AIDS PATIENT CARE and STDs

Volume 13, Number 3, 1999

Mary Ann Liebert, Inc.

HIV-Associated Dementia: New Insights into Disease

Pathogenesis and Therapeutic Interventions
SUSAN SWINDELLS, M.D., JIALIN ZHENG, M.D., Ph.D.,
and HOWARD E. GENDELMAN, M.D.

ABSTRACT

Remarkable progress was made in recent years in the therapeutics of HIV-1-associated de-
mentia (HAD) and in unraveling the complex pathophysiology that follows viral invasion of
the central nervous system (CNS). Viral replication in and outside of the CNS was
signifi-
cantly reduced in HIV-1 infected subjects by new potent antiretroviral therapies. This has re-
sulted in partial repair of cellular immune function with improvement in, and the
prevention
of, neurologic deficits associated with progressive HIV-1 disease. In regard to HAD patho-
physiology, it is now known that CNS damage induced by HIV-1 infection occurs indirectly.
Neuronal loss is mediated through immune activation and viral infection of mononuclear
phagocytes (MPs) (brain macrophages and microglia). Cellular and viral factors secreted by
brain MPs produce, over time, neuronal damage and drop out. Viral growth in the brain
ap-
pears necessary, but not sufficient, to produce cognitive and motor impairments in affected
individuals. Indeed, the best predictor for neurologic impairment following HIV-1 infection
is the absolute number of immune-competent macrophages; not the level of viral
production
in affected brain tissue. As yet, an understanding of macrophage-related neurodegeneration
has not translated into significant improvements in the treatment of this
devastating com-
plication of HIV disease. Nonetheless, adjunctive antiinflammatory and neuroprotective ther-
apies are being developed. New ideas regarding HAD neuropathogenesis, and implications
for the diagnosis and treatment of HAD are summarized in this article.

CHANGING TRENDS IN 1-infected individuals experience significant

EPIDEMIOLOGY neurologic impairment.1'3'4 Cognitive and mo-
tor impairments occur commonly in patients

Progressi
vdisorders
e HIV
immunosuppression
ated with
and its associated
disease

commonly associ-
are
of the central nervous sys-
tem (CNS), the most frequent being HIV-1-as-
sociated dementia (HAD).1-2 Estimates of
disease prevalence vary among centers, but dur-
ing advanced disease more than 20% of HIV-
with symptomatic HIV disease, rising with ad-
vancing immunosuppression.2'4-7 Antiretroviral
therapy has impacted both the incidence and
prevalence of HAD in the United States, begin-
ning during the era of zidovudine monotherapy.
Data from cohort studies dem-onstrated a clear
decrease in the incidence of HAD following the

Center for Neurovirology and Neurodegenerative Disorders and Departments of Internal Medicine and Pathology
and Microbiology, University of Nebraska Medical Center, Omaha, Nebraska.
154
general use of zidovudine.8 This is more pro-
nounced now that potent combination anti-
retroviral therapy is the standard of care.9'10 Ex-
tensive use of protease inhibitors is associated
with dramatic declines in overall morbidity and
mortality of HIV disease, including HAD.11'12
The incidence of HAD is projected at 3-5%. Im-
plications for the long-term prevalence of HAD
are less clear. Specific concerns are emerging for
HAD and other HIV-associated clinical seque-
lae following antiretroviral treatment failure in
some patients, associated with the emergence of
drug-resistant viral isolates.13
PATHOPHYSIOLOGY
HAD occurs as an indirect consequence of
viral infection of the CNS. Virus appears nec-
essary, but not sufficient, to induce neurologic
impairments. Although HIV enters the brain
soon after viral infection and is present
throughout the course of the disease, the re-
sulting neurologic dysfunction occurs years
later and is associated with immunosuppres-
sion. Interestingly, HAD is not caused by ac-
tive HIV replication in neurons. Instead, the
target cells are mononuclear phagocytes (MPs)
(brain macrophages and microglia).14'15 These
phagocytes, once infected, become immunoac-
tive and produce a variety of secretory prod-
ucts, many of which are neurotoxic. The break-
down of homeostatic mechanisms in the brain
causes a cascade of inflammatory events lead-
ing to upregulation of cytokines and other im-
munomodulators, including nitric oxide (NO),
platelet activating factor (PAF), arachidonic
acid metabolites, quinolinic acid, leukotrienes,
and the proinflammatory cytokines tumor
necrosis factor-a, interferon-a and -ß, and
interleukin-1/3 and -6.16"21 A metabolic en-
cephalopathy ensues that damages, and even-
tually destroys, neurons. High levels of neuro-
toxic factors have been found in the CSF and
brains of HIV-infected patients with neurologic
impairment, although their precise roles in
HAD pathogenesis remain unclear.18-20'22 Our
group previously demonstrated that many
HIV-1-induced neurotoxins are present at high
levels in plasma, exceeding those in CSF with
severe cognitive/motor impairment.23 Down-
SWINDELLS ET AL.
regulation of neurotrophic factors also may con-
tribute to decreased neuronal differentiation
and survival.24 The pathophysiologic effects of
this metabolic cascade are believed to lead to
neuronal death mediated through the CXCR4
receptor on neurons.25 In this manner, the lig-
and for CXCR4, stromal derived growth factor
la, produced by glial cells or gpl20 secreted by
virus-infected macrophages, binds to this neu-
ronal receptor and affects a series of intracellu-
lar signaling events that ultimately lead to
changes in calcium channels at the N-methyl-D-
aspartate (NMDA) receptor.26 Such effects on
neuronal viability suggest a potential role for
NMDA or calcium channel antagonists in the
treatment of HAD. To this end, alternative/ad-
junctive therapies such as memantine or ni-
modipine are currently under study.
Importantly, other neurodegenerative condi-
tions, for example Alzheimer's disease (AD)
and Parkinson's disease (PD) appear to share
common pathogenetic pathways with HAD for
brain disease.27 Similar to HAD, for AD and PD
neuronal damage is caused indirectly, induced
through microglial activation by the accumu-
lation of abnormal host proteins or HIV. The
resulting inflammatory response in the brain
affects neuronal function and influences dis-
ease pathogenesis. Clearly, the direct effects of
neuronal dysfunction by the underlying patho-
logic process represents only a part of the over-
all disease. Glial activation in neurodegenera-
tive disorders can occur regardless of the
pathogenic insult, whether it be a microbial
pathogen, trauma, or an abnormal host protein.
Neuronal dysfunction is mediated by glial se-
cretory neurotoxic factors that amplify one an-
other through CNS autocrine immune path-
ways, creating a vicious cycle of neural damage
that becomes self-sustaining. Figure 1 illus-
trates the complex neurotoxin circuits produc-
ing the inflammatory cascade that leads to neu-
ronal damage and death in HAD. The best
characterized HAD-associated neurotoxic fac-
tors are summarized in Table 1.
NEUROPATHOLOGY
Neuropathologic abnormalities seen in the
patients with HAD are usually
brain tissue of
HIV-1-ASSOCIATED DEMENTIA 155
Activated brain macrophages
and/or microglia Neuron
Priming TOXINS
(NO, QUIN, PAF, glutamate,
arachidonic acid)_^

Activation by secondary
infection and immune
stimuli (pro-inflammatory
cytokines, chemokines)

Peripheral infection
Monocyte and immune activation Tcell
FIG. 1. The pathophysiologic process of HIV-1-associated dementia. During monocyte maturation, macrophages
acquire ability to be productively infected by HIV-1. These virus-infected macrophages become immune activated
the
by a process that remains incompletely understood, but likely relates to peripheral cytokines or opportunistic infec-
tion. The HIV-1-infected activated brain macrophage secretes a variety of factors that affect neural function and CNS
inflammation, including but not limited to proinflammatory cytokines, chemokines, arachidonic acid and its metabo-
lites, quinolinic acid (QUIN), nitric oxide (NO), and platelet activating factor (PAF). Such immune factors induce ad-
hesion molecule (VCAM-1 and E-selectin) expression on brain microvascular endothelium. The breakdown of the
blood-brain barrier by immune factors and adhesion molecules affects the transendothelial migration into the brain
of leukocytes, predominantly macrophages. The release of progeny virions results in infection of the brain's en-
dogenous macrophage, the microglia, perpetuating the inflammatory cascade. Brain macrophages and microglia are
primed by HIV-1 and activated by secondary infection and immune stimuli (proinflammatory cytokines, chemokines)
or by T-cells trafficking in and out of the nervous
system. Activated brain macrophages/microglia induce a profound
astrogliosis through proinflammatory cytokines. These same cytokines may affect astrocyte function or lead to neu-
rotoxicity. Inevitably, the macrophage/microglial toxins cause neuronal death possibly through two mechanisms,
necrosis and apoptosis, by inducing excessive Ca2+ influx and affecting N-methyl-D-aspartate (NMDA) receptor ac-
tivations. The degree of neurotoxic macrophage responses are regulated by astrocytes. These cells may suppress
macrophage secretory activities or increase it, depending on their own functional status and whether they are ex-
posed to virus. Astrocytes are a major source of NO and stromal cell-derived factor 1 alpha (SDF-la). SDF-la and
progeny HIV-1 released from infected macrophages/microglia may bind CXCR4 expressed on neurons or astrocytes
and influence neural function.

diffuse and predominantly localized to the composed of elongated microglia, macro-

white and deep grey matter regions. Myelin phages, multinucleated giant cells, and occa-
pallor and inflammatory infiltrates composed sional lymphocytes29; some may have
even
of macrophages and multinucleated giant cells necrotic centers. Leukencephalopathy is
dif- a
are the hallmarks of this disease process, al- fuse process, often without prominent signs of
though a spectrum of lesions has been identi- inflammation. Myelin pallor is often seen, with
fied from encephalitis to leukoencephalopa- actual myelin breakdown and debris, as well
thy.26'28'29 HIV-1 encephalitis is characterized as macrophages, including multinucleated gi-

by disseminated, multifocal microgranulomas ant cells and reactive astrogliosis.

156 SWINDELLS ET AL.

Table 1. Putative Neurotoxic Factors 1-associated cognitive motor complex," was in-
Secreted by Brain Macrophages and
Microglia in HIV-Associated Dementia
troduced. This refers to mild signs or symptoms
and minimal functional impairment, analogous
Proinflammatory cytokines: IL-1, IL-2, IL-6, TNF to stage 1 in AIDS dementia complex staging.
Excitatory amino acids: quinolinic acid and glutamate The terms "HIV-associated dementia" and
Eicosanoids/arachidonic acid metabolites
Quindinic acid "HIV-associated myelopathy" are used under
Nitric oxide this classification system for patients with more
Platelet activating factor
Intracellular calcium induced by N-methyl-D-aspartate severe impairment, analogous to AIDS demen-

(NMDA)
Viral coat protein (gpl20), Tat, Nef, rev
tia complex greater or equal to stage 2. Al-
though the newer classifications have utility for
experts in the field and in research, most clini-
cians continue to use the simpler original AIDS
CLINICAL MANIFESTATIONS dementia complex staging. HIV encephalitis or
encephalopathy are pathological terms rather
The characteristic clinical feature of HAD is than clinical terms, although HIV encephalopa-
disabling cognitive impairment, often accom- thy remains the preferred term for children.33
panied by motor dysfunction, behavioral Although synonymous, in adults, "HIV-1-asso-
changes, or both.1'3 Degrees of impairment in ciated dementia" is now the preferred term
each of these categories may vary between rather than "AIDS dementia complex."
patients, and within patients over time. Pa-
tients with cognitive impairment typically have
difficulty with concentration and memory, DIAGNOSTIC EVALUATIONS
progressing to frank confusion. Behavioral
changes observed include apathy, lethargy, HAD remains a diagnosis of exclusion.
and social withdrawal. Differentiating this syn- Nonetheless, there are a variety of evaluations
drome from depression can be difficult. Addi- that, taken together, permit a definitive diag-
tional information from friends or family close nosis. In addition to a careful history and phys-
to the patient can be helpful to the clinician. ical examination including neurologic assess-
Early motor symptoms include unsteady gait, ment, the following analyses have utility in this
limb weakness, and tremor. Late changes are setting.
indicative of global neurologic dysfunction:
weakness (prominently of the lower extremi- Neuropsychological testing
ties, causing an inability to walk unassisted)
lack of coordination, hyperreflexia, and blad- Neuropsychological testing is helpful as an
to the neurologic examination in that it
der or bowel dysfunction. Frontal lobe release adjunct
signs such as the snout reflex also may be de-
tected. Table 2. Staging for the AIDS Dementia Complex

Stage 0 (Normal): Normal mental and motor function

DISEASE CLASSIFICATION Stage 0.5 (Subclinical): Minimal symptoms of cognitive
or motor dysfunction, but no impairment of work or

performance of daily duties

When first identified, the etiology of HAD
was unclear. The constellation of signs and
symptoms was termed the AIDS dementia com-
plex. A five-part staging system was subse-
quently developed1-30 (Table 2). Since then, the
World Health Organization and the American
Academy of Neurology have each generated
more complex and detailed classifications.31'32
Separate categories for myelopathy and de-
mentia were included, and a new term, "HIV-
Stage 1 (Mild): Unequivocal evidence of functional,
intellectual, or motor impairment, but able to
perform all but the more demanding aspects of
work; fully ambulatory
Stage 2 (Moderate): Unable to perform demanding
daily activities, but can accomplish basic self-care
activities; ambulatory, but may require a single prop
Stage 3 (Severe): Major intellectual incapacity or motor
disability
Stage 4 (End stage): Nearly vegetative and/or mute;
paraparetic or paraplegic with double incontinence
Adapted with permission.30
HIV-1-ASSOCIATED DEMENTIA 157

better defines areas involved and improves of detecting early functional abnormalities be-
quantification of cognitive/motor impairment. fore structural changes occur, and several pre-
Neuropsychological testing generally demon- liminary studies have been reported.39 Meta-
strates deficits in areas of attention, memory, bolic neuroimaging using magnetic resonance
mental flexibility, and motor speed. Tests that spectroscopy (MRS) is a noninvasive method
have been found to be diagnostically sensitive of quantitating neuronal loss performed with
in patients with HAD include trail making B, conventional magnetic resonance imagers. As-
digit symbol, grooved pegboard, and the com- sessment of in vivo metabolism provides bio-
puterized reaction time.34-37 The influence of chemical information that can complement the
age, education, and comorbid conditions such morphologic information from the MRI exam-
as systemic illness, previous head trauma, or ination, and in a quantitative fashion. Several
substance abuse must be considered when in- groups, including our own, have reported a re-
terpreting results of neuropsychological test- duction in N-acetylaspartate, a marker for neu-
ing. Moreover, although complete neuropsy- ronal loss, using in vivo proton aH MRS in pa-
chological testing remains the gold standard, tients with advanced HIV disease41^14 (Fig. 2B
such an evaluation is not always readily avail- and C). aH MRS has demonstrated progressive
able or practical. Simpler test batteries that can neuronal loss over time in HIV-infected indi-
be administered by clinicians at the bedside viduals, and the degree of neuronal loss ob-
also may be used when necessary. For exam- served correlates directly with neurologic im-
ple, the HIV Dementia Scale (HDS) is a reliable pairment.44'45
and quantifiable adaptation of the Mini-Men- These metabolic changes often precede clin-
tal Status Examination.38 The HDS consists of ical or morphologic changes and may have
four subtests: timed written alphabet, recall of utility as markers for early diagnosis. Evidence
four items at 5 minutes, cube copy time, and is accumulating that 1H MRS may be help-
antisaccadic errors. This instrument reliably ful in the differential diagnosis of HAD and
detects HAD, distinguishes HIV-infected pa- other neurologic disorders associated with HIV
tients with and without mild dementia, and is disease, for example, other metabolic en-
not unduly influenced by depression. HDS cephalopathies,46 or in the differentiation of
scores are relatively independent of age and ed- CNS toxoplasmosis and lymphoma.39 Improv-
ucation, and the test is simple to administer, ing technology has translated into increasing
with minimal instrumentation required. sensitivity of MRS to detect metabolic alter-
ations and neuronal loss much earlier than con-
ventional neuroimaging, and in a quantitative
Neuroimaging studies fashion. Although available at an increasing
Radiographic and functional imaging stud- number of referral centers, MRS is still
ies can often further delineate the structural primarily a research tool and not generally re-
and metabolic effects of HIV on the brain. Com- imbursable by third-party payors.
puterized tomography of the brain characteris-
tically shows cerebral atrophy in the majority Cerebrospinal fluid analysis
of patients with dementia, but atrophie changes
also may be present in asymptomatic individ- Abnormalities of the cerebrospinal fluid
uals.39 Magnetic resonance imaging (MRI) of (CSF) in HAD are generally nonspecific, with
the brain is generally more sensitive and often mild elevations in protein or pleocytosis.47
shows additional white matter abnormalities40 Plasma HIV RNA assays are now used in both
(Fig. 2A). clinical trials and treatment of patients.48 It ap-
Several advanced neuroimaging techniques pears that HIV-1 RNA levels in CSF correlate
that measure functional changes are under with the presence of cognitive impairment
study in HAD. These include single-photon rather than with plasma HIV-1 RNA levels.49"54
emission computed tomography, positron- The precise relationship of HIV-1 RNA values
emission tomography, and functional MRI. in CSF and the risk of development or pro-
These techniques have the potential advantage gression of neurologic disease has not yet been
158 SWINDELLS ET AL.

FIG. 2. A: Magnetic resonance imaging (MRI) of the brain with gadolinium in a patient with HAD showing mild
generalized atrophy. Diffuse increased signal intensity is present throughout the white matter of the cerebral hemi-
spheres on proton density and T2 weighing consistent with diffuse leukoencephalopathy (arrows). (Reprinted with
permission.23) B: Myo, myoinositol; Cho, choline; Cr, crearme; NAA, N-acetylaspartate. Single-voxel (8 cm3) magnetic
resonance spectroscopy (MRS) in the parietal lobe of a patient with HAD exhibiting a decreased N-acetylaspartate
peak (indicative of neuronal loss). Brain MRI was normal. Creatine and choline peaks are within the normal range.
The NAA-to-creatine ratio is 1.38, choline-to-creatine ratio 0.96, and myoinositol-to-creatine ratio 0.73. C: Normal MRS
for comparison from a similar area of brain in an HIV-infected patient without dementia of the same age, sex, and
stage of disease. The NAA-to-creatine ratio is 1.56, choline-to-creatine ratio 0.66, and myoinositol-to-creatine ratio
0.63.

elucidated. Monitoring of HIV-1 RNA levels in
CSF and other body fluids or tissues is not gen-
erally available or currently recommended. Be-
cause the amount of HIV in the CNS is small,
even in patients with neurologic disease, CSF
RNA levels are also relatively small (usually
about 1 log lower than plasma levels). Unfor-
tunately, the false-negative rate of CSF RNA
values is high, and minor neurologic dysfunc-
tion is not associated with high CSF HIV RNA
levels.49'50 Although probably more biologi-
cally relevant, HIV RNA levels in the brain are
obviously difficult to ascertain. Most human
studies to date have been performed post-
mortem and are too small to subject to statisti-
cal analysis.55'56
HIV-1-ASSOCIATED DEMENTIA 159

THERAPY able assays, increase CD4 T-lymphocyte

The critical role of productive viral infection
in the neuropathogenesis of HAD supports the
notion that antiretroviral agents should be used
to treat dementia. This approach is both logi-
cal and well-founded in disease mechanisms.
However, despite remarkable recent advances
in the treatment of HIV disease, effective ther-
apies for dementia still lacking. Zidovudine
are
is the only drug to date with any demonstrated
efficacy in the treatment of dementia.57-61
However, optimal dosing and the potential role
of ZDV in the prevention of dementia remain
unclear. Despite its relatively poor CSF pene-
tration, didanosine has been shown to be use-
ful for AIDS dementia in children.62
Because of the critical role of cytokines and
other immunomodulators in the pathogenesis
of HAD, some investigators postulate that ad-
junctive therapies may be necessary to treat de-
mentia effectively. AIDS Clinical Trial Group
protocol 162 examined 41 HIV-1-infected pa-
tients with varying degrees of dementia to
evaluate the effects of nimodipine, a calcium
channel blocker.63 The results showed that ni-
modipine was safe, but the study was not suf-
ficiently powered for efficacy, and only a trend
toward improvement in neuropsychological
performance was seen. Clinical trials of other
compounds are underway.
Potent antiretroviral combination therapies
have been shown to suppress plasma viremia
below the limit of detection of currently avail-
counts, and improve morbidity and mortal-
ity.9'10 Nevertheless, less is known about the ef-
ficacy of potent antiretroviral therapy in the
prevention and treatment of CNS complica-
tions of HIV-1 disease, particularly dementia.
Many available antiretroviral compounds are
highly plasma protein bound, and therefore do
not have good penetration of the blood-brain
barrier. The degree to which some newer drugs
in development infiltrate CSF is not known,
and the CSF penetrance of available protease
inhibitors is negligible (Table 3). Whether or
not CNS penetrance is critical or if suppression
of plasma viremia may be adequate, preven-
tion and treatment for CNS HIV-related dis-
eases is not known. Some experts believe that
inclusion of a drug with good CSF penetrance
in antiretroviral combinations is an important
component, although there are no data to guide
the practicing physician.
Concern exists that HIV may continue to
replicate (albeit at low levels) in anatomic com-
partments such as lymphoid tissue, the genital
tract, and the CNS, to which most antiretrovi-
ral agents have limited access. These potential
viral sanctuaries may act as reservoirs, con-
tributing to antiretroviral drug resistance and
possibly leading to reinfection of plasma if an-
tiretroviral therapy is discontinued. HAD may
still emerge as a consequence of inadequate tis-
sue penetrance of medications, or back-muta-
tion of HIV-1 from CNS to the blood compart-
ment. Autopsy studies suggest that only tissues

Table 3. CSF Penetration of Approved Antiretroviral Agents

Expressed as Mean Ratio of CSF to Plasma Concentrations
Mean ratio of Plasma:
Drug CSF concentrations References
Zidovudine (Retrovir) 60% 69
Didanosine (Videx) 20% 70
Zalcitabine (Hivid) 20% 71
Stavudine (Zerit) 40% 72
Lamivudine (Epivir)
Abacavir
10%
36%
45%
•

11
76,75
Nevirapine (Viramune)
Delavirdine (Rescriptor) 0.4% 77
Efavirenz (Sustiva) 1%
Saquinavir (Invirase) Negligible 79
Ritonavir (Norvir) Negligible Personal communication,
Abbott Laboratories
Indinavir (Crixivan) 2.2%-76% (60% plasma protein bound) 81,80
Nelfinavir Not known (>98% plasma protein bound) 82
160
in which macrophages are the main target for
HIV-1 harbor viral sequences that cluster sep-
arately. In other words, the brain may be the
only true viral reservoir.64
Encouragingly, anecdotal reports already
suggest improvement in severe HAD after ini-
tiation of therapy with protease inhibitors.65
Profoundly demented patients treated with Connell, M.D., Associate Professor of Radiology,
triple antiretroviral therapy at our center have University of Nebraska Medical Center, Depart-
evidenced dramatic reversal of cognitive im- ment of Radiology, for providing the illustra-
pairment, concomitant with reduction of viral tions of magnetic resonance spectroscopy.
load in plasma, and to a lesser degree in CSF.23
Significantly, this was accompanied by a more
profound reduction in levels of potential neu-
rotoxins in plasma and in CSF. Taken together,
these data strongly support the notion that
HAD is a metabolic encephalopathy and may
be reversed by potent antiretroviral therapy.
Some new investigational agents may hold
promise for the treatment of HAD. For exam-
ple, 1592 U89 (abacavir) is a nucleoside reverse-
transcriptase inhibitor with good CSF penetra- 3. Price R, Brew B, Sidtis J, Rosenblum M, Scheck A,
tion and potent in vitro antiretroviral activity.
In a recently developed animal model, abacavir
reduced spreading infection in brains of severe 4. Bacellar
combined immunodeficiency (SCID) mice with
encephalitis to a greater degree than observed
with zidovudine, stavudine, and lamivu-
dine.66'67 Abacavir is currently under study in
patients with HAD and appears to be well tol-
erated.68
DISCUSSION
New insights from basic and clinical research
efforts have enabled significant progress in un-
raveling the complex pathogenesis of HAD.
The indirect mechanisms by which HIV causes
neurologic disease present opportunities for in-
novative therapeutic interventions. Despite the
fact that the brain is well protected and a dif-
ficult compartment to penetrate with antiviral
agents, effective suppression of viral replica-
tion in the periphery may be adequate to pre-
vent or treat disease. Careful epidemiologic
studies still are needed to fully assess the im-
pact of new therapies on HAD. Moreover, fur-
ther study is required to guide management
decisions in selecting antiretroviral combina-
tions that will prevent HIV-mediated CNS dis-
SWINDELLS ET AL.
ease. Effective and durable therapies for de-
mentia remain in development.
ACKNOWLEDGMENT
The authors are deeply indebted to James Mc-
REFERENCES
1. Navia BA, Jordan BD, Price RW. The AIDS dementia
complex: I. Clinical features. Ann Neurol 1986;19:
517-524.
2. McArthur JC, Seines OA, Glass JD, Hoover DR, Ba-
cellar H. HIV dementia. Incidence and risk factors. In:
Price RW, Perry SW, eds. HIV, AIDS and the Brain.
Vol. 72. New York: Raven, 1994:251-272.
Cleary P. The brain in AIDS: central system
nervous
HIV-1 infection and AIDS dementia complex. Science
1988;239:586-592.
H, Munoz A, Miller E. Temporal trends in
the incidence of HIV-1-related neurological disease;
multicenter AIDS cohort study, 1985-1992. Neurol-
ogy 1994;44:1892-1900.
5. Miller EN, Seines OA, McArthur JC, et al. Neuropsy-
chological performance in HIV-1 infected homosex-
ual men: the Multicenter AIDS Cohort Study (MACS).
Neurology 1990;40:197-203.
6. Seines OA, Miller E, McArthur JC, et al. HIV-1 infec-
tion: no evidence of cognitive decline during the
asymptomatic stages. Neurology 1990;40:204-208.
7. McArthur JC, Cohen BA, Seines OA, et al. Low preva-
lence of neurological and neuropsychological abnor-
malities in otherwise healthy HIV-1-infected indi-
viduals: Results from the Multicenter AIDS Cohort
Study. Ann Neurol 1989;26:601-611.
8. Portegies P, Enting RH, de Gans J, et al. Presentation
and course of AIDS dementia complex: 10 years of
follow-up in Amsterdam. AIDS 1993;7:669-675.
9. Deeks ST, Smith M, Holodniy M, Kahn JO. HIV-1 pro-
tease inhibitors. JAMA 1997;277:145-153.
10. Carpenter CJ, Fischl MA, Hammer SM, et al. Anti-
retroviral therapy for HIV infection in 1998, updated
recommendations of the international AIDS society—
USA panel. JAMA 1998;280:78-86.
11. Palella FJ, Delaney KM, Moorman AC, et al. Declin-
ing morbidity and mortality among patients with ad-
vanced human immunodeficiency virus infection. N
Engl J Med 1998;338:853-860.
12. Centers for Disease Control and Prevention. HIV/
AIDS Surveillance Reports. 1997;9:1-44.
HIV-1-ASSOCIATED DEMENTIA
13. Hirsch MS, Conway B, D'Aquila RT, et al. Antiretro-
viral drug resistance testing in adults with HIV in-
fection: implications for clinical management. Inter-
national AIDS Society—USA panel. JAMA 1998;279:
1984-1991.
14. Koenig S, Gendelman HE, Orenstein JM, et al. Detec-
tion of AIDS virus in macrophages in brain tissue
from AIDS patients with encephalopathy. Science
1986;233:1089-1093.
15. Wiley CA, Schrier RD, Nelson JA, Lanpert PW, Old-
stone MBA. Cellular localization of human immun-
odeficiency virus infection within the brains of ac-
quired immune deficiency syndrome patients. Proc
Nati Acad Sei USA 1986;83:7089-7093.
16. Bukrinsky MI, Nottet HSLM, Schmidrmayerova H,
Cheng-Mayer C, Dubrovsky L, Gendelman HE. Reg-
ulation of nitric oxide synthase activity in human im-
munodeficiency virus type 1 (HlV-l)-infected mono-
cytes: implications for HIV-associated neurological
disease. J Exp Med 1995;181:735-745.
17. Heyes MP, Brew BJ, Marin A, et al. Quinolinic acid
in cerebrospinal fluid and serum in HIV-1 infection:
relationship to clinical and neurological status. Ann
Neurol 1991;29:202-209.
18. Griffin DE, Wesselingh SL, McArthur JC. Elevated
central nervous system prostaglandins in human im-
munodeficiency virus-associated dementia. Ann
Neurol 1994;35:592-607.
19. Gelbard HA, Nottet HSLM, Swindells S, et al. Platelet-
activating factor: a candidate human immunodefi-
ciency virus type-1 induced neurotoxin. J Virol
1994;68:4628-4635.
20. Guilian D, Yu J, Li X, et al. Study of receptor-medi-
ated neurotoxins released by HIV-1-infected mono-
nuclear phagocytes found in human brain. J Neurosci
1996;18:3139-3153.
21. Epstein L, Gendelman HE. Human immunodeficiency
virus type I infection of the nervous system: pathogenic
mechanisms. Ann Neurol 1993;33:429-436.
22. Glass JD, Fedor H, Wesselingh SL, McArthur JC. Im-
munocytochemical quantitation of human immuno-
deficiency virus in the brain: correlations with de-
mentia. Ann Neurol 1995;38:755-762.
23. Gendelman HE, Zheng J, Coulter CL, et al. Suppres-
sion of inflammatory neurotoxins by highly active an-
tiretroviral therapy in HIV dementia. J Infect Dis
1998;178:1000-1007.
24. Zheng J, Thylin M, Persidsky Y, Ghorpade A, Leis-
man GB, Gendelman HE. Unraveling the mechanisms
for neuronal dysfunction in HIV-1 dementia. Fifth
Conference on Retroviruses and Opportunistic Infec-
tions, Chicago, IL, February 1-5, 1998 [Session 56, p.
163].
25. Zheng J, Thylin MR, Ghorpade A, et al. Linkages be-
tween intracellular CXCR4 signaling, neuronal apop-
tosis, and the neuropathogenic mechanisms for HIV-
1 associated dementia. J Clin Invest (in press).
26. Lipton SA, Gendelman HE. Dementia associated with
the acquired immunodeficiency syndrome. N Engl J
Med 1995;332:934-940.
161
27. McGeer PL, McGeer EG. The inflammatory response
system of brain: implications for therapy of Alz-
heimer and other neurodegenerative diseases. Brain
Res Rev 1995;21:195-218.
28. Navia BA, Cho ES, Petito CK, Price RW. The AIDS
dementia complex: II. Neuropathol Ann Neurol
1986;19:525-535.
29. Budka H. HIV-Associated Neuropathology. In: Gen-
delman HE, Lipton SA, Epstein L, Swindells S, eds.
The Neurology of AIDS. New York: Chapman & Hall,
1998:241-260.
30. Price R, Brew B. The AIDS dementia complex. J Infect
Dis 1988;158:1079-1083.
31. World Health Organization. 1990 World Health Or-
ganization consultation on the neuropsychiatrie as-
pects of HIV-1 infection. AIDS 1990;49:935-936.
32. Nomenclature and research case definitions for neu-
rologic manifestations of human immunodeficiency
virus-type 1 (HIV-1) infection. Report of a Working
Group of the American Academy of Neurology
AIDS Tast Force [review]. Neurology 1991;41:
778-785.
33. Mintz M. Clinical features of HIV infection in chil-
dren. In: Gendelman HE, Lipton SA, Epstein L,
Swindells S, eds. The Neurology of AIDS. New York:
Chapman & Hall, 1998:385-407.
34. Goethe KE, Mitchell JE, Marshall DW, et al. Neu-
ropsychological and neurological function of human
immunodeficiency virus seropositive asymptomatic
individuals. Arch Neurol 1989;46:129-133.
35. Wilkie F, Eisdorfer C, Morgan R, Lowenstein DA, Sza-
pocznik J. Cognition in early human immunodefi-
ciency virus infection. Arch Neurol 1990;47:433^40.
36. Worth JL, Savage CR, Baer L, Esty E, Navia BA. Com-
puter-based screening for AIDS dementia complex.
AIDS 1993;7:677-681.
37. McArthur JC, Grant I. HIV Neurocognitive disorders:
In: Gendelman HE, Lipton SA, Epstein L, Swindells
S, eds. In: The Neurology of AIDS. New York: Chap-
man & Hall, 1998:499-523.
38. Power C, Seines OA, Grim JA, McArthur JC. HIV de-
-
mentia scale: a rapid screening test. J AIDS Hum
Retrovir 1995;8:273-278.
39. Gonzalez RG, Ruiz A, Tracey I, McConnell J. Struc-
tural, functional and molecular neuroimaging in
AIDS. In: Gendelman HE, Lipton SA, Epstein L,
Swindells S, eds. The Neurology of AIDS. New York:
Chapman & Hall, 1998:333-352.
40. Jernigan TL, Archibald S, Hesselink JR, et al. MRI
morphometric analysis of cerebral volume loss in HIV
infection. Arch Neurol 1993;50:250-255.
41. Jarvik J, Lenkinski R, Grossman R, Gomori J, Schnall
M, Frank I. Portón MR spectroscopy of HIV-infected
patients: characterization of abnormalities with imag-
ing and clinical correlation. Radiology 1993;186:
739-744.
42. Chong W, Paley M, Wilkinson I, et al. Localized cere-
bral proton MR spectroscopy in HIV infection and
AIDS. AJNR 1994;15:21-25.
43. Menon DK, Ainsworth JG, Cox IJ, et al. Proton MR
162
spectroscopy of the brain in AIDS dementia complex.
J Comput Assist Tomogr 1992;16:538-542.
44. McConnell JR, Swindells S, Ong CS, et al. Prospective
utility of cerebral proton magnetic resonance spec-
troscopy in monitoring HIV infection and its associ-
ated neurological impairment. AIDS Res Hum Retro-
1994;10:977-982.
vir
45. Meyerhoff DJ, Mackay S, Bachman L, et al. Reduced
brain N-acetylaspartate suggests neuronal loss in
cognitively impaired human immunodeficiency
virus-seropositive individuals. Neurology 1993;43:
509-515.
46. Swindells S, McConnell JR, McComb RD, Gendelman
HE. Utility of cerebral proton magnetic resonance
spectroscopy in differential diagnosis of HIV-related
dementia. J Neurovirol 1995;1:268-274.
47. Navia BA, Price RW. Clinical and biologic features of
the AIDS dementia complex. In: Gendelman HE, Lip-
ton SA, Epstein L, Swindells S, eds. The Neurology of
AIDS. New York: Chapman & Hall, 1998:229-240.
48. Saag MS, Holodniy M, Kuritzkes DR, et al. HIV viral
load markers in clinical practice: recommendations of
an International AIDS Society—USA Expert Panel.
Nat Med 1996;2:625-629.
49. Brew BJ, Pemberton L, Cunningham P, Law MG. Lev-
els of human immunodeficiency virus type 1 RNA in
cerebrospinal fluid correlate with AIDS dementia
stage. J Infect Dis 1997;175:963-966.
50. McCutchan JA, Ellis R, Hsia K, et al. Relationship of
cerebrospinal fluid HIV-1 RNA levels (CSF RNA) to
plasma RNA, CSF pleocytosis, stage of HIV infection
and cognitive functioning. Fourth Conference on
Retroviruses and Opportunistic Infections Meeting,
Washington, DC, January 1997 [Abstract 7].
51. Robertson K, Fiscus S, Kapoor C, et al. CSF and
plasma viral load and AIDS dementia: a preliminary
study. Fourth Conference on Retroviruses and Op-
portunistic Infections Meeting, Washington, DC, Jan-
uary 1997 [Abstract 349].
52. Bossi P, Dupin N, Coutellier A, et al. The level of hu-
man immunodeficiency virus (HIV) type 1 RNA in
cerebrospinal fluid as a mark of HIV encephalitis. Clin
Infect Dis 1998;26:1072-1073.
53. Lanier ER, McClernon D, Stocum M, et al. CSF HIV
RNA: relationship to HIV-associated neurological
disease. XI International Conference on AIDS, 1996
[Abstract We.B.3287].
54. LeTendre S, Ellis R, LeTendre SL, Heaton RK, Atkin-
son JH, McCutchan JA. Change in CSF RNA level cor-
relates with the effects of maximal antiretroviral ther-
apy on HIV associated neurocognitive disorder. 12th
World AIDS Conference, Geneva, June 28-July 3,1998
[Abstract 32198].
55. Fujimura RK, Eler M, Goodkin K, Meinz S, Gonzalez
J, Shapshak P. Mapping of distribution of HIV DNA
load in brains of AIDS patients with HIV associated
dementia. Xlth International Conference on AIDS,
1996.
56. Wiley C, Soontornniyomkis V, Achim C, et al. Distri-
bution of brain HIV load in AIDS. 12th World AIDS
SWINDELLS ET AL.
Conference, Geneva, Switzerland, June 28-July 3,
1998 [Abstract #32203].
57. Yarchoan R, Berg G, Brouwers P, et al. Response of
human immunodeficiency virus-associated neuro-
logical disease to 3'-azido-3'-deoxythymidine. Lancet
1987;8525:131-135.
58. Schmitt FA, Bigleg JW, McKinnis R, et al. Neuropsy-
chological outcome ofazidothymidine (AZT) in the
treatment of AIDS and AIDS-related complex: a dou-
ble-blind placebo-controlled trials. N Engl J Med
319:1573-1578,1988.
59. Pizzo PA, Eddy J, Falloon J, et al. Effect of continu-
ous intravenous infusion of zidovudine (AZT) in chil-
dren with asymptomatic HIV infection. N Engl J Med
1988;319:889-896.
60. Portegies P, de Gans J, Lange JM, et al. Declining in-
cidents of AIDS dementia complex after introduction
of zidovudine treatment. Br Med J 1989;299:819-821.
61. Sitdis JJ, Gatsonis C, Price RW, et al. Zidovudine treat-
ment of the AIDS dementia complex: results of a
placebo-controlled trial. Ann Neurol 1993;33:343-349.
62. Butler KM, Husson RN, Balis FM, et al. Dideoxyino-
sine in children with symptomatic human immuno-
deficiency virus infection. N Engl J Med 1991;324:
137-144.
63. Navia BA, Dafni U, Simpson D, et al. A phase I/II
trial of nimodipine for HIV-related neurologic com-
plications. Neurology 1998;51:221-228.
64. van't Wout AB, Ran LJ, Kuiken CL, Kootstra NA, Pals
ST, Schuitemaker H. Analysis of the temporal rela-
tionship between human immunodeficiency virus
type 1 quasispecies in sequential blood samples and
various organs obtained at autopsy. J Virol 1998;72:
488-496.
65. Filippi CG, Sze G, Färber SJ, Shahmanesh M, Selwyn
PA. Regression of HIV encephalopathy and basal gan-
glia signal intensity abnormality at MR imaging in pa-
tients with AIDS after the initiation of protease in-
hibitor therapy. Radiology 1998;206:491-198.
66. Persidsky Y, Limoges J, McComb R, et al. Human im-
munodeficiency virus encephalitis in SCID mice. Am
J Pathol 1996;149:1027-1053.
67. Limoges J, Persidsky Y, Rasmussen J, Lanier R, Rein-
hard J, Gendelman HE. 1592 prevents viral spread in
brain tissue of SCID mice with HIV-1 encephalitis.
Studies of HIV-1 neurovirulence in a SCID mouse
model of viral encephalitis. Fifth Conference on Retro-
viruses and Opportunistic Infections, Chicago, IL,
February 1-5, 1998 [Abstract 56].
68. Brew BJ, Brown SJ, Catalan J, et al. Safety and efficacy
of Abacavir (ABC, 1592) in AIDS dementia complex
(Study CNAB 3001). 12th World AIDS Conference,
Geneva, June 28-July 3, 1998 [Abstract 32192].
69. Retrovir (zidovudine) package insert. Glaxo Well-
come, Research Triangle Park, NC, revised Septem-
ber 1996.
70. Videx (didanosine) package insert. Bristol-Myers
Squibb, Princeton, NJ, revised March 1995.
71. Hivid (zalcitabine) package insert. Roche Laborato-
ries, Nutley, NJ, June 1994.
HIV-1-ASSOCIATED DEMENTIA
72. Zerit (stavudine) package insert. Bristol-Myers
Squibb, Princeton, NJ, revised October 1996.
73. Epivir (lamivudine) package insert. Glaxo Wellcome,
Research Triangle Park, NC, August 1996.
74. Price RW, Aweeka F, Bellibas SE, Staprans S, No-
vakovic-Agopian T, Mahoney W, Symonds W. Phar-
macokinetics of abacavir (1592U89) in cerebrospinal
fluid (CSF) using multiple lumbar punctures of few
subjects and "sparse sampling" techniques. Int Conf
AIDS 1998;12:827-8 [Abstract 42271].
75. Silverstein H, Riska P, Johnstone JN, et al. Nevirap-
ine, a nonnucleoside reverse transcriptase inhibitor,
freely enters the brain and crosses the placental bar-
rier. Fourth Conference on Retroviruses and Oppor-
tunistic Infections, Washington, DC, January 1997
[Abstract 567].
76. Glynn SL, Yazdanian M. In vitro blood-brain barrier
permeability of nevirapine compared to other HIV an-
tiretroviral agents. J Pharm Sei 1998;87:306-310.
77. Rescriptor (delavirdine) package insert. Pharmacia
Upjohn, Kalamazoo, MI, April 1997.
78. Fiske WD, Nibbelink DW, Brennan JM, Mutlib AE,
Ruiz NM. DMP 266 Cerebrospinal fluid concentra-
tions (CSF) after oral administration. 37th Annual
Meeting of the Interscience Conference on Antimi-
163
crobial Agents and Chemotherapy (ICAAC), Toronto,
Canada, September 28, 1997 [Abstract A12].
79. Invirase (saquinavir mesylate) package insert.
Roche Laboratories, Nutley, NJ, revised September
1996.
80. Collier AC, Marra C, Coombs RW, Zhong L, Stone J,
Nguyen B-Y. Cerebrospinal fluid (CSF) indinavir
(IDV) and HIV RNA levels in patients on chronic in-
dinavir therapy. 35th Annual Meeting of the Infec-
tious Diseases Society of America (IDSA), San Fran-
cisco, CA, September 13-16,1997 [Abstract 22].
81. Crixivan (indinavir sulfate) package insert. Merck &
Co., West Point, PA, March 1996.
82. Viracept (nelfinavir) package insert. Agouron Phar-
maceuticals, La Jolla, CA, April 1997.
Address reprint requests to:
Susan Swindells, M.D.
University of Nebraska Medical Center
Department of Internal Medicine
Section of Infectious Diseases
985400 Nebraska Medical Center
Omaha, NE 68198-5400
E-mail: sswindel@aries.unmc.edu