1319 neurologic complications of hiv-1 infection

Neurologic Complicationsof HIV-1 Infection and ItsTreatment in the Era ofAntiretroviral Therapy

Sarah M. Kranick, MD; Avindra Nath, MBBS, FAAN

ABSTRACTPurpose of Review: Neurologic complications of HIV infection are unfortunatelycommon, even in the era of effective antiretroviral treatment (ART). The consultingneurologist is often asked to distinguish among neurologic deterioration due to op-portunistic infection (OI), immune reconstitution, or the effect of the virus itself, andto comment on the role of immunomodulatory agents in patients with HIV infection.Additionally, as successful virologic control has extended the life span of patients withHIV infection, neurologists are called upon to manage long-term complications, suchas neurocognitive disorders and peripheral neuropathy.Recent Findings: Despite the use of ART, significant numbers of patients con-tinue to be affected by HIV-associated neurocognitive disorders, although withmilder forms compared to the pre-ART era. Regimens of ART have been rankedaccording to CNS penetration and are being studied with regard to neuro-psychological outcomes. Nucleoside analogs with the greatest potential for pe-ripheral neurotoxicity are no longer considered first-line agents for HIV treatment.Efavirenz, a non-nucleoside reverse transcriptase inhibitor, has the greatest fre-quency of neurologic side effects among newer ART regimens. The spectrum ofclinical manifestations of immune reconstitution inflammatory syndrome (IRIS) con-tinues to grow, including IRIS without underlying OI. A greater understanding ofpathophysiology and risk factors has shown that while HIV should be treated earlyto prevent severe immunocompromise, delayed initiation of ART may be helpfulwhile treating OIs.Summary: This article reviews the neurologic complications of HIV infection, or itstreatment, most commonly encountered by neurologists.

Continuum Lifelong Learning Neurol 2012;18(6):1319–1337.

INTRODUCTIONSince the advent of combination anti-retroviral therapy (ART) in 1996, theneurologic complications associated withHIV infection have shifted from thoseassociated with severe immunocom-promise, such as opportunistic infec-tions (OIs) of the CNS, to complicationsrelated to treatment. Some neurologiccomplications, such as HIV-associated

neurocognitive disorder (HAND), havecontinued to affect patients despitesatisfactory virologic control, althoughwith less severity. This review focuses onthese complications that have remainedprevalent despite the use of ART, as wellas the complications of ART attributableto both medication side effects andimmune reconstitution inflammatorysyndrome of the CNS (CNS-IRIS).

Address correspondence toSarah Kranick, NationalInstitutes of Health, Building10, 6-5700, Bethesda, MD20892, [email protected].

Relationship Disclosure:Dr Kranick reports nodisclosure. Dr Nath hasserved as an expert witnessin a brain infection case.

Unlabeled Use ofProducts/InvestigationalUse Disclosure:

Drs Kranick and Nath discussexperimental therapies forneuroprotection for HIV andperipheral neuropathy.

* 2012, American Academyof Neurology.

1319Continuum Lifelong Learning Neurol 2012;18(6):1319–1337 www.aan.com/continuum

Review Article

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

NEUROLOGIC COMPLICATIONSOF HIV-1 VIRAL INFECTIONHIV-AssociatedNeurocognitive DisorderHANDs encompass a range of cognitiveimpairment from asymptomatic cogni-tive decline to dementia in patients withHIV infection. HAND is the most prev-alent neurologic complication in this pop-ulation, and as patients continue to livelonger on ART, this disabling cognitivedisorder is likely to demand greater at-tention from the neurologic community.

TerminologyTerminology for cognitive change in HIVhas previously included HIV encepha-lopathy, minor cognitivemotor disorder,or AIDS dementia complex. Current no-sology rates the impairment using neu-ropsychological testing (if available) ormental status testing and assigns it to oneof three categories: asymptomatic neuro-cognitive impairment, mild neurocogni-tive disorder (MND), and HIV-associateddementia (HAD).1 This categorizationrecognizes the importance of using dem-ographically appropriate means for com-parison, as well as the possible contributionfrom confounding conditions such as de-pression, opportunistic CNS disease, orcoinfection with hepatitis C virus. As theclinical picture of HAND has changedover time, a standardized approach todiagnosis is necessary in order to un-derstand the burden of these disorders.The recognition of asymptomatic neuro-cognitive impairment requires detailedneuropsychological testing, which is notreadily available in some clinical settings.A screening test such as the Memorial-Sloan Kettering scale or 2007 consensusFrascati rating can be used to determinewhether patients need further neuropsy-chological testing.2

Clinical FeaturesAsymptomatic neurocognitive impairmentis characterized by poor performance in

two or more domains on neuropsycho-logical or mental status testing in pa-tients who do not report or otherwisedemonstrate cognitive decline. Thesepatients may go on to develop symp-tomatic impairment (MND or HAD), butthe time course of cognitive change inHIV is not predictable or linear in manycases. Even with consistent treatmentwith ART, cognitive performance mayfluctuate over time, making diagnosismore difficult; in some cases completerecovery occurs after initiation of ART.3

Patients with MND report or demon-strate mild functional decline not ex-plained by a confounding condition, andon neuropsychological or mental statustesting perform at least one SD belowan appropriate normative mean in atleast two cognitive domains. These pa-tients are likely to be able to continueworking, although at a reduced level ofproductivity or efficiency. Movement dis-orders, such as gait disturbance, tremor,and impairment of fine manual dexter-ity, may be present.4 HAD describes apattern of cognitive loss greater thantwo SDs below the mean in at least twocognitive domains (Case 4-1). With greatimpairment in daily function, these pa-tients are unlikely to be able to liveindependently.

The cognitive impairment classicallydescribed in HIV infection is a sub-cortical dementia, more similar to thecognitive deficits seen in Parkinson orHuntington disease than the corticaldementia of Alzheimer disease. Pref-erentially affecting the fronto-striato-thalamo-cortical circuits, patients withHIV infection have traditionally beenfound to have greater difficulty in ab-straction, rapid information processing,verbal fluency, decision making, andmaintaining working memory.5 Prospec-tive memory has been demonstrated tobe impaired in HIV; this type of memory,called ‘‘remembering to remember,’’is critical to intentional and planning

KEY POINTS

h HIV-associatedneurocognitive disordercomprises threeentities: asymptomaticneurocognitiveimpairment, mildneurocognitive disorder,and HIV-associateddementia.

h The cognitiveimpairment traditionallyassociated with HIVinfection is that of asubcortical dementia,with difficulties in thespeed of informationprocessing and verbalfluency, although thismay be changing inthe era of antiretroviraltherapy.

1320 www.aan.com/continuum December 2012

HIV and Antiretroviral Therapy


tasks such as remembering to take dailymedications.6 Memory domains relianton posterior neocortical and temporo-limbic systems, such as object namingand memory consolidation, were tradi-

tionally thought to be spared in HIV-infected individuals,5 although evidencesuggests that the cognitive domainsaffected may be changing in the eraof ART.7 The large National Institute

Case 4-1A 56-year-old man with known HIV infection of 2 years’ duration was brought to the hospital byhis brother, who reported that the patient was not acting like himself and was quieter than usual.The patient had been lost to follow-up after a prolonged hospitalization in which he was found tohave HIV/AIDS (CD4+ T-cell count of 6). His brother stated that they had recently moved to a newapartment and that the patient was unable to determine how to unpack the boxes or put awayhis belongings. He had little spontaneous speech, inattention, apathy, disorientation to time, and0/3 recall at 3 minutes. He had mild retropulsion, decreased arm swing, a broad-based gait, anddecreased sensation to temperature and vibration in a length-dependent, stocking-glove pattern,with positive sway on Romberg testing.

MRI of the brain showed a symmetric leukoencephalopathy and generalized cerebral atrophy(Figure 4-1). Lumbar puncture showed a white blood cell (WBC) count of 2/2L and normal proteinand glucose concentration; infectious studies for cryptococcal antigen, JC virus, Epstein-Barr virus,and cytomegalovirus were negative by PCR. Serum antibody testing was negative for hepatitis B and

C viruses; vitaminB12 level was650 pg/mL, andrapid plasma reaginwas nonreactive.The CD4+ T-cellcount was0 cells/2L, andthe HIV viral loadwas 286,270 RNAcopies/2L. HIVgenotypingindicated possibleresistance tosaquinavir andritonavir, likelyconsistent withpoor medicationadherence.

Comment.At the time ofhospitalizationfor personality

change, the patient met criteria for HAD as he had significant cognitive loss with multiple domainsaffected, particularly psychomotor speed and executive function, and was unable to liveindependently. While his ventricles were enlarged on MRI, he had no urinary symptoms and alarge volume lumbar puncture resulted in no improvement of his gait. After re-initiating treatmentwith ART and OI prophylaxis, he showed some improvement in his independent function on a visitmore than 6 months later, as evidenced by regular checks of his pillboxes to ensure that he wasfilling them properly and adhering to his regimen.

FIGURE 4-1 Fluid-attenuated inversion recovery sequence of MRI shows hydrocephalusex vacuo, diffuse leukoencephalopathy (A), and diffuse, generalized atrophy(B). This degree of leukoencephalopathy is not a cardinal imaging feature

of HIV encephalopathy but is commonly found in patients with advanced HIV infection and likelyrepresents axonal injury.



on Mental Health cohort CNS HIV Anti-Retroviral Therapy Effects Research(CHARTER) study used standardizedneuropsychological tests to evaluate 857patients from 1988 to 1995, and 937patients from 2000 to 2007. Between thepre-ART and ART eras, fewer patientsperformed poorly in verbal fluency (eg,number of animals named in timed set-ting), speed of information processing(digit vigilance time), and motor do-mains (grooved pegboard test), whilemore patients showed deficits in learn-ing and memory (story memory test)and executive functioning (Wisconsincard-sorting test). The domains of recall,working memory, and attention werestable across the two time periods.7 Thisshift in cognitive domains most affectedin patients with HIV infection has impor-tant implications for trials of adjunctivetherapy, such as antioxidants or neuro-protective agents, in determining whichcognitive tests to study. Whether thesedata reveal a greater degree of corticalimpairment, rather than subcortical pa-thology, in HAND in the era of ART willrequire neuroimaging and neuropatho-logic correlation.

EpidemiologyPrior to the introduction of combina-tion ART in 1996, dementia in patientswith HIV infection was described as aconsequence of profound immunosup-pression as reflected by the CD4 nadir.8

The number of HIV-infected patients withmoderate to severe dementia has beendramatically reduced by the use of ART,with one study describing a decreasein incidence from 7% in 1989 to 1% in2000.9 In patients with sustained viro-logic control, however, the continuedprevalence of the milder subtypes ofHAND has led to the question of whetherneurocognitive decline could be treat-ment resistant in some patients. In theCHARTER study, despite a significant de-crease in HAD from rates reported in

the pre-ART era, 45% of patients hadneuropsychological impairment diag-nosed either as asymptomatic neurocog-nitive impairment (33%) orMND(12%).10

Low CD4+ T-cell count nadir has con-tinued to be a risk factor for the devel-opment of HAND in the era of ART,suggesting that severe immunosuppres-sion may lead to irreversible brain pa-thology.7 Alternatively, the brain may actas a reservoir for HIV replication due tovariable blood-brain barrier penetrationby ART as well as viral sequestration inCNS macrophages.11

Despite the decline of the mostsevere HAND, HAD, cognitive disor-ders in HIV remain a significant sourceof morbidity for patients in the UnitedStates and elsewhere. While the de-gree of cognitive impairment may bemilder in patients with HAND in theART era, these ‘‘mild’’ or ‘‘minor’’ cog-nitive syndromes have nonethelessbeen associated with low antiretro-viral adherence and thus represent asignificant risk factor for decreasedsurvival.12

ManagementAs HAND has been shown to directlycorrelate with CD4 nadir, the primarygoal in management of HAND is to pre-vent HIV viral replication in the CNS. Inevaluating the efficacy of new ART agents,however, demonstrating the penetrationof any drug into brain parenchyma ischallenging. CSF drug levels are usedas a proxy of CNS penetration, but therelationship of CSF to brain levels isunknown. The difficulty of achieving sat-isfactory CSF levels is a well-recognizedproblem, as plasma drug levels are usu-ally much higher than those in the CSF,and protein pumps such as P-glycoproteinmay eliminate protease inhibitors fromthe brain.Macrophages, the primary targetof HIV in the brain, require much higherconcentrations of antiretrovirals for effec-tive control of viral replication than that

KEY POINTS

h The use of antiretroviraltherapy has decreasedthe prevalence of themost severe dementiaassociated with HIV,but the milder formsof HIV-associatedneurocognitive disorderhave remainedhighly prevalent.

h Low CD4+ T-cell nadircontinues to be asignificant risk factorfor the developmentof HIV-associatedneurocognitive disorderin the era ofantiretroviral therapy.

h The primary goal inmanagement ofHIV-associatedneurocognitive disorderis to prevent HIVreplication in the CNS.




required by T lymphocytes.13 Addition-ally, resistance mutations in the viruscan differ between the CSF and plasma,contributing to the potential reservoirof viral replication in the CNS.14

Previously, guidelines from the USDepartment of Health and Human Ser-vices (DHHS) regarding antiretroviral usein adults recommended initiation of ARTin patients with a history of an AIDS-defining illness or CD4+ T-cell countless than 350 cells/2L; while HIV-relatedencephalopathy is listed on the Centerfor Disease Control and Prevention guide-lines as an AIDS-defining illness, HANDhas not been specifically addressed. The2012 DHHS Guidelines for HIV Treat-ment now recommend ART for all HIV-infected patients regardless of CD4+T-cell counts.15 Given the difficulty in en-suring CNS penetration and that higherlevels are required to control replicationin brain macrophages, early initiation ofART in patients with neurocognitive ab-normalities is paramount.

Regimens associated with betterCNS penetration. A CNS penetration-effectiveness (CPE) index recently hasbeen proposed to guide the choice ofantiretrovirals in patients with HAND.Each antiretroviral drug is given a scoreof 1 to 4 (4 = much above averageCNS penetration, 3 = above average,2 = average, 1 = below average), and thesum of the individual agents’ scores in acombination regimen provides the CPEscore for that regimen. Individual ARTagents with scores of 4 include zidovu-dine, nevirapine, and indinavir/ritonavir.16

While ART regimens with higher CPEscores have been shown to correlatewith improved CSF viral suppression,in one study patients on higher CPE-scoring regimens also demonstratedpoorer neuropsychological performance,raising the possibility that some drugsmay be neurotoxic.17 Similar studies haveshown improved performance on neuro-psychological testing with better CNS

penetration of ARTs18 or no effect.19 Amulticenter, randomized, single blindtrial is currently recruiting participantsin order to compare ART regimens ofdifferent CNS penetration with the pri-mary outcome measurement of neuro-psychological testing after 16 weeks oftreatment.20

Other agents. Because of concernthat HAND may be resistant to effec-tive virologic control, a number of agentshave been tested in previous or ongoingtrials with regard to controlling inflamma-tion in the brain or providing neuropro-tection. Trials have included memantine,nimodipine, selegiline, minocycline,atorvastatin, lithium, valproic acid, andselective serotonin reuptake inhibitors(SSRIs), such as citalopram.21 Someagents have shown benefits on bio-markers of neurotoxicity, such as brainlevels of N-acetylaspartate in magneticresonance spectroscopy, or clinical ben-efit in smaller studies; but larger studieswith adequate power have not yieldedpositive results for these adjunctiveagents. Evidence from small studies,however, and pharmacologic mecha-nisms support the continued investiga-tion of psychiatric medications such asSSRIs and glycogen synthase kinase 3-"inhibitors, as well as antioxidants, in re-ducing inflammation or other toxic ef-fects of HIV replication in the brain.22

Other CNS SyndromesAssociated With HIV InfectionIn some patients with HIV infection, theviral infection presents with a multiplesclerosisYlike syndrome with either arelapsing-remitting or a progressive, fulmi-nant course, correlating with leukoen-cephalopathy on MRI.23 Pathology inthese patients is consistent with de-myelination but without identificationof JC virus or the pathologic changescharacteristic of progressive multifocalleukoencephalopathy (PML). A syn-drome of fulminant encephalopathy,

KEY POINT

h The CNSpenetration-effectivenessindex ranks eachantiretroviral agentcompared to ‘‘average’’CNS penetration.



frequently accompanied by renal fail-ure and seizures, has been identified inHIV-infected patients using IV drugs(Case 4-2).24 These patients invariablyhave T2-signal abnormalities in the bi-lateral basal ganglia and occasionally

also present with movement disorders.In HIV-infected children, basal gangliacalcification is a common finding amongpatients with abnormal imaging.25 Thesepatients almost uniformly have devel-opmental delay.

Case 4-2A 51-year-old woman with HIV infection of at least 3 years’ duration presented to the emergencydepartment with dizziness. In the weeks prior to admission she was frequently using cocaine andheroin, and both drugs were found in her urine. She had been noncompliant with ART, and herCD4+ T-cell count was 5 cells/2L with HIV viral load of 209,374 copies/mL. She had an encephalopathyand a shuffling gait with postural instability. CSF showed 0 WBCs but elevated protein concentration(76 mg/dL). She was found to be in acute renal failure with a serum creatinine level of 1.1 mg/dL.Bacterial, fungal, and mycobacterial cultures from the CSF were negative, as were toxoplasmaantibodies, Venereal Disease Research Laboratory testing, cryptococcal antigen, and PCRs forEpstein-Barr virus, JC virus, herpes simplex virus type 1 and type 2, cytomegalovirus, varicella-zoster virus(VZV), and arboviruses. MRI of the brain showed diffuse hyperintense T2/fluid-attenuated inversionrecovery (FLAIR) signal abnormalities in the bilateral basal ganglia and periventricular whitematter (Figure 4-2). Despite supportive care, her mental status deteriorated over the following6 months and she died.

Comment. A syndrome of acute encephalopathy and renal dysfunction, frequently accompanied byseizures and often progressing rapidly to death, has been described in HIV-infected drug abusers withpoor virologic control.24 CSF is typically acellular with elevated protein concentration. Even thoughthe basal gangliaare frequentlyaffected in HIVinfection and showatrophy in HANDand calcification inperinatally acquiredHIV infection, thissyndrome of fulminantencephalopathy inHIV-infected drugabusers shows verycharacteristic diffuse,bilateral basalganglia abnormalitieson T2/FLAIR imaging.Two patients whosurvived had receivedantiretroviral therapyduring admission,suggesting a possibleneuroprotective effectof ART in these cases.

FIGURE 4-2 Fluid-attenuated inversion recovery sequence of MRI shows diffuse bilateralhyperintense signal in the basal ganglia (A) and periventricular whitematter (B).




HIV-ASSOCIATED DISTALSYMMETRIC POLYNEUROPATHYHIV infection has been associated withnumerous syndromes in the peripheralnervous system, most commonly a dis-tal symmetric polyneuropathy. Peripheralneuropathy in patients with HIV infectionmay be related to neurotoxic ART (anti-retroviral toxic neuropathy [HIV-ATN])or to the viral infection itself (HIV dis-tal sensory polyneuropathy [HIV-DSP]).While the neurotoxicity of certain anti-retrovirals has led to declining use in fa-vor of other agents, many patients alivetoday have been on numerous thera-peutic regimens and thus may continueto be affected by irreversible peripheralneuropathy.

EpidemiologyIn the era preceding ART, HIV-DSPwas associated with profound immu-nosuppression. Peak HIV viral load andCD4+ T-cell nadir were found to berisk factors for the development ofHIV-DSP and also correlated with theseverity of symptoms.26 In studiescomparing rates of HIV-DSP in cohortsstudied before and after the introduc-tion of ART, the prevalence of HIV-DSP is quite similar: 55% of pre-ARTpatients affected in the Dana cohort27

and 53% of post-ART patients affectedin the Manhattan HIV Brain Bank.28 Inthe studies of HIV-DSP in patients inthe ART era, the degree of immuno-suppression no longer predicts thedevelopment or the severity of neuro-pathy.28 Similar to the theory for thecontinued prevalence of HAND despitebetter virologic control, the continuedprevalence of HIV-DSP may reflect lowlevels of viral replication or subsequentchronic inflammation that are below thelevel of detection and yet sufficient tocause neurotoxicity. Another theory forthe persistence of HIV-DSP is an im-mune reconstitution mechanism of dam-age to peripheral nerves. With longer life

spans, patients with HIV infection areat greater risk for other medication-induced (and illicit drugYinduced) neu-rotoxicity and other conditions, such asdiabetes, that lead to peripheral neu-ropathy. These additional risk fac-tors have been shown to contribute toHIV-DSP.28

Clinical Features andManagementHIV-DSP, a small fiber sensory neuro-pathy, typically manifests as painful par-esthesia or painless numbness in the feet.While symptoms may spread duringweeks to months, symptoms in HIV-DSP are typically confined to the lowerextremities below the knees, with lossof deep tendon reflexes at the ankles.Painful paresthesia is common in thissyndrome; in one study of 100 patientsscreened during 2 weeks in Australia,42% of patients had a distal sensory neu-ropathy, and of these, 93% reportedpainful symptoms.29 As a small fiber sen-sory neuropathy, pathologic changes maybe absent on EMG and nerve conduc-tion studies, thus requiring skin biopsyfor definitive diagnosis. In most cases,classic symptoms in the appropriate clin-ical context are sufficient to formulate adiagnosis and treatment plan.

Given that many patients with HIV-DSP would be otherwise healthy, withtheir HIV well controlled on ART, thesepainful symptoms have a significant im-pact on quality of life. Lamotrigine hasshown benefit in treating painful HIV-DSP in one randomized controlled trialin patients previously exposed to neu-rotoxic ART, although this benefit wasonly seen in a secondary outcome mea-sure using a visual analog scale.30 Otheragents traditionally used to treat neuro-pathic pain, such as amitriptyline, pre-gabalin, and gabapentin, have not shownefficacy in treating painful HIV-DSP inlarge randomized controlled trials.31

Mixed data exist for agents such as

KEY POINTS

h Distal sensorypolyneuropathy isthe most commonneurologic manifestationof HIV infection andremains highly prevalentdespite the use ofantiretroviral therapy.

h Neuropathic pain is acommon finding inHIV distal sensorypolyneuropathy. Agentstraditionally used totreat neuropathic painare frequently used,although none hasshown benefit inrandomized controlledtrials of painful HIVdistal sensorypolyneuropathy.



high concentration topical capsaicin.Injections of recombinant human nervegrowth factor have shown potential forbenefit in studies, although this agentis not yet approved by the US Food andDrug Administration. Smoked cannabishas been shown to effectively controlpain in HIV-DSP but cannot be recom-mended for routine therapy as the men-tal side effects are disturbing for some,preparations are not standardized, andlong-term risks of lung cancer are asso-ciated with inhalation of cannabinoids.31

NEUROLOGIC COMPLICATIONSOF TREATMENT FOR HIVWhile HIV has morphed from a rapidlyfatal infection into a chronic, treatabledisease because of the efficacy of ART,the range of possible complications re-lated to immune recovery, rather thanimmune deficiency, continues to grow.Additionally, side effects of chronictreatment with ART, such as peripheralneuropathy, are gaining in importanceas patients live longer and the cumu-lative toxicity over time is greater.

NEUROLOGIC SIDE EFFECTS OFANTIRETROVIRAL THERAPYNeurologists are unlikely to be pre-scribers of ART, yet the neurologicside effect profiles of these medica-tions are likely relevant to the consult-ing neurologist. Side effects of ARTtherapy are frequently in the differ-ential diagnosis of neurologic presen-tations of patients with HIV infection,such as the patient with acute mentalstatus change after initiating therapy,or patients on chronic ART with periph-eral neuropathy.

Nucleoside AnalogsThe potential of any nucleoside reversetranscriptase inhibitor to cause periph-eral neuropathy is directly related tothe mitochondrial toxicity of the spe-cific agent. Dideoxynucleoside agents

(‘‘d drugs’’) such as didanosine (ddI),stavudine (d4T), and zalcitabine (ddC)are therefore considered among themost peripheral neurotoxic agents andrarely used as first-line agents whenother options exist. Combination ther-apy with didanosine and stavudine isespecially avoided.32 HIV-ATN usuallyappears 2 to 3 months after initiationof treatment and may be more likely toaffect the hands earlier than would beexpected in HIV-DSP. The time coursemay be most helpful in distinguishingHIV-ATN from HIV-DSP, as these twoentities cannot be differentiated elec-trophysiologically.33 Stavudine has alsobeen associated with a neuromuscularsyndromeof acute, progressive ascendingweakness, similar to Guillain-Barre syn-drome. This HIV-associated neuromus-cular weakness syndrome (HANWS)usually includes lactic acidosis and hepa-tomegaly, invoking mitochondrial toxicityin its pathogenesis, although it has beenreported at a delay (up to 90 days) afterdiscontinuation of ART.34

Non-nucleoside ReverseTranscriptase InhibitorsOf the non-nucleoside reverse tran-scriptase inhibitors, efavirenz is mostcommonly associated with neurotoxicity,characterized by specific neuropsychiatricsymptoms that occur in more than 50%of patients on the medication. Typicallyoccurring at the onset of efavirenz use, thepatient may experience vivid dreams ornightmares, headaches, or psychiatric phe-nomena such as dissociative symptoms,depression, anxiety, or more rarely para-noia or psychosis. However, these symp-toms abate within 1month of treatment.35

Protease InhibitorsOne study of patients on protease in-hibitors between 1998 and 2004 sug-gested a link between these agents andthe development of peripheral neuro-pathy.36 The early data regarding the

KEY POINTS

h Peripheral neuropathyis most commonlyassociated with the‘‘d drugs’’ amongnucleoside analogs,such as didanosine,stavudine, andzalcitabine. Thesemedications are nowrarely used as first-lineagents when otheroptions exist.

h More than 50% ofpatients on efavirenzexperience neurologicside effects, usuallylimited to the firstmonth or weeks oftreatment.




possible peripheral neurotoxicity of pro-tease inhibitors are complicated by thefact that many of the patients reportedwere also likely exposed to nucleosidereverse transcriptase inhibitors, such asstavudine, and that the specific prote-ase inhibitors implicated are no longeramong the first-line ART agents. A muchlarger study of 1159 patients with HIVinfection found no increased risk forneuropathy associated with proteaseinhibitor use after adjusting for otherrisk factors.37

COADMINISTRATION OFANTIEPILEPTIC ANDANTIRETROVIRAL MEDICATIONSRecently, a joint panel of the AmericanAcademy of Neurology and the Inter-national League Against Epilepsy issuedguidelines for treatment of seizures inpatients with HIV infection.38 Given thehigh comorbidity of these conditions aswell as the greater availability world-wide of antiepileptics that are enzymeinducers, these guidelines help neurol-ogists in making appropriate adjustmentsto maximize therapeutic efficacy of anti-epileptic drugYART regimens.

IMMUNE RECONSTITUTIONINFLAMMATORY SYNDROMEIRIS is the phenomenon of clinical de-terioration despite immune recoveryfrom an immunodeficient state. Sincethe first report of acute Mycobacteriumavium-intracellulare infection in a pa-tient with HIV-1 infection treated withzidovudine in 1992,39 there has beenongoing research regarding the patho-physiology of the detrimental immuneresponse in IRIS as well as the range ofassociated infections and clinical mani-festations. While CNS-IRIS in HIV is mostfrequently linked to Cryptococcus or PML,other pathogens are continuing to bereported in association with CNS-IRISafter ART administration. Clinical dete-rioration despite effective virologic con-

trol on ART can also occur without anidentifiable pathogen, such as in theacute syndrome of fulminant HIV en-cephalitis after ART. While IRIS canaffect any organ system, this review fo-cuses on IRIS in the CNS in the contextof HIV infection.

TerminologyWhen IRIS is related to an OI, it istypically categorized in the literature as‘‘unmasked’’ IRIS versus ‘‘paradoxic’’IRIS. In unmasked IRIS, the immune-suppressed individual is already knownto have an OI but is unable to mount animmune response against it. Once ARTis initiated, the recovering immune sys-tem produces a robust immune responseagainst the pathogen, but the inflamma-tion thus produced is clinically detri-mental to the patient. In paradoxic IRIS,the patient has been treated for an OIbut suffers a clinical relapse once ART isinitiated. These terms can be confusingsince all IRIS is paradoxic in the sensethat immune recovery, not immune de-ficiency, produces the characteristic clin-ical worsening in these patients; similarly,whether an infection has been treatedor untreated, it is revealed (or unmasked)by the process of immune reconstitution.We prefer terms that describe the tempo-ral course of the OI with regard to IRIS:‘‘simultaneous’’ IRIS when the OI andIRIS clinically manifest at the same timein a patient, and ‘‘delayed’’ IRIS whenIRIS follows initiation of ART in a patientknown to be infected with an oppor-tunistic pathogen.

Epidemiology and Risk FactorsWhile IRIS in any organ system maycomplicate ART initiation in as many asone-third of patients with HIV infec-tion,40 CNS-IRIS is much rarer, occur-ring in only 1% of patients after startingART.41 The major risk factors identi-fied for the development of CNS-IRISare the degree of immunosuppression

KEY POINTS

h Guidelines are availablefor using the combinationof antiretroviral andantiepileptic medications;drug levels may besignificantly altereddepending on whichmedications are usedtogether.

h Immune reconstitutioninflammatory syndromeis characterizedby clinical deteriorationdespite immunerecovery from deficiency.



at the time of ART initiation as re-flected by CD4+ T-cell nadir, the pres-ence of any underlying OI, and therate of immune recovery as demon-strated by precipitous decline in viralload after starting ART.42 In one study,the incidence of CNS-IRIS rose from0.9% of all HIV-infected patients start-ing ART to 1.5% in those with CD4+nadir less than 200 cells/2L.41 Since pa-tients in resource-limited settings mayonly receive ART once they presentwith an AIDS-defining illness, it maybe that CNS-IRIS rates are actually higherworldwide. Genetic factors are likely alsorelevant as some patients with severeimmunosuppression and rapid immunerecovery do not develop IRIS. Expres-sion of proinflammatory cytokines suchas interleukin (IL)-6, as well as certaincytokine polymorphisms such as tumornecrosis factor !-308*2, may be greaterin the subpopulation of HIV-infectedpatients who develop IRIS.43,44 Addi-tionally, an anti-inflammatory cytokinepolymorphism in the IL-12B gene hasbeen negatively associated with non-CNS herpes-associated IRIS.44

PathophysiologyGiven the range of IRIS cases reportedin the literature, the picture that hasemerged is not of a unifying IRIS phys-iology but rather a myriad of diverseclinical, radiologic, and immunologic syn-dromes that likely encompass multipledisease mechanisms. What unifies thesecases is the trafficking of activated lym-phocytes into the brain. It is unclear whysome patients aremore likely than othersto develop IRIS during immune recon-stitution. Several studies of immunologicprofiling have shown that IRIS patientshad greater numbers of activated CD4+T cells of the effector memory subtype45

with increased expressionof interferon-+46

andprogrammed-deathmolecule 1 (PD-1)prior to ART initiation, suggesting thatthey are primed for immune activation.47

While PD-1 expression has been linkedto failure of immune restoration in ARTand T-cell exhaustion,47 its presence inpatients with IRIS may imply a disor-dered process of immune restoration.The abnormal presence of activatedlymphocytes in the brain may also occuron a chronic basis in HAND that devel-ops despite adequate virologic control.10

After initiating ART in the patientwith HIV infection, immune reconstitu-tion occurs in a biphasic manner. First,memory T cells increase in concert withthymic production of naıve T cells. Sec-ondary lymphoid organs that may havebeen damaged because of HIV-mediatedinflammation begin to recover and con-tribute to the rising numbers of CD4+T cells by releasing these, usually within3 to 6 months of ART initiation.48 CD8+T cells may increase more rapidly andthus may be a better indicator of earlyrisk for IRIS than CD4+ T-cell countsalone; CD8+ T-cell levels tend to returnto baseline after 4 months of ART, whileCD4+ T cells undergo a more gradualand longer-lasting increase.49,50 Patientswith IRIS have higher numbers of ac-tivated effector memory CD4+ T cellsand lower numbers of naıve CD4+ Tcells and central memory CD8+ T cells.45

The long-term outcome of patientswho recover from IRIS is unknown, butautopsy studies show that the inflam-matory infiltrates associated with op-portunistic infections consist of largenumbers of HIV-infected cells. Hencethe possibility exists that they may es-tablish a reservoir in the brain with long-term sequelae.

Opportunistic InfectionThe spectrum of viral, bacterial, fun-gal, and parasitic OIs, well describedfor their catastrophic consequencesfor patients with HIV infection priorto the advent of ART, have now beenassociated with IRIS. When an under-lying infection is present, initiating

KEY POINT

h Risk factors for thedevelopment of CNSimmune reconstitutioninflammatory syndromeare a low CD4+ T-cellnadir, the presenceof an underlyingopportunistic infection,and the rapid rate ofdecline in the HIVviral load.




ART in a patient with HIV infectioncauses a rise in the circulating T-cell pop-ulation, which may lead to an enhancedor severe inflammatory response to theinfection. The clinical manifestations ofCNS-IRIS, as well as in the latency ofsymptom onset after ART, vary signifi-cantly depending on the OI precipitat-ing the immune response. This suggeststhat while all CNS-IRIS may share gen-eral phenomena, such as blood-brainbarrier breakdown and leukocyte infil-tration, pathogen-specific disease mech-anisms may be responsible for IRIS seenin the context of each OI.

Viral PathogensJC Virus. Spread of the JC virus, a ubiq-uitous polyoma virus, to the glia in im-munosuppressed patients may lead todevelopment of PML, a frequently dev-astating disease characterized by pro-found destruction of the white matterand rapidly progressive neurologic de-cline corresponding to the brain lesions.As no antiviral is available for JC virus,PML is only treatable in the context ofHIV, since ART may enable the immunesystem to recover from the viral infec-tion; however, these patients are sub-sequently at risk for PML-IRIS. Ofapproximately 5% of HIV-infected pa-tients who develop PML, 19% may de-velop PML-IRIS after beginning treatmentwith ART.51 The development of IRIS inthe HIV-infected patient with PML can attimes present a diagnostic challenge.While some patients develop contrastenhancement of the white matter le-sions on MRI, in others the enhance-ment may be subtle. The uptake ofgadolinium in these lesions is associatedwith predominantly CD8+ T cells infiltrat-ing the perivascular spaces and in somecases the parenchyma, macrophages,and CD4+ T cells.52 Additionally, sig-nificant neurologic deficits, such as apha-sia and apraxia, associated with PML inthe HIV-infected patient may make de-

tection of any further clinical deteriora-tion after initiation of ART quite difficult.PML-IRIS typically develops within 4 to8 weeks after the initiation of ART buthas been reported to occur as late as 2years.53 The mortality rate may be 42%or higher in PML-IRIS,54 with a worseprognosis associated with HIV-infectedpatients diagnosed first with PML andsubsequently developing PML-IRIS afterinitiation of ART.55 Clinical response tosteroids confirms the diagnosis of PML-IRIS, as steroids will have no effect onPML but will help control the inflamma-tory response and frequently producesome clinical improvement in PML-IRIS.55

Herpes viruses. Multiple viruses fromthe Herpesviridae family have been de-scribed as the underlying OI in CNS-IRIS.VZV is the causative agent of varicella(chicken pox) and herpes zoster (shingles)with a prevalence above 90% in adults.56

A ubiquitous latent virus, VZV remainsdormant in the dorsal root ganglia afterthe primary infection characterized byfever and vesicular rash. Clinical syn-dromes resulting from reactivation ofVZV in the immunocompetent patientmay range from painful dermatomal rash(zoster) to transverse myelitis, but in theimmunocompromised patient also in-clude encephalitis and vasculitis, whichmay lead to cerebral infarcts. VZV-IRIShas been described as presenting 4 to10 months after the initiation of ARTand manifesting clinically as encepha-litis and vasculitis leading to cerebralinfarcts.57 In one report, the VZV-IRISwas not controlled by acyclovir alonebut required concurrent acyclovir andcorticosteroids.57 Cytomegalovirus hasalso been associated with IRIS in pa-tients with HIV infection, although ittypically manifests as retinitis and onlyrarely as CNS-IRIS. These few cases ofCNS-IRIS presented as ventriculitis andpolyradiculopathy and responded to treat-ment with ganciclovir with or withoutfoscarnet.58 One case of Epstein-Barr

KEY POINTS

h Progressive multifocalleukoencephalopathyCNS immunereconstitutioninflammatory syndromefrequently presentswith enhancement ofthe white matterlesions on MRI, andimmunomodulatorytherapy may berequired.

h Varicella-zoster virusCNS immunereconstitutioninflammatory syndromemay present withencephalitis orcerebral vasculitisleading to infarcts.



virus CNS-IRIS is reported with themanifestation of vision loss 6 weeksafter ART, with corresponding T2/FLAIRhyperintensities in the optic chiasm andEpstein-Barr virus detected in the CSF byPCR, without recovery of vision. Herpessimplex virus is suspected to have causedtemporal lobe encephalitis in one reportof CNS-IRIS that responded to acyclovir,although the pathogen could not befound in the CSF by PCR.59

Fungal PathogensCryptococcus neoformans is the mostcommon fungal infection associated withCNS-IRIS.60 Cryptococcal infections maybe associated with 10% to 30% of allCNS-IRIS,53 most often clinically manifestas aseptic recurrence of prior meningitisand rarely as intracranial cryptococcoma.

Several factors may make the diag-nosis of cryptococcal CNS-IRIS chal-lenging. Cryptococcal CNS-IRIS has alonger range of possible latency ofonset than that associated with PML ortuberculosis. While typically present-ing 3 to 20 months after the initiationof ART, it has been reported as soonas 14 days and as long as 2 years afterstarting ART.61 The clinical manifesta-tions of cryptococcal CNS-IRIS can bedifficult to detect as the acute devel-opment of aseptic meningitis may bemistaken for postinfectious hydrocepha-lus; both are characterized by headache,nausea, and vomiting. Additionally, be-cause tests for the cryptococcal antigenremain positive for several months afteradequate treatment at slowly decreas-ing dilutions, the possibility of crypto-coccal CNS-IRIS should be consideredif sterile inflammation of the CSF ispresent and no viable yeast is found inculture. Neuroimaging shows new men-ingeal or choroid plexus enhancementor perivascular enhancement in the sulci,indicating a cellular immune response tothe underlying infection.62 HIV-infectedpatients with cryptococcal CNS-IRIS com-

pared to non-IRIS cryptococcal men-ingitis most often have higher openingpressures, CSF white cell counts, andglucose concentrations.40

One case has been reported of Can-dida sp meningitis with acute deterio-ration after initiating treatment withART. This patient, who presented withsubacute meningitis during a period ofnoncompliance with ART, rapidly suc-cumbed despite ART initiation andevidence of improved immune func-tion. Postmortem examination revealedbasilar Candida sp meningitis as well asCD8+ T-cell perivascular infiltrates,particularly in the brainstem.63

Bacterial PathogensAntituberculous therapy has been rec-ognized for its potential to cause clinicaldeterioration, long before the era of ART.Mycobacterial CNS-IRIS typically presents5 to 10 months after ART initiation, mostcommonly with Mycobacterium tuber-culosis manifesting tuberculosis men-ingitis and rarely as tuberculoma.64,65

Because mycobacterial CNS-IRIS occursmuch later than mycobacterial IRIS inother organ systems, and also becauseof the difficulty of establishing a diagnosisin tuberculous meningitis, the prevalenceof mycobacterial CNS-IRIS is unknown.While cases of clinical deterioration afterART initiation due to expansion of in-tracranial tuberculomas have been de-scribed,64 fewer reports are availableregarding CNS-IRIS with recrudescenceor development of tuberculous menin-gitis. According to one study, as manyas 21% of patients in South Africa havedeveloped Mycobacterium-relatedCNS-IRIS.60 Given that one-third of theworld’s population has been infectedwith tuberculosis, a greater evidence ofmycobacterial CNS-IRIS may emerge asART increases in availability worldwide.If a patient presents with clinical deteri-oration and meningeal enhancement onMRI or communicating hydrocephalus

KEY POINTS

h Cryptococcus is themost common fungalinfection in immunereconstitutioninflammatory syndrome.A longer range oflatency, the clinicalmanifestation ofmeningitis only, and adelay in the normalizationof the cryptococcalantigen after treatmentmay contribute todifficulty in diagnosisand treatment.

h Clinical deteriorationis common with theinitiation of treatmentfor tuberculosis butcan occur much laterin the CNS than inother organ systems.




in the context of ART initiation withadequate treatment for mycobacteria,IRIS should be strongly considered inthe differential diagnosis.

Parasitic PathogensParasitic infections have not been foundto contribute significantly to CNS-IRISin patients with HIV infection. Severalcases of toxoplasmosis-associated CNS-IRIS have been described, manifestas encephalitis66; in one case elevatedCD8+ T-cell infiltrates were found in theCNS tissue.67 As parasitic infections havegreatest morbidity in resource-limitedsettings, recognition may be challeng-ing without access to neuroimaging;similarly, prevalence may be increas-ing with greater availability of ART.

Management of OpportunisticInfection–Related CNSImmune ReconstitutionInflammatory SyndromeChallenges in management. Since thevarious clinical scenarios of HIV CNS-IRIS are characterized by different OIs,no single treatment modality can re-verse the detrimental immune response.In the case of CNS-IRIS following ARTinitiation in patients with HIV infec-tion, the best therapy is prevention.The greatest predictor of whether anHIV-infected patient will develop IRIS isthe CD4 nadir68; therefore, if HIV isdetected and treatment initiated earlyin the disease course, CNS-IRIS is lesslikely to develop. While no immune-specific therapy exists, discovering asmuch as possible regarding the pres-ence of opportunistic organisms in theimmunocompromised patient may helpguide treatment decisions. In HIV-infectedpatients with known OIs, delaying ARTin certain circumstances while the in-fection is treated has shown benefit,although this has not been reproducedconsistently and may be etiology spe-cific. In patients in whom treatment

with ART cannot be delayed who arereceiving treatment for OIs, steroidsmay also be used, although little evi-dence exists to guide physicians in thisrecommendation. Steroids are mostoften used in cases of profound immu-nodeficiency, such as PML, in which noeffective antimicrobial treatment is avail-able for the infection(s) present andincreasing inflammation related to IRISpresents the risk of herniation or vascularcompromise. While no large trials havebeen done on which to base these treat-ment decisions, reports have recentlybeen published regarding which subpo-pulations are more or less likely to ben-efit from high-dose steroids in IRIS.69

In the absence of randomized treat-ment trials, management of patientswith HIV infection who develop CNS-IRIS after initiating ART must be guidedby the nature of the underlying infec-tion, the acuity of the clinical presenta-tion, and the tolerability of corticosteroids.

Antiretroviral therapy. Because a lowCD4+ T-cell nadir remains the mostconsistently identified predictor of thedevelopment of CNS-IRIS, early diagno-sis and early intervention with ART con-tinue to be paramount in importance.In one study in which patients werestarted on ART when CD4+ T-cell countswere above 400, the incidence of IRISwas 8%, significantly lower than whathas been traditionally reported in theliterature.70 ART may be delayed, how-ever, when an HIV-infected patient hasa known underlying OI for which ef-fective antimicrobial treatment is avail-able. The optimal timing of delayed ARTinitiation is unclear. In one study of cryp-tococcal meningitis, delaying ART for upto 1 month while antifungal therapy wasgiven improved outcomes.71 In a studyof severely immunocompromised pa-tients in Cambodia (CD4+ T-cell countless than 200 with a mean of 25 cells/2L),delaying ART for only 2 weeks instead of8 weeks during antituberculosis therapy

KEY POINT

h Early diagnosis in the HIVdisease course and earlyintervention withantiretroviral therapyare the best waysto prevent CNSimmune reconstitutioninflammatory syndrome.There is a rationale fordelaying antiretroviraltherapy in patients withan underlyingopportunistic infectionwhile the infection istreated.



reduced mortality by 34%.72 Thus de-laying ART is recommended over dis-continuing ART once it has beeninitiated, as stopping therapy increasesthe risk for drug resistance and diseaseprogression.

Immunomodulatory therapy. In HIV-infected patients with CNS-IRIS due toOI in whom massive inflammation hasresulted in impending herniation, high-dose corticosteroids are required.73 Weprefer 1 g/d of methylprednisolone IV,or an equivalent dose of any other cor-ticosteroid, for 5 days with a gradualtaper over 4 to 6 weeks. While reason-able concern exists about the complica-tions of further immunosuppression inthese patients, it is our experience thatstopping steroids after 3 to 5 days willoften lead to recurrence of IRIS symp-toms. This may be explained by re-bound of cytotoxic T cells in responseto the underlying OI once steroids havebeen removed. When effective antimi-crobial therapy exists for the underlyingOI, steroids may not be necessary forsuch a prolonged duration since the in-fection may respond to treatment. Incases such as CNS-IRIS due to PML,steroids may be required for a prolongedtime until memory T cells recover andcan mount a response to JC virus, usuallywithin 4 to 6 weeks.55 Such prolongeduse of steroids requires prophylaxis forpneumocystis pneumonia and fungalinfections, and for tuberculosis in ende-mic regions.

In HIV-infected patients with CNS-IRIS due to OI who are symptomaticbut without impending brain hernia-tion, the use of corticosteroids is farmore controversial. While modulatingthe immune response to an underlyinginfection may seem at odds with thegoal of clearing the infection, evidencesuggests that the immune response inthese patients is dysfunctional or in ex-cess of what is required to control theinfection. Historically, IRIS-related syn-

dromes such as the leprosy reactionhave been treated with adjunctive cor-ticosteroid therapy; the efficacy of thistreatment has been demonstrated inclinical trials. Without careful prospec-tive trials in CNS-IRIS, physicians mustuse their own clinical judgment in thecontext of each symptom and OI. InCryptococcus-related CNS-IRIS, ste-roids may be helpful in concert withCSF drainage when inflammation leadsto obstruction of CSF pathways andintracranial hypertension.74 In HIV-infected patients with CNS-IRIS asdemonstrated by enhancement onMRI without clinical symptoms, therisk to benefit ratio is unlikely to favorsteroid treatment.

HIV CNS–ImmuneReconstitution InflammatorySyndrome WithoutOpportunistic InfectionDuring immune reconstitution, patientswith HIV infection are at risk for otherdisease manifestations of IRIS unrelatedto underlying infection. These noninfec-tious IRIS syndromes range from fulmi-nant encephalitis to chronic syndromessuch as autoimmune disease in HIV.

Fulminant HIV encephalitis withimmune reconstitution inflammatorysyndrome. Initiation of ART has re-sulted in severe, progressive encepha-litis in some patients with HIV infection.Series of patients have been describedwith either worsening of preexisting en-cephalitis or dementia, or the new devel-opment of such after initiation of ART.74

The clinical deterioration in mental sta-tus in these patients may be accom-panied by leukoencephalopathy on MRI,with involvement of the uncinate fibers.The histopathologic correlate shows amass inflammatory infiltrate made upof predominantly CD8+ T cells in peri-vascular regions as well as the paren-chyma, in the absence of any detectablepathogen. CD8+ T cells have been

KEY POINTS

h High-dose corticosteroidsare required in patientswith immunereconstitutioninflammatory syndromein whom brain herniationis impending but mayalso be helpful in less diresituations when patientsare symptomatic fromCNS immunereconstitutioninflammatory syndrome.

h CNS immunereconstitutioninflammatory syndromemay exist without anidentifiable underlyingopportunistic infection,in which case theCD8+ T cells may beattacking viral reservoirswithin the brain, or thebrain itself. Thesesyndromes are usuallymanifest as fulminantencephalitis with whitematter changes on MRI.




found immediately adjacent to neu-rons in HIV encephalitis, implicatingcytotoxic T cells in neuronal toxicity.75

Alternatively, an HIV-infected patientwith encephalitis after ART has beendescribed with a fulminant focal area

Case 4-3A 59-year-old man with newly diagnosed HIV infection had a CD4+ T-cell count of 37 cells/2L and an HIVviral load of 95,710 copies/mL. Six days after initiation of ART, he developed an acute onset of profoundencephalopathy. He had a transcortical sensory aphasia, prolonged speech latency, apraxia, and diffuse,small amplitude myoclonus. MRI of the brain showed the new development of symmetric areas ofabnormal T2/FLAIR signal in the medial temporal lobes, extending into the subinsular white matteron the left, with subtle enhancement present in the left subinsula, increased vascular markings in theright basal ganglia, and diffuse dural enhancement from a lumbar puncture (Figure 4-3). CSF showed3 WBCs/2L, normal glucose concentration, and protein concentration elevated at 80 mg/dL. The CSFtest results were negative on flow cytometry and cytology, as well as for cultures for bacteria, fungi, andmycobacteria; PCRs for cytomegalovirus, Epstein-Barr virus, herpes simplex virus, human herpesvirus 6and 7, JC virus, and VZV were also negative. His serum showed slightly elevated ammonia at 53, normalthyroid function test, and B12 level of 993. His HIV viral load had fallen to 2858 copies/mL on the day of hismental status change. Over several days his mental status worsened and he was barely responsive tostimuli. High-dose corticosteroids (dexamethasone 10 mg twice daily) were started, with improvementin his speech and apraxia in the first 24 to 36 hours. Within 6 days he was close to his prior baseline andseveral weeks later was able to be discharged to home on a gradual taper of corticosteroids.

Comment. While no opportunistic organism could be identified as the etiology of his mental statuschange, this patient was at risk for IRIS given his low CD4+ T-cell count at the time of initiation ofART, as well as the rapid decrease in his HIV viral load after starting ART. The enhancement on MRI,albeit subtle, also provided a clue to the diagnosis. In a patient with clinical deterioration in thecontext of improved virologic control, the diagnosis of IRIS, and therefore treatment with adequatedoses of immunomodulatory agents, must be considered.

FIGURE 4-3 Fluid-attenuated inversion recovery (A, B) and T1 postgadolinium (C) sequences of MRI show hyperintensesignal in the bilateral mesial temporal cortex (A) and the periventricular white matter of the left subinsularcortex (arrow, B). The signal abnormality in the left subinsular cortex is found to have very subtle

enhancement on the postgadolinium image. Increased vascular markings are also seen in the right basal ganglia(arrow, C). While the enhancement is subtle, early intervention with corticosteroids reversed the encephalopathy inthis case, providing further evidence that early inflammation was developing.



of demyelination or tumefactive inflam-mation, similar to the Marburg variant ofmultiple sclerosis.76 The focal nature ofthis patient’s presentation invokes thequestion of whether, in the absenceof any pathogen, T cells are targetedagainst the viral reservoirs in the brainor the brain itself, and whether the na-ture of the encephalitis, focal or wide-spread, might imply different diseasemechanisms. Corticosteroids havebeen used in HIV-infected patients withnoninfectious encephalitis after ARTinitiation, sometimes with dramatic im-provement in clinical status (Case 4-3).74

REFERENCES1. Antinori A, Arendt G, Becker JT, et al.

Updated research nosology forHIV-associated neurocognitive disorders.Neurology 2007;69(18):1789Y1799.

2. Gandhi NS, Moxley RT, Creighton J, et al.Comparison of scales to evaluate theprogression of HIV-associatedneurocognitive disorder.HIV Ther 2010;4(3):371Y379.

3. Ellis R, Langford D, Masliah E. HIV andantiretroviral therapy in the brain:neuronal injury and repair.Nat Rev Neurosci 2007;8(1):33Y44.

4. McArthur JC, Brew BJ, Nath A. Neurologicalcomplications of HIV infection. LancetNeurol 2005;4(9):543Y555.

5. Reger M, Welsh R, Razani J, et al. Ameta-analysis of the neuropsychologicalsequelae of HIV infection. J Int NeuropsycholSoc 2002;8(3):410Y424.

6. Carey CL, Woods SP, Rippeth JD, et al.Prospective memory in HIV-1 infection.J Clin Exp Neuropsychol 2006;28(4):536Y548.

7. Heaton RK, Franklin DR, Ellis RJ, et al.HIV-associated neurocognitive disordersbefore and during the era of combinationantiretroviral therapy: differences in rates,nature, and predictors. J Neurovirol2011;17(1):3Y16.

8. Childs EA, Lyles RH, Selnes OA, et al.Plasma viral load and CD4 lymphocytespredict HIV-associated dementia andsensory neuropathy. Neurology1999;52(3):607Y613.

9. McArthur JC. HIV dementia: an evolvingdisease. J Neuroimmunol 2004;157(1Y2):3Y10.

10. Heaton RK, Clifford DB, Franklin DR Jr, et al.HIV-associated neurocognitive disorderspersist in the era of potent antiretroviraltherapy: CHARTER Study. Neurology2010;75(23):2087Y2096.

11. Finzi D, Hermankova M, Pierson T, et al.Identification of a reservoir for HIV-1 inpatients on highly active antiretroviraltherapy. Science 1997;278(5341):1295Y1300.

12. Hinkin CH, Castellon SA, Durvasula RS, et al.Medication adherence among HIV+ adults:effects of cognitive dysfunction and regimencomplexity. Neurology 2002;59(12):1944Y1950.

13. Aquaro S, Calio R, Balzarini J, et al.Macrophages and HIV infection:therapeutical approaches toward thisstrategic virus reservoir. Antiviral Res2002;55(2):209Y225.

14. Cinque P, Presi S, Bestetti A, et al. Effectof genotypic resistance on the virologicalresponse to highly active antiretroviraltherapy in cerebrospinal fluid. AIDS ResHum Retroviruses 2001;17(5):377Y383.

15. Panel on Antiretroviral Guidelines forAdults and Adolescents. Guidelines for theuse of antiretroviral agents in HIV-1-infectedadults and adolescents. Department ofHealth and Human Services. Availableat: http://aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf. AccessedMay 29, 2012.

16. Letendre SL, Ellis RJ, Ances BM,McCutchan JA. Neurologic complicationsof HIV disease and their treatment.Top HIV Med 2010;18(2):45Y55.

17. Marra CM, Zhao Y, Clifford DB, et al.Impact of combination antiretroviraltherapy on cerebrospinal fluid HIV RNAand neurocognitive performance.AIDS 2009;23(11):1359Y1366.

18. Smurzynski M, Wu K, Letendre S, et al.Effects of central nervous systemantiretroviral penetration on cognitivefunctioning in the ALLRT cohort. AIDS2011;25(3):357Y365.

19. Giancola ML, Lorenzini P, Balestra P, et al.Neuroactive antiretroviral drugs do notinfluence neurocognitive performancein less advanced HIV-infected patientsresponding to highly active antiretroviraltherapy. J Acquir Immune Defic Syndr2006;41(3):332Y337.

20. U.S. National Institutes of Health.Clinical trial of CNS-targeted HAART (CIT2).clinicaltrials.gov/ct2/show/NCT00624195?term=NCT00624195&rank=1. UpdatedOctober 6, 2010. Accessed February 22, 2012.




http://aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf

http://aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf

21. Tan IL, McArthur JC. HIV-associatedneurological disorders: a guide topharmacotherapy. CNS Drugs2012;26(2):123Y134.

22. Ances BM, Letendre SL, Alexander T, Ellis RJ.Role of psychiatric medications as adjuncttherapy in the treatment of HIV associatedneurocognitive disorders. Int Rev Psychiatry2008;20(1):89Y93.

23. Corral I, Quereda C, Garcia-Villanueva M,et al. Focal monophasic demyelinatingleukoencephalopathy in advanced HIVinfection. Eur Neurol 2004;52(1):36Y41.

24. Newsome SD, Johnson E, Pardo C, et al.Fulminant encephalopathy with basalganglia hyperintensities in HIV-infecteddrug users. Neurology 2011;76(9):787Y794.

25. Tahan TT, Bruck I, Burger M, Cruz CR.Neurological profile and neurodevelopmentof 88 children infected with HIV and 84seroreverter children followed from 1995to 2002. Braz J Infect Dis 2006;10(5):322Y326.

26. Simpson DM, Haidich AB, Schifitto G, et al.Severity of HIV-associated neuropathy isassociated with plasma HIV-1 RNA levels.AIDS 2002;16(3):407Y412.

27. Schifitto G, McDermott MP, McArthur JC,et al. Incidence of and risk factors forHIV-associated distal sensory polyneuropathy.Neurology 2002;58(12):1764Y1768.

28. Morgello S, Estanislao L, Simpson D, et al.HIV-associated distal sensorypolyneuropathy in the era of highlyactive antiretroviral therapy: the ManhattanHIV Brain Bank. Arch Neurol 2004;61(4):546Y551.

29. Smyth K, Affandi JS, McArthur JC, et al.Prevalence of and risk factors forHIV-associated neuropathy in Melbourne,Australia 1993Y2006. HIV Med 2007;8(6):367Y373.

30. Simpson DM, McArthur JC, Olney R, et al.Lamotrigine for HIV-associated painfulsensory neuropathies: a placebo-controlledtrial. Neurology 2003;60(9):1508Y1514.

31. Phillips TJ, Cherry CL, Cox S, et al.Pharmacological treatment of painfulHIV-associated sensory neuropathy: asystematic review and meta-analysis ofrandomised controlled trials. PLoS One2010;5(12):e14433.

32. Robbins GK, De Gruttola V, Shafer RW, et al.Comparison of sequential three-drugregimens as initial therapy for HIV-1infection. N Engl J Med2003;349(24):2293Y2303.

33. Gonzalez-Duarte A, Robinson-Papp J,Simpson DM. Diagnosis and management

of HIV-associated neuropathy. Neurol Clin2008;26(3):821Y832, x.

34. HIV Neuromuscular Syndrome Study Group.HIV-associated neuromuscular weaknesssyndrome. AIDS 2004;18(10):1403Y1412.

35. Clifford DB, Evans S, Yang Y, et al.Impact of efavirenz on neuropsychologicalperformance and symptoms in HIV-infectedindividuals. Ann Intern Med 2005;143(10):714Y721.

36. Pettersen JA, Jones G, Worthington C,et al. Sensory neuropathy in humanimmunodeficiency virus/acquiredimmunodeficiency syndrome patients:protease inhibitor-mediated neurotoxicity.Ann Neurol 2006;59(5):816Y824.

37. Ellis RJ, Marquie-Beck J, Delaney P,et al. Human immunodeficiency virusprotease inhibitors and risk forperipheral neuropathy. Ann Neurol2008;64(5):566Y572.

38. Birbeck GL, French JA, Perucca E, et al.Antiepileptic drug selection for peoplewith HIV/AIDS: evidence-based guidelinesfrom the ILAE and AAN. Epilepsia2012;53(1):207Y214.

39. French MA, Mallal SA, Dawkins RL.Zidovudine-induced restoration ofcell-mediated immunity to mycobacteriain immunodeficient HIV-infected patients.AIDS 1992;6(11):1293Y1297.

40. Shelburne SA, Visnegarwala F, Darcourt J,et al. Incidence and risk factors for immunereconstitution inflammatory syndromeduring highly active antiretroviral therapy.AIDS 2005;19(4):399Y406.

41. McCombe JA, Auer RN, Maingat FG, et al.Neurologic immune reconstitutioninflammatory syndrome in HIV/AIDS:outcome and epidemiology. Neurology2009;72(9):835Y841.

42. Manabe YC, Campbell JD, Sydnor E, Moore RD.Immune reconstitution inflammatorysyndrome: risk factors and treatmentimplications. J Acquir Immune Defic Syndr2007;46(4):456Y462.

43. Stone SF, Price P, Keane NM, et al. Levels ofIL-6 and soluble IL-6 receptor are increasedin HIV patients with a history of immunerestoration disease after HAART. HIV Med2002;3(1):21Y27.

44. Price P, Morahan G, Huang D, et al.Polymorphisms in cytokine genes definesubpopulations of HIV-1 patients whoexperienced immune restoration diseases.AIDS 2002;16(15):2043Y2047.

45. Antonelli LR, Mahnke Y, Hodge JN, et al.Elevated frequencies of highly activated



CD4+ T cells in HIV+ patients developingimmune reconstitution inflammatorysyndrome. Blood 2010;116(19):3818Y3827.

46. Worsley CM, Suchard MS, Stevens WS, et al.Multi-analyte profiling of ten cytokines inSouth African HIV-infected patients withimmune reconstitution inflammatorysyndrome (IRIS). AIDS Res Ther 2010;7:36.

47. Grabmeier-Pfistershammer K, Steinberger P,Rieger A, et al. Identification of PD-1 as aunique marker for failing immunereconstitution in HIV-1Yinfected patientson treatment. J Acquir Immune Defic Syndr2011;56(2):118Y124.

48. Bucy RP, Hockett RD, Derdeyn CA, et al.Initial increase in blood CD4(+) lymphocytesafter HIV antiretroviral therapy reflectsredistribution from lymphoid tissues. J ClinInvest 1999;103(10):1391Y1398.

49. Pakker NG, Notermans DW, de Boer RJ, et al.Biphasic kinetics of peripheral blood T cellsafter triple combination therapy in HIV-1infection: a composite of redistribution andproliferation. Nat Med 1998;4(2):208Y214.

50. Battegay M, Nuesch R, Hirschel B, KaufmannGR. Immunological recovery andantiretroviral therapy in HIV-1 infection.Lancet Infect Dis 2006;6(5):280Y287.

51. Cinque P, Bossolasco S, Brambilla AM, et al.The effect of highly active antiretroviraltherapy-induced immune reconstitution ondevelopment and outcome of progressivemultifocal leukoencephalopathy: study of43 cases with review of the literature.J Neurovirol 2003;9(suppl 1):73Y80.

52. Vendrely A, Bienvenu B, Gasnault J,et al. Fulminant inflammatoryleukoencephalopathy associated withHAART-induced immune restoration inAIDS-related progressive multifocalleukoencephalopathy. Acta Neuropathol2005;109(4):449Y455.

53. Torok ME, Kambugu A, Wright E. Immunereconstitution disease of the central nervoussystem. Curr Opin HIV AIDS2008;3(4):438Y445.

54. Wyen C, Hoffmann C, Schmeisser N, et al.Progressive multifocal leukencephalopathyin patients on highly active antiretroviraltherapy: survival and risk factors of death.J Acquir Immune Defic Syndr2004;37(2):1263Y1268.

55. Tan K, Roda R, Ostrow L, et al. PML-IRISin patients with HIV infection: clinicalmanifestations and treatment with steroids.Neurology 2009;72(17):1458Y1464.

56. Whitley RJ. Herpes simplex virus. In: KnipeDM, Howley PM, eds. Fields virology.

Philadelphia, PA: Lippincott Williams &Wilkins, 2001:2461Y2511.

57. Newsome SD, Nath A. Varicella-zoster virusvasculopathy and central nervous systemimmune reconstitution inflammatorysyndrome with human immunodeficiencyvirus infection treated with steroids.J Neurovirol 2009;15(3):288Y291.

58. Janowicz DM, Johnson RM, Gupta SK.Successful treatment of CMV ventriculitisimmune reconstitution syndrome. J NeurolNeurosurg Psychiatry 2005;76(6):891Y892.

59. French MA, Lenzo N, John M, et al. Immunerestoration disease after the treatment ofimmunodeficient HIV-infected patientswith highly active antiretroviral therapy.HIV Med 2000;1(2):107Y115.

60. Asselman V, Thienemann F, Pepper DJ,et al. Central nervous system disordersafter starting antiretroviral therapy inSouth Africa. AIDS 2010;24(18):2871Y2876.

61. French MA. HIV/AIDS: immunereconstitution inflammatory syndrome:a reappraisal. Clin Infect Dis 2009;48(1):101Y107.

62. Boelaert JR, Goddeeris KH, Vanopdenbosch LJ,Casselman JW. Relapsing meningitis caused bypersistent cryptococcal antigens and immunereconstitution after the initiation of highlyactive antiretroviral therapy. AIDS2004;18(8):1223Y1224.

63. Berkeley JL, Nath A, Pardo CA. Fatal immunereconstitution inflammatory syndrome withhuman immunodeficiency virus infectionand Candida meningitis: case report andreview of the literature. J Neurovirol2008;14(3):267Y276.

64. Crump JA, Tyrer MJ, Lloyd-Owen SJ, et al.Military tuberculosis with paradoxicalexpansion of intracranial tuberculomascomplicating human immunodeficiencyvirus infection in a patient receiving highlyactive antiretroviral therapy. Clin Infect Dis1998;26(4):1008Y1009.

65. Dhasmana DJ, Dheda K, Ravn P, et al.Immune reconstitution inflammatorysyndrome in HIV-infected patients receivingantiretroviral therapy: pathogenesis,clinical manifestations and management.Drugs 2008;68(2):191Y208.

66. Martin-Blondel G, Alvarez M, Delobel P,et al. Toxoplasmic encephalitis IRIS inHIV-infected patients: a case seriesand review of the literature. J NeurolNeurosurg Psychiatry 2011;82(6):691Y693.

67. Tsambiras PE, Larkin JA, Houston SH. Casereport. Toxoplasma encephalitis after




initiation of HAART. AIDS Read2001;11(12):608Y610, 615Y616.

68. Grant PM, Komarow L, Andersen J,et al. Risk factor analyses for immunereconstitution inflammatory syndrome ina randomized study of early vs. deferredART during an opportunistic infection.PLoS One 2010;5(7):e11416.

69. Clifford DB, De Luca A, Simpson DM,et al. Natalizumab-associated progressivemultifocal leukoencephalopathy in patientswith multiple sclerosis: lessons from 28 cases.Lancet Neurol 2010;9(4):438Y446.

70. Lihana RW, Khamadi SA, Lubano K, et al.Genetic characterization of HIV type1 among patients with suspected immunereconstitution inflammatory syndromeafter initiation of antiretroviral therapy inKenya. AIDS Res Hum Retroviruses2010;26(7):833Y838.

71. Makadzange AT, Ndhlovu CE, Takarinda K,et al. Early versus delayed initiation ofantiretroviral therapy for concurrent HIVinfection and cryptococcal meningitis insub-saharan Africa. Clin Infect Dis2010;50(11):1532Y1538.

72. Blanc FX, Sok T, Laureillard D. Significantenhancement in survival with early

(2 weeks) vs. late (8 weeks) initiation ofhighly active antiretroviral treatment(HAART) in severely immunosuppressedHIV-infected adults with newly diagnosedtuberculosis. Presented at: 16thInternational AIDS Society Conference;July 2010; Vienna, Austria.

73. Lipman M, Breen R. Immune reconstitutioninflammatory syndrome in HIV. Curr OpinInfect Dis 2006;19(1):20Y25.

74. Venkataramana A, Pardo CA, McArthur JC,et al. Immune reconstitution inflammatorysyndrome in the CNS of HIV-infectedpatients. Neurology 2006;67(3):383Y388.

75. Petito CK, Torres-Munoz JE, Zielger F,McCarthy M. Brain CD8+ and cytotoxicT lymphocytes are associated with, and maybe specific for, human immunodeficiencyvirus type 1 encephalitis in patients withacquired immunodeficiency syndrome.J Neurovirol 2006;12(4):272Y283.

76. Lindzen E, Jewells V, Bouldin T, et al.Progressive tumefactive inflammatorycentral nervous system demyelinatingdisease in an acquired immunodeficiencysyndrome patient treated with highlyactive antiretroviral therapy. J Neurovirol2008;14(6):569Y573.



1319 neurologic complications of hiv-1 infection

Documents

hiv treatment

use of art

neurologic complicationsof

regimens of art

hiv andperipheral neuropathy

preart era

cns cnsiris

newer art regimens