Cerebral Infarction in Adult AIDS Patients Observations From the Edinburgh HIV Autopsy Cohort

M.D. Connor, G.A. Lammie, J.E. Bell, C.P. Warlow, P. Simmonds and R.D. BrettleAutopsy Cohort

Cerebral Infarction in Adult AIDS Patients: Observations From the Edinburgh HIV

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 2000 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/01.STR.31.9.2117

2000;31:2117-2126Stroke.

http://stroke.ahajournals.org/content/31/9/2117World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer process is available in the

Request Permissions in the middle column of the Web page under Services. Further information about thisOnce the online version of the published article for which permission is being requested is located, click

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Strokein Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on November 24, 2014http://stroke.ahajournals.org/Downloaded from by guest on November 24, 2014http://stroke.ahajournals.org/Downloaded from

http://stroke.ahajournals.org/content/31/9/2117

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://stroke.ahajournals.org//subscriptions/

http://stroke.ahajournals.org/


Cerebral Infarction in Adult AIDS PatientsObservations From the Edinburgh HIV Autopsy Cohort

M.D. Connor, FCP(SA), FCP(Neurol)(SA); G.A. Lammie, MD, MRCPath; J.E. Bell, MD, MRCPath;C.P. Warlow, MD, FRCP; P. Simmonds, PhD, MRCPath; R.D. Brettle, MD, FRCP(Ed)

Background and Purpose—Autopsy series of patients with AIDS have found a 4% to 29% prevalence of cerebralinfarction. Little is known of the prevalence of cerebral infarction when not associated with non-HIV central nervoussystem (CNS) infection, lymphoma, or cardioembolic sources. Clinical correlation has seldom been available. Wedescribe the pathological and clinical features of patients from the Edinburgh HIV Cohort Study found to have hadcerebral infarcts without evidence of non-HIV CNS infection, CNS lymphoma, or cardioembolic sources at autopsy.

Methods—From 183 autopsy cases, 26 without evidence of opportunistic cerebral infection or lymphoma were selected.These 26 cases went through a second selection process in which the presence of cerebral infarction, in the absence ofthe conditions mentioned, was verified. Histology and clinical records for the remaining patients were reviewed.

Results—Ten (5.5%) cases fulfilled the inclusion criteria and demonstrated similar hypoxic-ischemic lesions. Small-vesselthickening was seen in all cases, and perivascular space dilatation, rarefaction, and pigment deposition, with vessel wallmineralization and perivascular inflammatory cell infiltrates, were seen in some cases. Vasculitis was not found. Onepatient had had a transient ischemic attack, and no patient had had a stroke.

Conclusions—Cerebral infarcts in HIV-infected patients are not common in the absence of cerebral non-HIV infection,lymphoma, or embolic sources. We found an HIV-associated vasculopathy with similar features in all risk groups. InAIDS patients presenting with stroke or transient ischemic attack, potentially treatable causes, such as cerebralcoinfection or tumor, should be sought.(Stroke. 2000;31:2117-2126.)

Key Words: acquired immunodeficiency syndromen cerebral infarctionn pathology

Human immunodeficiency virus (HIV) invades the centralnervous system (CNS) early.1,2 Any neurological man-

ifestations may result from opportunistic infections, neopla-sia, the immunological and metabolic response to HIV,iatrogenic causes, or corisk factors (eg, injection drug use), orthey may be related directly to HIV itself.2,3 Indeed, postmor-tem neuropathological series reveal that CNS abnormalitiesoccur in up to 90% of patients with advanced AIDS.2,4,5 In aclinical series, up to 70% of HIV-infected patients eventuallydevelop neurological manifestations.6 Cerebral infarction inan HIV-infected patient may, at least in theory, result fromany one of the causes listed in Table 1,7,8 but a strokelikeclinical presentation is unusual, occurring in'1%, withcerebral hemorrhage constituting 32% of the cases.9 On theother hand, autopsy series in adults have found a much higherprevalence of cerebral infarction (4% to 29%),9,12,16,28,43–46

but the majority of studies do not distinguish betweencerebral infarcts associated with, and therefore potentiallycaused by, non-HIV CNS infection, CNS lymphoma, or

cardioembolic sources and cerebral infarcts occurring in theabsence of these conditions.5,11,12,14,44–50Furthermore, patientrisk factors for cerebral infarction, including drug misuse, areseldom described,17,22,45,47,48,50and no clinical correlation hasbeen available or very little is known about the patient’sclinical course before death.11,12,14,16,17,22,28,44–47,49–51There-fore, it is far from clear whether HIV itself causes cerebralinfarction and (if it does) whether the patients experience astroke, transient ischemic attack (TIA), or other neurologicalsymptoms that are indicative of neurovascular disease beforedeath.8

The principal aims of the present study were (1) toestablish whether cerebral infarcts occurred in the EdinburghHIV autopsy cohort, in the absence of causes other thanpossibly HIV infection itself; (2) to describe the pathologicalcharacteristics of any cerebrovascular lesions; and (3) toretrospectively correlate the autopsy findings with relevantclinical features (in particular, symptoms and signs thatwould have been in keeping with neurovascular disease) andvascular risk factors.

Received February 16, 2000; final revision received June 5, 2000; accepted June 6, 2000.From the Neurology Unit (M.D.C.), Department of Medicine, University of the Witwatersrand, Johannesburg, South Africa; the Department of

Pathology (G.A.L.), University of Wales College of Medicine, Cardiff, Wales; and the Department of Pathology (Neuropathology) (J.E.B.), theDepartment of Clinical Neurosciences (C.P.W.), the Virus Evolution Group, Laboratory for Clinical and Molecular Virology (P.S.), and the RegionalInfectious Disease Unit (R.D.B.), University of Edinburgh, Edinburgh, Scotland.

Reprint requests to M.D. Connor, FCP(SA), Neurology Unit, Department of Medicine, Faculty of Health Sciences, Medical School, 7 York Rd,Parktown 2193, South Africa. E-mail [email protected]

© 2000 American Heart Association, Inc.

Stroke is available at http://www.strokeaha.org

2117 by guest on November 24, 2014http://stroke.ahajournals.org/Downloaded from


Subjects and MethodsStudy Design and PopulationWe performed an observational pathology study of patients from theEdinburgh HIV Cohort Study, with clinical correlation. The Edin-burgh HIV Cohort52,53 at the time of the study consisted of 1372HIV-infected patients (28% homosexual patients, 48% intravenousdrug users, 4% patients with blood product–related HIV, 17%heterosexual patients, and 3% patients with other or unknown riskfactors for HIV infection), all prospectively followed up, of whom535 had died. One hundred eighty-three (34%) of the patients (25%homosexual patients, 65% intravenous drug users, 5% heterosexualpatients, and 5% other) had a full neuropathological post mortem.Forty-four of the 183 cases in the autopsy cohort were pre-AIDScases; ie, they had not developed AIDS. Autopsies were requested onall patients, and results were not biased by dementia or otherunderlying clinical illness. There is virtually no overlap between the2 major “at-risk” groups, with only 1 homosexual patient havingbeen an injection drug user.

MethodsAfter formalin fixation, the brain was coronally sliced and blockedaccording to an established protocol to include all areas of the

cerebrum, brain stem, and cerebellum.52 Sections of the frontal lobe(in all cases) and additional blocks including temporal and occipitalcortex and basal ganglia (in most cases) were stained immunohisto-chemically by use of an antibody to HIV p24 core protein (Du Pont;diluted 1:200 and with microwave pretreatment, immersed in 0.01mol/L citric acid for 335 minutes).

Case selection was performed in 2 stages. First, all 183 autopsyreports were screened, and 157 cases were not selected for furtherdetailed review if (1) the report clearly documented normal braintissue; (2) an opportunistic cerebral infection was detected, such astoxoplasmosis, fungal infection, cytomegalovirus, varicella-zostervirus, herpes simplex virus, or mycobacterium; or (3) cerebrallymphoma was found. Cases with criteria 2 and 3 above wereexcluded because these conditions are all accepted causes of cerebralinfarction in HIV-infected patients.4,11,12,16–23,26,27,54Second, theCNS histology in the remaining 26 cases then went through anotherselection process, in which we checked that cases indeed hadpathologically verified cerebral infarcts without pathological evi-dence of non-HIV cerebral infection, lymphoma, hemorrhage, trau-matic injury, or potential sources of cerebral emboli. Sixteen caseswere excluded for the following reasons (number of cases inparentheses): no infarcts or ischemic lesions seen on CNS pathologyreview (n59), primary intracerebral hemorrhage (n52), non-HIV

TABLE 2. Summary of Number and Distribution of Cerebral Infarcts, Small Cerebral Vessel Lesions, and HIV Pathology

CaseRisk

Group

No. of Cerebral Infarcts

Total No. ofInfarcts

Small-Vessel Pathology

CX BG Th BrS CBM WMHyaline

Thickening

Perivascular SpaceDilation/

RarefactionVessel

MineralizationPerivascular

Pigment

PerivascularLymphocytes

and Macrophages

1 H 1 0 0 0 0 0 1 Mild Mild 1 1 2

2 H 95 34 1 2 2 0 134 Mild Mild 2 1 2

3 H 13 27 2 0 0 0 42 M/S M/S 2 1 1

4 H 38 13 0 0 0 0 51 M/S M/S 2 1 2

5 DU 17 45 10 0 11 0 83 M/S M/S 2 1 1

6 DU 34 19 0 0 0 0 53 M/S Mild 1 1 1

7 DU 3 3 0 0 0 0 6 M/S M/S 1 1 1

8 DU 7 6 0 0 0 0 13 M/S M/S 2 1 1

9 DU 69 32 7 0 0 0 108 M/S Mild 1 1 2

10 Hem 10 2 0 0 0 1 13 2 Mild 2 1 2

H indicates homosexual; DU, drug user; Hem, hemophiliac; CX, cortex; BG, basal ganglia; Th, thalamus; BrS, brain stem; CBM, cerebellum; WM, white matter;MNGC, multinucleate giant cell; IHC, immunohistochemistry; PCR, polymerase chain reaction; GM, grey matter; M/S, moderate/severe; 1, present; 2, absent; andND, not determined.

*Proviral load/mg DNA in frontal lobe.†In case 6, PCR was not performed separately in GM and WM.

TABLE 1. Potential Causes of Cerebral Infarction in HIV-Infected Patients

General Specific

Cardiac Nonbacterial thrombotic endocarditis,4,10–13 infective endocarditis14 (injection druguse), HIV myocarditis15 with thrombus

Vascular (vasculitis/vasculopathy) Cytomegalovirus,11,16–18 varicella-zoster virus,12,13,19–24 herpes simplex virus,24

mycobacterium tuberculosis,16,20 syphilis,16,25 cryptococcosis,16,20 mucormycosis,16

aspergillosis,16 candida albicans,17 coccidioidomycosis,26 toxoplasmosis,16,20,27

trypanosomiasis26

?HIV vasculopathy26,28–30 or vasculitis31 (eosinophilic,32 necrotizing,33

granulomatous34)

Abnormalities of coagulation Protein S deficiency,27,35,36 antiphospholipid antibodies,37–40 disseminatedintravascular coagulation4,11,12

Cerebral opportunistic infection/neoplasms Toxoplasmosis,20 lymphoma17,27,41

Injection drug use Cocaine, heroin14,15,24

Other associations Hepatitis B antigenemia42

2118 Stroke September 2000

by guest on November 24, 2014http://stroke.ahajournals.org/Downloaded from


viral encephalitis (n51), basal meningitis (n51), cerebral abscess(n51), bacterial endocarditis with vegetations (n51), and extensivehead trauma (n51). All histology on the remaining 10 cases wasreviewed in detail, with the site and number of infarcts, theirmorphological features, any associated vessel abnormalities, andevidence of HIV encephalitis documented.

HIV-1 proviral load was determined in fresh, frozen, frontal lobetissue from 9 of the 10 cases by nested polymerase chain reactionwith the use of conserved primers from thepol region, followed bycompetitive polymerase chain reaction with the use of templatesconstructed frompol sequences.52 In 8 cases, this was performedseparately on gray and white matter.

Clinical ReviewFor the 10 selected cases, all hospital and clinical records werereviewed, including general practitioner (family doctor) letters,clinical notes, inpatient notes, nursing notes, and laboratory andradiology reports. The following clinical details were elicited: patientdemographic details, length of clinical follow up, risk factors forHIV, intravenous and other drug misuse at the time of death andbefore death, prescribed medications, disease stage, vascular riskfactors including family history of vascular disease, documenteddiagnosis of ischemic heart disease, cardiac failure or arrhythmia,peripheral vascular disease, hypertension (defined as requiring treat-ment or based on a clinical diagnosis made during the finaladmission), diabetes mellitus (based on treatment or diagnosisdocumented in the clinical record), and hypercholesterolemia (re-quiring treatment, dietary modification, or documented in the clinicalrecord). Furthermore, any history of pulmonary embolism, deepvenous thrombosis, emboli to any site, or significant hypoxia (PaO2

,50 mm Hg on arterial blood gas measurement) during the lastadmission was noted. Detail was then sought on any history ofneurological problem at any time, including the nature of the onset,history and examination findings, investigations performed, neuro-logical and psychiatric opinion sought, and relation to the onset of adiagnosis of AIDS dementia complex. Records of investigationswere reviewed with emphasis on inflammatory markers (erythrocytesedimentation rate and C-reactive protein), tests of blood coagula-tion, cholesterol levels, serology for any of the infective conditionslisted in Table 1, and chest x-ray reports. ECGs were reviewed, andall neuroradiological reports performed at any time were reviewed inthe context of any documented neurological signs and symptoms.

ResultsTen cases out of 183 autopsies (5.5%) fulfilled all theinclusion criteria. Relevant pathological findings and frontallobe viral load estimates are shown in Table 2.

Gray Matter LesionsThe histological appearance of gray matter lesions (Figure 1)was similar in all but one case (case 1). Typical lesions weremultiple and small, composed of palely stained and rarefiedbrain tissue, with loss of neurons and variable numbers ofassociated reactive astrocytes and microglial cells. Somelesions were noncavitated, whereas others were completelycavitated, with other lesions exhibiting a spectrum of degreesof cavitation between these 2 extremes. Foamy macrophageswere seen in small numbers in a proportion of lesions,depending on the histological age of the lesion and the degreeof tissue damage. Thus, macrophages were most conspicuousin subacute or resolving lesions with extensive tissue necrosisand cavitation. The histological age of lesions varied from old(at least 3 months) to acute (#1 day); this variation wasapparent both between and within cases. Thus, in some cases,lesions were of a similar histological age; in some, theappearances suggested$2 “crops” of lesions; and in others,a wide range of lesion ages was seen. The number of lesionsper brain varied, with a clear predilection for neocortex andbasal ganglia. Neocortical lesions were relatively evenlydistributed throughout all 4 lobes. In 2 cases, there was anapparent clustering of lesions in superficial cortical layers 2and 3 (cases 4 and 9), whereas in others, there was anapparent preference for deeper cortical layers (cases 3 and 7).Less commonly, lesions were identified in the thalamus (4cases), cerebellum (2 cases), and brain stem (1 case). In case1, the appearances were somewhat different, in that there wasa small, old, solitary, linear, cavitated infarct in the parietalcortex.

In no case was there evidence of global, hemispheric, orregional vessel territory infarction, wedge-shaped corticalinfarction, or intracerebral or subarachnoid hemorrhage. In 3cases (cases 1, 6, and 10), focal lesions were superimposed onrecent, more widespread, hypoxic-ischemic changes, whichwere interpreted as being due to agonal global cerebralcirculatory insufficiency. These were most severe in case 6,in which a dilated cardiomyopathy, congestive cardiac fail-ure, and hypoperfusion were dominant clinical featureslatterly.

White Matter LesionsGeneralized white matter rarefaction and gliosis with U-fibersparing was a conspicuous feature in case 8, whereas whitematter pallor and edema were seen in case 6 as a manifesta-tion of agonal hypoperfusion. All other cases exhibited a mildto moderate degree of focal white matter rarefaction. Therewas a single focal white matter infarct in case 10.

Vessel PathologyLeptomeningeal vessels and main arteries on the circle ofWillis showed either no atherosclerosis (8 cases) or mildintimal thickening only (cases 1 and 10). Penetrating intra-cerebral vessels showed a mild to moderate degree of

HIV Pathology

MNGCsp24 IHC(GM/WM)

PCR*(GM/WM)

2 2/2 0/0

1 1/1 2000/0

2 2/2 0/60

– 2/2 0/0

1 2/1 ND

1 2/1 2000†

2 2/2 0/60

1 1/1 200/200

1 1/1 2000/200

1 2/1 ND

Connor et al Cerebral Infarction in AIDS Patients 2119



concentric hyaline thickening in 9 of the 10 cases; thickeningwas most severe in case 5. This was patchily distributed: in 3cases, it was most conspicuous in the basal ganglia (cases 1,4, and 9); in 2 cases, it was most conspicuous in the deepwhite matter (cases 3 and 8); and in 4 cases, it appearedevenly distributed in the cortex, white matter, and deep gray

structures (cases 2, 5, 6, and 7). The cerebellum in all caseswas relatively spared, and in case 10, there was no significantsmall-vessel thickening. These relatively mild structural ves-sel wall changes were accompanied by variable degrees ofperivascular space dilatation and parenchymal rarefaction, theseverity of which paralleled the severity and distribution ofsmall-vessel thickening. Similarly, small perivascular depos-its of pigment were a feature in all cases; their presence wasagain correlated with a small-vessel thickening and perivas-cular space dilatation and rarefaction. Microvessel mineral-ization was observed in 4 cases (cases 1, 6, 7, and 9),particularly in the basal ganglia, with 3 of these 4 patientsbeing drug users. Small numbers of perivascular inflamma-tory cells, both lymphocytes and macrophages, were presentin 5 cases, with 4 of them being drug users. Representativevessel pathology is shown in Figure 2.

In no case was there evidence of vasculitis. Small-vesselocclusion was not observed, with the exception of onepossible intravascular thrombus in a small penetrating arteryin the basal ganglia of case 2 and in a single neocorticalpenetrating artery in case 5. Fibrinoid necrosis of occasionalneocortical and basal ganglia microvessels was a feature onlyin case 6; this brain was unusual in that it showed severeagonal global hypoxic-ischemic damage.

Other Brain PathologyBy definition, no brain harbored pathology due to trauma,tumor, or opportunistic infection. Multinucleate giant cellswere identified in the brains of 1 homosexual patient, 4 drugusers, and the sole hemophiliac patient. In these 6 cases, HIVwas identified by use of p24 immunohistochemistry, thusfulfilling the criteria for the histological diagnosis of HIVencephalitis.55 Other features included focal leptomeningealthickening and varying degrees of cerebral atrophy, whichwere present in all cases.

Clinical FeaturesPatient characteristics are summarized in Table 3. In case 10,the only hemophiliac patient, despite a detailed generalpractitioner’s summary of the past medical history, hospital/clinic follow-up notes were available for only 2 weeks beforedeath, whereas for the remaining 9 cases, lengthy follow-up(20 to 224 months, mean 75 months) was available. Vascularrisk factors were not common for the group as a whole (Table3). Cardiac disease was present in only 1 patient (case 6), inwhom there was no evidence of intracardiac thrombus orvegetation at post mortem. Despite 5 patients having beenprevious injection drug users, only 2 (cases 6 and 8) weremisusing drugs at the time of death. One patient (case 6) hadbeen using oral dihydrocodeine and diazepam, and another(case 8) had been using injection as well as oral drugs(dihydrocodeine, temazepam, diazepam, and methadone).None of the 10 patients had used heroin or cocaine during the6 months before death. All patients died before the introduc-tion of protease inhibitors into clinical practice in Edinburgh,and anti-retroviral medication was limited to zidovudine anddidanosine, although only 2 patients were on anti-retroviralmedication at the time of death (cases 1 and 7). Patients wereon numerous prescribed medications in the months before

Figure 1. Photomicrographs illustrating the typical, small, predomi-nantly gray matter, hypoxic/ischemic brain lesions. A, Small well-circumscribed basal ganglia infarct (case 2). The bulk of the lesionis composed of necrotic tissue debris and macrophages, sur-rounded by a thin rim of edema with scattered reactive astrocytes.Had the patient survived, this lesion would have evolved into aclassic cavitated (type Ia) lacunar infarct. Original magnification395 (hematoxylin and eosin). B, Small area of cortical rarefactionand neuronal loss, of presumed hypoxic/ischemic pathogenesis(case 9). Original magnification 3190 (hematoxylin and eosin). C,Detail of small thalamic rarefied lesion (case 5) similar to that illus-trated in panel B. The tissue is rarefied, gliotic, and devoid of neu-rons but does not show frank cavitation. Such lesions have beentermed incomplete (type Ib) lacunar infarcts. Original magnification3380 (hematoxylin and eosin).




death, but none of these have been associated with cerebralinfarction or vasculopathy.

All patients had advanced AIDS (Table 3). Two patientshad positive syphilis serology (although neither had evidenceof vasculitis or meningitis at autopsy), and 7 had positivecytomegalovirus serology, although only 1 patient had circu-lating IgM antibodies. Hepatitis B or C serology was positivein 7 patients, and low platelet counts were present in 5, but

none had histological evidence of intravascular thrombi,hemorrhage, or other tissue changes in keeping with cryo-globulinemia or disseminated intravascular coagulation. Afull coagulation profile, including proteins C and S, wasavailable for 1 patient (case 4) only. Plasma fibrinogen wasraised in 1 patient (case 9) but was measured only in 5.Cholesterol levels were not available for any of the patients.No obvious source of emboli was found at post mortem in anypatient.

Five of the 10 patients had no neurovascular symptoms orsigns. In the remaining 5 patients (Table 4), a clinicallyevident stroke did not occur, and a diagnosis of TIA wasmade in only 1 patient (case 4). One other patient (case 9) hadradiological evidence of basal ganglia infarction but noclinical stroke syndrome. Cognitive changes were noted in 3patients (cases 6, 8, and 9), and diagnoses of AIDS dementiacomplex were made 2, 1, and 4 years before death,respectively.

DiscussionWe found histopathological evidence of cerebral infarction in'1 in 20 of a large autopsy series of adult HIV cases (7% ifone considers only AIDS patients), after patients with cere-bral opportunistic infections, lymphoma, non-HIV infections,or sources of cerebral emboli (all potential causes of cerebralinfarction) had been excluded. They all (drug users, homo-sexuals not using drugs, and 1 hemophiliac patient) hadevidence of a nonvasculitic vasculopathy with the samehistopathologic features. The presence of the vasculopathyand the extent of cerebral infarction were not associated withviral load. Despite these pathological findings, only 1 patienthad had a TIA clinically, and no patient had had a clinicallyevident stroke. All patients with evidence of cerebral infarc-tion had AIDS and were thus severely immunocompromised.

Reviews of AIDS neuropathology have documented cere-bral infarction in 4% to 29% of HIV-infected pa-tients,9,12,16,17,28,44–46but few studies have assessed how manypatients had cerebral infarction in the absence of potentialcauses, such as cerebral opportunistic infections, lymphoma,or emboli from nonbacterial infective endocarditis.9,23 Somestudies, however, have described their histopathologic find-ings, patient characteristics, and selection criteria in sufficientdetail to enable us to work out the proportion of their patientswith “isolated” cerebral infarcts and would have fulfilled (asfar as we can assess) our selection criteria.16,28,43,51 Thecombined total of the 4 studies51 (22 of 356 patients fulfillingour selection criteria) gives a proportion of HIV-associatedcerebral infarct cases (6%) that closely approximates ourown. Variations in individual studies may reflect populationdifferences, variation in patient selection for autopsy, or thesmall number of cases.

The vascular lesions in all our 10 cases were characterizedby hyaline small-vessel thickening, perivascular space dila-tation, rarefaction, and pigment deposition, with vessel wallmineralization and perivascular inflammatory cell infiltratesalso seen in some cases. There were only occasional intra-vascular thrombi. Such changes are similar to those describedby Mizusawa et al,28 although in their study opportunisticcerebral infection was a feature in 9 of 24 brains. Similar

Figure 2. Photomicrographs illustrating spectrum of cerebralsmall-vessel changes. A, Small penetrating artery in basal gan-glia showing concentric wall thickening (case 5). The surround-ing perivascular space is mildly dilated and contains occasionalmacrophages (arrow). Original magnification 3380 (hematoxylinand eosin). B, Small penetrating thalamic vessel without con-spicuous mural thickening (case 5) but showing moderatelydilated perivascular space containing pigment-laden macro-phages (arrow). Original magnification 3380 (hematoxylin andeosin). C, Fibrinoid material (asterisk) around a small penetratingbasal ganglia vessel (case 6). Original magnification 3190(hematoxylin and eosin).




microvessel changes have also been described in an AIDS-related leukoencephalopathy attributed to HIV-1 infection.56

Mineralization of vessels has been noted previously in AIDSbrains,22,57 particularly in children,58 and again has beensuggested to be a direct effect of HIV on vessel walls.57 Wedid not see convincing evidence of vasculitis. This mayreflect a different prevalence of varicella-zoster or herpessimplex infection because those viruses are said to mediatevasculitic pathologies,19,22,26but this may also reflect differ-ent pathological criteria for the diagnosis of vasculitis. Thus,although vasculitis has been reported in the nervous system inHIV,28,17,30,31,34,59its relation to HIV itself is unclear.55,60

Vasculitis must be distinguished from perivascular lympho-cyte cuffing55 and should be diagnosed only when inflamma-tory cells are seen within the vessel wall and are associatedwith mural damage.

Several of the neuropathological features we describe areconsistent with the effects of focal or diffuse blood-brainbarrier breakdown, such as hyaline vessel wall thickening,perivascular space dilatation, rarefaction, and pigment depo-sition. Damage to the blood-brain barrier in HIV brains withleakage of serum protein has been reported,61–63usually as achronic diffuse phenomenon but also as a fatal, acute,relapsing variant.64 These vessel features are also commonlyseen in aging non-HIV brains (cerebral arteriolosclerosis),exacerbated principally by hypertension and diabetes melli-tus. However, none of the patients in this series were elderly,hypertensive, or diabetic.

Although small-vessel pathology similar to what we de-scribe may be seen in HIV-negative drug-abuser brains(authors’ unpublished data, 1999), the fact that no vascularlesion was seen exclusively in any one HIV risk groupsuggests that in this series the changes are more likely aconsequence of HIV infection itself. It has been claimed thatHIV infects vascular endothelial cells,61,65,66although this isdisputed.56,67,68 Indeed, infection of endothelial cells withconsequent blood-brain barrier breakdown has been advancedas at least a theoretically possible mechanism of HIV entryinto the CNS.23,69–71 More specifically, HIV infection ofvessel walls has been linked to focal blood-brain barrierbreakdown56 and to vessel mineralization.57 Similar, thoughnot identical, vascular changes have been found in HIV-infected patients in the pre-AIDS stage72,73; these changes arepossibly the result of a lymphocyte T-cell reaction duringearly infection with HIV. In our series, neither the presenceand distribution of HIV encephalitis nor the viral load wascorrelated with the number or distribution of the cerebralinfarcts, although the small number of cases makes thiscorrelation uncertain.

Altered vasoreactivity, alone or in combination with any ofthe above-mentioned factors, may, at least in theory, havecontributed to the development of microinfarcts. Disturbedvasoreactivity in patients with HIV has recently been report-ed.29 Temporary vasospasm was one of the possible mecha-nisms involved in a recent study describing incompletelacunar infarcts74 in non-HIV brains; incomplete lacunar

TABLE 3. Patient Details

PatientNo.

Age atdeath, y Sex

HIVRisk

Risk Factors forCerebralInfarction

Duration KnownHIV Positivity/

AIDS, moDisease Stage*(CD4 Count)†

Stroke/TIASymptoms Laboratory Data‡

1 39 M H CS, alcohol 59/59 C3 (7) 1? Syphilis serology positive, evidence of past VZVexposure

2 41 M H CS 67/15 C3 (10) 1 Syphilis serology positive, HBV surface and coreantibody positive. CMV IgG positive

3 47 M H Alcohol 22/22 C3 (35) 1? CMV IgG positive

4 46 M H None 88/13 C3 (0) 1 Platelet count 813109/L, HBV surface antigen andcore antibody positive, CMV IgG positive

5 28 F DU CS, familyhistory

58/18 C3 (38) 2 Platelet count 343109/L, HBV core and surfaceantigen positive, CMV IgG and IgM positive

6 33 M DU CS, congestiveheart failure

63/14 C3 (31) 2 HCV positive, VZV IgG positive

7 30 M DU CS 93/19 C3 (40) 2 Platelet count 743109/L, HCV positive, CMV IgGpositive

8 35 F DU CS, alcohol 105/6 C3 (23) 2 HBV core antibody positive, HCV positive, CMV IgGpositive

9 34 M DU CS 114/24 C3 (8) 1? Platelet count 763109/L, CMV IgG positive, plasmafibrinogen elevated (5.3 g/L)

10 22 M Hem CS 122/36 C3 (0) 2 Platelet count 703109/L, HCV positive

CS indicates cigarette smoking; alcohol, alcohol excess; VZV, varicella-zoster virus; HBV, hepatitis B virus, CMV, cytomegalovirus; HCV, hepatitis C virus; NA, notavailable; 1, present; and 2, not present.

*1993 classification by Centers for Disease Control and Prevention.†At time of death.‡Results are listed if available and abnormal: full blood count, international normalized ratio, partial thromboplastin time, plasma fibrinogen levels (all performed

during last admission or clinic visit), antinuclear factor, syphilis serology, HBV and HCV serology, CMV serology, VZV serology, and toxoplasmosis serology (allperformed at some time before death).




infarcts are small foci of incompletely infarcted brain in deepgray structures that are similar in many respects to themicroinfarcts that we describe in the brains of AIDS patients.

What else may have contributed to the cerebral infarcts inour patients? Two patients had positive syphilis serology, butin the absence of meningitis or vasculitis, this is unlikely to berelevant. None of the patients had significant atherosclerosisof their medium or large vessels, and there were no otherarterial or cardiac sources of cerebral emboli detected duringlife or at autopsy. Other investigators have associated cerebralinfarcts in HIV-infected patients with hepatitis B42 (hepatitisB or C serology was positive in 6 of our patients), as well aswith a prothrombotic state related to protein S deficiency,35,36

antiphospholipid antibodies,39,40 or disseminated intravascu-lar coagulopathy.4,11,12We have protein S levels, which werenormal, for only 1 patient, and none of our patients hadantiphospholipid antibodies measured. Thus, the theoreticalpossibility of “hypercoagulability” remains, although intra-vascular thrombi, recent or organized, were seen only rarelyin this and other autopsy series, even in those cases with veryrecent infarcts. The fact that “crops” of small cerebral infarctsof similar histological age were seen in some brains mightsuggest showers of microemboli or a transient hypercoagu-lable state, but this remains speculative. Infarcts were not afeature in other organs at autopsy, militating against cardiacembolism or a generalized hypercoagulable state.

We could not demonstrate a clear correlation between thepathological changes and clinical features of stroke or TIA in

our patients. Five patients had no neurovascular symptoms atany stage. One patient had experienced recurrent anteriorcirculation TIAs that started 11 months before death, but noetiology was found on investigation on 2 separate occasions,although an angiogram was not performed. The diagnosis waschanged to migraine on review by a neurologist because theattacks were slow in onset and were associated with head-ache, but the diagnosis was reversed again after a typical leftpartial anterior circulation TIA. As the TIAs resolved afterthe introduction of ganciclovir, one might speculate that thepatient had a cytomegalovirus-induced vasculitis that wastreated with ganciclovir. Alternatively, the cessation of theTIAs may have been entirely coincidental. None of ourpatients received highly active anti-retroviral therapy becausethey came to autopsy before the introduction of the neweranti-retroviral agents.

Several patients presented diagnostic difficulty not only tothe attending physicians (cases 1, 2, 3, and 9) but also to theneurologists to whom they were referred (cases 1, 3, and 9).All 4 patients (cases 1, 2, 3, and 9) had recurrent transientneurological symptoms. One patient was diagnosed as havingatypical migraine, and in the other 3 patients, the symptomswere thought to be seizure-related, although in 2 of them, adiagnosis of atypical migraine was entertained at some point.Recently, recurrent transient neurological deficits similar tothose noted in our patients have been found to occur inclinical cohorts27,75,76and have been linked to antiphospho-lipid antibodies27,77 or protein S deficiency.27

Scans CT/MRI (mo before death) Cause of Death at Autopsy and Other Autopsy Findings

Both: mild atrophy (12) Bronchopneumonia, paraspinal Staphyloccus aureusabscess

Both: mild atrophy (4) Bronchopneumonia, pancreatitis, CMV adrenalitis

NA Bronchopneumonia, pulmonary lymphoma

MRI: mild global increase in white matter signal (8) CMV adrenalitis, Kaposi’s sarcoma of lung,gastrointestinal tract, and pancreas

MRI: normal (21), CT: atrophy (8) CMV retinitis, pancreatitis, vacuolar myelopathy

CT: atrophy (6) Cardiomyopathy (no evidence of clot/vegetation),pulmonary edema

CT: normal (19) Lung mycobacterium avium complex

MRI: atrophy (2) Bronchopneumonia

CT: atrophy (20), MRI: ? small basal ganglia infarctsbilaterally (8)

Bronchopneumonia, pancreatitis, hepatic fibrosis

NA Bronchopneumonia, acute on chronic pancreatitis, acutetubular necrosis, hepatitis C




In conclusion, the present study suggests that cerebralinfarcts in HIV-infected patients are not common in theabsence of cerebral opportunistic infection, lymphoma, non-HIV infection, or embolic sources. We confirm the existenceof an HIV-associated vasculopathy but highlight that thisvasculopathy shows similar histopathologic features in allrisk groups irrespective of drug use, hepatitis serology, oreven syphilis serology. The possible relation of this vascu-lopathy to direct HIV infection of vessel walls should befurther investigated. Finally, the present study supports theapproach that in AIDS patients presenting with a stroke orTIA, potentially treatable causes, such as cerebral coinfectionor tumor, be excluded before assuming that the cause is HIVitself, be that as a result of a vasculopathy or some as yetunrecognized pathogenic mechanism.

AcknowledgmentsDr Lammie is funded by the Medical Research Council (MRC), UK,and the Edinburgh Brain and Tissue Resource is funded by a grant toDr Bell (9708080) from the MRC, UK. The authors would like to

thank Frances Brannan for providing expert technical support for theMRC HIV Brain Bank, Dr Alison Richardson for help with theEdinburgh HIV register, and Prof Peter Sandercock forconstructive criticism.

References1. Price RW, Brew BJ. Central and peripheral nervous system compli-

cations. In: Devita V, Hellman S, Rosenberg SA, eds.AIDS: Etiology,Diagnosis, Treatment and Prevention. Philadelphia, Pa: Lippincott-Raven; 1997:331–353.

2. Said G, Saimot AG, Tardieu M, Lacroix C. Introduction. In: Warlow CP,ed. Neurological Complications of HIV and AIDS. London, UK: WBSaunders; 1997:1–3.

3. Simpson DM, Berger JR. Neurologic manifestations of HIV infection.Med Clin North Am. 1996;80:1363–1395.

4. Gray F, Gherardi R, Scaravilli F. The neuropathology of the acquiredimmune deficiency syndrome (AIDS): a review.Brain. 1988;111:245–266.

5. Kure K, Llena J, Lyman WD, Soeiro R, Weidenheim KM, Hirano A,Dickson D. Human immunodeficiency virus-1 infection of the nervoussystem: an autopsy study of 268 adult, pediatric, and fetal brains.HumPathol. 1991;22:700–710.

6. Berger JR, Moskowitz L, Fischl M, Kelley RE. Neurologic disease as thepresenting manifestation of acquired immunodeficiency syndrome.S MedJ. 1987;80:683–686.

TABLE 4. Patients With Symptoms That Might Have Been Attributed to Stroke/TIA

Patient No. Clinical Features Clinical Diagnosis

1 Four years before death, the patient developed intermittent episodes of homonymous hemianopia affectingeither side. There was no associated headache, nausea, or positive visual phenomena, and the episodeslasted for '20 min at a time. Symptoms resolved on pizotifen.

Initially migraine, then after episodesof dysphasia, diagnosis as seizurepossibly related to benzodiazepine

withdrawal

Three years before death, 1 week after discontinuing benzodiazepines, there were 2 episodes of transientexpressive dysphasia lasting '1 h, preceded by flashing lights in visual field and acrid smell. Anassociated recurrence of previous homonymous hemianopia resolved off AZT. EEG showed generalizedparoxysmal abnormality, and patient was treated with carbamazepine.

One year before death, there was an episode of paresthesia radiating down the left arm. No abnormalitywas detected on examination.

2 Five months before death (witnessed) there was a generalized seizure, attributed to hypocalcemia. CT wasnormal.

Seizure related to hypocalcemia

Two months before death, there were 3 episodes of unilateral facial numbness, spreading to one or botharms, lasting 5 min to 5 h. Migraine was considered and sumatriptan was prescribed.

?Migraine

One month before death, a possible episode of amaurosis fugax lasted 30 min. Variable visual field losswas found on formal testing, with no active CMV retinitis clinically.

Amaurosis Fugax

3 Eleven months before death, there was a 2-mo history of intermittent visual disturbance lasting 20minutes, diagnosed as migrainous aura without headache or other features of migraine. It resolvedspontaneously.

Possible migraine

4 Eleven months before death there were 2 episodes of acute onset dysarthria, followed by numbness ofthe whole left side of the body, which resolved over 2 h. CT brain, EEG, ECG, carotid and vertebralDoppler studies, transoesophageal echo with contrast, and full coagulation profile were all normal. CSFshowed no cells, and protein was raised at 0.72 g/L. Aspirin treatment yielded no resolution.

TIAs

Ten months before death, there was a neurological review. Because further similar attacks were thoughtto be of slow onset and were followed by headaches, a diagnosis of migraine was made. There was noresolution with sumatriptan and propranolol.

Nine months before death, there was an episode of expressive dysphasia and right hemiparesis lasting 20hours. Full workup again revealed only a raised CSF protein. The patient developed diarrhea andquestionable CMV retinitis and was started empirically on ganciclovir, after which, the above episodesresolved.

9 Six years before death, the patient developed transient numbness in the distribution of the right trigeminalnerve. MRI head scan was normal, and the numbness resolved spontaneously over a few weeks.

? Seizures related to drugwithdrawal

Five years before death, there was intermittent right-sided numbness, which lasted a year. No cleardiagnosis was made, but at the time of neurological review, patient was thought to be anxious or possiblyto have seizures related to drug withdrawal. Over the 4 y before death, the patient developed progressiveAIDS dementia complex.

Anxiety

AZT indicates azidothymidine; CSF, cerebrospinal fluid.




7. McArthur JC. Neurologic manifestations of AIDS.Medicine. 1987;66:407–437.

8. Simpson DM, Tagliati M. Neurologic manifestations of HIV infection:author’s reply.Ann Intern Med. 1995;122:884–884.

9. Pinto AN. AIDS and cerebrovascular disease.Stroke. 1996;27:538–543.10. Snider WD, Simpson DM, Nielsen S. Neurological complications of

acquired immune deficiency syndrome: analysis of 50 patients.AnnNeurol. 1983;14:403–418.

11. Guarda LA, Luna MA, Smith JL, Mansell PWA, Gyorkey F, Roca AN.Acquired immune deficiency syndrome: postmortem findings.Am J ClinPathol. 1984;81:549–557.

12. Anders KH, Guerra WF, Tomiyasu U, Verity MA, Vinters HV. Theneuropathology of AIDS: UCLA experience and review.Am J Pathol.1986;124:537–558.

13. Levy RM, Bredesen DE, Rosenblum ML. Neurological manifestations ofthe acquired immunodeficiency syndrome (AIDS): experience at UCSFand review of the literature.J Neurosurg. 1985;62:475–495.

14. Kibayashi K, Mastri AR, Hirsch CS. Neuropathology of human immu-nodeficiency virus infection at different disease stages.Hum Pathol.1996;27:637–642.

15. Brust JCM. AIDS and stroke. In: Welch KMA, Caplan LR, Seisjo BK,Reis DJ, Weir B, eds.Primer on Cerebrovascular Diseases.New York,NY: Academic Press Inc; 1997:423–425.

16. Berger JR, Harris JO, Gregorios J, Norenberg M. Cerebrovascular diseasein AIDS: a case-control study.AIDS. 1990;4:239–244.

17. Kieburtz KD, Thomas MD, Eskin TA, Ketonen L, Tuite MJ. Opportu-nistic cerebral vasculopathy and stroke in patients with the acquiredimmunodeficiency syndrome.Arch Neurol. 1993;50:430–432.

18. Guiloff RJ, Fuller GN. Other neurological diseases in HIV-1 infection:clinical aspects.Baillieres Clin Neurol. 1992;1:175–209.

19. Eidelberg D, Sotrel A, Horoupian DS, Neumann PE, Pumarola-Sune T,Price RW. Thrombotic cerebral vasculopathy associated with herpeszoster.Ann Neurol. 1986;19:7–14.

20. Engstrom JW, Lowenstein DH, Bredesen DE. Cerebral infarctions andtransient neurologic deficits associated with acquired immunodeficiencysyndrome.Am J Med. 1989;86:528–532.

21. Gray F, Mohr M, Rozenberg F, Belec L, Lescs MC, Dournons E, SinclairE, Scaravilli F. Varicella-zoster virus encephalitis in acquired immuno-deficiency syndrome: report of four cases.Neuropathol Appl Neurobiol.1992;18:502–514.

22. Petito CK, Cho E-S, Lemann W, Navia BA, Price RW. Neuropathologyof acquired immunodeficiency syndrome (AIDS): an autopsy review.J Neuropathol Exp Neurol. 1986;45:635–646.

23. Dal Canto MC. AIDS and the nervous system: current status and futureperspectives.Hum Pathol. 1989;20:410–418.

24. Scully RE, Mark EJ, McNeely WF, Ebeling SH. Case records of theMassachusetts General Hospital: case 36–1996.N Engl J Med. 1996;335:1587–1595.

25. Matlow AG. Neurosyphilis as a cause of cerebral infarction in AIDS.Am J Med. 1990;88:700–701. Letter.

26. Brannagan TH III. Retroviral-associated vasculitis of the nervous system.Neurol Clin. 1999;15:927–944.

27. Brew BJ, Miller J. Human immunodeficiency virus type 1-relatedtransient neurological deficits.Am J Med. 1996;101:257–261.

28. Mizusawa H, Hirano A, Llena JF, Shintaku M. Cerebrovascular lesions inacquired immune deficiency syndrome (AIDS).Acta Neuropathol. 1988;76:451–457.

29. Brilla R, Nabavi DG, Schulte-Altedorneburg G, Kemeny V, Reichelt D,Evers S, Schiemann U, Husstedt IW. Cerebral vasculopathy in HIVinfection revealed by transcranial Doppler: a pilot study.Stroke. 1999;30:811–813.

30. Cho E-S, Sharer LR, Peress NS, Little B. Intimal proliferation of lepto-meningeal arteries and brain infarcts in subjects with AIDS.J Neuro-pathol Exp Neurol. 1987;46:385–385. Abstract.

31. Scaravilli F, Daniel SE, Harcourt-Webster N, Guiloff RJ. Chronic basalmeningitis and vasculitis in acquired immunodeficiency syndrome: apossible role for human immunodeficiency virus.Arch Pathol Lab Med.1989;113:192–195.

32. Schwartz ND, So YT, Hollander H, Allen S, Fye KH. Eosinophilicvasculitis leading to amaurosis fugax in a patient with acquired immu-nodeficiency syndrome.Arch Intern Med. 1986;146:2059–2060.

33. Vinters HV, Guerra WF, Eppolito L, Keith PE. Necrotizing vasculitis ofthe nervous system in a patient with AIDS-related complex.NeuropatholAppl Neurobiol. 1988;14:417–424.

34. Yankner BA, Skolnik PR, Shoukimas GM, Gabuzda DH, Sobel RA, HoDD. Cerebral granulomatous angiitis associated with isolation of humanT-lymphotropic virus type III from the central nervous system.AnnNeurol. 1986;20:362–364.

35. Bissuel F, Berruyer M, Causse X, Dechavanne M, Trepo C. Acquiredprotein deficiency: correlation with advanced disease in HIV-1-infectedpatients.J Acquir Immune Defic Syndr. 1992;5:484–489.

36. Roquer J, Palomeras E, Pou A. AIDS and cerebrovascular disease.Stroke.1996;27:1694–1697.

37. MacLean C, Flegg PJ, Kilpatrick DC. Anti-cardiolipin antibodies andHIV infection. Clin Exp Immunol. 1990;81:263–266.

38. Abuaf N, Laperche S, Rajoely B, Carsique R, Deschamps A, RouquetteAM, Barthet C, Khaled Z, Marbot C, Saab N, et al. Autoantibodies tophospholipids and to the coagulation proteins in AIDS.Thromb Haemost.1997;77:856–861.

39. Brew BJ. Central and peripheral nervous system abnormalities.Med ClinNorth Am. 1992;76:63–81.

40. Naranjo IC, Santos JAT, Rodriguez MAA. Ischemic stroke as the solemanifestation of human immunodeficiency virus infection.Stroke. 1992;23:117–118.

41. Gonzales MF, Davis RL. Neuropathology of acquired immunodeficiencysyndrome.Neuropathol Appl Neurobiol. 1988;14:345–363.

42. Atalaia A, Ferro J, Antunes F. Stroke in an HIV-infected patient.J Neurol. 1992;239:356–357.

43. Kieburtz KD, Eskin TA, Ketonen L, Tuite MJ. Opportunistic cerebralvasculopathy and stroke in patients with the acquired immunodeficiencysyndrome.Arch Neurol. 1993;50:430–432.

44. Rhodes RH. Histopathologic features in the central nervous system of 400acquired immunodeficiency syndrome cases: implications of rates ofoccurrence.Hum Pathol. 1993;24:1189–1198.

45. Lanjewar DN, Jain PP, Shetty CR. Profile of central nervous systempathology in patients with AIDS: an autopsy study from India.AIDS.1998;12:309–313.

46. Moskowitz L, Hensley GT, Chan JC, Gregorios J, Conley FK. Theneuropathology of acquired immune deficiency syndrome.Arch PatholLab Med. 1984;108:867–872.

47. Budka H, Costanzi G, Cristina S, Lechi A, Parravicini C, Trabattoni GR,Vago L. Brain pathology induced by infection with the human immuno-deficiency virus (HIV).Acta Neuropathol. 1987;75:185–198.

48. De Girolami U, Smith TW, Henin D, Hauw J-J. Neuropathology of the acquiredimmunodeficiency syndrome.Arch Pathol Lab Med. 1990;114:643–655.

49. Lantos PL, McLaughlin JE, Scholtz CL, Berry CL, Tighe JR. Neuropa-thology of the brain in HIV infection.Lancet. 1989;1:309–310.

50. Chimelli L, Rosemberg S, Hahn MD, Lopes MBS, Barretto Netto M.Pathology of the central nervous system in patients infected with thehuman immunodeficiency virus (HIV): a report of 252 autopsy casesfrom Brazil. Neuropathol Appl Neurobiol. 1992;18:478–488.

51. Rosemberg S, Lopes MBS, Tsanaclis AM. Neuropathology of acquiredimmunodeficiency syndrome (AIDS): analysis of 22 Brazilian cases.J Neurol Sci. 1986;76:187–198.

52. Bell JE, Donaldson YK, Lowrie S, McKenzie CA, Elton RA, Chiswick A,Brettle RP, Ironside JW, Simmonds P. Influence of risk group andzidovudine therapy on the development of HIV encephalitis and cognitiveimpairment in AIDS patients.AIDS. 1996;10:493–499.

53. Brettle RP, McNeil AJ, Burns S, Gore SM, Bird AG, Yap PL, MacCallumL, Leen CS, Richardson AM. Progression of HIV: follow-up ofEdinburgh injecting drug users with narrow seroconversion intervals in1983–1985.AIDS. 1996;10:419–430.

54. Engstrom JW, Lowenstein DH, Bredesen DE. Cerebral infarctions andtransient neurologic deficits associated with acquired immunodeficiencysyndrome.Am J Med. 1989;86:528–532.

55. Budka H, Wiley CA, Kleihues P, Artigas J, Asbury AK, Cho ES,Cornblath DR, Dal Canto MC, DeGirolami U, Dickson D. HIV-associated disease of the nervous system: review of nomenclature andproposal for neuropathology-based terminology.Brain Pathol. 1991;1:143–152.

56. Smith TW, DeGirolami U, Henin D, Bolgert F, Hauw J-J. Human immu-nodeficiency virus (HIV) leukoencephalopathy and the microcirculation.J Neuropathol Exp Neurol. 1990;49:357–370.

57. Belman AL, Lantos G, Horoupian D, Novick BE, Ultmann MH, DicksonD, Rubinstein A. AIDS: calcification of the basal ganglia in the infantsand children.Neurology. 1985;36:1192–1190.

58. Sharer LR, Epstein LG, Cho E-S, Joshi VV, Meyenhofer MF, Rankin LF,Petito CK. Pathologic features of AIDS encephalopathy in children.HumPathol. 1986;17:271–284.




59. Gherardi R, Florea-Strat A, Authier FJ, Baudrimont M. Lack of detectionof varicella-zoster virus in systemic HIV-associated vasculitis.Acta Neu-ropathol. 1994;87:655–655.

60. Budka H. Neuropathology of human immunodeficiency virus infection.Brain Pathol. 1991;1:163–175.

61. Rhodes RH. Evidence of serum-protein leakage across the blood-brainbarrier in the acquired immunodeficiency syndrome.J Neuropathol ExpNeurol. 1991;50:171–183.

62. Petito CK, Cash KS. Blood-brain barrier abnormalities in the acquiredimmunodeficiency syndrome: immunohistochemical localization ofserum proteins in postmortem brain.Ann Neurol. 1992;32:658–666.

63. Power C, Kong P-A, Crawford TO, Wesselingh S, Glass JD, McArthurJC, Trapp BD. Cerebral white matter changes in acquired immunodefi-ciency syndrome dementia: alterations of the blood-brain barrier.AnnNeurol. 1993;34:339–350.

64. Gray F, Belec L, Chretien F, Dubreuil-Lemaire ML, Ricolfi F,Wingertsmann L, Poron F, Gherardi R. Acute, relapsing brain oedemawith diffuse blood-brain barrier alteration and axonal damage in theacquired immunodeficiency syndrome.Neuropathol Appl Neurobiol.1998;24:209–216.

65. Wiley CA, Schrier RD, Nelson JA, Lampert PW, Oldstone MBA.Cellular localization of human immunodeficiency virus infection withinthe brains of acquired immune deficiency syndrome patients.Proc NatlAcad Sci U S A. 1986;83:7089–7093.

66. Ward JM, O’Leary TJ, Baskin GB. Immunohistochemical localization ofhuman and simian immunodeficiency viral antigens in fixed tissuesections.Am J Pathol. 1920;127:199–205.

67. Budka H. Human immunodeficiency virus (HIV) envelope and coreproteins in CNS tissues of patients with the acquired immune deficiencysyndrome (AIDS).Acta Neuropathol. 1990;79:611–619.

68. Kure K, Lyman WD, Weidenheim KM, Dickson DW. Cellular local-ization of an HIV-1 antigen in subacute AIDS encephalitis using animproved double-labeling immunohistochemical method.Am J Pathol.1990;136:1085–1092.

69. Ho DD, Pomerantz RJ, Kaplan JC. Pathogenesis of infection with humanimmunodeficiency virus.N Engl J Med. 1987;317:278–286.

70. Rhodes RH, Ward JM. AIDS meningoencephalomyelitis: pathogenesisand changing neuropathologic findings.Pathol Annu. 1991;26:247–276.

71. Budka H. Human immunodeficiency virus (HIV)-induced disease of thecentral nervous system: pathology and implications for pathogenesis.Acta Neuropathol. 1989;77:225–236.

72. Gray F, Scaravilli F, Everall I, Chretien F, An S, Boche D, Adle-BiassetteH, Wingertsmann L, Durigon M, Hurtrel B, et al. Neuropathology of earlyHIV-1 infection. Brain Pathol. 1996;6:1–15.

73. Gray F, Lescs M-C, Keohane C, Paraire F, Marc B, Durigon M, GherardiR. Early brain changes in HIV infection: neuropathological study of 11HIV seropositive, non-AIDS cases.J Neuropathol Exp Neurol. 1992;51:177–185.

74. Lammie GA, Brannan F, Wardlaw JM. Incomplete lacunar infarction(type1b lacunes).Acta Neuropathol. 1998;96:163–171.

75. Rinaldi R, Manfredi R, Azzimondi G, Rodorigo G, Tonon C, De Rosa V,Mastroianni A, Coronado OV, Legnani C, Chiodo F, et al. Recurrent‘migrainelike’ episodes in patients with HIV disease.Headache. 1997;37:443–448.

76. Baily GG, Mandal BK. Recurrent transient neurological deficits inadvanced HIV infection.AIDS. 1995;9:709–712.

77. Card T, Wathen CG, Luzzi GA. Primary HIV-1 infection presenting withtransient neurological deficit.J Neurol Neurosurg Psychiatry. 1990;64:281–282.




Cerebral Infarction in Adult AIDS Patients Observations From the Edinburgh HIV Autopsy Cohort

Documents

Transcript of Cerebral Infarction in Adult AIDS Patients Observations From the Edinburgh HIV Autopsy Cohort