2. HIV-Associated Opportunistic Infections of the CNS

8/11/2019 2. HIV-Associated Opportunistic Infections of the CNS

1/13www.thelancet.com/neurology Vol 11 July 2012 605

Review

Lancet Neurol2012; 11: 60517

HIV Neuroscience Program,

Johns Hopkins University,

Baltimore, MD, USA

(I L Tan MBBS, B R Smith MD,

G von Geldern MD,

F J Mateen MD,

Prof J C McArthur MBBS)

Correspondence to:

Prof Justin C McArthur,

Department of Neurology, Johns

Hopkins University, Meyer 6113,

600 North Wolfe Street,

Baltimore, MD 21287, [email protected]

HIV-associated opportunistic infections of the CNS

Ik Lin Tan, Bryan R Smith, Gloria von Geldern, Farrah J Mateen, Justin C McArthur

Survival in people infected with HIV has improved because of an increasingly powerful array of antiretroviraltreatments, but neurological symptoms due to comorbid conditions, including infection with hepatitis C virus,malnutrition, and the effects of accelerated cardiovascular disease and ageing, are increasingly salient. A therapeuticgap seems to exist between the salutary effects of antiretroviral regimens and the normalisation of neurologicalfunction in HIV-associated neurocognitive disorders. Despite the advances in antiretroviral therapy, CNS opportunisticinfections remain a serious burden worldwide. Most opportunistic infections can be recognised by a combination ofcharacteristic clinical and radiological features and are treatable, but some important challenges remain in thediagnosis and management of HIV-associated opportunistic infections.

IntroductionHIV infection leads to substantial morbidity and

mortality worldwide. In 2009, 333 million adults andchildren were living with HIV, two-thirds of whom werein sub-Saharan Africa. Overall, the annual incidence of new infections hasdeclined by 19% since the peak of the worldwide HIVepidemic in 1999, in line with the Millennium DevelopmentGoal. Nevertheless, about 26 million individuals werenewly infected in 2009, and the incidence continues toincrease in some regions.17000 new HIV infections occurdaily, 95% of which are in low-income and middle-incomecountries, where only about a third of patients who requireantiretroviral drugs have access to them.1

In high-income countries, the introduction of

combination antiretroviral therapy (cART) in 1996 greatlychanged the incidence of neurological opportunisticinfections, from 131 per 1000 patient-years in 199697 to10 per 1000 in 200607 (tables 1, 2).2,7 Many of theopportunistic infections that affect the CNS are AIDS-defining conditions, including progressive multifocalleukoencephalopathy (PML), CNS cytomegalovirus, CNStuberculosis, cryptococcal meningitis, and cerebraltoxoplasmosis, including toxoplasmic encephalitis(table 2), and all have high associated mortality.8Treatmentof CNS opportunistic infections in conjunction withcART improves survival, but such infections continue tobe important, especially where access to cART is limited.

In this Review we focus on the most common CNS

opportunistic infections associated with HIV infectionworldwide, and provide a summary of each in terms ofepidemiology, clinical presentations, diagnosis, andtreatment.

Biology of HIV-1HIV-1 infection accounts for most of the global HIVpandemic. Only 12 million of the 33 million HIVinfections are caused by HIV-2. HIV-1 belongs to thefamily Retroviridae and genus Lentivirus.9 It is a single-stranded, positive-sense RNA virus that contains a reversetranscriptase, which transcribes viral RNA into DNA that isintegrated into the hosts genome as a provirus. HIV-1primarily targets CD4 receptors and infects CD4-positiveT lymphocytes and cells of the monocyte or macrophage

lineages. HIV-1 also infects CD4-negative cells, includingastrocytes, but in a restrictive manner. HIV-1 subtypes are

defined by chemokine co-receptors: T-tropic viruses useCXCR4 (receptor for SDF1, also termed CXCL12) to infectlymphocytes, and M-tropic viruses use CCR5 (receptor forRANTES, also termed CCL5, and MIP1 or CCL3/4) toinfect macrophages.10,11

HIV infection is characterised by three stages: acuteprimary infection, an asymptomatic (latent) stage, andsymptomatic chronic illness. Disease progression ishighly variable: from 6 months after seroconversion tomore than 2030 years, or minimal progression might beseen in elite suppressors. In the absence of cART, themean time to the development of AIDS is 1011 years, 12and median survival after AIDS develops is 1337 years,

depending on the CD4-cell count.

13

After primary infection, acute disseminated viraemia isseen.14 The initial massive depletion of gut-associatedmemory T cells15 leads to physical and immunologicalbreaches of the gut mucosa, as well as expansions of someHIV-specific CD8-positive T-cell responses,16 the crucial

Incidence per

1000 person-years

Overall 10*

Progressive multifocal leukoencephalopathy 07

Toxoplasmic encephalitis 04

Cryptococcal meningitis 02

Data are from 200607.2

*The sum of the individual opportunistic infections ismore than 10 because some individuals have more than one infection.

Table : Incidence of HIV-associated CNS opportunistic infections

Common CNS opportunistic infections

Asian and Pacific regions3 Cryptococcal meningitis, cerebral toxoplasmosis, tuberculous meningitis,

Japanese encephalitis B

Sub-Saharan Africa4 Tuberculous meningitis, cryptococcal meningitis, cytomegalovirus, malaria

Europe and North America2 PML, toxoplasmic encephalitis, cryptococcal meningitis

South America5 Cerebral toxoplasmosis, tuberculous meningitis, cryptococcal meningitis;

Chagas disease is reported in southern US states and South America6

PML=progressive multifocal leukoencephalopathy.

Table :Incidence of HIV-associated CNS opportunistic infections by geographical region


2/13

606 www.thelancet.com/neurology Vol 11 July 2012

Review

host immune response against HIV. This robust immune

response leads to virus being trapped within dendriticcells in lymphoid tissue and a marked reduction ofviraemia.17 The virus, however, is not completelyeliminated from the body and chronic, persistent viralreplication leads to systemic immune activation.Additionally, quiescent CD4-positive memory T cells andmacrophages serve as long-term reservoirs for latent HIVinfection.18 Persistent viral replication, chronic immuneactivation, and progressive deterioration of immunefunction result in symptomatic disease with severeimmune deficiency in advanced HIV infection.19Typically,CNS opportunistic infections occur during this stage ofthe HIV infection; waning immunity and high HIV load,both systemically and in the CNS, create a favourable

milieu. Apart from immune deficiency, other features ofHIV-1 might directly facilitate CNS opportunisticinfections, for example in PML. Most CNS opportunisticinfections result from reactivation of latent pathogens,including PML, toxoplasmic encephalitis, and primaryCNS lymphoma. HIV infection produces substantialdepletion of the CD4-cell count and preferentially destroysthe cellular immune system, which is crucial for defenceagainst viral, fungal, and parasitic infections.

Clinical features of CNS opportunistic infectionsCNS opportunistic infections should be suspected in allpeople with advanced HIV infection. Individuals whoare unaware of their HIV status can first present withCNS opportunistic infections. Additionally, in patientsin whom cART is started, immune reconstitution in-flammatory syndrome (IRIS) might unmask previouslyunsuspected CNS opportunistic infections. The maindiagnostic features of CNS opportunistic infections areclinical presentation, temporal evolution, and CSF andradiographic features. These infections typically developwhen the CD4-cell count is lower than 200 cells per L.One important principle is that up to 15% of CNSopportunistic infections involve multiple concurrent

processes (panel 1) and, therefore, unexpected

worsening after treatment for an opportunistic infectionis started should prompt consideration of a secondprocess. Individuals who have had immunologicalrecovery from cART might still be at risk of developingCNS opportunistic infections.

Some infections, such as toxoplasmic encephalitis andcryptococcal meningitis, evolve over hours, whereasothers, such as PML and CNS lymphoma, typically havea more indolent course, with development often takingweeks to months. Many exceptions to these patterns may,however, be seen. Although many CNS opportunisticinfections are associated with non-specific symptoms,such as fever and lethargy, the combination of symptomssuch as a new pattern of headache or headache lasting

longer than 3 days, new-onset seizures, or altered mentalfunction strongly suggest an acute focal brain lesion.20

The incidence, temporal pattern, and typical CD4-cellcount for common CNS opportunistic infections arepresented in table 3. CSF and radiographic patterns arepresented in tables 4 and 5, respectively. The neuro-imaging features provide a guide, but diagnosis shouldnever be made on neuroimaging findings alone.

A diagnostic and management algorithm for theassessment and management of intracranial masslesions in patients with AIDS was issued by the QualityStandard Subcommittee of the American Academy ofNeurology in 1998.38 We present an algorithm for themanagement of patients presenting with headaches andsymptoms suggestive of CNS infections (figure 1).Although many advances have been made in thediagnosis and treatment of CNS opportunistic infections,the algorithm approach remains applicable, especially inresource-poor settingsfor example, in patients withmultiple CNS lesions, empirical treatment withantitoxoplasmosis therapy might be appropriate. Patientswith severe immune suppression (CD4-cell count lowerthan 200 cells per L) are at risk of toxoplasmicencephalitis, cryptococcal meningitis, cytomegalovirusinfection, primary CNS lymphoma, and PML, whereaspatients with moderate immune suppression (CD4-cellcounts of 200500 cells per L) are at risk of tuberculous

meningitis and PML.In resource-rich settings, a battery of investigations is

used to diagnose CNS opportunistic infections(panel 2). Assessment of CSF with antibody testing orPCR detection can be especially helpful in the definitiveidentification of the causative organism. However, theavailability and sensitivity of these tests vary, and insome cases diagnosis remains a challenging task.Stereotactic brain biopsy might lead to a definitivediagnosis in such cases, or sometimes after failedempirical therapy.39,40 Image-guided stereotactic brainbiopsy, particularly with MRI-compatible stereotacticsystems that are associated with high diagnostic yields(up to 8890%) for focal CNS lesions and low mortal-ity (23%), facilitates early detection, diagnosis,

Panel : Principles of HIV-associated CNS opportunistic

infections

CNS opportunistic infections typically occur when the

CD4-cell count is less than 200 cells per L

Diagnosis should be based on clinical presentation, temporal

evolution, CSF, and radiographic features

Multiple infections are present in 15% of cases and some

infections might be revealed only after combination

antiretroviral therapy is started

Combination antiretroviral therapy should be started,

modified, or continued with appropriate antimicrobial

therapy

Antimicrobial treatment is generally required until immune

recovery (CD4-cell count more than 200 cells per L) is

achieved with antiretroviral therapy


3/13

www.thelancet.com/neurology Vol 11 July 2012 607

Review

appropriate treatment, and improved prognosis.3941The

opportunistic infections most frequently diagnosedwith stereotactic biopsy are PML (2930%), primaryCNS lymphoma (2325%), and toxoplasmosis (1516%).MRI-guided stereotactic brain biopsy might beindicated in patients with solitary lesions on neuro-imaging, those who have multiple lesions that have notresponded to antitoxoplasmosis treatment, and thosewith multiple lesions and neurological progression.40

cART and management of CNS opportunisticinfectionsIn patients with CNS opportunistic infections, the timingof cART initiation is pertinent to the management oftreatment-naive patients as well as those who have

previously received cART. The use of cART to achieveimmune restoration is especially effective in patientswith opportunistic infections for which no effectiveantimicrobial therapy is available, including PML, butseveral complications are of concern, including druginteractions, toxic effects, and IRIS.

CNS opportunistic infections that develop in patients

who are already taking cART might be due to IRIS,especially within the first 12 weeks after cART is started(table 6). Alternatively, infection could be due to cARTtreatment failure, and the patients adherence should beassessed and antiretroviral resistance should be tested.In both scenarios, cART should be continued andmodified if virological failure (inability to achieve ormaintain viral replication to an HIV RNA level lowerthan 200 copies per mL), immunological failure (inabilityto achieve or maintain an adequate CD4 response despitevirological suppression), or adverse reactions develop, inaddition to antimicrobial therapy.

In resource-poor countries where cART is not available,a prophylactic antimicrobial regimen might help to

prevent CNS opportunistic infections. Region-specificpolicies on the timing and type of prophylaxis areneeded.4Guidelines from the Centers for Disease Controland Prevention (CDC) on the prevention and treatmentof opportunistic infections in adults and adolescentsinfected with HIV provide useful recommendations.6

CD4-cell count at

presentation (cells per L)

Time from symptom onset to

presentation

Change in mental

status

Seizures Headache Fever Focal

deficits

Cranial

neuropathies

Toxoplasmic encephalitis21,22


4/13


Review

Tuberculous meningitis and brain abscessesWHO estimates that a third of the worlds population isinfected with Mycobacterium tuberculosis and thatindividuals with HIV co-infection are at increased risk ofdisseminated, active forms of disease, includingtuberculous meningitis.42 The exact incidence of HIV-associated tuberculous meningitis is uncertain becauseepidemiological data are limited, especially for areas whereco-infection rates are highest, such as sub-Saharan Africa.42

Infection with M tuberculosisoccurs after inhalation ofairborne bacilli that traverse the alveoli into thebloodstream. Intact T-cell-mediated immunity helps tocontrol haematogenous spread. The disease disseminatesreadily in patients infected with HIV who have reducedCD4-cell counts. In the brain, granulomas or Rich focican form within the subpial and subependymal layers43and these can expand to create tuberculomas orparenchymal brain abscesses (table 5), or, morecommonly, rupture to cause meningitis.

Rapid detection is crucial in patients with HIV-associatedtuberculous meningitis, but detection of acid-fast bacilli orthe causative organism by positive culture or PCR is oftendiffi cult (table 4).44 Sensitivities for these tests havetraditionally been poor at about 50%, although patientsinfected with HIV have higher yield rates of up to 80%,possibly owing to incomplete immune responses andincreased bacterial loads.45 Neither tuberculin skin tests(sensitivity 31%) nor the interferon--release assayQuantiFERON-TB Gold (Cellestis, Valencia, CA, USA;sensitivity 60%) can precisely exclude tuberculosis inindividuals with HIV, especially when CD4-cell counts arelower than 200 cells per L.46 Repeated, large-volumelumbar punctures might improve the yield.44CSF tests thatinclude a nucleic-acid amplification test have substantially

shortened detection times and can improve testsensitivities when used in combination with acid-fast-bacillus microscopy and repeated sample testing (table 4).45

Treatment for tuberculosis does not differ significantlybetween patients with and without HIV infection. Thestandard regimen begins with isoniazid, pyrazinamide,ethambutol, and rifampicin for 2 months. Rifampicinnotably lowers levels of protease inhibitors and nevirapinein plasma, but rifabutin is an appropriate alternative.47After this initial phase, isoniazid and rifampicin orrifabutin are continued for at least 912 months.48The roleof corticosteroids and the optimum timing of the initiationof cART in conjunction with antituberculosis therapyremain controversial issues. In a randomised, controlledtrial of 545 patients with tuberculous meningitis inVietnam, an adjuvant corticosteroid (dexamethasone) wasassociated with a 30% reduction in the relative risk ofdeath, but not a significant reduction in the proportion ofseverely disabled patients.49 A smaller study assessed

253 patients who were treated with standardised anti-tuberculosis treatment, adjuvant dexamethasone, andprophylactic co-trimoxazole and who were followed-up for12 months. Immediate start of cART did not improveoutcomes compared with cART deferred for 2 months.32Thus, delayed initiation of cART is recommended forpatients with HIV-associated tuberculous meningitis.

Mortality from HIV-associated tuberculous meningitisoften exceeds 50%, which is roughly double the rate inpatients without HIV.43Even with the addition of cART,mortality is not improved.32

Multidrug-resistant tuberculosis (resistant to at leastisoniazid and rifampicin) is estimated to account for36% of incident tuberculosis diagnoses worldwide, andextensively drug-resistant disease (also resistant to

Mass effect Proportion of

solitary lesions (%)

Typical locations Enhancement Other

Toxoplasmic

encephalitis21,22,36

Frequent 3 cm diameter

Cytomegalovirus

encephalitis27,28None NA Periventricular


5/13


Review

fluoroquinolones and second-line injectable drugs) has

been documented in at least 58 countries since 2010.

1

Case series of multidrug-resistant HIV-associatedtuberculous meningitis suggest that mortality isextremely high.50,51This disease is of notable concern inresource-poor settings because the cost of treatment is100 times greater than that for susceptible tuberculousmeningitis.52 Further studies are needed to bettercharacterise HIV-associated multidrug-resistant andextensively drug-resistant tuberculous meningitis.

Toxoplasmic encephalitisToxoplasmic encephalitis remains the most commonlyreported neurological opportunistic infection since cARTwas introduced, although declines in incidence have

been seen: in 2007, the incidence in the UK was 04 per1000 person-years, which was a notable decline from 32only 10 years earlier.7Because toxoplasmic encephalitis iscaused by reactivation of encysted bradyzoites ratherthan primary infection, rates vary according to theseroprevalence of Toxoplasma gondii in the population.Risk factors for toxoplasmic encephalitis include thedegree of immunosuppression and whether or notprophylaxis for Pneumocystis jirovecipneumonia is beingused, since trimethoprim is also an effective preventivetherapy for toxoplasmic encephalitis.

After human ingestion of oocysts containing oocytes,usually from feline definitive hosts, the active tachyzoitesreplicate and transform into latent bradyzoites thatpersist in the brain and muscle.53 In immunodeficientindividuals, CD4-positive T cells are unable to suppressthis latent infection, and the tachyzoites re-emerge andreplicate.53 The tachzyoites commonly produce focalnecrotising cerebritis surrounded by a thick wall ofhistiocytes and inflamed vessels, although a more diffuseencephalitis, known as microglial nodular encephalitis,can occur in people with severe immunosuppression.53

The diagnosis of toxoplasmic encephalitis is oftenestablished by clinical and radiographic improvementsafter empirical treatment and by a positive test for IgGantibodies to T gondii in serum. Only about 3% ofpatients have negative serology.54When clinically feasible,

PCR of CSF obtained by lumbar puncture can be helpfulto detect the parasite. The specificity of PCR is excellent,although the sensitivity is only about 50% when tested ator near the time of treatment initiation (table 4).55NewPCR techniques and testing for CSF excretory-secretoryantigens offer increased sensitivity, but the costs of thesetests remain prohibitively high in resource-poor areas.5658Primary CNS lymphoma is often the principal differentialdiagnosis. MRI, F-fluorodeoxyglucose PET, and SPECTmight help to distinguish specific features,59,60but a brainbiopsy could be necessary if equivocal or worsenedresponse to empirical treatment is seen.

Combined pyrimethamine, folinic acid, andsulfadiazine has traditionally been used, althoughtrimethoprim-sulfamethoxazole was equally effective in a

Figure : Algorithm of diagnostic principles for HIV-associated CNS opportunistic infections

Headache, seizure, altered mental state

+/ fever, focal deficits

Neuroimaging (MRI/CT brain)

Lesions on MRI or CT

No sign of herniation

Single lesion Multiple lesions

Lumbar punctureor brain biopsy

Toxoplasma serologypositive

Toxoplasma serologynegative

Trial of empiricalantitoxoplasmosistreatment

If patient responds,continue

If no response,consider brain biopsy

Brain biopsy

Signs of herniation

Neurosurgical referralfor decompressionand brain biopsy

No mass lesion

Lumbar puncture formicroscopy and cultureEmpirical antimicrobial therapy

Panel : Diagnostic work-up for patients with HIV

infection and presumed CNS opportunistic infections

Imaging

Neuroimaging with contrast-enhanced MRI or CT

Chest radiography or CT for tuberculosis

CSF

Bacterial, fungal, and mycobacterial culture

White-blood-cell count and differential protein and

glucose concentrations

Cytology and flow cytometry

PCR for JC virus, Epstein-Barr virus, cytomegalovirus,

Toxoplasma gondii, herpes simplex virus types 1 and 2, and

varicella zoster virus

Cryptococcal antigen

Microscopy for India ink stain (Cryptococcus) and acid-fast

bacilli (Mycobacterium tuberculosis)

Other laboratory tests

CD4-positive T-cell count

IgG antibodies to Toxoplasma gondiiin CSF, blood, and urine

Serum QuantiFERON-TB Gold test*

*Cellestis, Valencia, CA, USA.


6/13


Review

small randomised trial, and is perhaps more economicalfor resource-poor regions.61 In a large clinical study ofpatients treated presumptively for toxoplasmicencephalitis, the median time to treatment response was5 days; 74% improved within 7 days and 91% respondedwithin 14 days. Failure to improve within 1014 days ofstarting treatment should, therefore, prompt reassess-ment of the diagnosis, with either thallium SPECT orbrain biopsy.38Corticosteroids are indicated only when asubstantial mass effect is seen, and should bediscontinued as soon as possible. Induction treatmentshould be continued for at least 6 weeks, until substantialradiographic improvement is recorded, including the lossof contrast enhancement,6 although in a few patientscontrast enhancement might persist for months or evenyears. Maintenance therapy should be continued in allpatients until immune reconstitution is achieved(persistent CD4-cell count of more than 200 cells per L).Anticonvulsants are indicated only in patients with ahistory of seizures related to toxoplasmic encephalitis.Despite the introduction of cART, mortality fortoxoplasmic encephalitis remains high at 2060% within

1 year of diagnosis.21,62Low CD4-cell counts at the time ofdiagnosis, a history of other AIDS-defining illnesses, ageolder than 45 years, and the presence of encephalopathyportend a poor prognosis.21

Cryptococcal meningitisBefore the introduction of cART, 510% of patients withAIDS developed cryptococcal meningitis. Although theincidence has fallen, this disease remains a major concernin sub-Saharan Africa and south and southeast Asia.63

Cryptococcus neoformans is an encapsulated yeastcommonly found in soil and in excrement from pigeons.Cryptococcus gattii is generally found in tropical andsubtropical regions, and causes disease in immuno-competent individuals.64 Exposure to Cryptococcus spp

probably occurs via inhalation of spores, which leads to

primary pulmonary infection, latent infection, ordisseminated disease. Cryptococcal meningitis is thoughtto be the result of reactivation of latent infection inimmunocompromised patients.65Although described asmeningitis, the yeast typically forms cystic accumulationsin Virchow-Robin spaces, around the deep-penetratingblood vessels in the brain and cranial nerves, termedsoap-bubble cysts.

Lumbar puncture with manometry is especially helpfulfor the diagnosis of cryptococcal meningitis, as the CSFopening pressure is typically raised (table 4).66The CSFprofile at presentation is normal in about 30% ofpatients.67Microscopic detection with India ink stainingand fungal culture of the CSF are diagnostic.

Immunoassays of cryptococcal antigens are rapid,sensitive, and specific, and confer prognostic value.68Urinary antigen detection has high sensitivity, and canbe useful, especially in resource-poor settings, in theappropriate clinical context.69

Clinically, cryptococcal meningitis is most notable forits propensity to cause communicating hydrocephaluswith increased intracranial pressure, which develops inmore than 15% of patients.70Aggressive management ofraised intracranial pressure is crucial to lessen the risk ofearly death. Impaired reabsorption of CSF by thearachnoid villi is thought to be due to sticky polysaccharidecapsules. Patients with increased intracranial pressuretypically present with headache, vomiting, visualchanges, hearing loss, palsy of the abducens nerve, andimpaired consciousness; however, this complication canbe silent and is potentially fatal. Dysfunction in multiplecranial nerves might occur owing to associated basalmeningitis. Repeated high-volume lumbar puncturefrequently leads to immediate symptomatic relief, andcan reverse neurological morbidity, such as blindness ordeafness.71 Data on the management of raised intra-cranial pressure in cryptococcal meningitis are limited.Guidelines recommend serial daily lumbar punctures ifthe opening pressure is persistently greater than 25 cmH2O.

72Drainage of CSF to reduce the pressure by 50% orto a pressure in the normal range (less than 20 cm H 2O)

is recommended. Lumbar drains or even ventriculo-peritoneal shunts have been used if increased intracranialpressure remains diffi cult to control.73

The recommended treatment for cryptococcalmeningitis is intravenous amphotericin B in combinationwith oral flucytosine for a minimum of 2 weeks, followedby oral fluconazole for at least 8 weeks.72 Liposomalamphotericin B is associated with lower risks of renaltoxic effects and other side-effects than conventionalamphotericin B and has similar effi cacy, but it is moreexpensive.74Combined flucytosine with amphotericin Bleads to faster and greater sterilisation of CSF than doesamphotericin B alone.75Prophylaxis with fluconazole isrequired until a durable immune reconstitution with aCD4-cell count higher than 200 cells per L is achieved.

IRIS Worsening or recurrent

opportunistic infection

Context Onset early (average 2 months)

after start of cART, generally in

patients with good adherence

to treatment

Generally not closely related to start of

cART, often seen in patients with a

history of non-adherence to treatment

Plasma HIV RNA concentration Decreased Increased

CD4-positive T-cell count Increased Decreased

CSF profile Lymphocytosis, raised protein

concentrations, and, frequently,

sterile culture

Lymphocytosis less intense than with

IRIS, raised protein concentrations,

possibly low glucose, and generally

positive culture

Neuroradiology Worsening lesion with cerebral

oedema and contrast

enhancement

May have worsening lesions, but

surrounding cerebral oedema might be

less intense than with IRIS; contrast

enhancement

Treatment Corticosteroid (anecdotal) Antimicrobial

IRIS=immune reconstitution inflammatory syndrome. cART=combination antiretroviral therapy.

Table : Distinguishing features of IRIS and worsening or recurrent opportunistic infections


7/13


Review

In resource-limited settings, only amphotericin B and

fluconazole are generally available.

75

A large trial in Africasupports the use of fluconazole as primary prophylaxis,irrespective of whether patients are receiving cART.76

IRIS affects 1040% of patients with cryptococcalmeningitis, and the risk is increased in individuals whoare antiretroviral naive and who have high concentrationsof HIV RNA. Mortality for IRIS associated withcryptococcal meningitis is 3366%.77 Appropriatemanagement of IRIS includes continuation of cART,antifungal therapy, and a course of corticosteroids.6

Cytomegalovirus encephalitisNeurological diseases caused by cytomegalovirus are rarebut very serious. Infection can lead to encephalitis,

retinitis, radiculomyelitis, or mononeuritis multiplex.Although cytomegalovirus encephalitis seems to occur atsimilar rates worldwide, retinitis has a distinctgeographical distributionit is the predominant HIV-associated eye disease in Asia, where up to a third ofHIV-infected patients are affected. 78

The ubiquitous herpes virus, cytomegalovirus isendemic worldwide and usually causes asymptomatic orclinically benign infections. Most people have beeninfected by the time they reach adulthood. The highneurovirulence of some cytomegalovirus strains andhost genetic risk factors might play a part in thedevelopment of neurological disease.79

In cases of suspected cytomegalovirus encephalitis,brain imaging with contrast should be done. Meningealenhancement or periventricular inflammation can beseen in infected patients, but these findings are notspecific to cytomegalovirus encephalitis (table 5).80PCRof CSF is highly sensitive and specific in the diagnosis ofcytomegalovirus encephalitis (table 4) and might also bepositive in patients with cytomegalovirus radiculomyelitis.Quantitative PCR results can also help in the assessmentof disease severity and in monitoring response toantiviral therapy, having a sensitivity of 250 copies per mLand a diagnostic specificity of 9099%.81 A neutrophil-predominant CSF pleocytosis with negative bacterialculture is highly suggestive of cytomegalovirus

infection.82 Retinitis associated with cytomegalovirusinfection is typically diagnosed by areas of haemorrhagicinfarction, perivascular sheathing, and retinalopacification on fundoscopy.

Patients with cytomegalovirus encephalitis commonlypresent with rapidly progressive encephalopathy, oftenwith hyponatraemia due to adrenal involvement.83Brainstem symptoms and seizures can also occur.Patients with extraocular complications frequently haveprevious or concurrent cytomegalovirus retinitis andshould undergo fundoscopy, as blindness can develop ifretinal disease remains untreated. Visual symptoms caninclude floaters, loss of peripheral vision, or centralscotoma, but not all patients are symptomatic. The visualloss in cytomegalovirus retinitis is usually the result of

retinal necrosis, but macular oedema, retinal detachment,

and papillitis might also reduce visual acuity.

84

Radiculomyelitis typically affects the lumbosacral regionsand can present as a rapidly progressive cauda equinasyndrome. The clinical presentation resembles Guillain-Barr syndrome, except that the CSF generally has aneutrophilic predominance and MRI imaging showscontrast-enhancing and greatly thickened nerve roots.

Little evidence is available to definitively guide thetreatment of cytomegalovirus encephalitis. A combinationof ganciclovir and foscarnet was generally used before theintroduction of cART, but this regimen is associated withsubstantial adverse effects. The use of cART plus oneanticytomegalovirus treatment is favoured. Ganciclovirwas the first available agent and remains the first-line

treatment, although mutations in the viral genes UL97orUL54 can lead to drug resistance. Ganciclovir is onlyavailable intravenously, but its prodrug, valganciclovir, canbe given orally. While a sustained-release intraocularimplant treats cytomegalovirus retinitis in the affected eye,oral ganciclovir is recommended to prevent cytomegalo-virus retinitis or extraocular disease.85 Foscarnet is usedwhen patients develop dose-limiting leucopenia whiletaking ganciclovir or in ganciclovir-resistant cases,although some cross-resistance has been shown. Cidofoviris a competitive inhibitor of the viral DNA polymeraseafter conversion to an active form. As the activation ofcidofovir does not rely on viral kinases, it retains its activityfor cytomegalovirus with UL97mutations, but mutationsin DNA polymerase genes, such as UL54, lead toresistance.86 Secondary prophylaxis with long-term oralanticytomegalovirus therapy should be continued untilsustained immune recovery is achieved with cART.87Drugsusceptibility testing based on screening for knownresistance-inducing mutations in the viral genome canhelp to guide antiviral therapy.88

Progressive multifocal leukoencephalopathyPML is a rare but severe CNS opportunistic infection thatis caused by the reactivation of latent JC virus, apolyomavirus, in immunosuppressed individuals.Unlike most of the other HIV-associated CNS

opportunistic infections, which are very rare when theCD4-cell count is higher than 100200 cells per L, PMLoccasionally occurs in patients with much higher CD4-cell counts (table 3).89,90The seroprevalence of JC virus inthe general population is 5090%,9193and about 30% ofpeople shed the virus in urine.94The incidence of PMLhas declined from seven cases per 1000 patient-yearsbefore the introduction of cART to 07 per 1000 (table 1).7

The mechanisms underlying the pathogenesis of PML,especially the mode of acquisition, latency, anddissemination of JC virus, site of reactivation, and theescape of CNS immunosurveillance, remain unclear.95Primary JC-virus infection is typically asymptomatic93andthe virus remains in the kidneys, bone marrow, andlymphoid tissue.96 JC-virus DNA is detectable in brain


8/13


Review

tissue from individuals without PML, but with variable

frequency (1168%).

97,98

The pathogenic sequence of eventsin the development of PML is uncertain. The archetypalvirus is non-pathogenic, and for neurovirulence and PMLto occur, mutational changes are required in the non-coding control region and in the capsid viral protein VP1,which modulates ganglioside binding.99 Whether PMLdevelops from a primary brain ingress of neurovirulent JCvirus from circulating lymphocytes after reactivation inthe bone marrow96,100,101 or whether the latent virusundergoes secondary reactivation within the brain isunclear. In the brain, JC virus infects mainly oligo-dendrocytes102 and astrocytes,103 and, occasionally,cerebellar granular cells104 and cortical pyramidalneurons.105,106HIV-1 may act directly as a cofactor for JC-

virus replication via the HIV-1 TAT protein, throughtransactivation of the JC-virus late promoter in glial cells,which provides an additional pathogenic mechanism.107

PML typically causes multifocal demyelinating lesionsin white matter, and patients often present with focalneurological deficit that has developed over severalweeks. Rarely, the deficit will develop acutely and couldbe mistaken for stroke.108Other, less common diseasesassociated with JC-virus infection have been reported,including JC-virus granule cell neuronopathy101 withcerebellar ataxia, JC-virus encephalopathy, and JC-virusmeningitis.105.106

Unifocal or multifocal areas of hyperintensity insubcortical white matter on T2-weighted, fluid-attenuatedinversion recovery MRI associated with well-demarcatedhypointense lesions on T1-weighted MRI are highlysuggestive of PML. The cortical ribbon is classicallyspared (table 5, figure 2). Minimal tissue oedema or nocontrast enhancement is usual unless PML iscomplicated by IRIS.109,110 Rarely, the only radiologicalfinding is severe cerebellar atrophy.111

A definitive diagnosis of PML is established by thedetection of JC virus in CSF by PCR. The diagnosticsensitivity varies across laboratories, but specificity ishigh (table 4).112,113 Brain biopsy might occasionally berequired to confirm the diagnosis.

The mainstay of treatment for PML in patients infected

with HIV is immune reconstitution with cART. Thisapproach has improved survival from 10% before theintroduction of cART to 5075% since.23,114116The presenceof JC-virus-specific CD8-positive cytotoxic T lymphocytes inplasma is associated with improved survival.117No antiviraltherapy with proven effi cacy is available. Various agents,including interferon alfa,118 cytarabine,119 mefloquine,mirtazapine (a 5-HT2A-receptor blocker that can prevent JC-virus entry into glial cells),120and cidofovir,121have shownlittle or no effi cacy. The low effi cacy of cidofovir might bedue to poor CNS penetration and dose-limiting side-effects,but a lipid derivative, CMX001 (Chimerix, Durham, NC,USA), has shown promising results in vitro.122A commoncomplication, the exaggerated inflammatory response,IRIS, that is associated with cART, might unmask or

paradoxically worsen PML, with variable outcomes.109

Corticosteroids administered at various doses and fordifferent treatment durations have had some positiveeffects on brain inflammation. The CDC guidelinessuggest the use of corticosteroids for PML-associatedIRIS, but not in individuals without any evidence ofinflammation.6

Primary CNS lymphomaHIV infection is a well-established risk factor for primaryCNS lymphoma.123 Since cART was introduced, theincidence among people with HIV infection hassubstantially declined,124 and it is now a rare disease.Between 2001 and 2007 in the USA, around 26 new casesof primary CNS lymphoma per 100 000 person-years

were reported among people with AIDS compared with038 cases among people without AIDS.125

Primary CNS lyphoma must be distinguished fromsecondary involvement of the CNS in systemiclymphoma. The pathogenesis of primary CNSlymphoma in HIV/AIDS is variably associated withmultiple genetic alterations, including proto-oncogeneactivation (MYC, PIM1, RHOH, also known as TTF, andPAX5),126and monoclonal Epstein-Barr virus infection.127Most primary CNS lymphomas are high-grade B-celltumours that are multicentric and express markers ofthe germinal centre, including BCL-6 and IRF-4 (alsoknown as MUM1) and pan-B-cell markers, includingCD19, CD20, CD22, and CD79a.128

Clinical features of primary CNS lymphoma are non-specific, and may include lethargy, cognitive changes,headache, and focal neurological symptoms from anintracranial mass lesion. Presentations may also belimited to the leptomeninges and, exceptionally, to thespinal cord. Lymphoma cells are seen in the anteriorchamber of the eye in 20% of patients.128 TypicalB systemic symptoms are not usually present.

Contrast-enhanced CT or MRI usually revealsmultifocal lesions, predominantly supratentorial, thatspread along the white-matter tracts, which showheterogeneous contrast enhancement with gadolinium(table 5). Use of contrast-enhanced body imaging, such as

CT of the chest, abdomen, and pelvis, or whole-body PETscanning, is imperative to exclude systemic lymphoma.The gold standard for diagnosis of primary CNSlymphoma is brain biopsy. PCR is highly sensitive(table 4) and should be used to test for Epstein-Barr virusDNA in CSF unless lumbar puncture is contraindicated.The positive predictive value of Epstein-Barr virus DNAmight be as low as 29%, and specificity is also low.129Quantification of Epstein-Barr virus DNA present in theCSF can improve the specificity for primary CNSlymphoma compared with qualitative detection (96% vs66% when a cut-off DNA load of 10 000 copies per mL isused).130CSF cytopathology is almost never informative.Thallium SPECT can help to distinguish primary CNSlymphoma from other infections, such as toxoplasmic


9/13


Review

encephalitis, with sensitivity and specificity of about

90%.

131

We recommend thallium SPECT imaging whenempirical treatment for toxoplasmic encephalitis seemsto be failing, as tracer uptake generally indicates neoplasmrather than infection.

Corticosteroids can obscure biopsy results and,therefore, they should be avoided unless required toprevent imminent herniation. Patients with HIVinfection and primary CNS lymphoma generally havepoor outlook despite treatment,132with a median survivalof 2 months.133The level of evidence for the treatment ofprimary CNS lymphoma is low; people with HIVinfection have been excluded from two phase 3 trials andone randomised phase 2 chemotherapy trial for primaryCNS lymphoma.128 cART is the definitive treatment for

HIV-related primary CNS lymphoma, and radiotherapymight improve survival.134

Herpes simplex virus encephalitisHerpes simplex virus is an infrequent cause of CNSopportunistic infections. In studies from before and afterthe widespread use of cART, PCR detection in CSF ofDNA for herpes simplex virus type 1 or 2 was only 2%,135,136whereas detection of cytomegalovirus was 74% in adultswith HIV infection and neurological symptoms.35 Co-infection of herpes simplex virus 2 and Epstein-Barrvirus has been reported.136

Herpes simplex virus is a neurotropic virus. Type 1 iscommonly acquired in early life and may remain latentin specific CNS sensory ganglia. Type 2 is acquiredthrough contact with infected mucosal surfaces, andseroprevalence increases with age and number of sexualpartners.35 Necrotising encephalitis, which frequentlyaffects the temporal and inferior frontal lobes, occurs inpatients with HIV infection, although fatal encephalitiscan occur even in the absence of a vigorous inflammatoryresponse. Cutaneous herpetic lesions are not predictiveof CNS infection by herpes simplex virus.

Clinical presentation of CNS herpes simplex virusinfection in the setting of HIV/AIDS varies widely, andincludes fever, headache, neck stiffness, vomiting,disorientation, memory loss, dysphasia, depression, con-

fusion, personality change, seizures, visual hallucinations,and photophobia.35Herpes simplex virus type 1 typicallycauses encephalopathy that might develop subacutelyover several weeks. Herpes simplex virus type 2 typicallycauses a diffuse meningoencephalitis that can recur. PCRof the CSF is highly sensitive (100%) and specific (996%)for herpes simplex virus DNA, as confirmed by autopsy,135and can help to identify cases of mild encephalitis,although false-negative results are possible when testingis done less than 72 h after the onset of symptoms.137Electroencephalography most commonly shows periodiclateralised epileptiform discharges in frontotemporalregions.35 T2-weighted brain imaging shows hyper-intensities in most cases and sometimes gadolinium-enhancing lesions involving the inferomedial temporal

lobes,35 but lesions can also involve the brainstem,cerebellum, diencephalon, and periventricular zones.

The treatment of choice is 30 mg/kg aciclovir daily,given intravenously for 1421 days. The drug is convertedto aciclovir triphosphate, a potent inhibitor of herpessimplex virus DNA polymerase, which is required for viralreplication. Progression can occur despite treatment,especially if CD4-cell counts are low.135 Ganciclovir andfoscarnet have also been used with some success,135,136andaciclovir resistance has been described.138 Relapses within3 months of treatment have been reported, and a trial of90 days of valaciclovir after induction treatment has been

Figure : Axial MRI of the brain in patients with HIV and CNS opportunistic infections

(A) Cryptococcal meningitis in a man aged 58 years who presented with headache, weight loss, anorexia, and a

generalised seizure, with a CD4-cell count of 10 cells per L. The arrow shows gadolinium enhancement of the

posterior meninges on T1-weighted MRI. A cryptococcoma is present in the right basal ganglion. (B) Cerebral

toxoplasmosis in a man aged 39 years who presented with generalised seizure and nadir in CD4-cell count of

2 cells per L. Diagnosis was confirmed by brain biopsy. Multiple hyperintense lesions with surrounding oedema

(arrow), appearing as a hypointense rim, with mass effect can be seen on T2-weighted, fluid-attenuated inversion

recovery MRI. (C) Primary CNS lymphoma in a man aged 21 years who presented with acute-onset left-sided

weakness and a CD4-cell count of 0 cells per L. Diagnosis was confirmed by brain biopsy. A rim-enhancing lesion

(arrow) is visible in the right temporoparietal lobe on gadolinium-enhanced MRI. (D) Progressive multifocal

leukoencephalopathy in a man aged 42 years who presented with lethargy, left hemiparesis, dysphagia, seizures,

and visual hallucinations. T2-weighted, fluid-attenuated inversion recovery MRI shows asymmetric, confluent

hyperintensities involving the cerebral white matter (arrow) without mass effect. All patients were started oncombination antiretroviral therapy and appropriate antimicrobials and supportive treatment, and all survive.

A B

C D


10/13


Review

completed, but the results are awaited (ClinicalTrials.gov,NCT00031486).

ConclusionsAlthough CNS opportunistic infections are decreasing infrequency, they continue to have a devastating effect onHIV-positive individuals, especially those in whomdiagnosis is delayed or HIV treatment is inadequate.Mortality is often high even with appropriate treatment,and recurrences and residual neurological deficits arecommon. For some disorders, such as PML, primaryCNS lymphoma, and multidrug-resistant tuberculosis,curative treatments are not yet available, whichunderscores the urgent need for prevention and earlydetection of HIV and associated CNS opportunisticinfections. Access to cART can be lifesaving, and globalprogrammes to roll out this treatment in resource-poorcountries are clearly major factors in reducing theincidence of CNS opportunistic infections.

Concurrent infection with more than one CNSopportunistic infection is an important consideration inthe care of patients infected with HIV, which adds to thechallenge of managing the neurological complications ofHIV infection. The relations between HIV and infectiousdiseases such as neurosyphilis, varicella zoster, and somethat are uncommon in high-income settings (eg, cerebralmalaria, Chagas disease, and neurocysticercosis) are notdiscussed here, but they deserve further research andattention. Vigilance is required in patients with HIV

infection receiving cART because CNS-associated IRISmust always be considered when a patient clinicallyworsens after treatment is started (table 6). Owing to thedearth of data from clinical trials, the treatment of nearlyall CNS opportunistic infections requires a therapeuticchoice based on data from patients without HIV infectionor indirect evidence from cohort studies or case series.

Overall, the burden of CNS opportunistic infections inpeople with HIV/AIDS has declined substantially, whichis a measure of widespread, successful treatment withcART. Knowledge acquired as the HIV epidemic unfoldshas led to improved understanding of multiple infectionsthat were previously rare. This knowledge can now beused to treat patients in other settings, such as in thecontext of immunosuppression for transplanted

allografts and in the testing of new immunomodulatory

drugs for autoimmune diseases. Nevertheless, patientswith unrecognised, and thus untreated, HIV infectionswill continue to present with CNS opportunisticinfections. Whether due to stigma associated with adiagnosis of HIV, poor access to medical care,unavailability of cART, non-adherence, or other factors,the persistence of this vulnerable group of individualsmust prompt efforts to improve the outcomes for patientswith CNS opportunistic infections in the context of HIV.

Contributors

All authors contributed to the literature search, the writing, and thereview of the paper.

Conflicts of interest

We declare that we have no conflicts of interest.

AcknowledgmentsThis work is supported by a grant from the Johns Hopkins NationalInstitute for Mental Health Research Center (P30MH075673). ILT issupported by a Whitehurst foundation gift. BRS is supported by theJohns Hopkins University T32 Training Program in Hematology. GvG issupported by John Hopkins University Project RESTORE and MultipleSclerosis Center. FJM is supported by an American Academy ofNeurology Practice Research Fellowship grant.

References1 UNAIDS. Global report: UNAIDS report on the global AIDS epidemic,

2010. Geneva: Joint United Nations Programme on HIV/AIDS, 2010.

2 Garvey L, Winston A, Walsh J, et al. HIV-associated central nervoussystem diseases in the recent combination antiretroviral therapyera. Eur J Neurol2011; 18:52734.

3 Wright EJ, Nunn M, Joseph J, et al. NeuroAIDS in the Asia PacificRegion.J Neurovirol2008; 14:46573.

4 Holmes CB, Losina E, Walensky RP, et al. Review of human

immunodeficiency virus type 1-related opportunistic infections insub-Saharan Africa. Clin Infect Dis2003; 36:65262.

5 Eza D, Cerrillo G, Moore DA, et al. Postmortem findings andopportunistic infections in HIV-positive patients from a publichospital in Peru. Pathol Res Pract2006; 202:76775.

6 Kaplan JE, Benson C, Holmes KH, et al. Guidelines for preventionand treatment of opportunistic infections in HIV-infected adultsand adolescents: recommendations from CDC, the NationalInstitutes of Health, and the HIV Medicine Association of theInfectious Diseases Society of America. MMWR Recomm Rep2009;58:1207.

7 dArminio Monforte A, Cinque P, Mocroft A, et al. Changingincidence of central nervous system diseases in the EuroSIDAcohort. Ann Neurol2004; 55:32028.

8 Mocroft AJ, Lundgren JD, dArmino, et al. Survival of AIDS patientsaccording to type of AIDS-defining event. The AIDS in EuropeStudy Group. Int J Epidemiol1997; 26:40007.

9 Chiu IM, Yaniv A, Dahlberg JE, et al. Nucleotide sequence evidencefor relationship of AIDS retrovirus to lentiviruses. Nature1985;317: 36668.

10 Choe H, Farzan M, Sun Y, et al. The -chemokine receptors CCR3and CCR5 facilitate infection by primary HIV-1 isolates. Cell1996;85:113548.

11 Dragic T, Litwin V, Allaway GP, et al. HIV-1 entry into CD4+cells ismediated by the chemokine receptor CC-CKR-5. Nature1996;381:66773.

12 Rutherford GW, Lifson AR, Hessol NA, et al. Course of HIV-Iinfection in a cohort of homosexual and bisexual men: an 11 yearfollow up study. BMJ1990; 301:118388.

13 Fauci AS, Pantaleo G, Stanley S, Weissman D. Immunopathogenicmechanisms of HIV infection. Ann Intern Med1996; 124:65463.

14 Daar ES, Moudgil T, Meyer RD, Ho DD. Transient high levels ofviremia in patients with primary human immunodeficiency virustype 1 infection. N Engl J Med1991; 324:96164.

15 Brenchley JM, Price DA, Douek DC. HIV disease: fallout from a

mucosal catastrophe? Nat Immunol2006; 7:23539.

Search strategy and selection criteria

We chose the infections to include in this Review on the basis of the frequency and

importance of the infection in the CNS in people infected with HIV. References were found

by searches of PubMed for papers published from January, 1990, to February, 2012. We

used the search term human immunodeficiency virus AND primary central nervous

system lymphoma OR Epstein Barr virus OR JC virus OR progressive multifocal

leukoencephalopathy OR tuberculosis OR toxoplasmosis OR cytomegalovirus OR

herpes simplex virus OR cryptococcus. We only reviewed papers published in English.

Additional materials were found by manual searches of the reference lists of selected

articles, textbooks, and relevant disease-specific guidelines.


11/13


Review

16 Koup RA, Safrit JT, Cao Y, et al. Temporal association of cellularimmune responses with the initial control of viremia in primary

human immunodeficiency virus type 1 syndrome.J Virol1994;68:465055.

17 Safrit JT, Andrews CA, Zhu T, et al. Characterization of humanimmunodeficiency virus type 1-specific cytotoxic T lymphocyteclones isolated during acute seroconversion: recognition ofautologous virus sequences within a conserved immunodominantepitope.J Exp Med1994; 179:46372.

18 Finzi D, Hermankova M, Pierson T, et al. Identification of areservoir for HIV-1 in patients on highly active antiretroviraltherapy. Science1997; 278:1295300.

19 Clark SJ, Saag MS, Decker WD, et al. High titers of cytopathic virusin plasma of patients with symptomatic primary HIV-1 infection.N Engl J Med1991; 324:95460.

20 Rothman RE, Keyl PM, McArthur JC, et al. A decision guideline foremergency department utilization of noncontrast head computedtomography in HIV-infected patients. Acad Emerg Med1999;6:101019.

21 Antinori A, Larussa D, Cingolani A, et al. Prevalence, associatedfactors, and prognostic determinants of AIDS-related toxoplasmicencephalitis in the era of advanced highly active antiretroviraltherapy. Clin Infect Dis2004; 39:168191.

22 Pereira-Chioccola VL, Vidal JE, Su C. Toxoplasma gondiiinfectionand cerebral toxoplasmosis in HIV-infected patients.Future Microbiol2009; 4:136379.

23 Berger JR, Pall L, Lanska D, Whiteman M. Progressive multifocalleukoencephalopathy in patients with HIV infection.J Neurovirol1998; 4:5968.

24 Engsig FN, Hansen AB, Omland LH, et al. Incidence, clinicalpresentation, and outcome of progressive multifocalleukoencephalopathy in HIV-infected patients during the highlyactive antiretroviral therapy era: a nationwide cohort study.J Infect Dis2009; 199:7783.

25 Haldorsen IS, Krakenes J, Goplen AK, et al. AIDS-related primarycentral nervous system lymphoma: a Norwegian national survey19892003. BMC Cancer2008; 8:225.

26 Bayraktar S, Bayraktar UD, Ramos JC, et al. Primary CNSlymphoma in HIV positive and negative patients: comparison ofclinical characteristics, outcome and prognostic factors.J Neurooncol2011; 101:25765.

27 Anduze-Faris BM, Fillet AM, Gozlan J, et al. Induction andmaintenance therapy of cytomegalovirus central nervous systeminfection in HIV-infected patients. AIDS2000; 14:51724.

28 Holland NR, Power C, Mathews VP, et al. Cytomegalovirusencephalitis in acquired immunodeficiency syndrome (AIDS).Neurology1994; 44:50714.

29 Graybill JR, Sobel J, Saag M, et al. Diagnosis and management ofincreased intracranial pressure in patients with AIDS andcryptococcal meningitis. The NIAID Mycoses Study Group andAIDS Cooperative Treatment Groups. Clin Infect Dis2000;30:4754.

30 van der Horst CM, Saag MS, Cloud GA, et al. Treatment ofcryptococcal meningitis associated with the acquiredimmunodeficiency syndrome. National Institute of Allergy and

Infectious Diseases Mycoses Study Group and AIDS Clinical TrialsGroup. N Engl J Med1997; 337:1521.

31 Antinori S, Radice A, Galimberti L, et al. The role of cryptococcalantigen assay in diagnosis and monitoring of cryptococcalmeningitis.J Clin Microbiol2005; 43:582829.

32 Torok ME, Yen NT, Chau TT, et al. Timing of initiation of antiretroviraltherapy in human immunodeficiency virus (HIV)-associatedtuberculous meningitis. Clin Infect Dis2011; 52:137483.

33 Katrak SM, Shembalkar PK, Bijwe SR, Bhandarkar LD. The clinical,radiological and pathological profile of tuberculous meningitis inpatients with and without human immunodeficiency virusinfection.J Neurol Sci2000; 181:11826.

34 Thwaites GE, Duc Bang N, Huy Dung N, et al. The influence ofHIV infection on clinical presentation, response to treatment, andoutcome in adults with tuberculous meningitis.J Infect Dis2005;192:213441.

35 Tyler KL. Herpes simplex virus infections of the central nervoussystem: encephalitis and meningitis, including Mollarets. Herpes

2004; 11(suppl 2):57A64A.

36 Antinori A, Ammassari A, De Luca A, et al. Diagnosis of AIDS-related focal brain lesions: a decision-making analysis based on

clinical and neuroradiologic characteristics combined withpolymerase chain reaction assays in CSF. Neurology1997; 48:68794.

37 Mathews VP, Alo PL, Glass JD, et al. AIDS-related CNScryptococcosis: radiologic-pathologic correlation.AJNR Am J Neuroradiol1992; 13:147786.

38 Quality Standards Subcommittee of the American Academy ofNeurology. Evaluation and management of intracranial masslesions in AIDS. Report of the Quality Standards Subcommittee ofthe American Academy of Neurology. Neurology1998; 50:2126.

39 Skolasky RL, Dal Pan GJ, Olivi A, et al. HIV-associated primaryCNS lymorbidity and utility of brain biopsy.J Neurol Sci1999;163:3238.

40 Levy RM, Breit R, Russell E, Dal Canto MC. MRI-guided stereotaxicbrain biopsy in neurologically symptomatic AIDS patients.J Acquir Immune Defic Syndr1991; 4:25460.

41 Corti M, Metta H, Villafane MF, et al. Stereotactic brain biopsy inthe diagnosis of focal brain lesions in AIDS. Medicina (B Aires)2008; 68:28590 [in Spanish].

42 Kwan CK, Ernst JD. HIV and tuberculosis: a deadly humansyndemic. Clin Microbiol Rev2011; 24:35176.

43 Simmons CP, Thwaites GE, Quyen NT, et al. Pretreatmentintracerebral and peripheral blood immune responses in Vietnameseadults with tuberculous meningitis: diagnostic value and relationshipto disease severity and outcome.J Immunol2006; 176:200714.

44 Marais S, Thwaites G, Schoeman JF, et al. Tuberculous meningitis:a uniform case definition for use in clinical research.Lancet Infect Dis 2010; 10:80312.

45 Thwaites GE, Caws M, Chau TT, et al. Comparison of conventionalbacteriology with nucleic acid amplification (amplifiedmycobacterium direct test) for diagnosis of tuberculous meningitisbefore and after inception of antituberculosis chemotherapy.J Clin Microbiol2004; 42:99602.

46 Syed Ahamed Kabeer B, Sikhamani R, Swaminathan S, et al. Roleof interferon gamma release assay in active TB diagnosis amongHIV infected individuals. PLoS One2009; 4:e5718.

47 Luetkemeyer AF. Current issues in the diagnosis and managementof tuberculosis and HIV coinfection in the United States.Trop HIV Med2010; 18:14348.

48 American Thoracic Society, CDC, Infectious Diseases Society ofAmerica. Treatment of tuberculosis. MMWR Recomm Rep2003;52:177.

49 Thwaites GE, Nguyen DB, Nguyen HD, et al. Dexamethasone forthe treatment of tuberculous meningitis in adolescents and adults.N Engl J Med2004; 351:174151.

50 Torok ME, Chau TT, Mai PP, et al. Clinical and microbiologicalfeatures of HIV-associated tuberculous meningitis in Vietnameseadults. PLoS One2008; 3:e1772.

51 Daikos GL, Cleary T, Rodriguez A, Fischl MA. Multidrug-resistanttuberculous meningitis in patients with AIDS. Int J Tuberc Lung Dis2003; 7:39498.

52 Patel VB, Padayatchi N, Bhigjee AI, et al. Multidrug-resistanttuberculous meningitis in KwaZulu-Natal, South Africa.Clin Infect Dis2004; 38:85156.

53 Nath A, Sinai AP. Cerebral toxoplasmosis. Curr Treat Options Neurol2003; 5:312.

54 Raffi F, Aboulker JP, Michelet C, et al. A prospective study ofcriteria for the diagnosis of toxoplasmic encephalitis in 186 AIDSpatients. The BIOTOXO Study Group. AIDS1997; 11:17784.

55 Cingolani A, De Luca A, Ammassari A, et al. PCR detection ofToxoplasma gondiiDNA in CSF for the differential diagnosis ofAIDS-related focal brain lesions.J Med Microbiol1996; 45:47276.

56 Alfonso Y, Fraga J, Fonseca C, et al. Molecular diagnosis ofToxoplasma gondiiinfection in cerebrospinal fluid from AIDSpatients. Cerebrospinal Fluid Res2009; 6:2.

57 Nogui FL, Mattas S, Turcato Junior G, Lewi DS.Neurotoxoplasmosis diagnosis for HIV-1 patients by real-time PCRof cerebrospinal fluid. Braz J Infect Dis2009; 13:1823.

58 Meira CS, Vidal JE, Costa-Silva TA, et al. Immunodiagnosis incerebrospinal fluid of cerebral toxoplasmosis and HIV-infectedpatients using Toxoplasma gondiiexcreted/secreted antigens.

Diagn Microbiol Infect Dis2011; 71:27985.


12/13


Review

59 Pierce MA, Johnson MD, Maciunas RJ, et al. Evaluatingcontrast-enhancing brain lesions in patients with AIDS by using

positron emission tomography. Ann Intern Med1995; 123:59498.60 Skiest DJ, Erdman W, Chang WE, et al. SPECT thallium-201

combined with Toxoplasma serology for the presumptive diagnosisof focal central nervous system mass lesions in patients with AIDS.J Infect2000; 40:27481.

61 Dedicoat M, Livesley N. Management of toxoplasmic encephalitis inHIV-infected adults (with an emphasis on resource-poor settings).Cochrane Database Syst Rev2006; 3:CD005420.

62 Mayor AM, Fernandez Santos DM, Dworkin MS, et al. Toxoplasmicencephalitis in an AIDS cohort at Puerto Rico before and afterhighly active antiretroviral therapy (HAART).Am J Trop Med Hyg2011; 84:83841.

63 Park BJ, Wannemuehler KA, Marston BJ, et al. Estimation of thecurrent global burden of cryptococcal meningitis among personsliving with HIV/AIDS. AIDS2009; 23:52530.

64 Brizendine KD, Pappas PG. Cryptococcal meningitis: currentapproaches to management in patients with and without AIDS.Curr Infect Dis Rep2010; 12:299305.

65 Jarvis JN, Lawn SD, Vogt M, et al. Screening for cryptococcalantigenemia in patients accessing an antiretroviral treatmentprogram in South Africa. Clin Infect Dis2009; 48:85662.

66 Bicanic T, Brouwer AE, Meintjes G, et al. Relationship ofcerebrospinal fluid pressure, fungal burden and outcome inpatients with cryptococcal meningitis undergoing serial lumbarpunctures. AIDS2009; 23:70106.

67 Garlipp CR, Rossi CL, Bottini PV. Cerebrospinal fluid profiles inacquired immunodeficiency syndrome with and withoutneurocryptococcosis. Rev Inst Med Trop Sao Paulo 1997; 39:32325.

68 Cohen DB, Zijlstra EE, Mukaka M, et al. Diagnosis of cryptococcaland tuberculous meningitis in a resource-limited African setting.Trop Med Int Health2010; 15:91017.

69 Jarvis JN, Percival A, Bauman S, et al. Evaluation of a novelpoint-of-care cryptococcal antigen test on serum, plasma, and urinefrom patients with HIV-associated cryptococcal meningitis.Clin Infect Dis 2011; 53:101923.

70 Chen YY, Lai CH. Nationwide population-based epidemiologicstudy of cryptococcal meningitis in taiwan. Neuroepidemiology2011;36:7984.

71 Wijewardana I, Jarvis JN, Meintjes G, et al. Large volume lumbarpunctures in cryptococcal meningitis clear cryptococcal antigen aswell as lowering pressure.J Infect2011; 63:48486.

72 Perfect JR, Dismukes WE, Dromer F, et al. Clinical practiceguidelines for the management of cryptococcal disease: 2010 updateby the infectious diseases society of america. Clin Infect Dis2010;50:291322.

73 Macsween KF, Bicanic T, Brouwer AE, et al. Lumbar drainage forcontrol of raised cerebrospinal fluid pressure in cryptococcalmeningitis: case report and review.J Infect2005; 51:e22124.

74 Hamill RJ, Sobel JD, El-Sadr W, et al. Comparison of 2 doses ofliposomal amphotericin B and conventional amphotericin Bdeoxycholate for treatment of AIDS-associated acute cryptococcalmeningitis: a randomized, double-blind clinical trial of effi cacy andsafety. Clin Infect Dis2010; 51:22532.

75 Sloan D, Dlamini S, Paul N, Dedicoat M. Treatment of acutecryptococcal meningitis in HIV infected adults, with an emphasison resource-limited settings. Cochrane Database Syst Rev2008;4:CD005647.

76 Parkes-Ratanshi R, Wakeham K, Levin J, et al. Primary prophylaxisof cryptococcal disease with fluconazole in HIV-positive Ugandanadults: a double-blind, randomised, placebo-controlled trial.Lancet Infect Dis2011; 11:93341.

77 Boulware DR, Meya DB, Bergemann TL, et al. Clinical features andserum biomarkers in HIV immune reconstitution inflammatorysyndrome after cryptococcal meningitis: a prospective cohort study.PLoS Med2010; 7:e1000384.

78 Pathai S, Gilbert C, Weiss HA, et al. Differing spectrum ofHIV-associated ophthalmic disease among patients startingantiretroviral therapy in India and South Africa. Trop Med Int Health2010; 16:35659.

79 Steininger C, Schmied B, Sarcletti M, et al. Cytomegalovirusgenotypes present in cerebrospinal fluid of HIV-infected patients.AIDS2005; 19:27378.

80 Fink KR, Thapa MM, Ishak GE, Pruthi S. Neuroimaging ofpediatric central nervous system cytomegalovirus infection.

Radiographics2010; 30:177996.81 Gaeta A, Verzaro S, Cristina LM, et al. Diagnosis of neurological

herpesvirus infections: real time PCR in cerebral spinal fluidanalysis. New Microbiol2009; 32:33340.

82 Granter SR, Doolittle MH, Renshaw AA. Predominance ofneutrophils in the cerebrospinal fluid of AIDS patients withcytomegalovirus radiculopathy. Am J Clin Pathol1996; 105:36466.

83 Pirskanen-Matell R, Grutzmeier S, Nennesmo I, et al. Impairmentof short-term memory and Korsakoff syndrome are common inAIDS patients with cytomegalovirus encephalitis. Eur J Neurol2009;16:4853.

84 Au Eong KG, Beatty S, Charles SJ. Cytomegalovirus retinitis inpatients with acquired immune deficiency syndrome.Postgrad Med J1999; 75:58590.

85 Biron KK. Antiviral drugs for cytomegalovirus diseases.Antiviral Res2006; 71:15463.

86 Razonable RR. Antiviral drugs for viruses other than human

immunodeficiency virus. Mayo Clin Proc2011; 86:100926.87 Holbrook JT, Colvin R, van Natta ML, et al. Evaluation of the UnitedStates public health service guidelines for discontinuation ofanticytomegalovirus therapy after immune recovery in patients withcytomegalovirus retinitis. Am J Ophthalmol2011; 152:62837.e1.

88 Posthuma CC, van der Beek MT, van der Blij-de Brouwer CS, et al.Mass spectrometry-based comparative sequencing to detectganciclovir resistance in the UL97 gene of human cytomegalovirus.J Clin Virol2011; 51:2530.

89 Mascarello M, Lanzafame M, Lattuada E, et al. Progressivemultifocal leukoencephalopathy in an HIV patient receivingsuccessful long-term HAART.J Neurovirol2011; 17:19699.

90 Di Giambenedetto S, Vago G, Pompucci A, et al. Fatal inflammatoryAIDS-associated PML with high CD4 counts on HAART: a newclinical entity? Neurology2004; 63:245253.

91 Padgett BL, Walker DL. Prevalence of antibodies in human seraagainst JC virus, an isolate from a case of progressive multifocalleukoencephalopathy.J Infect Dis1973; 127:46770.

92 Knowles WA, Pipkin P, Andrews N, et al. Population-based study ofantibody to the human polyomaviruses BKV and JCV and thesimian polyomavirus SV40.J Med Virol2003; 71:11523.

93 Egli A, Infanti L, Dumoulin A, et al. Prevalence of polyomavirus BKand JC infection and replication in 400 healthy blood donors.J Infect Dis2009; 199:83746.

94 Lednicky JA, Vilchez RA, Keitel WA, et al. Polyomavirus JCVexcretion and genotype analysis in HIV-infected patients receivinghighly active antiretroviral therapy. AIDS2003; 17:80107.

95 Berger JR. The basis for modeling progressive multifocalleukoencephalopathy pathogenesis. Curr Opin Neurol2011; 24:26267.

96 Houff SA, Major EO, Katz DA, et al. Involvement of JC virus-infected mononuclear cells from the bone marrow and spleen inthe pathogenesis of progressive multifocal leukoencephalopathy.N Engl J Med1988; 318:30105.

97 Delbue S, Branchetti E, Boldorini R, et al. Presence and expressionof JCV early gene large T antigen in the brains ofimmunocompromised and immunocompetent individuals.J Med Virol2008; 80:214752.

98 White FA 3rd, Ishaq M, Stoner GL, Frisque RJ. JC virus DNA ispresent in many human brain samples from patients withoutprogressive multifocal leukoencephalopathy.J Virol1992; 66:572634.

99 Sunyaev SR, Lugovskoy A, Simon K, Gorelik L. Adaptive mutations inthe JC virus protein capsid are associated with progressive multifocalleukoencephalopathy (PML). PLoS Genet2009; 5:e1000368.

100 Ciappi S, Azzi A, De Santis R, et al. Archetypal and rearrangedsequences of human polyomavirus JC transcription control regionin peripheral blood leukocytes and in cerebrospinal fluid.J Gen Virol1999; 80:101723.

101 Fedele CG, Ciardi MR, Delia S, et al. Identical rearranged forms ofJC polyomavirus transcriptional control region in plasma andcerebrospinal fluid of acquired immunodeficiency syndromepatients with progressive multifocal leukoencephalopathy.J Neurovirol2003; 9:55158.

102 Merabova N, Kaniowska D, Kaminski R, et al. JC virus agnoprotein

inhibits in vitrodifferentiation of oligodendrocytes and promotesapoptosis.J Virol2008; 82:155869.


13/13

Review

103 Sabath BF, Major EO. Traffi c of JC virus from sites of initialinfection to the brain: the path to progressive multifocal

leukoencephalopathy.J Infect Dis2002; 186(suppl 2):S18086.104 Koralnik IJ, Wuthrich C, Dang X, et al. JC virus granule cell

neuronopathy: a novel clinical syndrome distinct from progressivemultifocal leukoencephalopathy. Ann Neurol2005; 57:57680.

105 Wuthrich C, Dang X, Westmoreland S, et al. Fulminant JC virusencephalopathy with productive infection of cortical pyramidalneurons. Ann Neurol2009; 65:74248.

106 Tan CS, Koralnik IJ. Progressive multifocal leukoencephalopathyand other disorders caused by JC virus: clinical features andpathogenesis. Lancet Neurol2010; 9:42537.

107 Tada H, Rappaport J, Lashgari M, et al. Trans-activation of the JC viruslate promoter by the tat protein of type 1 human immunodeficiencyvirus in glial cells. Proc Natl Acad Sci USA1990; 87:347983.

108 Kumwenda JJ, Mateyu G, Kampondeni S, et al. Differentialdiagnosis of stroke in a setting of high HIV prevalence in Blantyre,Malawi. Stroke2005; 36:96064.

109 Tan K, Roda R, Ostrow L, et al. PML-IRIS in patients with HIV

infection: clinical manifestations and treatment with steroids.Neurology2009; 72:145864.

110 Post MJ, Yiannoutsos C, Simpson D, et al. Progressive multifocalleukoencephalopathy in AIDS: are there any MR findings useful topatient management and predictive of patient survival? AIDSClinical Trials Group, 243 Team. AJNR Am J Neuroradiol1999;20:1896906.

111 Tan IL, Brew BJ. Possible JCV granular cell neuronopathy in apatient with HIV infection. Neurology2009; 73:159899.

112 Weber T, Klapper PE, Cleator GM, et al. Polymerase chain reactionfor detection of JC virus DNA in cerebrospinal fluid: a qualitycontrol study. European Union Concerted Action on ViralMeningitis and Encephalitis.J Virol Methods1997; 69:23137.

113 Hammarin AL, Bogdanovic G, Svedhem V, et al. Analysis of PCR asa tool for detection of JC virus DNA in cerebrospinal fluid fordiagnosis of progressive multifocal leukoencephalopathy.J Clin Microbiol1996; 34:292932.

114 Gasnault J, Costagliola D, Hendel-Chavez H, et al. Improved

survival of HIV-1-infected patients with progressive multifocalleukoencephalopathy receiving early 5-drug combinationantiretroviral therapy. PLoS One2011; 6:e20967.

115 Marzocchetti A, Tompkins T, Clifford DB, et al. Determinants ofsurvival in progressive multifocal leukoencephalopathy. Neurology2009; 73:155158.

116 Antinori A, Cingolani A, Lorenzini P, et al. Clinical epidemiologyand survival of progressive multifocal leukoencephalopathy in theera of highly active antiretroviral therapy: data from the ItalianRegistry Investigative Neuro AIDS (IRINA).J Neurovirol2003;9(suppl 1):4753.

117 Koralnik IJ, Du Pasquier RA, Kuroda MJ, et al. Association ofprolonged survival in HLA-A2+progressive multifocalleukoencephalopathy patients with a CTL response specific for acommonly recognized JC virus epitope.J Immunol2002;168:499504.

118 Huang SS, Skolasky RL, Dal Pan GJ, et al. Survival prolongation inHIV-associated progressive multifocal leukoencephalopathy treated

with alpha-interferon: an observational study.J Neurovirol1998;4:32432.

119 De Luca A, Giancola ML, Cingolani A, et al. Clinical and virologicalmonitoring during treatment with intrathecal cytarabine in patientswith AIDS-associated progressive multifocal leukoencephalopathy.Clin Infect Dis1999; 28:62428.

120 Cettomai D, McArthur JC. Mirtazapine use in humanimmunodeficiency virus-infected patients with progressivemultifocal leukoencephalopathy. Arch Neurol2009; 66:25558.

121 De Luca A, Ammassari A, Pezzotti P, et al. Cidofovir in addition toantiretroviral treatment is not effective for AIDS-associated

progressive multifocal leukoencephalopathy: a multicohort analysis.AIDS2008; 22:175967.

122 Jiang ZG, Cohen J, Marshall LJ, Major EO. Hexadecyloxypropyl-cidofovir (CMX001) suppresses JC virus replication in human fetalbrain SVG cell cultures. Antimicrob Agents Chemother2010;54:472332.

123 Gerstner ER, Batchelor TT. Primary central nervous systemlymphoma. Arch Neurol2010; 67:29197.

124 Bower M, Powles T, Nelson M, et al. Highly active antiretroviraltherapy and human immunodeficiency virus-associated primarycerebral lymphoma.J Natl Cancer Inst2006; 98:108891.

125 Shiels MS, Pfeiffer RM, Hall HI, et al. Proportions of Kaposisarcoma, selected non-Hodgkin lymphomas, and cervical cancer inthe United States occurring in persons with AIDS, 19802007.JAMA2011; 305:145059.

126 Montesinos-Rongen M, Van Roost D, Schaller C, et al. Primarydiffuse large B-cell lymphomas of the central nervous system aretargeted by aberrant somatic hypermutation. Blood2004;103:186975.

127 Ballerini P, Gaidano G, Gong J, et al. Molecular pathogenesis ofHIV-associated lymphomas. AIDS Res Hum Retroviruses1992;8:73135.

128 Deckert M, Engert A, Bruck W, et al. Modern concepts in thebiology, diagnosis, differential diagnosis and treatment of primarycentral nervous system lymphoma. Leukemia2011; 25:1797807.

129 Ivers LC, Kim AY, Sax PE. Predictive value of polymerase chainreaction of cerebrospinal fluid for detection of Epstein-Barr virus toestablish the diagnosis of HIV-related primary central nervoussystem lymphoma. Clin Infect Dis2004; 38:162932.

130 Corcoran C, Rebe K, van der Plas H, et al. The predictive value ofcerebrospinal fluid Epstein-Barr viral load as a marker of primarycentral nervous system lymphoma in HIV-infected persons.J Clin Virol2008; 42:43336.

131 Kasamon YL, Ambinder RF. AIDS-related primary central nervoussystem lymphoma. Hematol Oncol Clin North Am2005;

19:66587, vivii.132 Levine AM, Sullivan-Halley J, Pike MC, et al. Human

immunodeficiency virus-related lymphoma. Prognostic factorspredictive of survival. Cancer1991; 68:246672.

133 Kreisl TN, Panageas KS, Elkin EB, et al. Treatment patterns andprognosis in patients with human immunodeficiency virus andprimary central system lymphoma. Leuk Lymphoma2008;49:171016.

134 Newell ME, Hoy JF, Cooper SG, et al. Human immunodeficiencyvirus-related primary central nervous system lymphoma: factorsinfluencing survival in 111 patients. Cancer2004; 100:262736.

135 Cinque P, Vago L, Marenzi R, et al. Herpes simplex virus infectionsof the central nervous system in human immunodeficiencyvirus-infected patients: clinical management by polymerase chainreaction assay of cerebrospinal fluid. Clin Infect Dis1998; 27:30309.

136 Grover D, Newsholme W, Brink N, et al. Herpes simplex virusinfection of the central nervous system in human immunodeficiencyvirus-type 1-infected patients. Int J STD AIDS2004; 15:597600.

137 Domingues RB, Tsanaclis AM, Pannuti CS, et al. Evaluation of therange of clinical presentations of herpes simplex encephalitis byusing polymerase chain reaction assay of cerebrospinal fluidsamples. Clin Infect Dis1997; 25:8691.

138 Bacon TH, Levin MJ, Leary JJ, et al. Herpes simplex virus resistanceto acyclovir and penciclovir after two decades of antiviral therapy.Clin Microbiol Rev 2003; 16:11428.

Documents

2. HIV-Associated Opportunistic Infections of the CNS