Abstract
Background: Campylobacter jejuni enteritis is the most common antecedent infection in Guillain–Barré syndrome (GBS). C. jejuni-related GBS is usually acute motor axonal neuropathy (AMAN), but previous reports described many cases of the demyelinating subtype of GBS (acute inflammatory demyelinating polyneuropathy [AIDP]) after C. jejuni infection.
Objective: To investigate whether C. jejuni infection elicits AIDP.
Methods: In 159 consecutive patients with GBS, antibodies against C. jejuni were measured using ELISA. Antecedent C. jejuni infection was determined by the strict criteria of positive C. jejuni serology and a history of a diarrheal illness within the previous 3 weeks. Electrodiagnostic studies were performed weekly for the first 4 weeks, and sequential findings were analyzed.
Results: There was evidence of recent C. jejuni infection in 22 (14%) patients. By electrodiagnostic criteria, these patients were classified with AMAN (n = 16; 73%) or AIDP (n = 5; 23%) or as unclassified (n = 1) in the first studies. The five C. jejuni-positive patients with the AIDP pattern showed prolonged motor distal latencies in two or more nerves and had their rapid normalization within 2 weeks, eventually all showing the AMAN pattern. In contrast, patients with cytomegalovirus- or Epstein–Barr virus-related AIDP (n = 13) showed progressive increases in distal latencies in the 8 weeks after onset.
Conclusion: Patients with C. jejuni-related Guillain–Barré syndrome can show transient slowing of nerve conduction, mimicking demyelination, but C. jejuni infection does not appear to elicit acute inflammatory demyelinating polyneuropathy.
Guillain–Barré syndrome (GBS) is classified into demyelinating and axonal categories by electrophysiologic and pathologic criteria.1–3⇓⇓ In Western countries, GBS is usually caused by acute inflammatory demyelinating polyneuropathy (AIDP) with electrodiagnostic and pathologic evidence of peripheral nerve demyelination.4–6⇓⇓ In contrast, a pure motor axonal form of GBS, acute motor axonal neuropathy (AMAN), is frequently seen in eastern Asian countries such as China and Japan.3,7,8⇓⇓
Campylobacter jejuni enteritis is the most common antecedent infection in GBS8,9⇓ and is often associated with antiganglioside antibodies.8,10⇓ The C. jejuni lipo-oligosaccharides have a ganglioside GM1-like structure indicative of molecular mimicry between neural tissue and bacterial components.11 Ganglioside GD1a has been shown to be expressed on the lipo-oligosaccharides of C. jejuni.12 C. jejuni infection is likely to induce antiganglioside antibodies and thereby GBS.
Whereas electrodiagnostic and pathologic studies have shown that C. jejuni infection is significantly associated with primary axonal dysfunction, the relationship between C. jejuni infection and neurophysiology is still the subject of debate.8,13⇓ In northern China, positive C. jejuni serology was found for 76% of 21 AMAN patients and for 42% of 12 AIDP patients, suggesting that C. jejuni infection elicits AMAN more frequently than AIDP, but a considerable number of AIDP cases also occur after C. jejuni infection.3 In a large study conducted in North America and Europe involving 229 GBS patients, 52 patients had positive serology for C. jejuni, and 56% of them showed demyelinating neurophysiology.13 A study in Japan investigating 86 GBS patients showed that of the 20 C. jejuni-positive patients, 70% had AMAN and 15% had AIDP.8 These results raise the possibility that C. jejuni infection can elicit AMAN and AIDP.
To investigate whether C. jejuni infection really elicits AIDP, we conducted serial electrodiagnostic studies in C. jejuni-positive GBS patients who were selected by strict criteria and blind serologic assay and compared results with those of patients with typical AIDP after infection by cytomegalovirus (CMV) or Epstein–Barr virus (EBV).
Methods.
Patients.
One hundred fifty-nine consecutive patients with GBS who fulfilled the clinical criteria for GBS and were seen at Chiba University Hospital in Japan between 1994 and 2002 were studied.14 The first electrophysiologic studies were performed within 13 days of the onset of neurologic symptoms. The mean patient age was 38 years (range, 3 to 80 years). Pretreatment serum samples taken during the first 3 weeks after onset were frozen at −80 °C. Patients were considered positive for C. jejuni if they had serologic evidence of C. jejuni infection and a definite history of a diarrheal illness within the previous 3 weeks of GBS onset.15
Infection serologies.
Sera from the patients were tested for the presence of immunoglobulin (Ig) G antibodies against C. jejuni by ELISA, as described elsewhere.16 Serum was considered positive when the titer was ≥1:2,000. Anti-C. jejuni antibody titers were ≥1:2,000 in 14 (82%) of 17 GBS patients from whom C. jejuni had been isolated but had decreased to the normal range within 6 months of the onset of GBS.16 Antibodies to CMV and EBV were measured commercially (The Special Reference Laboratory Company, Tokyo, Japan). CMV infection was defined as the presence of IgM antibodies in the ELISA, and EBV infection was defined as the presence of high titers of antibodies against early antigens or IgM antibodies against the virus capsid antigen. The cutoff values for anti-CMV and -EBV assays were determined from data of 200 healthy Japanese subjects. We included CMV/EBV-related GBS because GBS subsequent to infections by these viruses has been proved to be typical AIDP.8,17⇓
Antiganglioside antibody testing.
The same serum samples were tested for the presence of antibodies to GM1, GM1b, GM2, GD1a, GalNAc-GD1a, GD1b, GT1a, and GQ1b by ELISA, as described elsewhere.18 IgG anti-C. jejuni antibody and antiganglioside antibodies were measured by one of authors (M.K.), who was blinded to the clinical information.
Electrophysiology.
Nerve conduction studies were done weekly for the first 4 weeks and monthly for the next 5 months by two examiners (S.K. and K.O.) using the same conventional procedures and the same EMG machine (Viking 4, Nicolet Biomedical Japan, Tokyo, Japan). Motor nerve studies were made of the median, ulnar, peroneal, and tibial nerves, including F-wave analyses. Amplitude and duration of compound muscle action potential (CMAP) were measured for the initial negative phase. Antidromic sensory nerve conduction studies were performed in the median, ulnar, and sural nerves. Patients were classified as having AMAN or AIDP based on the published electrodiagnostic criteria.3 Normal data of nerve conduction parameters were obtained from 101 healthy subjects.
Results.
Infection serology and initial electrodiagnoses.
Table 1 shows the relation between antecedent infections and electrodiagnoses in the first examination. Of the 159 GBS patients, 30 (19%) were positive for anti-C. jejuni antibody; of these 30 patients, 22 had a history of diarrhea before the onset of GBS and were determined as having recent C. jejuni infection. Only two patients had a positive stool culture for C. jejuni. Eleven (7%) were positive for anti-CMV antibody, and three (2%) were positive for anti-EBV antibody. None had positive serology for two or more organisms. By electrodiagnostic criteria, 159 patients were classified with AIDP (n = 55; 35%) or AMAN (n = 69; 43%) or as unclassifiable (n = 35; 22%) based on the results of the first electrodiagnostic studies performed 1 to 13 days (median, 6 days) after neurologic onset.
Table 1 Antecedent infections and initial electrodiagnoses in 159 patients with Guillain–Barré syndrome
Of the 22 C. jejuni-positive patients, 16 (73%) were diagnosed with AMAN (p < 0.01, compared with the CMV/EBV group), and 5 (23%) showed the AIDP pattern; of these 5 patients, 3 showed prolonged (>120% of upper limits of normal) motor distal latencies in the median and ulnar nerves, and 2 showed prolonged distal latencies in the median, ulnar, and peroneal nerves. None of the five patients had a decrease in conduction velocities and prolonged F-wave latencies, fulfilling the criteria for demyelination. There was no abnormal temporal dispersion or prolonged CMAP duration. Therefore, only prolonged distal latencies met the criteria for demyelination in these patients. One unclassified patient showed mildly reduced CMAP amplitudes, which did not meet the criteria for AMAN (<80% of lower limits of normal).
Of the 14 CMV/EBV-positive patients, 13 (93%) were classified with AIDP (p < 0.001, compared with the other patient groups). All 13 patients had prolonged motor distal latencies and F-wave latencies in almost of the nerves tested. The remaining one patient had mild prolongation of distal latencies and F-wave latencies, which did not meet the criteria for AIDP.
Time course of electrodiagnostic findings.
Because the electrodiagnosis of AIDP in all 5 C. jejuni-positive patients was based on the prolonged distal latencies, we analyzed serial findings of the median distal latencies; figure 1 shows the sequential data in the 22 C. jejuni-positive and 14 CMV/EBV-positive patients. The mean latency of the C. jejuni-positive group was slightly prolonged during week 1, but there was no significant prolongation in the 6 months after onset of GBS. In contrast, there was a progressive increase in the mean distal latency through month 2 or 3 in the CMV/EBV-positive group. Analysis of ulnar nerve studies showed similar differences in the time course of distal latencies.
Figure 1. Serial findings of motor distal latencies in the median nerves in patients with infection by cytomegalovirus (CMV)/Epstein–Barr virus (EBV) or Campylobacter jejuni. The data are given as mean ± SEM. A dotted line indicates the cutoff value of criteria for demyelination.
Figure 2 shows distal latency of the median nerve in patients who were classified with AIDP in the first studies. In all five C. jejuni-positive patients, the distal latency returned toward normal within 2 weeks. Representative waveforms of CMAPs are shown in figure 3. There were similar increases in distal latencies for C. jejuni-positive and CMV-positive patients during week 1. However, the time courses were distinct; the prolonged distal latencies quickly returned toward normal in C. jejuni-positive patients but progressively increased in the CMV-positive patient. In the C. jejuni-positive patients, CMAP amplitudes after distal stimulation were already low (<80% of lower limit of normal) during week 1, and after normalization of prolonged distal latencies in week 2, these patients were reclassified with AMAN according to the electrodiagnostic criteria.3
Figure 2. Serial findings of median distal latencies in patients with infection by cytomegalovirus (CMV)/Epstein–Barr virus (EBV) or Campylobacter jejuni. A dotted line indicates the cutoff value of criteria for demyelination.
Figure 3. Compound muscle action potentials after median nerve stimulation at the wrist in patients with positive serology for Campylobacter jejuni (A, B) or cytomegalovirus (CMV; C). Despite similar prolongations of distal latencies during week 1, note rapid shortening within days in C. jejuni-positive patients in contrast to a progressive increase in distal latency in a CMV-positive patient (C).
Features of C. jejuni-positive patients with the initial diagnosis of AIDP.
Table 2 shows clinical profiles of patients with C. jejuni infection who were classified with AIDP in the first studies. All five patients had diarrhea preceding the onset of GBS. Sensory nerve involvement was found for only one patient. These features were consistent with those of AMAN.19 In the follow-up electrodiagnostic studies during week 2 or 3, four (Patients 1 through 4) were diagnosed with AMAN, and their sensory nerve conduction was normal. Patient 5 was classified with acute motor-sensory axonal neuropathy (AMSAN) because, besides the AMAN pattern in motor nerve studies, the amplitudes of sensory nerve action potentials were decreased in the median, ulnar, and sural nerves. Therefore, none of the C. jejuni-positive patients had the time course of typical AIDP in serial electrodiagnostic studies, and all had the final electrodiagnosis of AMAN/AMSAN. All five patients had elevated serum antibodies against gangliosides GM1, GM1b, GD1a, or GalNAc-GD1a. In contrast, none of the 14 CMV/EBV-positive patients had these antibodies, but 3 CMV-positive patients were positive for anti-GM2 IgM antibodies.
Table 2 Profiles of patients with Campylobacter jejuni infection and initial electrodiagnosis of AIDP
Discussion.
Our results showed that in this Japanese series 23% of 22 C. jejuni-positive patients, who were selected using strict criteria, had the AIDP pattern in the first electrophysiologic studies. The electrodiagnosis was based on the prolonged distal motor latencies, but the distal conduction slowing rapidly returned toward normal, and sensory nerve conduction was normal in almost of them. The time course of nerve conduction abnormality was distinct from that of CMV/EBV-positive AIDP; therefore, demyelination is unlikely to be responsible for the conduction abnormalities. These results suggest that C. jejuni infection does not appear to elicit AIDP.
Isolation of C. jejuni from stool culture is the gold standard for the diagnosis of infection by this bacterium, but culture survey would underestimate the frequency of C. jejuni infection because the time between the infection and GBS onset often exceeds the duration of excretion of viable C. jejuni in stools.20 Serologic studies are more sensitive but less specific than culture-based methods. There are no standards for serologic testing for C. jejuni infection with regard to antigens used or cutoff values for the positivity, and the sensitivity and specificity of serologic assays vary considerably among laboratories.21 The specificity of our assay was 88%,16 and in combination of a clinical history of a definite diarrheal illness, the strict criteria in this study could reduce the false-positive cases.15 We suggest that the frequency of C. jejuni-related AIDP in previous studies could, in part, be affected by the low specificity of serologic assays and by the lack of analyses of serial electrodiagnostic findings. In a recent report investigating Chinese GBS patients with anti-GalNAc-GD1a or GM1b antibodies, C. jejuni serology measured by the same investigator (M.K.) was positive for 54% of 28 AMAN patients and for none of 9 AIDP patients, supporting the view that C. jejuni infection is not associated with AIDP.22
The blood-nerve barrier is anatomically deficient in the distal nerve terminals and nerve roots,23 and these regions are preferentially involved in AIDP and AMAN.24,25⇓ Therefore, conduction slowing or block in the distal nerve segments is the most frequent nerve conduction abnormality in GBS. A number of factors other than demyelination can cause slowing of nerve conduction, such as loss of the fastest fibers, membrane hyperpolarization or depolarization, and sodium channel inactivation.26 Axonal degeneration of the fast fibers is unlikely to explain prolonged distal latencies seen in our C. jejuni-positive patients because motor nerve conduction velocities were normal in those patients, and their rapid normalization is against this view. A previous report studying axonal excitability in GBS suggests that resting membrane potential is not altered in AIDP and AMAN.27 We speculate that sodium channel dysfunction explains the prolonged distal latencies and their quick reversal in C. jejuni-positive patients: in patients with poisoning of tetrodotoxin, which specifically inactivates voltage-gated sodium channels, profound slowing of nerve conduction became normal within days.28 Moreover, recent electrophysiologic studies in AMAN patients suggest sodium channel dysfunction in the motor nerve terminals,27,29⇓ possibly mediated by antibodies, cytokines, or other inflammatory substances, although our data could not provide direct evidence of physiologic conduction failure at the nodes of Ranvier.
In almost all of our C. jejuni-positive patients with prolonged distal latencies, sensory nerve conduction was normal, and this supports the view that these patients had axonal GBS rather than AIDP because sensory nerve fibers are usually involved in AIDP.6 Although our study included only Japanese patients and the possibility of host susceptibility factors could not be excluded, based on the clinical and electrophysiologic findings found in this study, we suggest that slowing of motor nerve conduction in C. jejuni-positive patients is not caused by segmental demyelination, and, therefore, C. jejuni infection does not cause typical AIDP.30 Patients with axonal GBS can show transient conduction slowing, mimicking one caused by demyelination in the early phase of the illness.31 In this regard, electrophysiologic classification of GBS would be better determined based on sequential findings rather than on results of single studies.
- Received January 20, 2004.
- Accepted March 26, 2004.
References
Disputes & Debates: Rapid online correspondence
- Reply to Berciano et al
- Does Campylobacter jejuni infection elicit "demyelinating" Guillain–Barré syndrome?
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.