Guillain Barré syndrome associated with normal or exaggerated tendon reflexes

Transcription

1 J Neurol (2012) 259: DOI /s ORIGINAL COMMUNICATION Guillain Barré syndrome associated with normal or exaggerated tendon reflexes Nobuhiro Yuki Norito Kokubun Satoshi Kuwabara Yukari Sekiguchi Masafumi Ito Masaaki Odaka Koichi Hirata Francesca Notturno Antonino Uncini Received: 29 August 2011 / Revised: 9 November 2011 / Accepted: 13 November 2011 / Published online: 6 December 2011 Ó Springer-Verlag 2011 Abstract Areflexia is part one of the clinical criteria required to make a diagnosis of Guillain Barré syndrome (GBS). The diagnostic criteria were stringently developed to exclude non-gbs cases but there have been reports of patients with GBS following Campylobacter jejuni enteritis with normal and exaggerated deep tendon reflexes (DTRs). The aim of this study is to expand the existing diagnostic criteria to preserved DTRs. From the cohort of patients referred for anti-ganglioside antibody testing from hospitals throughout Japan, 48 GBS patients presented with preserved DTR at admission. Thirty-two patients had normal or exaggerated DTR throughout the course of illness whereas in 16 patients the DTR became absent or diminished during the course of the illness. IgG antibodies against GM1, GM1b, GD1a, or GalNAc-GD1a were frequently present in either group (84 vs. 94%), suggesting a N. Yuki (&) Departments of Microbiology and Medicine, National University of Singapore, Block MD4A, Level 5, 5 Science Drive 2, Singapore , Singapore micyuki@nus.edu.sg N. Kokubun M. Ito M. Odaka K. Hirata Department of Neurology, Dokkyo Medical University, Tochigi, Japan S. Kuwabara Y. Sekiguchi Department of Neurology, Graduate School of Medicine, Chiba University, Chiba, Japan F. Notturno A. Uncini Department of Neurosciences and Imaging, University G. d Annunzio, Chieti, Italy F. Notturno A. Uncini Neurocenter of Southern Switzerland (EOC), Lugano, Switzerland close relationship between the two groups. We then investigated the clinical and laboratory findings of 213 GBS patients from three hospital cohorts. In 23 patients, eight presented with normal tendon reflexes throughout the clinical course of the illness. Twelve showed hyperreflexia, with at least one of the jerks experienced even at nadir, and exaggerated reflexes returning to normal at recovery. The other three had hyperreflexia throughout the disease course. Compared to 190 GBS patients with reduced or absent DTR, the 23 DTR-preserved patients more frequently presented with pure motor limb weakness (87 vs. 47%, p = ), could walk 5 m independently at the nadir (70 vs. 33%, p = ), more frequently had antibodies against GM1, GM1b, GD1a, or GalNAc-GD1a (74 vs. 47%, p = 0.014) and were more commonly diagnosed with acute motor axonal neuropathy (65 vs. 34%, p = ) than with acute inflammatory demyelinating polyneuropathy (13 vs. 43%, p = ). This study demonstrated that DTRs could be normal or hyperexcitable during the entire clinical course in approximately 10% of GBS patients. This possibility should be added in the diagnostic criteria for GBS to avoid delays in diagnosis and effective treatment to these patients. Keywords Guillain Barré syndrome Acute motor axonal neuropathy Acute inflammatory demyelinating polyneuropathy Anti-ganglioside antibody Campylobacter jejuni Tendon reflex Introduction Clinical diagnostic criteria for Guillain Barré syndrome (GBS) requiring universal areflexia or hyporeflexia, as well as progressive motor weakness of more than one limb,

2 1182 J Neurol (2012) 259: were proposed in 1978 by an ad hoc committee of the National Institute of Neurological and Communication Disorders and Stroke following the increased incidence of GBS associated with the swine flu vaccination program in the US, and these criteria were reaffirmed in 1981 and 1990 [1 3]. The criteria were devised to provide guidelines to non-neurologists investigating possible outbreaks of GBS, and were purposefully restrictive. The overall picture of GBS has changed over the last two decades, and GBS is currently classified into two major subtypes, acute inflammatory demyelinating polyneuropathy (AIDP) and acute motor axonal neuropathy (AMAN) on the basis of underlying pathogenesis [10, 11]. The Gram-negative bacterium Campylobacter jejuni is the most frequent infective antecedent in GBS, and it is strongly associated with AMAN and IgG autoantibodies against gangliosides GM1, GM1b, GD1a, or GalNAc-GD1a [39]. From a historical point of view, because of the areflexia/ hyporeflexia criterion, three North American men who presented with an acute paralytic syndrome and normal to brisk deep tendon reflexes (DTRs) after a gastrointestinal illness were thought not to be affected by GBS [15]. Similarly, four patients with acute motor neuropathy, C. jejuni-positive cultures and persistent preserved and even brisk DTRs could not be diagnosed as GBS [38]. However, all four patients had IgG anti-gm1 antibodies, axonal degeneration at electrophysiology, and were eventually diagnosed with AMAN. The findings suggest that some patients with acute paralytic syndrome subsequent to C. jejuni enteritis may have normal or even brisk DTRs; and although they do not fulfill the stringent diagnostic criteria for GBS, their condition is not distinct from the AMAN subtype. To expand the existing diagnostic criteria for GBS with preserved DTRs, we showed continuity between GBS patients with hypoexcitable or absent DTRs ( DTRreduced patients ) and those with normal or exaggerated DTRs ( DTR-preserved patients ), and demonstrated the clinical, serological, and electrophysiological features in the latter. Methods Patients The Japanese survey study Sera from GBS patients from hospitals throughout Japan were sent to the neuroimmunological laboratory at Dokkyo Medical University for serum anti-ganglioside antibody testing between the years 2001 and Detailed DTR findings throughout the disease course were available in 48 patients who fulfilled the clinical criteria for GBS [3] but had normal or brisk reflexes at first examination. As controls, 1,000 patients who presented with absent or diminished DTRs at admission and had the final diagnosis of GBS were randomly chosen. Primary physicians provided information regarding to the patients and obtained written consent from each of them. The three-hospital study Clinical, serological, and electrophysiological features of 68 GBS patients who attended Dokkyo Medical University Hospital between 1999 and 2010, 90 who attended at Chiba University Hospital between 2000 and 2009, and 55 GBS patients seen at University Hospital of Chieti, Italy, were studied retrospectively. Patients disabilities were evaluated with the Hughes functional grading scale [14]. CSF albuminocytological dissociation was defined as raised protein concentration ([45 mg/dl) associated with a count of ten or fewer leukocytes/ll [3]. The study protocol was approved by the local Ethics Committee of each participating institution. Serological studies Anti-ganglioside antibodies were assayed as previously described [35]. Electrophysiological studies Serial nerve conduction studies were performed as described elsewhere [12, 19, 36]. Patients were classified as AIDP, AMAN, and unclassifiable according to Ho s criteria [13]. AMAN was also diagnosed when there was evidence of reversible conduction failure [23, 36]. Patients were classified as acute motor-sensory axonal neuropathy when results of motor nerve conduction studies were as in AMAN and sensory nerve action potential amplitude was \50% of the lower limit of normal in at least two nerves [7]. The final electrodiagnosis was based on the results of sequential studies and took into consideration the complete electrophysiological history of each patient. Statistics Statistical analysis was done using SPSS 12.0J (SPSS Inc., Chicago, IL). Differences in medians were examined by the Mann Whitney U test and differences in frequencies between groups were compared using the v 2 or Fisher s exact test. A difference of p \ 0.05 was considered statistically significant.

3 J Neurol (2012) 259: Results Comparison between DTR-preserved and DTR-reduced GBS patients in the Japanese survey study Antecedent diarrhea was more frequent (63 vs. 36%, p \ 0.001) in the DTR-preserved patients as compared to the 1,000 GBS patients who showed absent or decreased DTRs at admission (Table 1). With respect to the initial symptoms, limb weakness was more common in DTRpreserved patients (92 vs. 62%, p \ 0.001) whereas distal paresthesia was less common (21 vs. 41%, p = ). With respect to the neurological findings during the illness course, the following symptoms were less common in DTR-preserved patients: ophthalmoparesis (0 vs. 13%, p = ), facial weakness (8 vs. 28%, p = ), oropharyngeal palsy (13 vs. 26%, p = 0.034) and sensory dysfunction (27 vs. 47%, p = ). At nadir, DTRpreserved patients more often had a mild functional grade (2 or less; 67 vs. 26%, p \ 0.001). IgG antibodies to GM1, GM1b, GD1a, or GalNAc-GD1a were more frequently found in the DTR-preserved patients (88 vs. 35%, p \ 0.001). Thirty-two of the 48 patients maintained normal or exaggerated DTRs throughout the course of the disease, while in the remaining 16 patients, DTRs became absent or diminished. The antecedent events, initial symptoms, neurological findings during the clinical course, functional grade at nadir and anti-ganglioside antibodies did not differ significantly between the two groups. Clinical features of DTR-preserved GBS patients in the three-hospital study At Dokkyo Medical University Hospital, nine (13%) of the 68 patients who otherwise fulfilled the GBS criteria [3] had normal or brisk DTRs in all limbs at nadir and throughout the course of the illness (Table 2). Despite the preserved DTRs, a diagnosis of GBS was still made in this group of Table 1 Clinical and serological features of deep tendon reflex-preserved Guillain Barré syndrome Group A Group B A versus B Group C (A? B) Group D C versus D Preserved throughout course Initially preserved but reduced later p value Initially preserved tendon reflexes Reduced throughout course p value Odds ratio (95% confidence interval) Total number of patients ,000 Mean age [range], years 36 [14 70] 39 [20 76] [14 76] 44 [1 93] Male-to-female sex ratio 22:10 9: :17 591: Antecedent symptom Diarrhea 19 (59) 11 (69) (63) 364 (36) \ ( ) Upper respiratory tract 10 (31) 4 (25) (29) 481 (48) ( ) infection Initial symptoms Limb weakness 30 (94) 14 (88) (92) 619 (62) \ ( ) Distal paresthesias 9 (28) 1 (6) (21) 414 (41) ( ) Neurological findings Ophthalmoparesis 0 (0) 0 (0) 0 (0) 129 (13) ( ) Facial weakness 2 (6) 2 (13) (8) 280 (28) ( ) Oropharyngeal palsy 3 (9) 3 (19) (13) 262 (26) ( ) Sensory sign 10 (31) 3 (19) (27) 473 (47) ( ) Autonomic nerve 2 (6) 1 (6) [ (6) 157 (16) 0.10 involvement Hughes grade at nadir Grade 2 or less 21 (65) 11 (69) [ (67) 255 (26) \ ( ) Grade 4 or more 5 (16) 1 (6) (13) 572 (57) \ ( ) IgG antibodies to GM1, GM1b, GD1a or GalNAc-GD1a 27 (84) 15 (94) (88) 345 (35) \ ( ) The data are given as number (percentage), unless otherwise indicated. Hughes grade is defined as follows; grade 1: minimal symptoms and signs, able to run, grade 2: able to walk 5 m independently, grade 3: able to walk 5 m with the use of aids, grade 4: chair or bed bound and grade 5: requires assisted ventilation

4 1184 J Neurol (2012) 259: patients due to the other supportive features. Six patients presented with hyperreflexia. Four had antecedent diarrhea: C. jejuni was isolated from Patient 2, and serologic evidence of C. jejuni infection was found in Patients 6 and 7, according to a method described elsewhere [18]. Patient 4 had had Mycoplasma pneumoniae. All patients had no or minimal sensory involvement. Five patients showed CSF albuminocytological dissociation. Patient 1 received plasma exchange, and the others were treated with intravenous immunoglobulin. The interval from disease onset to nadir ranged from 4 to 10 days (median: 5 days). At nadir, six patients were able to walk independently, and three were bedridden. Patients 2 and 3 showed rapid clinical recovery, which was defined as improvement by two or more Hughes grades within 2 weeks from the nadir [20]. All patients were able to walk by 6 months from onset. Ten (11%) of the 90 patients at Chiba University Hospital, and four (7%) of the 55 patients at University Hospital of Chieti did not fulfill the clinical criteria for GBS [3] because DTRs were persistently preserved in both arms and legs throughout the illness (Table 3). In 23 DTR-preserved patients at the three hospitals, eight presented with normal jerks in biceps, triceps, brachioradialis, knees, and ankles throughout the clinical course of the illness. Twelve showed hyperreflexia, with at least one jerk even at nadir, and exaggerated reflexes returned to normal at recovery. The other three had hyperreflexia throughout the disease course. In these DTR-preserved patients, compared to the 190 DTR-reduced GBS patients, antecedent diarrhea was more frequent and autonomic nerve involvement was less common (although they did not reach statistical significance); less common were also distal paresthesias as an initial symptom (9 vs. 38%, p = ), sensory sign during the disease course (13 vs. 53% p = ) and facial weakness (13 vs. 38%, p = 0.011) (Table 3). At nadir, the DTR-preserved patients could walk independently more frequently (70 vs. 33%, p = ). None of the DTRpreserved patients developed respiratory failure. In DTRpreserved patients, neither spasticity nor Babinski s sign was present, and CSF albuminocytological dissociation was found in 53% of the patients. Cranial and spinal MRI was performed in 11, and none of patients showed lesions. Anti-ganglioside antibodies in DTR-preserved GBS patients in the three-hospital study Six of the nine patients admitted to Dokkyo Medical University Hospital had IgG antibodies against GM1, GM1b, GD1a or GalNAc-GD1a (Table 2). Two other patients (Patients 4 and 7) had IgM antibodies but no IgG against GM1, GD1a or GalNAc-GD1a, and the class switching from IgM to IgG did not occur during the course of the illness. Data from all three hospitals showed that IgG anti-gm1 or -GD1a antibodies (48 vs. 36%) and IgG anti- GM1, -GM1b, -GD1a or -GalNAc-GD1a antibodies (61 vs. 45%) were more common in the 23 DTR-preserved patients than in the 190 DTR-reduced patients, although differences were not significant. In the DTR-preserved patients, IgM antibodies to GM1 or GD1a (13 vs. 2%) and to GM1, GM1b, GD1a or GalNAc-GD1a (13 vs. 2%) as well as either IgG or IgM antibodies to GM1 or GD1a (61 vs. 37%, p = 0.030) and to GM1, GM1b, GD1a or Gal- NAc-GD1a (74 vs. 47%, p = 0.014) were more frequent (Table 3). In contrast, no anti-ganglioside antibodies were detected in the sera from patients with acute transverse myelitis (n = 9), acute demyelinating encephalomyelitis (n = 46) or multiple sclerosis (n = 44). Serial nerve conduction studies in DTR-preserved GBS patients at the three-hospitals Figure 1 shows the results of serial motor nerve conduction studies of the DTR-preserved patients at Dokkyo Medical University Hospital. Prolonged distal motor latency (DML) fulfilling the criteria for demyelination [13] was present in the median nerve of one patient (Patient 3) and the tibial nerve of six patients (Patients 1, 2, 3, 5, 6 and 8). On serial recordings, DMLs reverted to normal within 5 weeks except in the tibial nerve of Patient 5. In the ulnar nerves, DMLs were within normal limits throughout the follow-up period in all patients. Low-amplitude distal compound muscle action potentials (CMAPs) matching the criteria for axonal degeneration were present in all nine patients in the tibial nerve, in seven patients in the ulnar nerve, and in four in the median nerve. Motor conduction velocities remained within normal range. Abnormal reduction of proximal CMAP amplitude was present in the elbow segment of the ulnar nerve in Patient 2 (see Fig. 1), Patients 3 and 5 (approximately 30%), and in the forearm segment of Patient 1 (46%). Based on Ho s electrodiagnostic criteria [13], six of the nine patients (Patients 1, 2, 4, 5, 6 and 7; Table 2) were initially diagnosed as having AMAN. Patient 3 was diagnosed with AIDP in the early stage of the illness because of prolonged DMLs in the median and tibial nerves. However, at follow-up he was re-diagnosed as AMAN with reversible conduction failure because of prompt normalization of DMLs and CMAP amplitudes without any remyelinating features. Patients 8 and 9 were unclassifiable. They had decreased CMAPs in the tibial nerve. Both had normal distal CMAPs in the median and ulnar nerves, but amplitude reduction of proximal CMAPs were present in the median nerve of Patient 8 (35%) and in the ulnar nerve of Patient 9 (36%). In the recovery stage, none of the nine patients showed conduction slowing or abnormal temporal

5 J Neurol (2012) 259: Table 2 Clinical features of nine patients with deep tendon reflex-preserved Guillain Barré syndrome at Dokkyo Medical University Hospital Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8 Patient 9 Age/Sex 41/M 31/M 68/M 29/M 42/M 70/M 45/M 20/M 40/M Antecedent symptoms Running nose Diarrhea No Pyrexia, Diarrhea Diarrhea Diarrhea No Headache cough Initial symptoms Limb weakness Yes Yes Yes Yes Yes Yes Yes Yes Yes Distal paresthesia No No No No No No No No No Neurological findings Cranial nerve involvement No Unilateral facial palsy No No Oropharyngeal palsy No No No No Deep tendon reflexes Exaggerated Exaggerated Normal Normal Normal Exaggerated Exaggerated Exaggerated Exaggerated Babinski sign No No No No No No No No No Sensory symptom No No Yes No Yes No Yes No Yes Sensory sign No No No No No No Yes No Yes Hughes grade at nadir CSF findings [cell (/ll)/protein (mg/dl)] Week 1 1/26 1/35 0/21 31/69 3/ 1/37 1/113 Not done Week 2 or later 1/39 4/125 5/122 4/63 Anti-ganglioside antibodies IgG antibodies to GM1 GM1 GM1b None None GD1a None GalNAc-GD1a GM1 GM1b GM1b GD1a GalNAc- GD1a GD1a GD1a GD1b GalNAc- GD1b GD1a GD1b IgM antibodies to None None GM1b GM1 None None GD1a GalNAc- GD1a GM1b GalNAc- GD1a Initial electrodiagnosis AMAN Unclassifiable AIDP AMAN AMAN AMAN AMAN Unclassifiable Unclassifiable Final electrodiagnosis AMAN AMAN AMAN AMAN AMAN AMAN AMAN Unclassifiable Unclassifiable None M male, AMAN acute motor axonal neuropathy, AIDP acute inflammatory demyelinating polyneuropathy

6 1186 J Neurol (2012) 259: Table 3 Clinical, serological, and electrophysiological features of deep tendon reflex-preserved and -reduced Guillain Barré syndrome patients at three hospitals Dokkyo Chiba G.d Annuzio Total patients p value Odds ratio (95% confidence interval) Preserved Reduced Preserved Reduced Preserved Reduced Preserved Reduced Total number of patients Mean age [range], years 43 [20 70] 49 [5 90] 39 [12 82] 28 [16 71] 31 [20 41] 50 [9 80] 38 [1 2 82] 42 [5 90] Male-to-female sex ratio 9:0 31:28 6:4 51:29 3:1 33:18 18:5 115: Antecedent infection Diarrhea 4 (44) 31 (53) 5 (50) 25 (31) 3 (75) 21 (41) 12 (52) 77 (41) 0.28 Upper respiratory tract infection 3 (33) 27 (59) 2 (20) 30 (38) 0 (0) 13 (26) 5 (22) 59 (31) 0.25 Initial symptoms Limb weakness 9 (100) 47 (80) 7 (70) 57 (71) 4 (100) 46 (90) 20 (87) 150 (79) 0.28 Distal paresthesias 0 (0) 23 (39) 2 (20) 20 (25) 0 (0) 29 (57) 2 (9) 72 (38) ( ) Neurological findings Ophthalmoparesis 0 (0) 3 (15) 0 (0) 8 (10) 0 (0) 4 (8) 0 (0) 15 (8) 0.39 Facial weakness 1 (11) 19 (32) 2 (20) 24 (30) 0 (0) 30 (59) 3 (13) 73 (38) ( ) Oropharyngeal palsy 0 (0) 6 (10) 1 (10) 21 (26) 0 (0) 18 (35) 1 (4) 45 (24) ( ) Upper limb weakness 9 (100) 59 (100) 8 (80) 72 (90) 3 (75) 48 (94) 20 (87) 179 (94) 0.18 Lower limb weakness 9 (100) 59 (100) 9 (90) 74 (93) 4 (100) 48 (94) 22 (96) 181 (95) 0.71 Sensory sign 2 (22) 27 (46) 1 (10) 42 (53) 0 (0) 32 (63) 3 (13) 101 (53) ( ) Autonomic nerve involvement 0 (0) 3 (5) 0 (0) 15 (19) 0 (0) 6 (12) 0 (0) 37 (29) Hughes grade at nadir Grade 2 or less 6 (67) 12 (20) 8 (80) 28 (35) 2 (50) 22 (43) 16 (70) 62 (33) ( ) Grade 4 or more 3 (33) 34 (58) 1 (10) 34 (43) 0 (0) 21 (41) 4 (17) 89 (47) ( ) CSF albuminocytological dissociation 5 (63) 30 (67) 1 (20) 27 (61) 3 (75) 44 (96) 9 (53) 101 (75) 0.29 Antibodies to GM1 or GD1a 7 (78) 32 (54) 4 (40) 28 (35) 3 (75) 17 (33) 14 (61) 71 (37) ( ) Antibodies to GM1, GM1b, GD1a, 8 (89) 41 (70) 6 (60) 28 (35) 3 (75) 19 (37) 17 (74) 90 (47) ( ) or GalNAc-GD1a Final electrodiagnosis AMAN 7 (78) 34 (58) 5 (50) 17 (20) 3 (75) 14 (28) 15 (65) 65 (34) ( ) AMSAN 0 (0) 1 (2) 0 (0) 0 (0) 0 (0) 4 (8) 0 (0) 5 (3) 0.58 AIDP 0 (0) 16 (27) 2 (20) 34 (43) 1 (25) 31 (61) 3 (13) 81 (43) ( ) Unclassifiable 2 (22) 8 (14) 3 (30) 29 (36) 0 (0) 2 (4) 5 (22) 39 (21) 0.53 The data are given as number (percentage) unless otherwise indicated. Hughes grade is defined as follows; grade 1: minimal symptoms and signs, able to run, grade 2: able to walk 5 m independently, grade 3: able to walk 5 m with the use of aids, grade 4: chair or bed bound and grade 5: requires assisted ventilation AMAN acute motor axonal neuropathy, AMSAN acute motor-sensory axonal neuropathy, AIDP acute inflammatory demyelinating polyneuropathy

7 J Neurol (2012) 259: Fig. 1 Serial changes in motor conduction abnormalities. Recordings show superimposed distal compound muscle action potential (CMAPs) from abductor pollicis brevis following stimulation of the median nerve at the wrist and elbow (as well as the axilla in Patients 2 and 7), and from the abductor digiti minimi stimulated at the wrist and below the elbow (as well as above the elbow and axilla in Patients 6 and 9) in the ulnar nerve. Patient 1. On day 5, distal CMAP amplitude was slightly reduced (4.0 mv) and there was a 46% reduction of proximal CMAP amplitude. By day 34, CMAPs normalized, without prolonged distal motor latency (DML) or remyelinating slow components of CMAPs. Similar findings were found in the tibial nerve and on the basis of these findings the electrodiagnosis of AMAN was made. The decrement of proximal CMAP amplitude at day 5 suggested some early reversible conduction failure, which, however, did not reach the 50% required decrement. Patient 2. On day 3, median DML was slightly prolonged (4.6 ms). Distal CMAPs in the ulnar and the tibial nerves were low in amplitude with normal DMLs. Conduction block (72%) was present across the elbow segment of the ulnar nerve. On day 6, distal CMAPs decreased in amplitude in the median (from 10.3 to 6.4 mv) and ulnar nerves (from 4.0 to 2.5 mv), and conduction block (79%) in the ulnar nerve remained. On day 10, median nerve CMAPs returned to 10.2 mv and DML shortened to 4.1 ms. Conduction block in the ulnar nerve remained but proximal CMAPs recovered slightly. On day 25, conduction block in the ulnar nerve resolved but distal CMAP amplitude remained reduced. No remyelinating slow components of CMAPs were seen in the recovery stage. The median, ulnar, and sural SNAPs were normal throughout the illness course. The final diagnosis of AMAN with reversible conduction failure was made. Patient 3. On day 7, CMAP amplitude in the median nerve was reduced and DML was prolonged (6.1 ms), fulfilling the criteria for demyelination. On the basis of prolonged DML (5.8 ms in the tibial nerve), the diagnosis of AIDP was made. However, prompt increase of CMAPs and normalization of CMAPs and DMLs was noted at follow-up and the patient was re-diagnosed as AMAN with reversible conduction failure in distal nerve segments. Patient 4. Progressive CMAPs reduction as typically seen in AMAN with axonal degeneration. Patient 5. Reduced CMAPs were recorded throughout the follow-up period in the ulnar and the median nerves. There were no demyelinating features in the upper limb nerves but DML was prolonged in the tibial nerve. The final diagnosis was AMAN. Patient 6. Progressive decrease of CMAPs as seen in axonal degeneration. The amplitude reduction of proximal CMAP (32%) across the elbow segment at day 5 with uniform reduction of all CMAPs at day 11 suggests a pseudo conduction block due advancing axonal degeneration. The final diagnosis was AMAN. Patient 7. Reduced CMAPs were recorded throughout the follow-up period. The final diagnosis was AMAN. Patients 8 and 9. Both patients were unclassifiable as their distal CMAPs were normal in amplitude and latency. However, the proximal CMAPs showed a 36% reduction in amplitude in Patient 8 and 35% in Patient 9, which later resolved without any demyelinating features, suggesting the presence of early reversible conduction failure, even though not reaching the required 50% decrement of amplitude. L left, R right

8 1188 J Neurol (2012) 259: dispersion to support a remyelinating process, and the final diagnosis was AMAN in seven (two with reversible conduction failure), and unclassifiable in two. Of the ten DTR-preserved GBS patients at Chiba University Hospital, five had AMAN (two with reversible conduction failure), two had AIDP, and three were unclassified at the final diagnosis (Table 3). Of the four at University Hospital of Chieti, three had AMAN (two with reversible conduction failure) and one had pure motor AIDP at the final diagnosis. The combined data of the three hospitals showed the AMAN subtype to be more common (65 vs. 34%, p = ) than AIDP (13 vs. 43%, p = ) in the 23 DTR-preserved GBS patients compared to the 190 DTR-reduced GBS patients. Discussion This is the first study that has investigated clinical, serological, and electrophysiological features of DTRpreserved GBS in large series of patients. In the Japanese survey study, GBS patients in whom DTRs were preserved at admission had antecedent diarrhea, pure motor limb weakness and IgG anti-gm1, -GM1b, -GD1a or -GalNAc- GD1a antibodies more often compared to GBS patients, in whom DTRs were either reduced or absent at admission. We then divided the GBS patients with preserved DTR at admission between those in whom DTRs were preserved throughout the course and those of in whom DTRs became reduced in the course of the illness. The clinical and serological features did not differ between the two subgroups, indicating that both form a continuous spectrum. In the second study, we examined the clinical and laboratorial findings of patients who presented consecutively at the three hospitals. It is notable that the findings are consistent in both Japan and Italy. About 10% of patients showed normal or hyperexcitable DTRs throughout the course of the illness. In DTR-preserved patients antecedent diarrhea and IgG anti-gm1, -GM1b, -GD1a, or -GalNAc- GD1a antibodies were frequent, although they did not reach statistical significance. Compared to DTR-reduced GBS, DTR-preserved patients had pure motor weakness more often, AMAN and better functional grade at nadir. Although C. jejuni-related GBS is considered to be severe [16, 32], the present study suggests that patients with C. jejuni-related AMAN can have a milder form of the disease and show normal or hyperactive DTRs. Therefore, AMAN should be considered in patients with acute pure motor quadriparesis with normal or brisk reflexes, especially they present with antecedent gastroenteritis as well. Retained reflexes were observed in some of the Chinese AMAN patients when the disease was mild [4]. In Dutch patients with pure motor GBS, DTRs were preserved in patients with paresis to the degree of Medical Research Council s Grade 3 [37]. However, all DTRs were brisk in a Slovenian AMAN patient, in whom muscle strength was graded 0 2 in the distal lower limb muscles [31]. In the present study, 7% of the 213 GBS patients DTR were exaggerated throughout the disease and 17% of the 23 DTR-preserved GBS patients were bedbound at nadir. Moreover, some AMAN patients showed areflexia during the acute phase and hyperreflexia at the recovery phase [22, 25]. The above observations suggest that preserved DTR are not limited to less severe cases and at the whole a spectrum of DTR excitability exists in GBS. Moderate weakness and normal afferent reflex arc could account for the preservation of DTRs but does not explain hyperreflexia. The mechanism underlying hyperreflexia in AMAN is unknown, but the abnormal reflex spreading to other segments as reported in some patients suggests central mechanisms [21, 22]. Increased lower motor neuron excitability in hyperreflexic patients was supported by increased soleus H-reflex amplitude and the possibility of recording H-reflexes in the hand or foot muscles. These patients, as well as the ones in the present study, did not have signs of upper motor neuron damage such as spasticity and the Babinski s sign, suggesting dysfunction of the inhibitory spinal interneurons, although a magnetic stimulation study suggested the involvement of corticospinal tract [29]. Hyperreflexia is occasionally seen in chronic motor neuropathy associated with IgM anti-gm1 antibodies. Anti-GM1 serum injected in the subarachnoid space caused damage to the central axons in the spinal cord as well as nerve roots [33]. The nerve roots where the blood nerve barrier is anatomically deficient [28] are preferentially affected in GBS [5]. Inflammation in the nerve roots could lead to disruption of the blood CNS barrier and allow anti-ganglioside antibodies to reach intramedullary collateral branches of the inhibitory interneurons causing, disinhibition of lower motor neurons. In the three-hospital study, repeated nerve conductions showed that DTR-preserved patients had AMAN more often than AIDP. Some patients that were eventually diagnosed with AMAN with reversible conduction failure showed, in the early stage of the disease, prolonged DML or conduction block, and therefore could have been fallaciously classified as AIDP [6, 23, 36]. Current electrodiagnostic criteria for AIDP and AMAN do not require serial nerve conduction studies [9, 13], but we deem that these studies are essential to avoid overestimation of AIDP [12]. For instance, Patient 3 was classified as AIDP according to Ho s criteria in the first study based on the DML prolongation in the two nerves, but the prolonged DML and reduced CMAPs resolved quickly without the development of remyelinating features. The presence of remyelinating

9 J Neurol (2012) 259: features, such as excessive temporal dispersion and conduction slowing at the recovery stage, are the most reliable electrophysiological evidence of a demyelination-remyelination process. In human and rabbit AMAN, the earliest and mildest pathological changes consisted of lengthening of the nodes of Ranvier with distortion of the paranodal myelin, and in some instances with breakdown of the outermost myelin terminal loops [8, 34]. In the spinal anterior roots of AMAN rabbits, IgG antibodies bind to nodes where GM1 is highly expressed. They activate complement, resulting in the formation of membrane attack complex. This is followed by the disappearance of sodium channels and detachment of terminal myelin loops [34]. This damage starting from the nodes and extending to the paranodes may explain conduction slowing or block in the early stage of AMAN. IgG anti-gm1, -GM1b, -GD1a, or -GalNAc-GD1a antibodies are serological markers of AMAN [26, 27]; whereas IgM anti-gm1 antibodies are markers of multifocal motor neuropathy [17, 30]. Unexpectedly, the threehospital study showed that the DTR-preserved GBS patients more often had IgM anti-gm1, -GD1a, or -GalNAc-GD1a, but not IgG antibodies, and that the isotype did not change from IgM to IgG in two patients with serial determinations. When we considered both IgG and IgM antibodies against GM1, GM1b, GD1a, or GalNAc-GD1a, 74% of the patients were positive. As neither GM1b nor GalNAc-GD1a is commercially available, both IgG and IgM antibodies against GM1 or GD1a could be used as screening being positive in 61% of the DTR-preserved patients. In conclusion, DTR-reduced and -preserved GBS form a continuous spectrum, and existing diagnostic criteria should be expanded for GBS patients who have normal or exaggerated DTRs. GBS patients benefit from plasmapheresis or intravenous immunoglobulin therapy. The presence of hyperreflexia in an appropriate clinical and electrophysiological context should not induce delay in treatment. It is important in differential diagnosis to exclude spinal lesion, although inflammatory polyradiculoneuropathy and myelitis, which are rare, but may coexist in the same patient [24]. The history of preceding infection, neurological examination, CSF analysis, repeated nerve conduction studies and anti-ganglioside antibody measurement in association with normal neuroimaging of the central nervous system, are helpful for supporting the diagnosis GBS. Acknowledgments We thank Ms. Chiaki Yanaka (Department of Neurology, Dokkyo Medical University) for her assistance in antiganglioside antibody measurement and Dr. Tsubasa Sada (Department of Neurology, Dokkyo Medical University) for his assistance in statistical analysis. This study was supported in part by the National Medical Council (IRG 10 Nov 086), Ministry of Health, Singapore. Conflict of interest None. References 1. Asbury AK (1981) Diagnostic considerations in Guillain Barré syndrome. Ann Neurol 9(Suppl): Asbury AK, Arnason BG, Karp HR, McFarlin DE (1978) Criteria for diagnosis of Guillain Barré syndrome. Ann Neurol 3: Asbury AK, Cornblath DR (1990) Assessment of current diagnostic criteria for Guillain Barré syndrome. Ann Neurol 27(Suppl):S21 S24 4. Asbury AK, Cornblath DR (1998) Preserved tendon reflexes in Campylobacter neuropathy: reply. Ann Neurol 43: Brown WF, Snow R (1991) Patterns and severity of conduction abnormalities in Guillain Barré syndrome. J Neurol Neurosurg Psychiatry 54: Capasso M, Caporale CM, Pomilio F, Gandolfi P, Lugaresi A, Uncini A (2003) Acute motor conduction block neuropathy: another Guillain Barré syndrome variant. Neurology 61: Capasso M, Notturno F, Mazoil C (2011) Involvement of sensory fibers in axonal subtypes of Guillain Barré syndrome. J Neurol Neurosurg Psychiatry 82: Griffin JW, Li CY, Ho TW, Tian M, Gao CY, Xue P, Mishu B, Cornblath DR, Macko C, McKhann GM, Asbury AK (1996) Pathology of the motor-sensory axonal Guillain Barré syndrome. Ann Neurol 39: Hadden RDM, Cornblath DR, Hughes RAC, Zielasek J, Hartung HP, Toyka KV, Swan AV (1998) Electrophysiological classification of Guillain Barré syndrome: clinical associations and outcome. Ann Neurol 44: Hafer-Macko C, Hsieh S-T, Li CY, Ho TW, Sheikh K, Cornblath DR, McKhann GM, Asbury AK, Griffin JW (1996) Acute motor axonal neuropathy: an antibody-mediated attack on axolemma. Ann Neurol 40: Hafer-Macko CE, Sheikh KA, Li CY, Ho TW, Cornblath DR, McKhann GM, Asbury AK, Griffin JW (1996) Immune attack on the Schwann cell surface in acute inflammatory demyelinating polyneuropathy. Ann Neurol 39: Hiraga A, Kuwabara S, Ogawara K, Misawa S, Kanesaka T, Koga M, Yuki N, Hattori T, Mori M (2005) Patterns and serial changes in electrodiagnostic abnormalities of axonal Guillain Barré syndrome. Neurology 64: Ho TW, Mishu B, Li CY, Gao CY, Cornblath DR, Griffin JW, Asbury AK, Blaser MJ, McKhann GM (1995) Guillain Barré syndrome in northern China: relationship to Campylobacter jejuni infection and anti-glycolipid antibodies. Brain 118: Hughes RAC, Newsom-Davis JM, Perkin GD, Pierce JM (1978) Controlled trial prednisolone in acute polyneuropathy. Lancet 2: Jackson CE, Barohn RJ, Mendell JR (1993) Acute paralytic syndrome in three American men: comparison with Chinese cases. Arch Neurol 50: Jacobs BC, van Doorn PA, Schmitz PIM, Tio-Gillen AP, Herbrink P, Visser LH, Hooijkass H, van der Meché FGA (1996) Campylobacter jejuni infections and anti-gm1 antibodies in Guillain Barré syndrome. Ann Neurol 40: Kinsella LJ, Lange DJ, Trojaborg W, Sadiq SA, Younger DS, Latov N (1994) Clinical and electrophysiologic correlates of elevated anti-gm1 antibody titers. Neurology 44: Koga M, Koike S, Hirata K, Yuki N (2005) Ambiguous value of Haemophilus influenzae isolation in Guillain Barré and Fisher syndromes. J Neurol Neurosurg Psychiatry 76: Kokubun N, Nishibayashi M, Uncini A, Odaka M, Hirata K, Yuki N (2010) Conduction block in acute motor axonal neuropathy. Brain 133: Kuwabara S, Mori M, Ogawara K, Hattori T, Yuki N (2001) Indicators of rapid clinical recovery in Guillain Barré syndrome. J Neurol Neurosurg Psychiatry 70:

10 1190 J Neurol (2012) 259: Kuwabara S, Nakata M, Sung JY, Mori M, Kato N, Hattori T, Koga M, Yuki N (2002) Hyperreflexia in axonal Guillain Barré syndrome subsequent to Campylobacter jejuni enteritis. J Neurol Sci 199: Kuwabara S, Ogawara K, Koga M, Mori M, Hattori T, Yuki N (1999) Hyperreflexia in Guillain Barré syndrome: relation with acute motor axonal neuropathy and anti-gm1 antibody. J Neurol Neurosurg Psychiatry 67: Kuwabara S, Yuki N, Koga M, Hattori T, Matsuura D, Miyake M, Noda M (1998) IgG anti-gm1 antibody is associated with reversible conduction failure and axonal degeneration in Guillain Barré syndrome. Ann Neurol 44: Martens-Le Bouar H, Korinthenberg R (2002) Polyradiculoneuritis with myelitis: a rare differential diagnosis of Guillain Barré syndrome. Neuropediatrics 33: McKhann GM, Cornblath DR, Griffin JW, Ho TW, Li CY, Jiang Z, Wu HS, Zhaori G, Liu Y, Jou LP, Liu TC, Gao CY, Mao JY, Blaser MJ, Mishu B, Asbury AK (1993) Acute motor axonal neuropathy: a frequent cause of acute flaccid paralysis in China. Ann Neurol 33: Ogawara K, Kuwabara S, Koga M, Mori M, Yuki N, Hattori T (2003) Anti-GM1b IgG antibody is associated with acute motor axonal neuropathy and Campylobacter jejuni infection. J Neurol Sci 210: Ogawara K, Kuwabara S, Mori M, Hattori T, Koga M, Yuki N (2000) Axonal Guillain Barré syndrome: relation to anti-ganglioside antibodies and Campylobacter jejuni infection in Japan. Ann Neurol 48: Olsson Y (1968) Topographical differences in the vascular permeability of the peripheral nervous system. Acta Neuropathol (Berl) 10: Oshima Y, Mitsui T, Endo I, Umaki Y, Matsumoto T (2001) Corticospinal tract involvement in a variant of Guillain Barré syndrome. Eur Neurol 46: Pestronk A, Cornblath DR, Ilyas AA, Baba H, Quarles RH, Griffin JW, Alderson K, Adams RN (1988) A treatable multifocal motor neuropathy with antibodies to GM1 ganglioside. Ann Neurol 24: Podnar S, Vodusek DB (2000) Hyperreflexia in a patient with motor axonal Guillain Barré syndrome. Eur J Neurol 7: Rees JH, Soudain SE, Gregson NA, Hughes RAC (1995) Campylobacter jejuni infection and Guillain Barré syndrome. N Engl J Med 333: Schwerer B, Lassmann H, Kitz K, Bernheimer H (1986) Ganglioside GM1, a molecular target for immunological and toxic attacks: similarity of neuropathological lesions induced by ganglioside-antiserum and cholera toxin. Acta Neuropathol 72: Susuki K, Rasband MN, Tohyama K, Koibuchi K, Okamoto S, Funakoshi K, Hirata K, Baba H, Yuki N (2007) Anti-GM1 antibodies cause complement-mediated disruption of sodium channel clusters in peripheral motor nerve fibers. J Neurosci 27: Tagawa Y, Yuki N, Hirata K (2002) High anti-gm1 and anti- GD1a IgG antibody titers are detected in Guillain Barré syndrome but not in chronic inflammatory demyelinating polyneuropathy. Eur Neurol 48: Uncini A, Manzoli C, Notturno F, Capasso M (2010) Pitfalls in electrodiagnosis of Guillain Barré syndrome subtypes. J Neurol Neurosurg Psychiatry 81: van der Meché FGA, Meulstee J, Vermeulen M, Kievit A (1988) Patterns of conduction failure in the Guillain Barré syndrome. Brain 111: Yuki N, Hirata K (1998) Preserved tendon reflexes in Campylobacter neuropathy. Ann Neurol 43: Yuki N, Kuwabara S (2007) Axonal Guillain Barré syndrome: carbohydrate mimicry and pathophysiology. J Peripher Nerv Syst 12: