ANTI-GM2 ANTIBODIES IN PATIENTS WITH GUILLAIN-BARRÉ SYNDROME CROSS-REACT WITH CYTOMEGALOVIRUS
Ang C.W., Jacobs B.C., Van Doorn P.A., Brandenburg A.F., Laman J.D., Osterhaus A.D.M.E., De Klerk, M.A., Van der Meché F.G.A. Departments of Neurology, Virology and Immunology, Erasmus University Medical Centre Rotterdam, The Netherlands
In the Guillain-Barré syndrome, patients with an antecedent infection with cytomegalovirus (CMV) frequently have antibodies to the ganglioside GM2. In the present study, we sought evidence for the hypothesis that these anti-GM2 antibodies arise through molecular mimicry between GM2 and CMV. Incubation of three anti-GM2 containing serum samples with CMV-infected fibroblasts resulted in a dose dependent inhibition of anti-GM2 reactivity. This effect was observed not only with a CMV strain isolated from a GBS patient but also with a CMV reference strain. Uninfected fibroblasts and fibroblasts that were infected with varicella zoster virus and herpes simplex virus type I had no or only a slight effect on anti-GM2 reactivity. Incubation of anti-GQlb containing serum samples with CMV-infected fibroblasts did not reduce antiGQlb reactivity. In conclusion, we found that CMV-antigens cross-react with anti-GM2 antibodies in Guillain-Barré patient. These findings indicate that molecular mimicry may play a role in the induction of anti-GM2 antibodies in CMV-associated GBS.
MYELINATED NERVE CONDUCTION BLOCK PRODUCED BY ANTIGANGLIOSIDE ANTIBODIES IN VITRO
Arasaki K.(1), Kusunoki S.(2), Kudo N.(3), Tamaki M.(1) (1)NTT Kanto Teishin Hospital, Tokyo 141-8625, Japan, (2)University of Tokyo, Tokyo 113-0033, Japan (3)University of Tsukuba, Ibaraki 305-0001, Japan
In spite of the well-documented association between inflammatory demyelinating polyneuropathies (IDPs) and antiganglioside antibodies in the patients' sera, the mechanism by which antiganglioside antibodies may cause the IDPs is unknown. Conduction block of myelinated nerve impulses is a clinical feature suggestive of myelin damage in the IDPs. We studied the role of antiganglioside antibodies in the pathogenesis of conduction block using an in vitro preparation of the rat sciatic-tibial nerve. The patients' sera with high titers of the antibodies were applied to a restricted segment of the sciatic-tibial nerve moun-ted in a recording chamber, and we examined effects of the sera on the compound nerve action potentials of the myelinated and unmyelinated nerve fibers. Serum samples containing IgM antibodies reacting with the terminal disaccharide of galactose (ß1-3) N-acetylgalactosamine most often blocked myelinated nerve conduction at the segment within a few hours, and coexistence of active complements was its prerequisite. The unmyelinated fiber conduction at the same segment, however, remained unchanged. These findings suggest that the antiganglioside antibodies damage the myelin sheath with the aid of complements and result in IDPs with myelinated nerve conduction block.
ANTEROGRADE TRANSPORT OF HSV-1 IN CULTURED HUMAN AND RAT DORSAL ROOT GANGLIA
Miranda-Saksena M.(1), Armati P.J.(2), Boadle R.A.(1), Holland D.J.(1), Cunningham A.L.(1) (1)Westmead Hospital and Westmead Institutes of Health Research, Westmead, NSW 2050, (2)University of Sydney, NSW 2006, Australia.
Herpes simplex virus (HSV-l) infects dorsal root ganglion neurons resulting in 'cold sores' and herpetic pain. The mechanism of viral transport in neurons is undefined. This study examines the mode of anterograde viral transport in cultured human and rat DRG. Neurons were infected with HSV-1 and the distribution of capsid, tegument and glycoprotein antigens detected by confocal microscopy employing antibodies to VP5, VP16 and gC/gB respectively. Cells were examined at 2,6,10,13,17 and 24 hours post infection (HPI) plus or minus the microtubule depolymeriser nocodazol or the golgi inhibitor brefeldin A (BFA). VP5 was detected in the nucleus nuclear membrane and cell body at 10HPI, the axon hillock at 13HPI and in the axon 15-17HPI. GC/cB were first detected in the cell body and axon hillock at 10HPI and the axon at 13HPI, always preceding VP5. VP16 was first detected in the cell body at 10-13 HPI and the axon at 16-17HPI. Nocodazole inhibited transport of all the antigens and the kinetics of VP5 could be dissociated from that of gC/gB. BFA inhibited transport into the axon of gC/gB and VP16 but not VP5. Transmission immuno-electronmicroscopy confirmed that unenveloped nucleocapsids were transported into the axon in BFA treated and control cultures. Nocodazole blocked such transport. These findings show that there is separate anterograde transport of HSV-l nucleocapsid and envelope and nucleocapsid/tegument transport in the absence of VP16. These data have important implications for designing therapeutic strategies to block viral transport.
VARIED FIBER DISTRIBUTION IN THE SURAL NERVE OF CIDP
Baba M., Miura H., Matsunaga M. Department of Neurology, Hirosaki University School of Medicine, Hirosaki, Japan
CIDP is acquired neuropathy characterized by multifocal demyelinating lesions in the peripheral nerves. There have been little morphometrical studies of biopsied nerves. We examined transverse sections of the sural nerves taken from 9 patients with idiopathic CIDP and 4 control subjects; number of myelinated fibers, external fiber diameter, axonal diameter and g-ratio were measured. Fiber distribution pattern in the sampled fascicles and flames was evaluated mathematically utilizing the coefficient of variation (CV) and index of dispersion (ID). Six patients showed decrease in density of myelinated fibres; the CD and the ID were significantly high for total myelinated fibers in 3 patients, for large fibers (external diameter >7 mm) in 5 patients compared with control subjects, which indicates the fiber distribution pattern is varied between fascicles. Markedly-varied distribution of thinly-myelinated large fibers (external diameter >7mm, g-ratio>0.7) among fascicles were seen in 5 patients. In conclusion, in addition to varied distribution of thinly-myelinated fibers among fascicles, varied density of myelinated fibers among fascicles is common in CIDP; multifocal inflammatory lesions in some fascicles just proximal to the examined part of the nerve would be responsible for the results.
LONG-TERM FOLLOW-UP OF CHILDREN WITH CHRONIC RELAPSING POLYNEUROPATHY
Barisic N. Department of Pediatrics, Medical School, Zagreb, Croatia
Relapsing course of chronic demyelinating polyneuropathy (CIDP) is more common in children than in adults. Long-term follow-up of 3 children with onset of relapsing CIDP at the age from 15 months to 4 years is presented. Pronounced slowing of motor conduction velocity (5-28 m/s) and conduction block of more than 75% were registered in two boys. Complete conduction block and neuromyotonia were obtained in a boy at the age of 3 year, which completely disappeared on intravenous immunoglobulin (IVIg) treatment. Sural nerve biopsy revealed signs of segmental demyelination and remyelinating onion bulbs in all three patients. Eight severe relapsing episodes of CIDP were observed in a boy with Arylsulphatase A (ASA) pseudodeficiency during 9 years of follow up. Complete remission occurred after introduction of cyclosporin (5 mg/ kg). He had no relapses for 4 years without medication. Two other boys manifested only one mild relapsing episode treated with fluocortolon, which was gradually completely tapered off. Improvement of motor conduction velocity (38-44 m/s) and of CMAP amplitude (1,5-3 mV) respectively was registered in 2 boys during 3 years of follow up. In the boy with several relapses of disease and ASA pseudodeficiency the motor conduction remained very slow (8-15 m/s) during 9 years of follow up, while CMAP amplitude showed mild increasing up to 1 mV. There is a positive correlation between electrophysiological abnormalities and neurological dysfunction in children with relapsing CIDP. Clinical outcome is excellent although electrodiagnostic signs of pronounced demyelination or axonal damage might remain unchanged during long-term follow-up.
ANTI-GM1 IgM ANTIBODIES IN MULTIFOCAL MOTOR NEUROPATHY: SLIGHTLY IMPROVED DETECTION BY COVALINK ELISA TECHNIQUE
Carpo M., Allaria S., Scarlato G., Nobile-Orazio E. Neuroimmunology Service, Institute of Clinical Neurology, Milan University, IRCCS Ospedale Maggiore Policlinico, Milan, Italy
IgM anti-GM1 antibodies have been associated with multifocal motor neuropathy (MMN) where they are found, according to different laboratories and techniques, in 20% to 80% of patients. Pestronk et al. (1997) recently reported a Covalink ELISA technique, based on the covalent binding of GM1 to secondary amino groups of ELISA plates, which allowed to reveal these antibodies in 85% of MMN patients but none with amyotrophic lateral sclerosis (ALS), chronic inflammatory demyelinating polyneuropathy (CI-DP) or other chronic neuropathies. To verify whether this technique may improve the identification of MMN patients, we compared the sensitivity, specificity and positive predictive value for MMN of anti-GM1 testing by our standard ELISA and Covalink ELISA techniques in 16 patients with MMN, 50 with other motor neuron/neuropathy syndromes (OMNS) including 15 with ALS, 14 with lower motor neuron disease (LMND), 21 with CIDP and in 23 normal controls. High anti-GM1 IgM were detected by standard (>1/320) and Covalink ELISA (>1/4000) in 25 and 37% respectively of MMN and in 10 and 6% respectively of OMNS including 19 and 16% of CIDP and 7% and none of LMND. The positive predictive value of these antibodies for MMN (i.e. the proportion of GM 1 -positive patients having indeed MMN) was 44 and 62% respectively with the two techniques. Even if the Covalink ELISA technique slightly improved the detection of anti-GM1 antibodies in MMN it did not permit to identify the majority of them.
DOES INTRAMYELINIC EDEMA PLAY A ROLE IN CHRONIC DEMYELINATING NEUROPATHY ASSOCIATED WITH IGM PARAPROTEINEMIA?
De Armas L., Sabatelli M., Mignogna T., Lippi G., Madia F., Mereu M.L., Tonali P. Istituto di Neurologia. Università Cattolica. Rome, Italy.
Intramyelinic edema has received little consideration in the literature as possible elementary pathologic lesion in chronic immunomediated demyelinating polyneuropathies. Intramyelinic edema is a prominent feature in EAN of chickens induced by passive transfer of human IgM anti-MAG (Tatum, 1993; Ann Neurol 33: 502). In human nerves this change has been described in CIDP (Sabatelli et al. Clin Neuropathol 1996, 1:17) but never in IgM paraproteinemic neuropathy. We describe a 64-year-old man affected by par-esthesias and numbness with glove-and-stocking distribution and slight distal weakness in his limbs since 5 years. Electrophysiological studies showed reduced conduction velocities with markedly increased distal motor latencies. Serum immuno-electrophoresis disclosed an IgM paraproteinemia without evidence of Waldenström disease. Sural nerve biopsy showed a reduced density of nerve fibers and segmental de-remyelination. Occasional fibers showed intramyelinic edema causing marked swelling of the fiber and axonal constriction. We suggest that edema of the myelin sheath may play a crucial role in the pathogenesis of demyelinating neuropathies.
DIPHTHERITIC POLYNEUROPATHY IN LATVIA: A CLINICAL STUDY OF 50 PATIENTS
Logina I., Donaghy M. Latvian Academy of Medicine, Riga and University of Oxford, UK
Diphtheria polyneuropathy (DP) returned to Eastern Europe in the mid-1990s because immunity falls in adults following vaccination in childhood. The Latvian epidemic which peaked in 1994-96 mainly affected urban dwellers (84%) aged 41-60 (80%) causing an estimated paralysis rate of 15.2%. 50 DP adults were compared with 21 GBS patients. 98% of DP was of bulbar onset, and 10% never developed limb symptoms. Whereas GBS reached peak severity at a median of 10d, with improvement commencing by 21d, DP evolved more slowly, cresting by 49d with improvement after 73d. A novel observation was of secondary deterioration in 66% of DP patients which required 4 patients to be ventilated for the first time, and we attribute it to systemic dissemination of toxin, following initially localised bulbar toxicity. 16% of DP patients died, not from neuropathic respiratory failure, but due to cardiomyopathy or other organ failure. Cardiovascular effects of diphtheritic cardiomyopathy could not be differentiated reliably from acute autonomic instability, itself a common feature of GBS. Motor conduction slowing was similar in DP and GBS. All DP survivors regained independent bulbar and respiratory function by l yr, but 6% remained unable to walk independently. No deaths occurred in 5 DP patients receiving antitoxin on days 1 or 2 of infection whereas death occurred in 18% of those receiving antitoxin on days 3-6, or not at all. DP is differentiated from GBS at the time of presentation by palatal pseudomembrane or bull neck, by bulbar onset with normal limbs and by associated internal organ involvement.
PROGNOSIS OF GUILLAIN-BARRÉ SYNDROME.
Elazouni O.M., Katifi H., Sedgwick E.M. Clinical Neurosciences, Southampton University, Southampton, UK
Guillain-Barré syndrome (GBS) is a complex syndrome in which axonal degeneration could herald a poor recovery, 55 GBS patients aged 17-82 years (54.51±17.37) were assessed up to 24 months (12.21±9.42) and classed by Hughes' disability score (Hughes et al. Lancet 1978; 2 : 750) as good (0-1) or poor (> 2) recovery. Good recovery occurred in 34, poor in 18, and 3 patients died. Clinical predictors of poor prognosis were age, disability score of 4 at nadir, short time to nadir, long nadir time, diarrhoea before onset, early muscle wasting (4 weeks), and admission to ICU for more than 2 weeks but respiratory and autonomic involvement in isolation did not influence prognosis. Mean recovery time was 116 days for demyelinating, 402 for axonal degeneration, and 364 for axonal and demyelinating group. Early denervation had little effect on prognosis at 2 years but prolonged the recovery time (250 vs 100) days, P<0.005. Low CMAP and SNAP amplitudes were accompanied by bad prognosis. CMAP amplitude of less than 25% of lower limit of normal (LLN) was a bad prognostic factor. Absent F-response in two or more nerves also predicted poor prognosis rather than prolonged or absent response in only one nerve. We conclude that there is no single reliable prognostic factor for GBS but prognosis should be assessed by multiple factors. Early reduction of CMAP and SNAP should not be conceived as predicting poor outcome unless they are less than 25% LLN.
CHEMOKINE mRNA EXPRESSION IN THE PERIPHERAL NERVES OF RATS DURING EXPERIMENTAL ALLERGIC NEURITIS
Fujioka T., Purev E., Rostami A. Department of Neurology, University of Pennsylvania, Philadelphia, PA USA.
We analyzed a sequential expression of several chemokine messages, interferon-g inducible protein (IP)-10, monocyte chemotactic protein (MCP)-1, macrophage inflammatory protein (MIP)-1a, the regulated upon activation normal T cell expressed and secreted chemokine (RAN-TES), and lymphotactin in the cauda equina (CE) of rats with experimental allergic neuritis (EAN) on days 0 (pre-immunization), 7 (preclinical stage), 10 (disease onset), 13 (clinical progression), 17 (disease peak), as well as 20, 24, and 34 post immunization (p.i.; recovery) using a quantitative competitive reverse transcriptase PCR method. MCP-1 message increased at preclinical stage and peaked at day 17 p.i. The increase in early stage was not detected in other tissues indicating the peripheral nerve-specific upregulation. MIP-1a and IP-10 messages surged at day 13, returned to low in recovery stage. RANTES message also increased at day 13 and peaked at day 17 p.i., however, unlike other chemokines, it showed second peak of expression on day 24. Lymphotactin message was undetectable at any time point. MCP-1 protein was detected immunohistologically in endothelia, macrophages, and Schwann cells in perivascular infiltration foci. These data suggest an important role of MCP-1 in initiating the EAN lesions and pro-inflammatory role of MIP-1a, RANTES and IP-10. Moreover unique yet undetermined role of RANTES for the recovery from EAN was speculated.
THE IMMUNISATION OF LABORATORY ANIMALS WITH CAMPYLOBACTER JEJUNI ISOLATES FROM GBS PATIENTS.
Gregson N.A., Linton D., Rees J.R., Hughes R. A.C. Section of Neurology, Department of Clinical Neurosciences, GKT Medical School,King's College, Guy's Hospital, London SE1 9RT.
Rabbits and mice have been immunized with preparations of Campylobacter
jejuni from GBS patients and enteritis controls. Immunisation
of both rabbits and mice produces antibody responses against bacterial
protein and LPS antigens. Acutely no animal developed signs of
neurological impairment nor were there any morphological signs
of nerve damage. Mice immunized with GBS isolates failed to gain
weight to the same extent as those with control isolates. Animals
chronically immunized with whole bacteria although appearing neurologically
normal showed morphological changes compatible with altered axonal
transport. Similar changes were not seen after immunisation with
LPS alone.
ATTEMPT TO INDUCE EXPERMENTAL AUTOIMMUNE NEUROPATHY BY SENSITIZATION WITH A GANGLIOSIDE, GALNAC-GD1A
Kaida K., Kusunoki S.(1), Kamakura K., Murayama S.(1), Kanazawa I.(1) National Defense Medical College, Tokorozawa, Saitama, JAPAN, (1)University of Tokyo, Bunkyo-ku, Tokyo, JAPAN
To investigate a pathogenic role of anti-GalNAc-GDla antibody in Guillain-Barré syndrome, we attempted to induce experimental autoimmune neuropathy by immunizing 3 rabbits with purified Ga1NAc-GDla. IgG anti-GalNAc-GDla antibody was raised in the sera of 2 rabbits (the titer; 1:51200 and 1:1600, normal range < 1:100). No rabbits revealed apparent clinical symptoms. Pathological examinations of the peripheral nerves were performed 6 months after first sensitization, showing slight loss of myelin in the ventral root fibers in the rabbit with higher IgG antiGalNAc-GDla antibody titer. Other rabbits had no pathological changes. Elevation of IgG anti-GalNAc-GDla antibody titer may be associated with demyelination of motor nerves in the rabbit. Quantity and localization of GaINAc-GDla on the nerves might influence the clinical symptoms and the intensity of pathologic changes in this study. The rabbit sera with anti-GalNAc-GDla antibody may be useful for investigating localization of GalNAc-GD1a in human peripheral nerves.
CLINICAL AND PATHOLOGICAL ASPECTS IN CHRONIC INFLAMMATORY DEMYELINATING POLYNEUROPATHY (CIDP) IN DIABETES
Kamijo M.(1), Baba M.(2), Ozeki N.(1), Yoshida Y.(1) (1)Aichi Medical University, Nagoya, (2)Hirosaki University School of Medicine, Hirosaki, Japan
Recently, CIDP occurring in diabetics has been reported and recognized as an important differential diagnosis, since CIDP is treatable. We report 5 patients of CIDP with diabetes (CD) comparing with their clinical and pathological findings to 5 age-matched CIDP patients (C). All 10 patients fulfilled the criteria for CIDP. CD patients had been suffered from type II diabetes (mean DM duration 12.7+-5.8 years). Four of 5 in CD had a successful effect to immune therapy. In C, all patients had motor and sensory symptoms; three of 5 had a good effect to immune therapies. In biopsied sural nerve, the mean myelinated fiber density of CD decreased significantly (C5927± 689/mm2 vs. CD2975±499, P<0.01). The variability of myelinated fiber density within fascicles was greater in C compared with CD. Thin myelinated fibers and small fiber clusters were more frequently found in C. In contrast, regenerated changes were less often and the axonal atrophy was frequently found in CD. Electron microscopy showed thickening of the basement membrane and endothelial cell proliferation of the endoneurial vessels in CD. In conclusion, despite the similarity of clinical aspects, the pathological findings were different between C and CD. Nerve regeneration might be suppressed in CIDP with diabetes.
ANTECEDENT SYMPTOMS IN GUILLAIN-BARRÉ SYNDROME: RELATIONSHIP TO CLINICAL AND SEROLOGICAL FEATURES.
Koga M., Yuki N., Hirata K. Dokkyo University School of Medicine, Tochigi, Japan
PURPOSE: To examine whether Guillain-Barré syndrome (GBS) could be classified into subgroup based on the prior symptoms. MATERIALS AND METHODS: Infectious symp-toms were reviewed in GBS (n=119) and its related disorders (n=57). Statistical analysis was done on the association of prior symptoms with infection serology (Campylobacter jejuni, cytomegalovirus, Epstein-Barr virus and Mycoplasma pneunoniae), anti-ganglioside antibodies and neurological findings. RESULTS: Frequent antecedent symptoms in GBS were fever (52%), cough (40%), sore throat (31%) and diarrhea (31%). The patients with sore throat or cough frequently showed opthalmoparesis and had IgG anti-GQlb antibody. Fever and headache respectively were associated with bulbar and facial palsy. The patients with diarrhea often had anti-ganglioside IgG (GM1 and GMlb) and IgM (GM1, GMlb and GalNAc-GDla) antibodies and rarely showed opthalmoparesis and bulbar palsy. No prior event was found in 12% of the patients, who rarely showed facial and bulbar palsy and often presented dysautonomia. Diarrhea was closely associated with C. jejuni serology, whereas other symptoms were not related to any pathogens. CONCLUSION: GBS with prior sore throat, cough, fever, headache or diarrhea respectively forms clinical and/or serological subgroups, not necessarily dependent on infection by frequent trigger pathogens in GBS.
ANTIBODY AGAINST A PERIPHERAL NERVE MYELIN-SPECIFIC GANGLIOSIDE LM1 IN GUILLAIN-BARRÉ SYNDROME
Kusunoki S., Yako K., Miyazaki T., Kanazawa I. University of Tokyo, Tokyo 113-8655, Japan.
Antiganglioside antibody titers are frequently elevated in the acute phase sera of Guillain-Barré syndrome (GBS). LM1 is the predominant ganglioside in human peripheral nerve myelin. We investigated the presence of anti-LM1 IgG antibody in GBS sera and the clinical and electrophysiological features of GBS patients with this antibody. Anti-LM1 IgG antibody was found in 7 of 140 patients with GBS in the acute phase, 2 of 47 with Miller Fisher syndrome, and 1 of 33 with chronic inflammatory demyelinating polyneuropathy. No anti-LM1 IgG antibody was detected in normal or other disease controls. Among the 7 patients with anti-LM1 IgG antibody, 6 patients recovered to clinical grade 1 (Hughes et al. 1978) within 1 month of the onset of neuropathy. Electrophysiological studies revealed demyelination in 5, of which 1 had axonal damage in addition, whereas sufficient evidence of demyelination or axonal degeneration was not observed in the remaining 2. Five had antecedent respiratory tract infection and also serum anti-GQlb IgG antibody. Four of those 5 had opthalmoplegia. Anti-LM1 IgG antibody might be involved in the pathogenetic mechanisms of GBS as a possible demyelinating factor. Presence of both antiGQ1b and anti-LM1 antibodies may be associated with some infectious agent(s) affecting the respiratory tract.
INFLAMMATORY NEUROPATHIES: CHARACTERIZATION OF SERUM IMMUNE RESPONSE TO SCHWANN CELLS, NERVE AND RECOMBINANT MYELIN COMPONENTS
Kwa M.S.G.(1), De Vlieger M.(1), Van Schaik I.N.(2) , Baas F.(1), Brand A.(1), Vermeulen M.(2) (1)LUMC, Immunohematology, Leiden, the Netherlands, (2)AMC, Neurology, Amsterdam, the Netherlands
Guillain-Barré Syndrome (GBS), chronic inflammatory demyelinating polyneuropathy (CIDP) and multifocal motor neuropathy (MMN) are thought to be caused by an auto-immune reaction to myelin components leading to myelin breakdown and nerve conduction block. We used in vitro cultured human Schwann cells to characterize candidate epitopes involved in demyelinating neuropathies. Eleven primary Schwann cell cultures were established from nervus suralis biopsies. Cells were screened for the presence of: 1) a panel of immunological markers and 2) epitopes recognized by sera from CIDP and GBS patients. We also tested serum immune response to known myelin components: myelin protein zero (P0), per-ipheral myelin protein 22 (PMP22) and connexin 32 (Cx32). Recombinant glycoproteins were produced in mammalian (CHO-K1) cells, using a transient expression vector based on Semliki Forest Virus (pSFV). Western blot analysis of a small panel of patient sera, showed so far no immunoreactivity towards recombinant P0, PMP22 and Cx32. However, several CIDP and GBS sera showed specific immunoreactive (lgM/G) bands in cultured human Schwann cells and nerve tissue. Immunoreactivity of patient sera towards cultured Schwann cells was also observed using immuno-fluorescence microscopy.
INHIBITION OF INDUCIBLE NITRIC OXIDE SYNTHASE REDUCES THE EFFECTS OF NUCLEUS PULPOSUS ON SPINAL NERVE
Larsson K., Brisby H., Byröd G., Aoki Y., Rydevik B., Olmarker K. Dept of Orthopaedics, Göteborg University, Göteborg, Sweden.
The aim of this study was to investigate if systemic treatment with aminoguanidine (AG), a selective NOS inhibitor, can reduce the negative effects on vascular permeability and nerve conduction velocity (NCV), previously observed after nucleus pulposus (NP) exposure to spinal nerve roots. Two sets of experiments were performed in the pig. 1. S2 and S3 nerve roots were exposed unilaterally for autologous NP. In the AG group (n=7), aminoguanidine (20 mg/kg in 50 ml NaCl) was injected intravenously before NP application. The control group (n=8) received the same amount of NaCl. The animals were kept anaesthetized for 2 hours, perfused with Evans blue solution and sacrificed. Five sections from each specimen were investigated by fluorescent microscopy and the edema in the nerve root was graded as none/slight, moderate or severe. 2. Spinal nerve roots at the level of cauda equina were exposed to autologous NP. In the AG group (n=5) aminoguanidine (20 mg/kg) was injected intravenously prior to application of NP and repeated intramuscularly the day after surgery. Seven animals served as controls. NCV over the exposed area were measured 7 days after surgery. 1. In the AG group, a severe spinal nerve root edema was observed in 12%, moderate edema in 38% and slight/none in 50% of the animals. In the control group the frequencies were 57%, 29 %, 14% respectively. 2. Nerve conduction velocity were significantly higher in the aminoguanidine treated animals than in control animals (71 ± 18 vs 48 ± 18 m/s, p <0.05 ). The results suggests that nitric oxide is involved in the pathophysiological effects of nucleus pulposus in disc herniation.
SCHWANN CELLS CAN STIMULATE RESTING P2 SPECIFIC T-CELLS
Lilje O., Armati P.J. University of Sydney, NSW, Australia
There is increasing evidence that Schwann cells are more than passive targets in inflammatory demyelinating diseases (IDNs) of the peripheral nervous system. These diseases include GBS and CIDP. To further study this possibility and to determine if Schwann cells can induce the proliferation of CD4+ cells raised against Schwann cell specific antigens, we exposed P2 specific, P2 peptide specific and Ovalbumen (OA) specific T-cell lines to replicate Schwann cell cultures in the presence or absence of Interferon gamma (IFNg). Unlike P2 and P2 peptide, OA is a non-mammalian antigen and OA specific T-cells and OA antigen were employed as a control treatment. Analysis of variance showed that Schwann cells exposed to IFNg induced significant P2 and P2 peptide T-cell proliferation in the absence of exogenous antigen whereas OA T cells proliferated only in the presence of exogenous OA. The experiments were carried out on live, fixed or irradiated cultures to determine if antigen was processed by the Schwann cells and if the MHC class 11 dependent T-cell lines recognised antigen processed prior to fixation or irradiation. The most significant proliferation occurred with live cells followed by irradiated and then fixed cells. This study indicates that endogenous Schwann cells antigens can be presented to antigen specific T-cells and that Schwann cells are antigen specific in their induction of proliferation as demonstrated by the lack of OA T-cell induction in the absence of OA. The results further demonstrate the ability of Schwann cells to participate in their own damage by immune system cells. Also activated T-cell cytokines could exacerbate blood nerve barrier breakdown - a factor increasingly implicated in IDNs.
DEVELOPMENT AND CLINIMETRIC EVALUATION OF A NEW SENSORY SCALE IN IMMUNUNE-MEDIATED POLYNEUROPATHIES: THE "INCAT" SENSORY SUMSCORE
Martina I.S.J., Schmitz P.I.M., van der Meché F.G.A., van Doorn P.A. The European Inflammatory Neuropathy Cause and Treatment (INCAT) group. Erasmus University Medical Centre Rotterdam The Netherlands
We investigated the clinimetric properties of a new sensory scale ("INCAT" sensory sumscore [ISS]) in immune-mediated polyneuropathies. The ISS comprises a vibration and pinprick sensation grade (tested at 8 locations in the arms and legs) plus a 2-point discrimination grade assessed at the 2nd digit (sumscore: 0 [normal] to 20 [maximal deficit]). A panel of neurologists with special interest in this field evaluated the ISS and felt that the scale has face and content validity. Consequently, other clinimetric properties of the ISS were assessed in 113 patients with a stable clinical condition (83 with Guillain-Barré syndrome (GBS), 22 with chronic inflammatory demyelinating polyneuropathy (CIDP) , 8 with a polyneuropathy associated with a gammopathy of undetermined significance). The scale was also longitudinally (40 weeks) utilised in 10 patients with recently diagnosed GBS or CIDP with changing clinical conditions. A disability sumscore (DSS) was also assessed in all patients. Good inter-/intra-observer reliabilities were demonstrated for the ISS (R=0.85-0.89) with significant correlation between this scale and the DSS (r= 0.50-0.83; p<0.0001). Predominantly good responsiveness was also obtained for the ISS. In conclusion, all essential clinimetric properties are provided for the "INCAT" sensory sumscore. The ISS seems to be a valuable instrument for the evaluation of sensory deficits in patients with immune-mediated polyneuropathies.
EFFECT OF IVIg ON ANTI-GLYCOLIPID REACTIVITY IN MULTIFOCAL MOTOR NEUROPATHY (MMN)
Meucci N., Terenghi F., Carpo M., Allaria S., Scarlato G., Nobile-Orazio E. Neuroimmunology Service, Inst. Clinical Neurology, Milan University, Ospedale Maggiore Policlinico, Milan, Italy
We evaluated the effect in vivo and in vitro of IVIg on anti-glycolipid reactivity in patients with MMN to clarify the possible mechanism of action of IVIg in these patients. Sera from 5 patients improving after IVIg and not receiving other immune therapies were tested by ELISA for anti- GM1, -GM2, -GDla reactivity before and after IVIg. Pre-treatment sera were also tested for anti-glycolipid reactivity after pre-incubation with increasing IVIg (1 to 10mg/ml) for 16 hours at 4°C, to evaluate the in vitro effect of IVIg on anti-glycolipid antibodies. Purified anti-glycolipid antibodies from positive sera were also pre-incubated with IVIg. Three patients had high anti-glycolipid antibodies including one with anti-GM2 IgM (1/327,680) and two with anti-GM1/ GM2 IgM (1/81,920 and 1/10,240 respectively in one and 1/20,480 for both in the other). Even if IgM antibody titers did not change after treatment in any patient, in two the optical density of anti-GM2 reactivity at the highest positive dilution decreased after IVIg. Pre-incubation with IVIg did not inhibit anti-glycolipid antibodies even if all reactivities tended to decrease at the highest IVIg concentrations. Similar results were observed with purified antibodies. In our patients with MMN and high anti-glycolipid IgM responding to IVIg, antibody reactivity was only marginally affected by IVIg both in vivo and in vitro, indicating that other mechanisms than the anti-idiotypic neutralization of anti-glycolipids antibodies may be involved in the improvement of MMN due to IVIg.
CLINICAL FEATURES OF GUILLAIN-BARRÉ SYNDROME PATIENTS WITH ANTIBODIES SPECIFIC TO GANGLIOSIDE GD1b
Miyazaki T., Kusunoki S., Kaida K.1, Kanazawa I. University of Tokyo, Tokyo 113-8655, Japan, 1National Defence Medical College, Saitama 359-0042, Japan
We investigated clinical features of patients with Guillain-Barré syndrome (GBS) with antibodies against ganglioside GD1b. Serum samples were obtained from several hospitals throughout Japan from December 1992 to June 1998. Antibody reactivities against ganglioside GM1, GM2, GM3, GDla, GDlb, GD3, GTlb, GQlb, GA1 and galactocerebroside were tested by ELISA. Clinical information was obtained by letters or specially designed questionnaire. Seventy-five of 379 GBS patients had anti-GDlb antibodies, and 22 of these 75 patients did not have anti-GM1 antibody activity. Within these 22 cases with anti-GDlb antibodies, only IgG class was seen in 20 cases, only IgM class in 1 and both classes in 1. Seven of 22 patients had only anti-GDlb IgG antibodies and did not react to other antigens shown above. All of these 7 patients had manifestation of upper airway tract infection before onset of disease, and clinically showed distal dominant muscle weakness and sensory disturbances. Three of them had electrophysiological findings of demyelination but no findings of axonal degeneration represented by denervation potential or significant reduction of CMAP. Localization of GDlb in the paranodal myelin and the primary sensory neurons of human peripheral nerve has been reported. Anti-GD1b antibodies may be involved in the pathophysiological mechanisms of demyelination and sensory disturbances in GBS by specifically binding to those regions where GDlb is localized.
POSSIBLE DETRIMENTAL EFFECT OF STEROID THERAPY IN DYSIMMUNE NEUROPATHIES
Nobile-Orazio E., Meucci N., Cappellari A., Carpo M., Di Troia A., Manfredini E., Scarlato G. Neuroimmunology Service, Institute of Clinical Neurology, Milan University, IRCCS Ospedale Maggiore Policlinico, Milan, Italy.
Some patients with multifocal motor neuropathy (MMN) were recently reported to deteriorate after steroid therapy. It is unclear whether this effect is restricted to MMN or involves other dysimmune neuropathies. Since 1990 we observed 6 patients (2 men and 4 women, aged 26-63 years, mean 49.7) with dysimmune neuropathies who worsened after steroids. Three patients had an asymmetric, step-wise, pure motor neuropathy lasting 2-25 years; nerve conduction studies (NCS) in two confirmed the diagnosis of MMN while no conduction block was detected in one; one patient had mildly increased CSF proteins. The other 3 patients had an asymmetric, step-wise or chronic progressive, predominantly motor (2) or sensory-motor neuropathy (1) lasting 0.5-10 years; in these patients NCS were consistent with a demyelinating neuropathy and CSF proteins were increased, supporting the diagnosis of CIDP. Of the three patients with pure motor neuropathy, one repeatedly worsened concomitantly with topical endonasal steroids for severe asthma, while the others, similarly to those with predominantly motor or sensory-motor neuropathy worsened 4-5 days after starting oral prednisone 50-125 mg/die. All patients subsequently improved with high-dose intravenous immunoglobulin (IVIg). This report draws attention on the possible untoward effect of steroids not only in MMN but also in other asymmetric predominantly or purely motor dysimmune neuropathies.
OXIDATIVE STRESS, INFLAMMATORY NEUTROPHIL, AND DEMYELINATION IN REPERFUSION NERVE INJURY
Nukada H., He K., Murphy M.P., McMorran P.D. University of Otago, Dunedin, New Zealand
We investigated the role of oxidative stress and inflammatory neutrophils and their role in the development of demyelination in reperfusion nerve injury. Biochemical and pathological studies of reperfused rat sciatic nerves were undertaken following severe 5-h ischaemia. Protein carbonyl content, measured by an ELISA assay, increased significantly by up to 1.5 fold after 6-12h reperfusion and carbonyl formation was blocked by allopurinol. Measurement of nitrotyrosine formation by immunoblotting of protein extracts showed that GFAP was the only major protein, situated on tyrosine residues. The amount of nitrated form of GFAP increased over 12-18h following reperfusion, suggesting that peroxynitrite contributes to reperfusion injury. At this stage of reperfusion, ICAM-1 expression on endothelial cells was rapidly increased, and PMN adhesion to the luminal surface of these endothelial cells was prominent. After 24h of reperfusion, the number of endoneurial PMNs (HIS48 immunohistochemistry) reached a peak. Subsequently monocytes accumulate and endoneurial mononuclear macrophage numbers (1C7 immunohistochemistry) elevated four fold at 48-72h of reperfusion. Demyelination then progressed by phagocytosis of myelin. We observed a direct invasion of phagocytic mononuclear cells into myelin lamellae. In conclusion, ROS formation and inflammatory neutrophil infiltration play a key role to initiate nerve lesion formation. This study also provided evidence of the common effector mechanism that induces inflammatory demyelination in immune-mediated neuropathies and reperfusion nerve injury.
ANTI-GQIB ANTIBODIES CAUSE MORPHOLOGICAL DESTRUCTION OF THE MOTOR NERVE TERMINAL
O'Hanlon G.M.(1), Goodyear C.S.(1,2), Plomp J.J.(3,4), Morrison I.(1), Wagner E.(1), Veitch J.(1), Conner J.(2), Molenaar P.C.(3), Willison H.J.(1) (1)Department of Neurology, University of Glasgow, (2) Department of Biological Sciences, Glasgow Caledonian University, UK & Departments of (3)Physiology, (4)Neurology, Leiden University Medical Centre, The Netherlands.
In our studies on Miller Fisher syndrome using the mouse phrenic nerve-hemidiaphragm preparation as a model system, we have demonstrated that human and murine anti-GQlb antisera and monoclonal antibodies cause a complement dependent latrotoxin-like block of neuromuscular transmission. Here we describe the pathological features at the neuromuscular junction (NMJ) in this experimental system. Hemidiaphragms were processed for immunofluorescence and ultrastructural analysis using standard methods. At the light microscopic level, extensive deposits of immunoglobulin and complement components C3c and membrane-attack complex were present at the NMJ, overlying both capping Schwann cells and the axon terminal. Using anti-neurofilament (NF) antibodies we observed a significant loss of NF immediately over the NMJ, but this did not extend into intramuscular nerve bundles, which were also free of immuno-deposits. At the EM level the nerve terminal was distended and depleted of synaptic vesicles. Schwann cell processes were encroaching the synaptic cleft. Similar pathological changes have been observed following treatment with latrotoxin. These studies demonstrate that clear immunopathological and ultrastructural defects occur in association with the electrophysiological lesion in this anti-GQ1b antibody-induced model of motor nerve terminal failure.
UNEXCITABLE SURAL NERVES IN ACQUIRED DEMYELINATING POLYNEUROPATHIES: A MORPHOMETRIC STUDY.
Said G., Guglielmi J.M., Planté V., Ropert A. Service de Neurologie, CHU de Bicêtre, Univ. Paris XI, France.
The amplitude of the sensory action potential (SAP), which
correlates well with the density of large myelinated fibers (MF),
is often used as a measure of fiber loss (Buchtal at al, 1984).
In some cases however, preservation of a large number of MF in
nerve biopsy specimens of unexcitable nerves comes as a surprise.
In order to learn more on these discrepancies we analyzed the
clinical and morphometric findings in a series of 114 patients
with acquired demyelinating polyneuropathy whose undetectable
sural nerve SAP contrasted with preservation of more than 3000
MF/mm(exp2) (N: 7630 ± 327 MF/mm(exp2)) in the sural nerve
biopsy specimen. All patients had a disabling polyneuropethy,
They had been referred to our center for a nerve biopsy. All underwent
a neurological examination, surface and near nerve needle measures
of sural nerve SAP and a biopsy of the unexcitable sural nerve,
under local anesthesia. Nine patients had a chronic inflammatory
demyelinating polyneuropathy, two a subacute demyelinating polyneuropothy.
Two patients had a benign, one a malignant IgM monoclonal gammopathy.
The total density of MF ranged from 3150 to 8240 per mm(exp2)
(5028 ± 1688 per mm(exp2)); that of MF> 8m
diameter were 1392 ± 813 per mm(exp2). On average, 22%
of the isolated fibers showed segmental abnormalities of the myelin
sheath, and 3% were undergoing axonal degeneration. One patient
had normal morphological findings coexisting with motor nerve
conduction blocks. When this patient recovered SAP gradually returned
to normal, which suggests the possibility of conduction blocks
on sensory nerves proximal to the site of nerve biopsy. Conclusion:
These findings show that absent SAP in the sural nerve is not
a reliable indicator of axon lose in patients with a demyelinating
neuropathy.
BLOOD gd T CELLS IN GUILLAIN-BARRÉ SYNDROME
Scelsa S.N.(1,2), Herskovitz S.(2), Ghali V.(1), Liau S.(1), MacGowan D.J.L.(1) Beth Israel Med Center,(1) NY, NY, Albert Einstein,(2) Bronx, NY, USA
gd T cells participate in the early microbial defense against various pathogens, including intestinal bacteria, are common in intestinal epithelial cells, and are elevated in several autoimmune diseases. We hypothesized that, in Guillain-Barré syndrome (GBS), gd T cells are activated in Campylobacter jejuni infection (Cj), resulting in an immune-mediated attack against peripheral ner-ves. We are prospectively studying whether gd T cells are elevated in peripheral blood (PB) of GBS patients (pts), and if elevations are associated with increased Cj or GM1 antibody titers. All 6 pts met established criteria for GBS and had neurologic symptoms less than 2 weeks (wks). Other con-founding causes of polyneuropathy excluded pts. Pts had PB studied at presentation and at 4 wks to determine the percentage of gd T cells (by flow cytometry), Vd1 and Vd2 subsets and Cj total antibody titers. GM1 titers were sent at presentation. Differences between the GBS group at presentation and controls (4 pts) were not significant in terms of percent gd T cells (mean, GBS vs controls, 2.6 vs 4.5, p=0.13), Vd1 (1.4 vs 2.2, p=0.31), Vdl/CD8 (median, 0.20 vs 0.3 5, p=0.9 1), Vd2 (mean, 1.3 vs 2.3, p=0.24), and Vd2/CD8 (median, 0.20 vs 0.20, p=1.0). Differences between the groups were also not significant at 4 wks- gd T cells (median, 4.1 vs 4.4, p=0.91), Vd1 (mean, 1.7 vs 2.2, p=0.67), Vdl/CD8 (0.80 vs 0.40, p=0.38), Vd2 (1.2 vs 2.3, p=0.16), and Vd2/CD8 (0.26 vs 0.20, p=0.70). Three of 6 pts had elevated GM1 titers. Cj titers were highly elevated in 1 pt with diarrhea and increased GM1 titers. Two (1 Cj+) of 3 pts with elevated GM1 titers had a >50% rise in the percentage of gd T cells (Vd1); the Vd1/CD8 subset was transiently elevated in the third pt. GBS does not appear to be associated with gd T cell elevations; further work is needed to assess if T cells are elevated in GBS pts with Cj and increased GM1 titers.
FINE SPECIFICITY OF IgG ANTI-GQlb ANTIBODIES
Susuki K., Yuki N., Odaka M., Koga M., Hirata K. Dokkyo University School of Medicine, Tochigi, Japan
PURPOSE: IgG anti-GQlb antibody is frequently detected in sera from patients with Miller Fisher syndrome, Bickerstaff's brainstem encephalitis, Guillain-Barré syndrome, acute ophthalmoparesis and so on. To investigate the fine specificity of IgG anti-GQlb antibodies, we performed ab-sorption study of the antibodies. MATERIALS AND METHODS: Serum samples with high titers of IgG anti-GQlb antibody were obtained from 56 patients, including 40 with Miller Fisher syn-drome, 8 with Bickerstaff's brainstem enceph-alitis, 7 with Guillain-Barré syndrome, 1 with acute ophthalmoparesis. The absorption study of IgG anti-GQlb antibody was made by using microtiter plates coated with 5 pmol portions of GQ1b, GT1a, GM2, GM1, GD3, GD2, GD1a, GD1b, or GT1b. The samples were added to the wells at each dilution that gave an optical density between 0.5 and 2.5. After being kept overnight at 4°C, the samples were used as the primary antibodies in the standard ELISA. Absorption rates were expressed as percentages of the optical densities obtained with and without absorption. RESULTS: IgG anti-GQlb antibodies were effectively absorbed by GT1a in 55 samples (98%). The antibodies were also absorbed by GD3 in 16 (28%), by GD2 in 7 (13%), by GDlb in 9 (16%), and by GTlb in 3 (5%). The anti-bodies were also absorbed by GM2 and GM1 in one sample, and by GD1a in two. CONCLUSIONS: Å In almost all samples, IgG anti-GQlb antibodies cross-reacted with GTla containing a disialosyl residue linked to the external galactose common to both GQlb and GT1a. Ç IgG anti-GQlb antibodies frequently cross-reacted with b-series gangliosides containing a disialosyl residue linked to the internal galactose common to both GQlb and b-series gangliosides. É In some samples, IgG anti-GQlb antibodies cross-reacted with structurally different ganglio-sides such as GM2, GM1, or GDla. Our results suggest that IgG anti-GQlb antibodies are heterogeneous.
IgM ANTI-MAG/SGPG ANTIBODY-ASSOCIATED POLYNEUROPATHY IS A SUBGROUP OF CHRONIC INFLAMMATORY DEMYELINATING POLYNEUROPATHY
Tagawa Y.(1,2), Yuki N.(1), Hirata K.(1), Touge T.(2) (1)Dokkyo University School of Medicine, Tochigi, Japan, (2)Kagawa Medical University, Kagawa, Japan
PURPOSE: Most neurologists consider polyneuropathy with monoclonal gammopathy as a nosological entity different from chronic inflammatory demyelinating polyneuropathy (CIDP). On the basis of the criteria proposed by the American Academy of Neurology AIDS Task Force (1991), monoclonal gammopathy is, however, a condition concurrent with CIDP. The aim of this study was to clarify the nosological relationship between CIDP and IgM anti-myelin-associated glycoprotein (MAG)/sulfated glucuronyl paragloboside (SGPG) antibody-associated polyneuropathy. MATERIALS AND METHODS: Serum samples were obtained from 85 patients with CIDP (definite, n = 19; probable, n= 38; possible, n= 28), 66 patients with Guillain-Barré syndrome, and 80 normal controls. We investigated IgM anti-SGPG antibody by enzyme-linked immunosorbent assay and thin-layer chromatography-immunostaining, and also IgM anti-MAG antibody by immunoblotting. We then examined the relation of M-protein in the presence of IgM anti-MAG/SGPG antibody. RESULTS: Seventeen (20%) of the 85 CIDP patients had positive IgM anti-SGPG serology. IgM anti-SGPG antibody was positive in CIDP more frequently than in Guillain-Barré syndrome and normal controls (p < 0.001). The result was almost confirmed by thin-layer chromatography-immunostaining for IgM anti-SGPG antibody and by immunoblotting for IgM anti-MAG antibody. Eleven patients (65%) without IgM M-protein included in the 17 CIDP patients with high titers of IgM anti-MAG/SGPG antibody. DISCUSSION: We confirmed the results of our preliminary study that the polyneuropathy associated with IgM anti-MAG/SGPG antibody is a subgroup of CIDP, not of IgM gammopathy. Because CIDP is heterogeneous clinically, each subgroup of CIDP should need more specific treatment. We therefore propose the polyneuropathy associated with IgM anti-MAG/SGPG antibody is a subgroup of CIDP.
FINE SPECIFICITY OF IgG ANTI-GM1 ANTIBODY AND ITS RELATIONSHIP TO CLINICAL FEATURE IN GUILLAIN-BARRÉ SYNDROME.
Tatsumo M., Koga M., Yuki N., Hirata K. Dokkyo University School of Medicine, Tochigi, Japan
PURPOSE: We investigated fine specificity of IgG anti-GM1 antibody and its relationship to clinical features in Guillain-Barré syndrome (GBS). MATERIALS AND METHODS: Serum samples with high titers of IgG anti-GM1 antibody were taken from 30 GBS patients. The absorption studies of IgG anti-GM1 antibody were made by using microtiter plates coated with 5 pmol portions of GMI, asialo-GM1 (GA1), GM1b, GD1a, GalNAc-GD1a, GD1b, GM2, GQ1b, and GT1b. The samples were added to the wells at each dilution that gave an optical density between 1.0 and 2.0. After being kept overnight, the samples were used as the primary antibodies in the standard ELISA. Absorption rates were expressed as percentages of the optical densities with and without absorption. RESULTS: IgG anti-GM1 antibodies were effectively absorbed by GM1 alone in 14 samples (47%). The antibodies were also absorbed by GD1b in 6 (20%), by GA1 in 13 (43%), and by GM1b in 11 (37%). The patients with IgG anti-GM1 antibody which cross-reacted with GD1b often showed cranial nerve palsy and sensory disturbance than those with the anti-GM1 antibody which did not. When IgG anti-GM1 antibody was absorbed with GD1b or GA1, the serological evidence of antecedent Campylobacter jejuni infection was rarely found. CONCLUSION: IgG anti-GM1 antibody in GBS had various fine specificities, which could be associated with clinical features.
HIGH DOSE ANTIGEN THERAPY IN EXPERIMENTAL AUTOIMMUNE NEURITIS (EAN): INHIBITION OF TNF ALPHA BY NEUTRALIZING ANTIBODIES REDUCES T-CELL APOPTOSIS AND PREVENTS LIVER NECROSIS
Gold R.(1), Weishaupt A.(1), Hartung T.(2), Gaupp S.(1), Brück W.(3) Toyka K.V.(1) (1)Department of Neurology, University of Würzburg, Germany, (2)Biochemical Pharmacology, University of Konstanz, Germany, (3)Department of Neuropathology, University of Göttingen, Germany
We investigated the role of TNF-a in antigen-specific therapy of adoptive transfer (AT-) EAN in-vivo. Antigen therapy has been successfully employed for the treatment of this animal model. Possible mechanisms of action comprise T-cell apoptosis leading to termination of inflammation in-situ. Neutralization of TNF-a alone (d 3-7) was only partially effective. When recombinant human P2 protein was administered i.v. as the therapeutic antigen together with neutralizing anti-TNF-a antibody (d 6+7), T-cell apoptosis in inflamed nerve was reduced to levels observed in control animals. Focal liver necrosis which had been observed in earlier studies after antigen therapy was prevented by passive immunization with neutralizing anti-TNF-a antibody. We also investigated Schwann cell apoptosis. Antigen therapy led to a slight, non-significant increase of Schwann cell apoptosis. These results show that, in the context of antigen therapy, TNF-a has the dual potential to augment beneficial apoptosis of inflammatory T-cells which is in principal beneficial for autoimmune diseases or to induce liver damage as a adverse effect. Optimal dosing of the antigen may be critical when using this approach for therapeutic trials.
OUTCOME IN GUILLAIN-BARRÉ SYNDROME (GBS) PATIENTS IN SMALLER VERSUS LARGER CENTRES
van Koningsveld R., van Doorn P.A., Schmitz P.I.M., van der Meché F.G.A. Erasmus Medical Centre Rotterdam, The Netherlands
PROBLEM: The results of a trial, comparing plasma exchange (PE) with intravenous immunoglobulin (IVIg) in patients with Guillain-Barré syndrome (GBS), resulted in a shift towards IVIg as treatment of choice. Since IVIg is easy applicable, there is no longer a need to transfer patients to trial centres for practical issues. The course of GBS is however unpredictable and autonomic dysfunction may occur. Therefore one must consider whether it is advisable to transfer patients to larger centres. We studied a possible change in referral pattern and analysed outcome in patients admitted in centres participating in GBS trials versus non-participating hospitals. PATIENTS AND METHODS: Records of 331 GBS patients unable to walk (functional grading score [fscore] at nadir >3) were selected from all GBS patients admitted in hospitals in the Southwest of the Netherlands between 1987 to 1996 (n=476). RESULTS: Before publication of the Dutch PE/IVIg trial, 63% of the patients was transferred from non-trial to trial centres compared with only 12% during the subsequent IVIg/methylprednisolon trial (p£ 0.001). There was no difference in percentage of patients able to walk at 8 weeks and 6 months between the centres. A larger percentage of complications was found in trial centres. In a multivariate analysis this was explained by significantly more patients with a f-score of 4 or 5 at nadir admitted to these trial centres. CONCLUSION: There was no difference found in outcome between patients admitted in trial and non-trial centres. This study does not support thoughts that all GBS patients are per se better off in larger centres.
COMPARISON OF MILDLY AND SEVERELY AFFECTED PATIENTS IN AN EPIDEMIOLOGICAL SURVEY ON GUILLAIN-BARRÉ SYNDROME (GBS)
van Koningsveld R., van Doorn P.A., Schmitz P.I.M., van der Meché F.G.A. Erasmus Medical Centre Rotterdam, The Netherlands
PROBLEM: Understanding pathophysiological mechanisms of a disease requires knowledge on the whole spectrum of an illness. However, most data published thus far on GBS concern mainly severely affected patients. METHODS: A clinical and epidemiological survey was performed on GBS patients admitted in hospitals in the Southwest of the Netherlands between 1987 and 1996. Data on mildly affected patients with the ability to walk at nadir (functional grading score [f-score]< 3) were compared with severely affected patients (f-score >3). RESULTS: 476 patients met the NINCDS criteria for GBS which resulted in an incidence rate (IR) for GBS of 1.18/100.000. This IR showed a linear increase with age (p=0.0001). Men were more frequently affected (p=0.0001). A seasonal distribution was not found for the IR nor for specific preceding infections. Patients under 50 years (p<0.001) and men (p=0.001) were more frequently found in the mild group. In both groups a clinically preceding infection was reported in 70%. In the severe group, serological evidence was mostly covered by infections with Campylobacter jejuni, CMV, EBV and Mycoplasma pneumoniae (42%). This is in contrast with the mildly affected group where significantly less serological evidence was found for the 4 pathogenes (25%) (p=0.02). CONCLUSIONS: The incidence rates in the Netherlands conform other studies. Patients of 50 years and younger and males predominantly experienced a mild form of GBS. The 4 described infections are less frequently found in mildly affected GBS patients. This suggests that we have to look for other preceding agents involved in the pathogenesis of mild forms of GBS.
DEMYELINATING ANTIBODIES IN CIDP
Yan W.X., Pollard J.D. Department of Neurology, Institute of Clinical Neurosciences Royal Prince Alfred Hospital and University of Sydney Sydney 2006, Australia
There is mounting evidence for the involvement of antibodies in inflammatory demyelinating neuropathy both GBS and CIDP. Our group has focused attention on patients with typical demyelinating disease e.g. AIDP and CIDP, and potentially demyelinating antibodies. In a group of patients with CIDP responsive to plasma exchange 7 of 12 patients showed C3 and IgG or IgM binding to the myelin/Schwann cell membrane, by indirect immunofluorescence. When sera or purified immunoglobulin was injected intraneurally only sera showing IgG antimyelin activity caused conduction block and demyelination. Because of controversy concerning the validity of this technique, in current studies we are attempting passive transfer in animals treated to induce blood nerve barrier leakage by activated T-cells. The results of these current studies will be presented.
IMMUNOPATHOGENESIS COMMON TO ACUTE MOTOR-SENSORY AXONAL NEUROPATHY AND ACUTE MOTOR AXONAL NEUROPATHY
Yuki N.(1), Kuwabara S.(2), Koga M.(1), Hirata K.(1) (1)Dokkyo University School of Medicine, Tochigi; (2)Chiba University School of Medicine, Chiba, Japan
BACKGROUND: Griffin and colleagues proposed that acute
motor axonal neuropathy (AMAN) and acute motor-sensory axonal
neuropathy (AM-SAN) are part of the spectrum of a single type
of immune attack on the axon. In contrast, IgG anti-GM1 antibody
is associated closely with AMAN, but whether other IgG anti-ganglioside
antibodies are associated with this neuropathy is not clear. METHODS:
To test the hypothesis of Griffin et al., we first investigated
whether IgG anti-ganglioside antibodies can be used as immunological
markers to differentiate AMAN from acute inflammatory demyelinating
polyneuropathy (AIDP), then whether they are present in AMSAN.
RESULTS: IgG anti-GM1 antibodies were positive in 17 (81 %) of
the 21 AMAN patients, but in only 1 (5%) of the 19 AIDP patients.
The frequency of positive IgG anti-GM1 serology in AMAN was significantly
higher than that in AIDP. IgG anti-GMlb antibodies were positive
in 14 (67%) of the AMAN patients, but in only 1 (5%) of the AIDP
patients. IgG anti-GD1a antibodies were positive in 7 (33%) of
the AMAN patients, but in only 1 (5%) of the AIDP patients. Our
2 AMSAN patients had anti-GM1, anti-GM1b, and anti-GD1a Ig-G antibodies.
These positive frequencies for AMSAN were significantly higher
than those for AIDP. CONCLUSION: Anti-GM1b and anti-GD1a IgG,
as well as anti-GM1 IgG antibodies, therefore are immunological
markers for AMAN. The patients with AMSAN had anti-GM1, anti-GMlb,
and anti-GDla IgG antibodies, indicative of a common immunopathogenesis
in AMAN and AMSAN.