05.


Brain HIV Versus the Nosological Diagnosis of HAND

Brain HIV RNA concentration between 4 neuropsychologically and neuropathologically classified groups was significantly different (P < 0.001) (Fig. 1A). The subjects with HAND plus HIVE had substantially higher brain HIV RNA relative to the subjects without HAND (delta = 2.48 log10 units, n = 36 versus 27, P < 0.001). Subjects with HAND and MGNE also had a slightly higher brain HIV RNA than those without HAND that was not significant (delta = 0.92 log10 units, n = 36 versus 12, P = 0.123). Subjects with HAND but without HIVE or MGNE had no substantial difference in their brain HIV RNA relative to the subjects without HAND (delta = 0.385 log10 units, n = 36 versus 69, P = 0.314). Results using brain HIV DNA concentration in these comparisons mirrored the HIV RNA results, but the group differences were less pronounced (Figs. 1A versus 1B).

Brain HIV RNA (A) and DNA (B) expressed as copies per gram of brain tissue in the neuropsychologically focused cohort. One-way analysis of variance showed a significant group effect (P < 0.0001). P values for the post-hoc Tukey-Kramer tests are shown. Both brain HIV RNA and DNA were significantly higher in subjects with HAND plus HIVE. Brain HIV RNA was marginally higher in HAND with MGNE. Brain HIV RNA and DNA were not significantly higher in HAND without HIVE or MGNE. The number of subjects without HAND shown here differs from Table 1 because 4 subjects with HIVE did not have HAND and could not be assigned to a group (see Methods). N, number of subjects in each group.

Brain HIV Versus Neuropsychological Test Scores

To determine if the severity of CNS impairment was related to brain HIV RNA, a correlation analysis was performed using the composite normalized NP T score. A higher level of brain HIV RNA was significantly correlated with having a worse composite NP T-score (ρ = −0.290, P = 0.001) (Table 2). Correlation coefficients for all of the NP T scores were closer to unity when the subjects with HIVE were considered separately relative to those without HIVE. Fisher r-to-Z transformations showed that the differences with HIVE versus without it were statistically marginal when corrected for multiple comparisons, with the sharpest contrast occurring in the abstract executive (P = 0.045) and attention working memory (P = 0.023) testing domains.

Spearman Correlation Between Log Brain HIV-RNA and Normalized Neuropsychological T Scores

Brain HIV DNA was less strongly correlated with neurocognitive test scores (see Table S1, Supplemental Digital Content, One of 7 NP testing domains (learning) was correlated significantly. All of the correlation coefficients were closer to unity in patients with HIVE versus without it, and the r-to-Z transformations were statistically significant in several test domains after correction for multiple comparisons, with P values ranging from 0.001 to 0.030.

Brain HIV Versus Neuropathological Diagnosis

The neuropathologically focused cohort contained 65 cases of HIVE, 25 cases of MGNE, and 105 cases with neither diagnosis (Fig. 2A). There were highly significant differences in the levels of brain HIV RNA between the groups (P < 0.001). Brain HIV RNA was substantially higher with HIVE versus without it (delta = 2.85 log10 units, n = 65 versus 105, P < 0.001). Brain HIV RNA also was higher in the subjects with MGNE versus no MGNE (delta = 1.10 log10 units, n = 26 versus 105, P < 0.023). The HIVE and MGNE groups were significantly different from each other (P < 0.001). Less pronounced differences were observed in these groups for brain HIV DNA (Figs. 2A versus 2B).

Brain HIV RNA and DNA expressed as copies per gram of brain tissue in the larger neuropathologically focused cohort. One-way analysis of variance showed a significant group effect (P < 0.0001). P values for the post-hoc Tukey-Kramer tests are shown. Brain HIV RNA was higher in the subjects with MGNE and HIVE versus neither (A). The group with MGNE had significantly lower HIV RNA than the group with HIVE (A). Results using HIV DNA (B) are similar with smaller effect sizes. N, number of subjects in each group.

Brain HIV Versus HIV in Blood Plasma, HIV in CSF and Plasma CD4+ Lymphocytes

Plasma HIV RNA, obtained close to the time of NP testing, was correlated weakly with NP T scores with marginal statistical significance (ρ = −0.180, n = 114, P = 0.055). Plasma HIV RNA was, however, significantly correlated with brain HIV RNA (ρ = 0.371, n = 160, P < 0.001). CSF HIV RNA obtained close to the time of NP testing was not correlated with NP T scores significantly (ρ = −0.109, n = 84, P = 0.327) but was correlated significantly with brain HIV RNA (ρ = 0.435, n = 104, P < 0.001). Low plasma CD4+ lymphocyte counts close to the time of testing were correlated with worse (lower) NT T scores (ρ = 0.234, P = 0.0089).

Brain HIV Versus Neuroinflammatory mRNA Markers

Brain HIV replication and HIVE are associated strongly with host inflammatory responses including type 1 interferon response genes.9,23,24 Both ISG15 and MX1 mRNAs in the brain specimens were significantly correlated with composite NP T scores (Fig. 3A–D). IRF1 mRNA, which is a predominantly type 2 interferon response gene,25 was more weakly correlated with NP T scores (Fig. 3E). A neuronally expressed mRNA (DRD2L) is also given (Fig. 3F).20 Correlations between inflammatory markers and brain HIV RNA were highly significant (see Figure S1, Supplemental Digital Content,

Composite performance on neuropsychological testing (ordinate) was plotted with neurovirological and neuroimmunological measurements in the brain specimens (abscissa) of 119 HIV-infected subjects. High load of brain HIV RNA was significantly correlated with lower composite neuropsychological T scores (A). High load of brain HIV DNA had a similar but weaker pattern that was not significant statistically (B). High levels of ISG15 (C) and MX1 (D) mRNAs were significantly correlated with low T scores. The levels of IRF1 mRNA had a weaker correlation that was marginal statistically (E). High expression of DRD2L mRNA, which is a dopamine receptor transcript expressed primarily in neurons, was correlated with worse impairment (F). Relative expression of mRNAs was normalized to GAPDH mRNA. Correlations used Spearman test and were log transformed on both axes to achieve a uniformly scaled comparison panel.


This is the largest neurovirological survey to date of HIV-infected human brain specimens from subjects who underwent formal neurocognitive testing. The survey confirms that the CNS is a unique compartment with regard to controlling the rate of virus replication and supports the hypothesis that virus replication in the CNS contributes to the pathophysiology of HAND in certain subjects. The broad clinical implication is that targeting and reducing HIV replication in the CNS compartment is a logical approach to treat HAND in certain types of neuropathologically characterized patients on HAART. Further decreasing brain HIV loads to levels lower than those currently achieved on HAART could produce beneficial albeit highly variable degrees of improvement in neurocognitive performance. Most of the variation in potential NP benefit segregated according to the neuropathological diagnoses of HIVE and MGNE. Thus, targeting CNS HIV replication therapeutically is likely to produce some NP benefit in patients who (1) have HIVE and MGNE neuropathologically, (2) have elevated brain type I interferon responses, which could drive neurocognitive dysfunction,24 and perhaps (3) have a brain HIV RNA that has “broken out” to above a critical (threshold) value that has not yet been established. In contrast, brain HIV RNA was correlated with NP impairment either minimally or not at all in the numerically largest group of subjects with HAND (i.e., patients lacking HIVE or MGNE) (Fig. 1 and Table 1). In those patients, the potential NP benefit of further reducing HIV RNA in the brain is not as clearcut, because they did not have more brain HIV RNA than the group without HAND. Autopsy surveys show that HAART-era subjects in unselected cohorts with HAND are not likely to have a very high prevalence of HIVE or MGNE and/or brain HIV loads in the higher ranges.26,27 In turn, reducing residual HIV RNA incrementally, by intensifying virus suppression in the CNS compartment, might not produce substantial NP improvement in subjects already on HAART.2

Most of the subjects with HAND had neither HIVE nor MGNE, and other neuropathological changes do not segregate with HAND.13 The reasons for that remain unknown.14 It is possible that the early stages of HAND produce neurophysiological changes that cannot be observed histopathologically.10,28 Alternatively, newly recognized histomorphological changes in brain cells remain to be discovered. At present there remains a “gap” between HAND and its putative neurovirological substrate. The disparity is analogous to the therapeutic gap that has been recognized between HAND and HAART.2

A biomarker capable of separating specific types of subjects with HAND could be used in clinical surveys to screen patients according to neuropathological risk categories, which would allow for individualizing treatment according to a patient’s likelihood of obtaining added NP benefit from targeting the reduction of CNS HIV loads. There is at present no available laboratory method to fill that niche. Obtaining a brain biopsy lacks feasibility in clinical practice. Measuring CSF HIV RNA is often suggested to be a viable marker for HAND because it was found to be a potentially useful surrogate for HIV in the CNS in a survey before HAART.29 In this survey, CSF HIV RNA did not recapitulate the significant neurovirological correlation observed using brain tissue, which agrees with HAART-era surveys which showed that CSF HIV RNA is not correlated with HAND.30 Notably, CSF HIV RNA in this survey was less informative than plasma HIV RNA with regard to neurocognitive dysfunction and the neurovirological correlation. That finding is consistent with studies which showed that HAND is correlated with other measurements in the blood compartment including anemia, endotoxin, and soluble CD14.31–33 Measurement of CSF HIV RNA still is held to be worthwhile in longitudinal research paradigms because it provides access to a subcompartment of the CNS, can be sampled multiple times, and showed linkage with neurocognitive impairment without HAART.29

The clinical and neurovirological implications of HIVE have undergone elucidation using animal models of lentivirus encephalitis and in vitro HIV-1 infection paradigms.8,11,12,34 The significant neurovirological correlation with HIVE and HAND quantified herein was anticipated from histomorphological observation.10,11 The clinical implications of other neuropathological outcomes including MGNE, diffuse microglial cell activation, and minimal changes are not as clear.10,14,35–37 MGNE and diffuse microglial cell activation in HIV/AIDS have been interpreted variously as: (1) responses to opportunistic infections with organisms such as cytomegalovirus or Toxoplasma ghodii,38 (2) an inflammatory pattern that should be classified together within the broad spectrum of HIVE-related effects,13 (3) a response to the interaction between HIV-associated neurodegeneration and substance use,39 or (4) CNS inflammation induced by lipopolysaccharide or other changes in blood plasma.31,40 The data presented show that brain HIV RNA was higher in subjects with MGNE versus without it (Fig. 2). That suggests that MGNE is a harbinger of unrestrained virus replication and worsening of HAND. Intensified therapy to control CNS viral replication in patients with MGNE would be a logical prophylactic measure to stem progression to more severe HAND and HIVE. Consistent with that suggestion, we note that subjects with MGNE were not as likely to be on HAART (Table 1), which could have increased their vulnerability to MGNE.

We explored whether there was a threshold value of brain HIV RNA that can discriminate between subjects likely to obtain NP benefit from targeting CNS HIV, from those less certain to obtain NP benefit. The 2 populations have overlapping distributions (Fig. 2). The best discrimination occurred at a value of about 15,000 copies per gram of brain neocortical tissue. Above that value, 73% of the subjects who are likely to obtain NP benefit are included, whereas 74% of the subjects with less certain NP benefit are excluded. Based upon the slope of the least squares regression line between brain and CSF HIV RNAs, the suggested “cutoff” value in CSF would be 1200 copies/mL. Great caution should be exercised regarding the clinical utility of using a cutoff value, because the interrelationship may not be linear and the NP data are quite variable.

Caveats of a retrospective clinicopathological autopsy study should be considered. (1) Clinical evaluations were obtained an average of 7 months before the brain specimens, and the plasma and CSF HIV loads could change with time (Table 1). Pre- versus postmortem CSF and plasma HIV RNA comparisons were available for just 28 and 12 subjects, respectively. CSF HIV RNA was indeed significantly higher in postmortem versus premortem samples (3.58 versus 2.51, P < 0.01); the plasma HIV RNA values were not significantly different. (2) Some subjects were not taking HAART when the brain specimen was obtained (Table 1). We evaluated the correlations between brain HIV RNA and NP T scores in the subset of 84 subjects on HAART and obtained the same result as with the entire NP cohort (ρ = −0.332, P = 0.0055). The same trend was present in the 41 subjects not on HAART (ρ = −0.223, P = 0.1617). (3) Comorbid conditions other than HIV infection such as stroke and cardiac disease could contribute to HAND. Pathology in organs other than the CNS including the heart was not significantly more prevalent in HAND (Table 1). Of interest was that brain infarcts tended to be more common with HAND versus without it (P < 0.11), which suggests that brain ischemia could have influenced the prevalence of HAND.


1. Heaton RK, Clifford DB, Franklin DR Jr, et al.. HIV-associated neurocognitive disorders persist in the era of potent antiretroviral therapy. Neurology. 2010;75:2087–2096.
2. McArthur JC, Steiner J, Sacktor N, et al. Human immunodeficiency virus-associated neurocognitive disorders: mind the gap. Ann Neurol. 2010;67:699–714.
3. Antinori A, Arendt G, Becker JT, et al.. Updated research nosology for HIV-associated neurocognitive disorders. Neurology. 2007;69:1789–1799.
4. Achim CL, Wang R, Miners DK, et al. Brain viral burden in HIV infection. J Neuropathol Exp Neurol. 1994;53:284–294.
5. Wiley CA, Soontornniyomkij V, RadhakrishnanL, et al.. Distribution of brain HIV load in AIDS. Brain Pathol. 1998;8:277–284.
6. Gendelman HE, Lipton SA, Tardieu M, et al.. The neuropathogenesis of HIV-1 infection. J Leukoc Biol. 1994;56:389–398.
7. Williams KC, Hickey WF. Central nervous system damage, monocyte and macrophages and neurological disorders in AIDS. Ann Rev Neurosci. 2002;25:537–562.
8. Kaul M, Garden GA, Lipton SA. Pathways to neuronal injury and apoptosis in HIV-associated dementia. Nature. 2001;410:988–994.
9. Glass JD, Fedor H, Wesselingh SL, et al. Immunocytochemical quantitation of human immunodeficiency virus in the brain: correlations with dementia. Ann Neurol. 1995;38:755–762.
10. Glass JD, Wesselingh SL, Selnes OA, et al. Clinical-neuropathologic correlation in HIV-associated dementia. Neurology. 1993;43:2230–2237.
11. Wiley CA, Achim CL. Human immunodeficiency virus encephalitis is the pathological correlate of dementia in acquired immunodeficiency syndrome. Ann Neurol. 1994;36:673–676
12. Anthony IC, Bell JE. The neuropathology of HIV/AIDS. Int Rev Psychiatry. 2008;20:15–24.
13. Everall I, Vaida F, Khanlou N, et al.. Cliniconeuropathologic correlates of human immunodeficiency virus in the era of antiretroviral therapy. J Neurovirol. 2009;15:360–370.
14. Gelman BB, Moore DH. HIV-1 neuropathology. In: Gendelman HE, Fox HS, Lipton SA, et al., eds. Neurology of AIDS. 3rd ed. New York: Oxford University Press, 2012;518–535.
15. Morgello S, Gelman BB, Kozlowski P, et al.. The national neuroAIDS tissue consortium: a new paradigm in brain banking with an emphasis on infectious disease. Neuropathol Appl Neurobiol. 2001;27:326–335.
16. Woods SP, Rippeth JD, Frol AB, et al.. Interrater reliability of clinical ratings and neurocognitive diagnoses in HIV. J Clin Exp Neuropsychol. 2004;26:759–778.
17. Janssen RS, Cornblath DR, Epstein LG, et al.. Nomenclature and research case definitions for neurological manifestations of human immunodeficiency virus type-1 (HIV-1) infection. Report of a working group of the American Academy of Neurology AIDS Task Force. Neurology. 1991;41:778–785.
18. Budka H, Wiley CA, Kleihues P, et al.. HIV-associated disease of the nervous system: review of nomenclature and proposal for neuropathology-based terminology. Brain Pathol. 1991;1:143–152.
19. Palmer S, Wiegand AP, Maldarelli F, et al.. New real-time reverse transcriptase-initiated PCR assay with single-copy sensitivity for human immunodeficiency virus type 1 RNA in plasma. J Clin Microbiol. 2003;41:4531–4536.
20. Gelman BB, Lisinicchia JG, Chen T, et al.. Prefrontal dopaminergic and enkephalinergic synaptic accommodation in HIV-associated neurocognitive disorders and encephalitis. J Neuroimmune Pharmacol. 2012;8:1–15.
21. Morgello S, Holzer CE 3rd, Ryan E, et al.. Interrater reliability of the psychiatric research interview for substance and mental disorders in an HIV-infected cohort: experience of the National NeuroAIDS Tissue Consortium. Int J Methods Psychiatr Res. 2006;15:131–138.
22. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B. 1995;57:289–300.
23. Masliah E, Roberts ES, Langford D, et al.. Patterns of gene dysregulation in the frontal cortex of patients with HIV encephalitis. J Neuroimmunol. 2004;157:163–175.
24. Sas AR, Bimonte-Nelson H, Smothers CT, et al.. Interferon-α causes neuronal dysfunction in encephalitis. J Neurosci. 2009;29:3948–3955.
25. Horiuchi M, Itoh A, Pleasure D, et al.. Cooperative contributions of interferon regulatory factor 1 (IRF1) and IRF8 to interferon-γ-mediated cytotoxic effects on oligodendroglial progenitor cells. J Neuroinflamm. 2011;8:. doi:10.1186/1742-2094-8-8.
26. Masliah E, DeTeresa RM, Mallory ME, et al.. Changes in pathological findings at autopsy in AIDS cases for the last 15 years. AIDS. 2000;14:69–74.
27. Morgello SM, Mahboob R, Yakoushina T, et al.. Autopsy findings in a human immunodeficiency virus-infected population over 2 decades. Influence of gender, ethnicity, risk factors, and time. Arch Pathol Lab Med. 2002;126:182–190.
28. Gelman BB, Soukup VM, Keherly MJ, et al.. Acquired neuronal channelopathies in HIV-associated dementia. J Neuroimmunol. 2004;157:111–119.
29. Ellis R, Hsia K, Spector SA, et al.. Cerebrospinal fluid human immunodeficiency virus type 1 RNA levels are elevated in neurocognitively impaired individuals with acquired immunodeficiency syndrome. Ann Neurol. 1997;42:679–688.
30. McArthur JC, McDermott MP, McClernon D, et al.. Attenuated central nervous system infection in advanced HIV/AIDS with combination antiretroviral therapy. Arch Neurol. 2004;61:1687–1696.
31. Ancuta P, Kamat A, Kunstman KJ, et al.. Microbial translocation is associated with increased monocyte activation and dementia in AIDS patients. PLoS ONE. 2008;3:e2516. doi:10.1371/journal.pone.0002516.
32. Gartner S. HIV infection and dementia. Science. 2000;287:602–604.
33. Lyons JL, Uno H, Ancuta P, et al.. Plasma sCD14 is a biomarker associated with impaired neurocognitive test performance in attention and learning domains in HIV infection. J Acquir Immune Defic Syndr. 2011;57:371–379.
34. Fox HS, Gold LH, Henriksen SJ, et al.. Simian immunodeficiency virus: a model for NeuroAIDS. Neurobiol Dis. 1997;4:265–274.
35. Gelman BB. Diffuse microgliosis associated with cerebral atrophy in the acquired immunodeficiency syndrome. Ann Neurol. 1993;34:65–70.
36. Bell JE, Brettel RP, Chiswick A, et al.. HIV encephalitis, proviral load, and dementia in drug users and homosexuals with AIDS: effect of neocortical involvement. Brain. 1998;121:2043–2052.
37. Xing HQ, Hayakawa H, Gelpi E, et al.. Reduced expression of excitatory amino acid transporter 2 and diffuse microglial activation in the cerebral cortex in AIDS cases with or without HIV encephalitis. J Neuropathol Exp Neurol. 2009;68:199–209.
38. Nebuloni M, Pellegrinelli A, Ferri A, et al.. Etiology of microglial nodules in brains of patients with acquired immunodeficiency syndrome. J Neurovirol. 2000;6:46–50.
39. Bell JE, Arango JC, Robertson R, et al.. HIV and drug misuse in the Edinburgh cohort. J Acquir Immune Defic Syndr. 2002;31(suppl 2):S35–S42.
40. Buttini M, Limonta S, Boddeke HW. Peripheral administration of lipopolysaccharide induces activation of microglial cells in rat brain. Neurochem Int. 1996;29:25–35.

dementia; encephalitis; HIVE; neurocognitive disorders; HAND; interferon

Supplemental Digital Content

© 2013 Lippincott Williams & Wilkins, Inc.