882 AD-A271 til Il llli llll llllf llllil lllll llll 22 December 1992 Reprint Role of Infection Due to Campylobacter Jejuni Army Project Order in the Initiation of Guillain-Barre Syndrome 90PP0820 Ban Mishu and Martin J. Blaser 93-26165 Veterans Administration Medical Center Research Service 1310 24th Avenue South Nashville, Tennessee 37212-2637 U.S. Army Medical Research & Development Command Fort Detrick Frederick, Maryland 21702-5012 Title of Project Order: Studies of the Outer Membrane Proteins of Campylobacter Jejuni for Vaccine Development Approved for public release; distribution unlimited SAccs;ý:n F~c-. ' 'T Campylobacter, Vaccines, Biotechnology, ID, Diseases C) RADI, Lab Animals, Volunteers Unclassified Unclassified Unclassified Unlimited Best Available Copy 104 Role of Infection Due to Campylobacterjejunii n the Initiation of Guillain-Barre Syndrome Ban Mishu and Martin J. Blaser Fromi,t ie Division / In/eunous Diseases. Department o/ Medic'/no. Vanderbilt I 'niversii v' SchoW (y Medicine: and ihe Medual Servi(e, l)cpartient o/I eterans A/fifirs Medical C'enter. Nashville, 7T'nnessee Recent reports suggest that infection with Campylobacterjejuni,a common enteric pathogen, may cause Guillain-Barrk syndrome (GBS) by triggering demyelination of peripheral nerves. GBS is preceded by an acute infectious illness (due to a variety of agents) in 50%-75% of cases; the onset of neurological symptoms is preceded by diarrhea in 10%-30% of cases. In the last decade, more than 20 published anecdotal reports and case series have described patients with C. jejuni infection documented 1-3 weeks before onset of GBS. Cultures of fecal samples obtained at the onset of neurological s) mptoms from patients with GBS have yielded C. jejuni in more than 25% of cases. A relatively rare serotype, Penner type 019, is overrepresented among isolates of C. jejuni from Japanese patients with GBS. Serological studies suggest that 20%-40% of patients with GBS have evidence of recent C. jejuni infection. In summary, infection with C. jejuni is a common antecedent to GBS and probably plays a role in initiating demyelination; although several pathogenic mechanisms are possible, none has been proven. Guillain-Barrý syndrome (GBS) is the most common ness and limb weakness and paralysis in 1892 [8]. Epidemio- cause of acute neuromuscular paralysis in both adults and logical studies throughout the world have confirmed the asso- children [ I, 2]. The annual incidence in the developed world ciation between GBS and preceding infectious illness [3, is between I and 2 cases per 100,0O0 pupuiatiýrn [3, 4]. In 9-16]: 50%-75% of patients have such an illness in the weeks recent years infection with Campyrlobacter jejuni has been preceding onset of GBS (table I). Specific infectious agents identified as a potentially important precipitating factor for associated with GBS include Mvcoplasinap pwumoniae, hepa- GBS. C jejuni is the most frequently identified bacterial titis B virus. cytomegalovirus, Epstein-Barr virus. varicella- cause ofgastroenteritis in the United States and in other in- zoster virus, rubeola virus, and (most recently) human immu- dustrialized countries [5]. Most infections with C.j ejuni pro- nodeficiency virus [1 7-2 1]. Although infections of the upper duce an acute self-limited diarrheal illness [6]. Although ex- respiratory tract are also frequently described as antecedent traintestinal complications of C. jejuni infection [71] suggest events, gastrointestinal infections including diarrheal illness the involvement of immunopathogenic mechanisms, GBS precede GBS in 10%-307c/ of cases [3. 12. 16]. associated with C. jejuni infection has not been well recog- nized. We review the existing evidence that infection due to C. jejuni plays a role in the pathogenesis of GBS, and we Association of C. jejuni with GBS discuss possible mechanisms for this association. In 1977. Skirrow reported that C. jejuni was a common cause of acute diarrheal illness in England and established methods that facilitated isolation of this organism [22]. Grad- GBS as a Postinfectious Syndrome ually, this advance was incorporated into the routines of Guillain-Barrý syndrome, sometimes called acute postin- many clinical microbiology laboratories. However, even at fectious polyneuritis, often develops 1-3 weeks after an present, many laboratories attempt isolation of C.jejuni from acute infection of the respiratory or gastrointestinal tract. fecal specimens only upon special request if at all, and few Osler first noted the assc-ciation between acute infectious ill- laboratories utilize appropriate techniques for isolation of all members of the rapidly expanding genus Camp vlobac- ter (and now Arcobacter). Thus. any estimate of the role of these organisms in the causation of human illness is likely to Received 9 October 1992: revised 22 December 1992. be low. Financial support: The Muscular Dystrophy Association and the In 1982, Rhodes and Tattersfield reported a case of GBS Thrasher Research Fund. Reprints or correspondence: Dr. Ban Mishu. Division of Infectious Dis- following enteric infection with C jejuni [23]. Because rou- eases. A 3310 Medical Center North, Vanderbilt University Medical tine culture of stool specimens for campylobacters was so Center. Nashville. Tennessee 37232. new, the authors suggested that previous cases of Campyrlo- Clinical Infectious Diseases 1993;07:104-8 bacter-associated GBS may have gone unrecognized. Since © 1993 by The University of Chicago. All rights reserved. 1058-4838/93/1701-0017S02.00 then, there have been 21 more reports of individual patients (I ) 1993:17 (July) C. Iywum and (Guillain-Barrý Sndrome 105 Table I. Frequency of antecedent intections among patients with 019 133]. Since Penner lype 019 accounts for fewer than 2, (.iuillain-Barr(cid:127) syndrome (BS). of('. jq/uni strains in Japan 1461, these studies suggest that an Percentage with uncommon serotype of C. jeuni may be overrepresented recognized among strains associated with GBS. I o)cation antecedent Despite the brief duration of convalescent excretion of('. oftstudy No. of patients inlfictious jejuni, serological tests may provide evidence of recent infec- Irelerence] Year with GBS illness tion even after the bacteria can no longer be isolated from Great Britain 1101 1964 52 69 stools. In both natural and experimental infections due to C. Europe. United States 191 1966 1.100 58 jejuni, specific IgM and IgG antibody responses peak at 2 Nlinnesota [31 1978 40 70 weeks, but titers may remain elevated tor 4-6 weeks: levels Norway I I11 1985 109 57 of IgA antibodies also peak at 2 weeks but fall to baseline England 1121 1988 100 55 values within 4 weeks [471. Thus it may be possible to deter- India 1131 1988 56 32 F~rance 1141 1989 71 50) mine serologically whether GBS patients have had anteced- Spain 1151 1990 15 53 ent C.jejuni infection. Indeed, several investigators have doc- Indonesia 1161 1990 28 75 umented high levels ofantibodies to C.jejuni in patients with GBS orGBS-like illnesses [ 12, 28-30]. Although theantibod- ies were assayed by a variety of techniques, most methods found serological evidence of preceding C.jejuni infection in and series of patients with GBS preceded by C.jejuni infec- 14%-409: of patients with GBS (table 3). tion [ 12, 24-44]: data from reports that described patient Taken together, the anecdotal reports. the studies of series characteristics are summarized in table 2. Although almost of GBS patients whose stools were cultured, and the serologi- three times as many cases of C. jejuni-associated GBS have cal investigations suggest that C. f'juni infection may com- been reported among male patients as among female pa- monly precede GBS. Nevertheless. GBS is a relatively rare tients, it is uncertain whether this difference reflects demo- disease and is probably an uncommon manifestation of this graphic reality or a selection bias in the reporting of cases. very frequent bacterial infection. The Centers for Disease Initial reports indicated that patients with C. jejuni-asso- Control and Prevention estimates an annual incidence of ciated GBS had severe disease, with extensive axonal dam- - 1.000 symptomatic infections with C. jejuni per 100.000 age and prolonged debilitation [23-25]. Subsequent reports. population-roughly 2 million cases-in the United States however, have also described milder cases of GBS occurring [48]. The annual incidence of GBS in this country has been in association with C jejuni infection [26, 27]. estimated at 1.7/100,000 population [3]. or 4.250 cases. IfC. After the first anecdotal reports. some investigators began jejuni infection is an important antecedent event in only to routinely culture the stools of patients with GBS for C. 10%-30% of GBS cases (the minimal estimate based upon jejuni. However. because the duration of convalescent excre- serological surveys), then between 425 and 1.275 cases of tion ofC. fjjuni is brief (mean, 16 days from onset of diar- GBS preceded by C jejuni infection occur in the United rheal symptoms) [45] and because neurological symptoms in States each year (table 4). Therefore, on the basis of the esti- C. jejuni-associated GBS usually develop 1-3 weeks after mated number of cases of GBS preceded by C.j ejuni ir.iec- the onset of diarrhea [23-26]. cultures of fecal samples ob- tion and the total number of C. jejuni infections, it can be tained at the onset of neurological symptoms may yield calculated that one in every 2,000-5,000 cases of C. jejuni "false" negative results that are misleading. Despite this limi- infection is followed by GBS. These numbers illustrate that a tation. Speed et al. isolated C. jejuni from stool cultures of common problem (C jejuni infection) with an uncommon one of four patients with GBS [401. Ropper found that 13 manifestation (GBS) may represent a common cause of an (12%) of 106 patients with GBS had diarrhea in the 10 days uncommon disease. preceding the onset of neurological symptoms: stool was cul- tured in nine of these cases, and C.jejuni was isolated in four Potential Mechanisms Involved in GBS Following instances (44%) [31]. Kuroki et al. isolated C. jejuni from seven (88%) of eight pediatric patients with GBS in Japan [321. Because the patients were not related to one another The hallmark of GBS is segmental demyelination of pe- and became ill in different years, it is not likely that these ripheral nerves, with mononuclear infiltrates and edema. My- strains represented a common-source outbreak. Four of these elin destruction may be mediated by a direct toxic effect or strains w(cid:127)c72 -f Pnner type 019: the remaining three isolates by an immunopathogenic mechanism: in the latter case. ei- were not available. Other Japanese investigators serotyped ther humoral or cellular immune mechanisms could be oper- four additional isolates of C. jejuni from patients with GBS alive, and found that these four strains were also of Penner type Because of the extensive axonal damage noted in the earli- 106 Mishu and Blaser CID 1993:17 (July) Table 2. Cases of*Guillain-Barrý syndrome (GBS) following infection with C. iiquii. as described in published reports. Location. )ear of stud'v Ireferencel: Days from onset of patient's age (y)/sex diarrhea to OBS Outcome England. 1982 1231: 45/M 15 Wheelchair-hound at 8 mo Finland. 1982 124): 42/M 9 Se\ ere deficits at 6 mo England, 1983 125]: 34/F 10 Intubated: severe deficits at 6 mo Australia. 1984 126] 34/M 23 Complete recos cr5 at 3 mo 22/F 21 Mild 6th-ner.e palks at 3 mo England. 1984 127]: 16/M 10 Complete recoer, at 6 mo England. 1985 1411: 27/M 7 Complete recoerry with minor delicits Australia, 1986 [371: 2/F 9 Intubated: hospitalized at 4 mo: complete recovers, at 6 mo New York. 1987 1381: 69/M 13 Wheelchair-bound at 10%% Swit/erland, 1988 1431: 19/M 13 Complete recover-, at 3 nio Boston. 1988 [311 63/F 7 Wheelchair-bound at 6 mo 74/F I I Complete reco%cr at 6 mo 32/M 7 Complete reco.ers at 6 mo 19/M 10 Complete recover. at 2 mo Switzerland. 1988 139] 38/M 6 Slow recoser over I 81/M I Complete recov.ery at 9 mo 6(/M 12 Complete recoers, at I France. 1989 1421* 30/M 10 Complete recovers at 28 mo 62/M 8 Complete recoverv at 3 mo 74/M 4 Not described Japan. 1990 135[* 25/M 14 Canes required for walking at 6 mo 83/F 7 Bedridden at 3 y Japan. 1991 1321 7/M ? Complete recovery with minor deficits 9/M 6 Complete recovery with minor deficits I O/F 15 Complete recovery I I/M 5 Complete recovery with minor deficits 13/M 6 Complete recovery 14/M 6 Complete recovery I 4/M 7 Improving at I mo * Diagnosis of campylobacter infection made by serological tests: diagnoses in all other reports were culture proven. est cases, direct toxic damage to neural structures-perhaps had this antigen [36]. Previous studies of GBS patients had attributable to a yet-unrecognized neurotoxin-has been not shown an association with a particular HLA type [51]. suggested [24]. However, the toxic activity of C. jejuni in (2) Perhaps only a few C. j(cid:127)juni strains are capable of trigger- vitro [491 has not been demonstrated in vivo [50]. The tim- ing immunologically mediated myelin destruction. The asso- ing of the onset of GBS (1-3 weeks after the peak of the ciation of Penner serotype 019 with the initiation of GBS diarrheal illness) also argues against a direct toxic effect of C. [32. 33] suggests that these strains may represent a particu- jejuni. larly virulent clone or that the 019 polysaccharide may be Two distinct possibilities for mechanisms of immunologic either a virulence determinant per se or a marker for other injury have been proposed. (I) Demyelination may result virulence factors. from a specific genetic predisposition. A Japanese investiga- The neurological target fbr immunologic injury also is un- tion showed that all of six patients with GBS who had ante- known. Serum from a patient with Cjejlni-associated GBS cedent infection due to C.jjuni possessed the HLA-B35 an- reacted strongly with PO, a peripheral nerve myelin-specific tigen, whereas only 14% of 3.090 healthy Japanese controls protein [34]I this observation suggested that C jejuni pos- 1I) 19 93:17 ()uhv ) ,wtin and (iuillain-Barrý Ssndrcrme 1o7 sesses antigens stimulating the production of'antibodies that Table 4. Estimated annual incidence,, of infection due to ( rweaich prpea L_e v ln n as iS inlr~ junm. (uillain-Barrý s~.ndrome (GBS). and (lBS preceded ký C. I,- o'tngliosides found in m~elin may be the ultimate target o mfn'toni h nie tts antibodies freinresponse to C. Jquni infection: OBS Aninual incidence has followed the administration of' parenteral ganoliosides (ca~.ei 1( 10.)48) 152. 531. DesPopulation) GBS ficlloswing C. /1//m infiection often is associated with Dsae [reference] I otal cases/V* extensive axonal damage. a relatively severe clinical courseI 41 .0008 and a poor recovery [29]. 11no ne study5 the documentation of' (BS high levels of' -gC antibodies recognizing GM Il in two pa All cases 1.7 131 4.2st0 tients with C. ftjitmi-associated GBS and extensive axonail Caises preceded hs damage suuested the involvement of' increased levels of' ( ,i,,in inflection' 0,17-0.S] 42 5-1 .2 7 these antibodies in pathogenesis [351. In another investigat Based on estimated U.S. population of 25 0 million. tion IgA antibodies to GM I vvere detected in 15 (28r) of'53 Estimated at I)0%-301, of all (iRS cases. patients with CBS and in only 1( 41() of 26 controls: the (iBS patients with antibodies to GM I had more extensive axonal involvement and more severe clinical courses [541. Koski et routine cultures of' stool for (Thnpi~obacier species should al. fiound that titers ot1gM antibodies to a neutral glveolipid beoeprofteitalvluinofainswthu- in peripheral nerve myelin were elevated in all of' 12 patients pected GBS. Further studies must elucidate the role of' C. with GBS but in none of 36 controls: the antibodies cross- Je.ju.itn in initiating CBS and illuminate the mechanisms in- reacted with Forssman antigen as well as with eukaryotic cell volved in pathogenesis. membranes [55]. However. earlier studies by Ilyas et al. showed no increase in levels of antibodies to GM I in associa- tion with CBS 1561. Furthermore. Yuki and colleagues re- References ,ported that antibodies to GM I were elevated in patients with 1 h ulanBrýS-vnrieSuv(ru.Pampeei n ct severe axnl0.adnoeiec f precedin- C. Jlflni(iuillain-Barrý ssndrome. Neurology 1985:35 1096-104. inf'ection [57]. In another investigation of 100 patients with 2. Briscoe D3M. McM. enamin JB. tYDanohac NV. Prognosis in (juillain- GBS. only 7 had anti bodies to peripheral nerve mvelin and Barrý, ssndrome. Arch tDts hild 1987:62:733-5. none had antibodies to galactocerebroside: howevesrm 3 Kennedy RH. tDanielson MA. Mulder l)W. Kurland LT. 6uillain- samples were collected up to 8 weeks after onset of~neurologi- Br~snrm:a4-ereieilgcadciia td.M~ cal symptoms 1581. 4. H(u hhn sPRrocC 1 978:u53la:9n3-B9.r snrm.Ldo:Sige-rag The exact contribution of these factors to the pathogenesis u1h990A. (u~anBr~snrm.Lno:Srne-el ol'(Jcjuni-associated CBS has not been determined. Never- 5. Blaser MJ. Reller LB. (asnpi-lohoiicr enteritis. N Eng'l J Med theless. clinical reports. serological surveys, and isolations of 1981:305:1444-52. C. iqn from series of painshave made it clear that a 6. Butzler.!P. Skirrow MB. mhss cretrtslinical C (jastroenter- subtania fpartpierinos o BarpecddbC.ologý 1979:8:737-65. subtaniafl pcrosoerstoonf BS re recdedby' 7. Pitkanen V. Ponka A. Pettersson 1. Kosunen TU. Campijuihbioier en- Jtjuni infection. and it is likely that this inflection plays a role teritis in 188 hospitalized patients. Arch Intern Med 1983:143:215- in the initiation of'demyel ination. 'Toward the goals ol'better 9). defining this conjunction and developing specific therapies. 8. Osler W. The principles and practice af medicine. Is t ed. New No rk: Appleton. 1892. 9. L~eneman FZ The (iuillain-Barr& ssndrome: definition, etialogs. and review at 1.1001e ases. Arch Intern Med 1966:118:139-44. 101. Melnick SC'. Hlewett TH. Role of intection in the (iuillain-Barrý ssn- Table 3. Scrolooical evidence of infection due to CJ.]euni amongdrm.IerilNusrgPchtr194239-(. patients with Culillain-Barrý syndrome (CBS) or CBS-like illness. ________________________________________________ 11I. Larsen itP. Kvale Gi. N'v land H. EpidemiologN of the (iuillain-Barrý syndrome in the count\ of Hordaland. western Norwas. Acta Neural Location No. ait' Percentage S.cn 957:37 ofstudvpatient with psitive12-. WineriB. Hughes RA(. Anderson Mi. Jones D)M. Kangro H akn I reeitrhe nieBl S seroagicl tst AsayRPIF. A prospectiv stud\v atacute idiopathic neuropath). 11. Ante- Unitd S7 ttes 28] 18Immuodotcedent events. J Netirol Neurosurg Psychtatry 1988:51:61 3-8. L ngland 1121 99 14 Complement 13. Kaur U. (hopra JS. Prahhakar S. Radhakrishnan K. Rana S. (juillain- fixatioin Barr.ý svndrome: a clinical. clectrophssiloigical and hiochemical Australia 1291 56 38 ELISA study. Acta Neuriil Scand 1986.73:394-4o2. China 1301l 32 41 LtISA 14. Tr ahelsi M. Mokrani R. Beninaceur B. Pok radiculon~vrites aigu6s de United States 1441 118 36 L L.ISA 1'enfant: A propcis de 71 cas. Pediatric 1989:44:413-8. 15. Cerehros-(iarctit M. Itorres-Mohedas J. Simon de las Heras R. Mateas- 108 Mishu and Bla,,er CID 1993:17 (July) Bcato F. Sanchei-lDia,' J1. Guillain Barriý sv ndrorne: I5 pediatric 39. Sosilla JY. Regli F. Franctoli PB. Guillain-Barrý ssndrome lolIlowing cases. Anl lsp Pedjair 1990:33:1201-3. ( umihhcrnn neii:report of three cases and res iew of the lI& Ismael S. Gufflain-Barrý' sv ndtomc in children: clinical manifestations, literature. Arch Intern Med 1988:148:739-4L. laboratorN findings. atnd the results of steroid treatment. Pacdiatr 40, Speed BR. Kaldor J. Watson J, et al. ( umipi/hoibcir Ityum/ Indones 1990:3t1:79-9ts. ( Coipi-lohicl' co/i-associated (juillain-Barre sx ndrome. immuno- 17. 1)owlin, PC, Cook SI). Role ol'infection in (iuillain-Barrý svn,!rome: Nlot confirmation of' the serological response Med J Aust lahoratory confirmation ot' herpcs% iruses in 41 cases- Ann Neurol 1987:147:13-6- 1981 :9bsuppl):44-SS. 4 1. Wroc Si. Blumhardt 1.1). Acute pol 'vneurits, with cranial nerve insolve' IS. Sanders EACM. Peters. AB Oratana JW. Hutghes RAC'. (juillain- ment fo~llowing ( *iipii/oýihoiticrjcjunii, infection [letter]. I Neurol Neu- Barr.ý svindrome after saricella-zoster infiection: report of two cases. J rosurg- Psychiatry 1985:48:593. Neurol 1987:234:437-9. 42. (laveloti 1P.B e~ tout J. Gourdiat A. (iarandeau A. Deffond D).T ournil- 19. Isukada N. Koh CS. Inoue A. Yanagisawa N. IDemyelinating neuropa- hac M. Atteintes neurologiques au d~cours d'infections a (aonihivl- thy associated with hepatitis B virus infection: detection of immune hociicr: cinq cas. Rev Neurol 1989:1I45:2(18-I 14. complex\es composed of' hepatitis B surtlice antigen. J Neurol Sci 43. Kohler A. de 'Iorrente A. Inderwildi B. Fisher's syndrome associated 1987:77:20)3-16. with Camnpihihoiiclr jciqin infection. Eur Neurol 1988:28:1 50-I1 0.Thorntorn C'A. Latif AS. Frmmandw tW. Gtiillain-Barrý syndrome asso- 44. Mishu B. Ilvas AA. Koski CL, et a]. Serologic evidence of' preceding ciated " ith human irinutnodeficienc.x viru~s infection in Zimbabwe. (uwoprlboau'rjeyioi infection in patients with Ciuillain-Barrý svn- Neurology 1991:41:812-S. drome. Ann Intern Med 1993 (in press). 21. vilaze D)G. Guillain-BarrK \sndrome. In: Feigin RD). (hem. JD. eds. 45. Svedhemn A. Kaijser B. Caimp h'hacwr /etu.s subspecies }itiini: a comn- Textbook of'pediatric infectious diseases. Vol 1. Philadelphia: WB mon cause of diarrhea in Sweden. I Infect IDis 1980;142:353-9. Saunders. 1992:464-74. 46. Nakanishi HI. Nukina M. Sakazaki R. Sero~groups of (ai~ipi/hobaierie.t 221. Skirrow MIB. (uiipihihoitr enteritis. a "new" disease. BMJ 1977:2:9- juni and C. co/i. Proceedings of the symposium of 100th annisersar-N II. of'Japaniese Societv of Veterinary Medicine. Veterinary medicine: its 23. Rhodes KMI. I attersfield AE. Guillain-Barrc svindrome associated with advance and view of' research [in Japanese]. Tokyo: Tokyo Press. Caipi-lohoiicri nfection. BMJ 1982:285:173-4. 1985:70-3. 24. Molnar GiK. Mertsola J. Erkko M. (iuiIlain-BarrCý syndrome associated 47. Blaser MI. [Duncan D)J. Human serum antibody response to ('uniprli- with (ooipilIioiacer infection [letterl. BN4J 1982:28i:652. baci'eijioii infection as measured in an enzyme-linked immunosor- 25ý. Constant (X'. Bentley ('C. Denman AM. Lehane JR. Larson HE. The bent assay. Infe~ct Immun 1984:44:292-8. G uil lain-B arrý s' ndrome following (aiirloi/ihiie' enteritis with re- 48. Tauxe RV. Epidemiology of ( niprIobacger jquioi infections in the covery after plasmapheresis. J Infect 1983:6:89-9 1. United States and other industrialized nations. In: Nachamkin 1. 26. Prvyor WM. Freiman JS. (iillies MIA. Tu ck RR. GuiIlain-Barrý syn- mlaser MI. Tompkins LS. eds. Guimpi'lohucierl qioii current status drome associated with ( upiii/ioibaiiir infection Iletter]. Ausi NZ J and future trends. Washington IX': American Society for Microbiol- Med 1984:14:687-8. ogv. 1992:9-19. 27. Speed B. Kaldor J, Cavanagh P. (iuillain-Barrý syndrome associated 49. Daikoku T. Kawaguchi M. Takama K. Suzuki S. Partial purification with (n/i//ccjimienteritis [letter]. J Infect 1984:8:85-6. and characterization of the enterotoxin produced by (oiipt/ohocu'r 28. (iruenewald R. Ropper All, Iior H. ('han J. Lee R. Molinaro VS. j'juiti. Infect Immun 1990;58:2414-9. Serologic evidence of (ompiipiobaciir lcjuiwlciwi enteritis in patients 5(1. Perez-1'erez GLi( 'ohn D)L. (luerrant RL. Patton ('M. Reller LB. Blaser with (idrJp dB.G.uillains-Bsadrro ýr hnye drome and91:48:1080-2. e Mi. (Clinical and immunologic significance of cholera-like toxin and 29K.a ld u I.r Spcd BR.(duylt nBarr syndome: nd88:1867-70.i Q/ cytotoxin productiI on by (Coiipvhiibacicr species in patients with eoao Viaiu -td.BI18:8:8771 acute inflammatory diarrhea in the LUSA. J Infect Dis 1989:160:460- 31). Blaser MI. Olivares A. Taylor D)N. 'ornblath D)R. McKhann GM. 8. (oopilohocft'r serology in patients with C htnese paralyttc syndrome 5I. Latovitzki N. Suciu-Fuica N. Penn AS. Olarte MR. (Chutorian AM. [letter]. L~ancet 1991:338:30)8. HLA typing and Guillain-Barrý syndrome. Nerlg 1979:29:743- 31. Ropper All. Camipi-lohoiu'r diarrhea and Guillain-Barrý syndrome. S 31 Arch Neurol 1988:45:65S-6. 52. Lain' N. Koski ('L. Walicke PA. Guillain-Barrý syndrome and paren- 3.Kuroki S. Haruta T. Yoshioka M. Koba'.ashi Y. Nukina M. Nakanishi trlgnloie lte] act19:3:5 tion Pedlin-atrInec l~synrm associated9- wih. mroatrif 53. ('arpo MI. Nobile-Orazio E. Meucci N Scarlato G. Anti-GiM I IgG tio. J 19P1d:i1tr: 19-ý1n.facncttiDboi dies in Guillain-Barrý snrm.Cin Neuropathol 1991: 33. Fujimoto S. Yuki N. Itoh T. Amakii K. Specific serotype of (uotpilo- s1ndome46. buiu'rjcjoni associated with (iuillain-Barr(ý synd rome [letter]. J In- 016 fieet IDis 1992:165:183. 54. Ilyas AA. Mithen FA. C~hen ZW. C'ook SD. Anti-GM I IgA antibodies 34. Fujimoto S. Amako K. (iuillain-Barrý snrmand (onprottibacter in Guillain-Barrý syndrome. J Neuroimmunol 1992:36:69-76. jquiii infection [letter]. Lancet 1990:335:1350. 55. Koski CL. C'hou l)KH. Jungalwala FB. Anti-peripheral nerve myelin 35. Yuki N. Yoshino H,. Sato S,.iatk T. Acute axonal polvneuropathy antibodies in Guillain-Barrý vdoebn a neutral glycolipid of associated with anti-GM I antibodies following C'unipi/lohocier enter- peripheral myelin and cross-react with Forssman antigen. J ('in In- itits. Neurology 1990:40:1900-2. vest 1989:84:280-7. 36. Yuki N. Sato S. Itohl T. Mivatake T. HLA-B335 and acute axonal poly- 56. Ilyas AA. Willison HI. Quarles RH. et al. Serum antibodies to ganglio- neuropathv following Ceont,,tlobacteri nfection. Neurology 1991:4'1: sides in Guillain-Barrý syndrome. Ann Neural 19W823:440-7. 156 1-3. 57. Yuki N. Yoshino H. Sato S, Shinozawa K. Miyatake T. Severe acute 37. DeBont B. Matthews N. Abbott K. Davidson GP. Guillain-Barri- syn- axonal form of Gufl~ain-Barrý syndrome associated with lgG anti- drome associated with ('oopr/ohoiactr enteritis in a child. J Pediatr GDIa antibodies. Muscle Nerve 1992:15:899-903, 1986:10)9:660-2. 58. Winer JB . Gray IA. Gregson NA. et al. A prospective study of acute 38. Kohler PC. G;oldblatt 1). (iuillain-Barrý syndrome following ('uoipi'o- idiopathic neuropathy. 111. Immunological studies. I Neurol Neuro- bacterjeiqui enteritis Iletteri. Arch Neurol 1987:44:1219. surg Psychiatry 1988:51:619-25.