Vol. 33 (Suppl). No. 1. 1993 Pnnred 1t7 L..S .-l

Pathogenesis of Immune-Mediated Neuropathies

A. M. RrnAMI Wwtrnmt qfNdogy, UUniveniry QPennryivmria &had QMediciue, fWadeiphia,Pennsylvania 19101

AB!3TRACT. A variety of peripheral neuropathies are MG, intravenous immune globulin believed to be imm~e-mediated.Acute Mammatory de- myeliarting or GuihbBarr6 syndrome (GBS) is the prototype of these neuropa&ks. GBS is characterized by acute progressive motor weakness of the extremities and of balk and facial musadatme. Deep tendon reflexes are reduced or abseat, and sensory symp GUILLAIN-BA& SYNDROME toms are mild. Respiratory failure and autonomic dysfunc- tion may be seen. The cerehpioal fluid shows increased Clinical mpects. GBS is a distinct clinical entity characterized protein and no or very few cells. The conduction by acute progressive motor weakness of the extremities and of velocity is slowed, and the pathology shows segmental bulbar and facial musculature. Deep tendon reflexes are reduced demyeli~mtionwith mononuclear cell Miltration. Studies or absent and sensory symptoms are mild. from man and experimental animals suggest an immune The CSF shows albuminocytologicdissociation with increased logic basis for demyelination of the peripheral in protein and no or very few cells. Electrophysiologically, the GBS, but the mechbm is not well Pmlerstood. Experi- disease has features of demyelination and pathology shows seg- mental allergic neuritis, an animal model of GBS, is in- mental demyelination with mononuclear cell infiltration. duced in laboratory animals by immaniPti011with myelin An ad hoc National Institute of Neurological Disorders and Pz protein, some peptides of PZprotein, and galadocere- Stroke committee produced guidelines for the diagnosis of GBS hide. The animals develop weakness and show ekctm (1). These criteria are somewhat restrictive and exclude certain physiologic and pathologic features similu to GBS. Pz- nosologic entities such as: I)Miller Fisher syndrome of ophthal- reactive T cells ad antigaktocerebmide antisera can moplegia, ataxia, and areflexia; 2) sensory loss and areflexia., 3) adoptively transfer experimental allergic neuritis. Various polyneuritis cranialis, and 4) pandysautonornia. These disorders antibodies to peripheral nerve myelin and circulating im- are generally believed to be "variants" of GBS (I). mme complexes have been found in patients with GBS. Motor weakness usually develops rapidly. More than 90% of The target antigen(s) for these antibodies are not well patients with GBS cease to progress within 4 wk into the disease. understood, but neutral glycolipids cross-reactive with Recovery usually begins 2 to 4 wk after the cessation of progres- Forssman antigen and gangliosides are possible candidates. sion but may be delayed for months (1). Many patients are left The mainstay of therapy is the management of the para- with a residual deficit after 1 y (2). In another study, the median lyzed patient. Steroids are ineffective. Pksmapheresis, time of recovery to independent waking was 85 d, and for those especially early in the course of the disease, can shorten on a respirator it was 169 d (3). Autonomic dysfunction such as the duration of and intubation. Results from a cardiac arrhythmias, tachycardia, postural hypotension, and hy- multicenter study in the Netherlands demonstrate the ef- pertension are seen and at times could be fatal. Conditions such fmcy of highdose immune globulin therapy in GBS. In as hexacarbon abuse, porphyria, , lead poisoning, bot- this review, Gllillrin-Barr6 syndrome as a putative im- ulism, toxic neuropathies, hysteria, tic paralysis, and poliomye- mane-mediited neuropathy is diiin detail. Chronic litis occasionally may be confused with GBS and need to be Mammatory demyeliarting polyneuropathy, neuropathks ruled out. associated with paraproteinemias, AIDS, and Lyme die Epidemiology. The incidence of GBS varies from 0.6 to 1.9/ ease are discussed briefly. Neuropathies with features of 100 000 per year (3). The disease is widely distributed throughout secondary to other and collagen the world. Children and adults of both sexes are affected. There vPscPlnr disorders are not the subject of this review. seems to be a bimodal distribution of patients, with the major (Pcdiatr Rm 33 (Suppl): S9eS94,1993) peak in patients aged 16 to 25 y and a smaller peak in those aged 45 to 60 y. Males and Caucasians seem to be most susceptible (4). GBS was the most likely etiology (15%) in 162 cases of Abbreviations neuropathy evaluated in the Children's Hospital of Philadelphia aDp9 cw demYeliartiagplmpa- (5\n a swey of more ths. 1000r~xa of GBS, approximately thy 66% had had a viral-like illness, mainly in the respiratory and m,eytomm* CSF, cerebrospinal fluid gastrointestinal tracts, within 8 wk of the onset of GBS and about EAN, experhntal allergic neuritis 90% had had a viral-like illness %thin 1 mo of expressing GBS EBV, Epstein-Barr virus symptoms. Five percent had a previous surgical procedure and GBS, GuillaibBarri syndrome 4.5% had received vaccinations. A cluster of GBS occurred in HIV, human immunodeficiency virus the United States in 1976-1977, after vaccination with A/New Jersey influenza. No similar outbreak occurred in a military Comspondence: A. M. Rostami, M.D., W.D., Department of Nc~~~logy,population dvingthe same vaccine (3,4). H@tal of the UnidW of Pen- 3400 SmStreet, -1- PA 19104. AnMent infections in GBS can be viral, wd,or para- supportedin part by hbNS08075 and AR39489 from the ~~hd~dmte~sitic. CMV and EBV are the most common viral infections of ~eelth. reported to be associated with GBS. High levels of IgM antibodies s90 5D NEUROPATHIES

to CMV re '-.i.10-151 of GBS Ta pacent have This form of EAN is clearly mediated through antibodies, and unt~bod~e\tow d 5% have sedu& avidGarce of M' gakkwxewseems to be the target antigen (15,27,28). pliii~ris1~1~wt~(6). An interdng ittmcWmbezw#nGBS Basedonthe~tionthattheherpesviruses,CMVIld andCatiigj/~( jiPfrmih~bllapQl0sdInoa~:study,%%EBV, are aso&ted with GBS, vimxs have been tho* a b of 56 CO~GBSwin m Atandim hospitdBad poadble etiologic agents for the disorder. Indeed, for some tias GBS was calkd acute hfkdivt?polynewitis (29). A herpa virrrr ofbirds,knownarMmk's~vinrs,producesalymabo/ptg- ~tivedisodeincbicheaaThisdiseaaecanbe~ withparalysisofthcwhgsandbgsinchidrerrsandaaslwt pathology simibr to GBS (30). A condition similar to GBS in patients with HIV infecton and bigh titer of anhito C. in jejuniorm~isalaotakcnascvideacttbotbodesiad ~maybein~intbepath~ofGBSdinctlyor arcthensultdoniamn~auackonthepadpbnrrlnavous indindlythtougbactivationof~intheimm~~~esystem. systemmytlinaparsiMySchwanncells.ThisevldcPlaisderivtd To date, no co-ve evideace as to the inkdous etiology of fromimmumkgk~~~~aathe~withGBS is availabk. GBsanddsdDfCoppm(DmmOddofthe~~. ThelandmPrkw~ofPwburyetd(31)onthe~of EANisindpgdmeariollsWoratorynnimrlsby~

wkPfterimm~theurin\ah~w&ht~~ tail,ataxia,and~ar,insmncsses,tetnple%iaTbe mrpri(rd(ht~~figgath.d;cl.M.m~l~ *. . . aaZthg10drfta~~there is ~theaciatiCmotorco~on,whkhis Tbenisdiqmimof

occurs in a perivenulor

pbgq tinssea intheinvdvedm(14-16). In whole nerve and Pr~EAN, in early stages of the dkae myelin nsicul.tion and endo& edema an be obaennd in the myalsnrbGathand/or§chmumodla Despltethwc absenceofm01~)nuckarcell~0lltact(l7).EANinLewis~~aodtPacatrodeforTcelbinthe~of be transfinrd with Prd~T &Is (18-20). T dbdve to documednd th~peptide53-78ofbavineP~proteiaalso~antnnafaEAN An-- cells in humen peripheral nemr art (2l).Thne----thatp- ~tobeSch~&Is(42);~,daEafrom~(we.I EAN is a cdCeimmune phenomenon. above) point to nsident xmmpbge rather than %hwam cells SchwmacdtshammwbomratshaveMHCctasSIIantigenS beingan~~cells. ontheirsurfir+mdpaentan~tosyneGadcmytlinbesic Antibodies to various peripheral maw components are rewe- protein mctbT cdk (22). Hawever,cells other tban Schwann ~~lltedinGBS.Onthcbroisoftheobarervationthatan~ 1 celk, and awd IkJy mscraphages, play on emmtial role as cdkintheperipherolnarw,asisshownby

of Lewis ad@ intb DA rats that have been tdbPlly irradiated and mm&mi with (Lewis x DA) M borrc mrriow cdl(23). ~oubkbdcxpcrhncnts using (~11tbdipto 5100 pro6ein d

pathology oampmd with mydin a P&adud EAN. *infiltrrtionofsnnall~isnot~butmrrp. (49) have bcen su%gated as candidatea ThcJe obsava roglobulinmb, heavy draia diserrse, cryoglobulinemia, nonma- tionsontherOaeof~~inGBSaadEAN~the~bt mwd -anafhy, and primary amyld- fa the uae and subsaqmt demdonof dlscacy d plat+ Sapmpobdn~firollrthesepatientsusuallyshows mqhmhin GBS. hom#mm W of abnormal mobility called paraproteins or fi).Elament.AhhoughaboutmofGBSpatients~#~)ver&omM prdeina Asymptomatic M pPoteiDs (not associated with thc~~~0~~~OfthG*canbestormy.Akmt30% matigaaplssms cell dyscmh) s~ecommon in adults (ammx- ofp.~inone~nq~in~nforana~ofSlhatdy 1%) end are nferrod to as benign or nonmalignant d;~~~s aywrrs61d(50).Tbemab&yofthcrapy moagmmop&y (62). NamnrligPant IgM monoclonal istht~cntaf~~'ts,whichisbcst9ccom- or piyqmtuncmta can be associated with distal plidLadin~vecpreunits. smmmowr, purely motor, or prabminantly sen- ~wsafoturdtobeeffedivein~mor-aaryoemopathirathatafe~dowty~ve.Walden- bidity md shorteaing the 60- of GES. In the United States sSrBan's~~should~ruIedoutbdorethedi~ multicmkr obatdlai stuay performed on 245 -ts witb nosis of noxudgmnt IgM gumnopathy and neuroprrthy is GBS (51X the patients treated with badbetta made. In same patients, M protein in the serum is shown to be outmms.whenm~ritb~=~pl, dircued qpkt mydinamchd glycoprotein. Serum hm mapbe& in tenns of time needed to improve ane ctinical odhcrpatieats~withchondroitin~or~~ grade, outcome at 4 wk and 6 mo, and time when the patieslt (62, 63). Malts on pmdomhntly aBctad by this dimdex. couldwaIkunaidcd.~alsoreduced~numberdNoamalignant~andI~mo~nalnnmlnoagthisamelso daysontherespirofor.In~ncrecrises,espacirrllywhenventilatory be associot#l with a chronic progmsive or recurrent polynamp ashmccisneedsdorwhenthe patient hasarapicnywve athy. The nave pathology in thest gammopathh is usually course, phmaphasis is indicated Omx the decision to treat demyalinative, sometime andatad with axonal -0% patients with has been made, it should begin as and shows no sieaifcant cellular Miltration. In o&w&m& early as possible. myeloma, which is almost always rllsociated with X light chains, In one randomized clinical trial of mticustcroid treatment for a syndrome oflJ~bJ=urmth~, ~omc%aly,enddopathy, GBS, no beneficial effect was observed (52). Another study, usiq M pnotein, and skin changes (POEMS)ha9 been reported (64). Am,reported &&cant reductions in duration of disease may be help11 in reducing M pr0t.h~in the (53). A British multicenter study is underway with -dose serum. By treathg mderlhplasma cell dyscrasia with chemo- rnethylpednisone as one treatment arm. There is no conclusive therapy and radiation therapy, the periphd nem deficit may evidence that azathioprine and cyc- can be of any improve. help in GBS (54). Tben mz reports of donof motor neuron-like dbse Results from a multicenter study in the N&hdands compar- ' In several of these amyotrophic lateral ing h@dose MGwith pIasma ex- in 150 patients dem- m-IgMpampm&hbindGMland oastrateabetteroutcomcinpatientsreceiviagMGannpared GDlb ~~(65). Selective degeneration of ventral roots with tho~etreated with plasma exchange. Beneficial outcome and chmatolytic dmga in the anterior horn cells ban been was defined as improvement in at kast one grade on a seven- 0~~~tSWiththe~mebaveimproved point functional scale 4 wk after randomization (55). with plasma exchange and immunmvetbapy (66).

NEUROPATHiES ASSOCIATED WITH HIV mPis~by~~~teanset(Irwrally0vermonths)Several paiphd nerve syndmmes are reported in patients of chro~ricpmgmsive, relapdng or chronic monophk course. with HIV Mection including 1) distal symmetrical sensorhotor Maximum dkidlity is usually reached after 6-12 mo, and neuropathy, 2) CIDP, 3) mononeuritis multiplex, 4) pmpdve peripberalnervedemyelinationcanbedcmonstratedbydectro- po1ydidopathy, 5) Guilbin-Badlike acute intlammatory physiology and ~ology(56, 57). Idammatory cells in polyneuropathy, and 6) . In AIDS pa- the nnves are less prominent than in GBS. Protein is ekvated tients, it is bdleved that acute polyn-y, in the CSF, but no or very few cells are found. No systemic CIDP, and pi mononeuritis multiplex have an autoim- disease or semm parapmtcins are pmmt In a large series of mune pathogamis. PolyradMop9thy- could be due to direct CIDP @mts (92 CAWS) (57, 58), 65% had a dapshg oourse inktion by CIMV, herpes virus, or HIY. The pathogenesis of and 35% had a progredve monophsk course. Thirty-five pa- other types of peripheral nerve disoPders in AIDS is even less cent had a history of an sntecedeat event, mainly infactious, u-. Pbmapkmis has been stKwvn to be effbcfh in with a significantly increased level of antibodies to CMV cam- AILSdated inflammatory neuropathies (67,68). patad with contrda A good merywas made by 73% of CIDP @en&, and they became iadependent (57, 58). The cxad WEUROPATHES ASSOCIATED WITH mechanism ofmye4in destruction is unkwrwn, althaugh im- mune-medhkd moGhanisms have bmn poatulased, an8 bi* Prripherd nerve involvements are rqmted in infectiotlJ by ofIgGtothemyelinsheathwithtbe~t~onofthe spiro&ete Boreiia (Lyme disease). Bilaferal facial myelSnsegrpeaQby-J--pw=dJ!=* ~,GGBS,m~n0n~m~lcx,Pad~~tieare1 ofcxdbmdatcd im sea8;andthoCSFofLym~diseasepatimtamayshwrPlympho- to a btugdot- -yeli- .bigRotnport idbmai~rydmga in the mws, The of periph- faitunwithpadnhoaeand 's, atha immunosup dnewedisordeninLymediseaseisnotwellundemtood. ~W~tS&P~~~amiAn@& therapy Is gmdyeWve in management of the IVIG may be used (61). puiphmd nave involvement in this disorder (68-70).

1. Ahfy AK 1981 Dia(plofflc aM&kbatbns in Guilhh-W syndrome. Ann Neurd ~~hl-5 ~~ycanoccurinassoci.tionwithplasma2. NkKlmnu GM 1990 ChdlnbBacr6 symJrome: clinical and thaaputc oh- oell such u multi* myeloma, Wfdhstmu's msc- ntiolls. Aan Nesrd 27(mppl)SlES16 s93

Saamemm SF. Wehwr HL -321321-

~inopabi.-bdyndromc.Aea~23~7 in 49.YuR,AripT.~T,~~~MvdilY,~N1990 AuWbmnm nurh.nimn rn PCriPbQPl nemopathia Ann Ned 27(suppl)5430435 50. RoppaAH,~BT1984~.ad~ofwtcanflexic pPrslysis witb rmpbrais on ~~syadnnne. In: Asbury K, OillisSt R (d)PeriphaPl Nave Disadaq A Prpctieal Approach, Vd 2. Butter-

of mydin P2 proCrin b a T- 53. autoimmune~cdl sides GMI and GDl b: improvement following immunothuapy. at least one grade at 1 mo was 59% in the group 38(suppl lP5YaW) 67. Cornblath DR, McArthur JC, Kennedy PGE, Witk AS, Griffin JW 1987 versus 39% in the group treated conventionally. The French In- demydinating pgipbaal ncuropathiaassociated with human study also showed signi6cant improvement with plasmapheresis T.cdllym~vinutypc~(~~ion.AMNcurd21:32-40compared with the conventional group. 68. VaUat JM, Hugon Lubeau ~,-~ebouteta, Dumas M, vnDr. Roifman: Effective in what? All they've shown are very R 1987 Tick-bite mahgodbb- dinial ~ysolog~~and marginal effects. bidogic finding in 10 Nevrology 37749-753 69. Halperin JJ, Little BW, Cojrk- PK, Danwykr RJ 1987 Lyme diserrpe: cause of Dr. Rostami: I disagree. As I showed in one of my slides, the a trertaMe peripheral aeuropathy. Neurdogy 371700-1706 time to reach one grade improvement was 17 d versus 34 d and 70. HP, Heiniager K, Schafa B, Fm W, Toyka KV 1988 Immune the duration of intubation decreased signiscantly, so I think mechanisms in inflammatory polyneuropathy. AM NY Acad Sci 540: 122- there is enough evidence to show that plasmapheresis is effective. 161 Whether antibody or something else is being removed from the circulation we don't know. It could be cytokines. We cannot rule FLOOR DISCUSSION out the Wbility of T cells, because it could be some cytokine that is being removed rather than antibody alone. Dr. Frank: As was mentioned during the talks, several authors Dr. Roifman: You mentioned the Dutch study and its prelim- have shown that there are high levels of antibody to myelin in inary results. Do we have information as to what type of patients patients with GBS. What was not mentioned was that in studies were involved? Are they patients with severe Guillain Ban+? done in our laboratory with Lee Koski and Moon Chen we also That is an important issue as well, because if these patients are showed very high levels of membrane attack complex in most of assessed 3 or 4 wk out, the milder cases are going to be very hard those patients. We published one paper in the Journal of Clinical to assess. Investigation and two in Blood. We believe the way that MG Dr. Rostami: I have spoken to Dr. Franz van der Mechk, the works is that it prevents complement from binding to targets, principal investigator in the Dutch study. That study was done and this mechanism was not even on your list. at nine centers with 150 consecutive patients, and they were Dr. Rostami: I apologize for that. I was attempting to expedite randomized. I have seen the manuscript that has been submitted the presentation. to the New England Journal of Medicine and have no doubt that Dr. Frank: Also, a recent review article in the New England the study was done correctly. They took patients in the first 2 Journal of Medicine doesn't mention high levels of membrane wk of the disease and followed them for 6 mo, comparing MG attack complex either. I believe that is a real possibility and and plasmapheresis. There was 53% improvement of strength, certainly worth further consideration. by one grade or more, with MGversus 34% with plasmapher- Dr. Rostami: Absolutely. esis. This study showed that IVIG was better than plasmapher- Dr. Roifman: You keep saying that plasmapheresis is effective esis, but there are a couple of criticisms of that study. First, more in Guillain-Ban-4, but in fact the multicenter studies showed than 150 patients should be studied, at least 245 as in the North almost no effect of plasmapheresis. At best, the results were American study. Second, the rate of improvement with plasma- marginal, maybe a difference of a few days under a respirator. pheresis was lower in this study than in the American study, That is an important issue because if plasmapheresis is effective, which points to the question of whether the plasmapheresis was one may postulate that there may be a humoral factor involved. comparable to that of the American study. It may be implicated in the pathogenesis of the disease. Dr. Siami: I would like to elaborate a little bit about plasma Dr. Rostami: Multicenter studies have clearly shown that exchange in GBS. The patient should not be able to walk more plasmapheresis is effective. In the American study, with 245 than 5 m, which indicates relatively severe disease, and the patients ( 123 in the conventionally treated group and 122 in the plasmapheresis should be started in the fvst week or 10 d to be plasmapheresis arm), the percentage of patients who improved useful.