Short Reports      73

Guillain–Barré syndrome: Association with Campylobacter jejuni and Mycoplasma pneumoniae infections
in India
S. P. GORTHI, LATA KAPOOR, RAMA CHAUDHRY, NIDHI SHARMA, GUILLERMO I. PEREZ-PEREZ,
PINAKI PANIGRAHI, MADHURI BEHARI

ABSTRACT
   Background. Guillain–Barré syndrome is the most common cause of acute neuromuscular paralysis and is considered a post-infectious disease.
   Methods. Twenty patients with Guillain–Barré syndrome admitted to the Neurosciences Centre at the All India Institute of Medical Sciences from November 1997 to August 1998 were investigated for evidence of antecedent infections. This case–control study included 2 controls for each patient, one a household control and the other an age- and sex-matched hospital control suffering from a neurological illness other than Guillain–Barré syndrome. Evidence of recent Campylobacter jejuni infection was investigated by culture and serology, and for Mycoplasma pneumoniae by serology.
   Results. There was evidence of recent C. jejuni infection in 35% of the patients compared with 25% of household controls and none of the hospital controls. M. pneumoniae infection was seen in 50% of patients compared with 25% of household controls and 15% of hospital controls. About one-third of the patients (30%) had evidence of both infections. The association of both infections in patients was found to be statistically significant as compared to hospital controls.
   Conclusion. C. jejuni and M. pneumoniae may be important antecedent illnesses in patients with Guillain–Barré syndrome in India.

Natl Med J India 2006;19:137–9

INTRODUCTION
Guillain–Barré syndrome (GBS) is the most common cause of acute neuromuscular paralysis.1 GBS is a disorder of the peripheral nervous system characterized by weakness, usually symmetrical, of the limbs and respiratory muscles and areflexia. The disease is self-limiting; however, 15%–20% of patients have severe residual neurological deficits. Electrophysiological and pathological studies have revealed several patterns of GBS, namely acute inflammatory demyelinating polyneuropathy (AIDP), acute motor axonal neuropathy (AMAN), acute motor and axonal neuropathy (AMSAN) and the Miller Fisher syndrome.2
   GBS is considered a post-infectious disease as approximately two-thirds of patients report some form of preceding infectious illness, especially of the respiratory or gastrointestinal tracts, in the previous 3 weeks.3 A number of specific infectious agents have been reported in patients with GBS including Campylobacter jejuni, Mycoplasma pneumoniae, cytomegalovirus, Ebstein–Barr virus, etc.4,5 The pathogenesis of GBS is still unknown but an autoimmune-mediated mechanism, possibly triggered by an infectious agent sharing epitopes with neural antigens (molecular mimicry) has been proposed.1 Kaldor and Speed reported C. jejuni infection as the most common antecedent illness in patients with GBS.1 The lipopolysaccharide coat of Campylobacter is rich in glycoconjugates which resemble human glycoconjugates. Antibodies against various gangliosides have been reported in a variable proportion of patients with GBS and their decrease during clinical improvement supports a possible pathogenetic role in the neuropathy.
   There is a paucity of data from India regarding antecedent infections in patients with GBS. A study by Hariharan et al.6 reported the role of precedent C. jejuni infections in this group of patients but the role of M. pneumoniae has not been studied. Moreover, their study was based on a select group of patients with GBS without age- and sex-matched controls and may therefore represent coincidental findings. We designed a case–control study to determine the association of GBS with precedent M. pneumoniae and C. jejuni infection in the Indian population.

METHODS
All patients admitted with a clinical diagnosis of GBS to the Neurosciences Centre of All India Institute of Medical Sciences, New Delhi from November 1997 to August 1998 were included. The diagnosis of GBS was defined clinically according to the criteria of Asbury and Cornblath.7 Briefly, the diagnosis was based on a history of progressive weakness of more than one limb over a period of <4 weeks, thought to be due to a neuropathy, in the absence of any identifiable genetic, metabolic or toxic cause. Electrophysiological studies were done which showed nerve conduction slowing with features of demyelination and/or axonal damage.
   At admission, each patient was asked about the occurrence of acute gastroenteritis or an acute respiratory tract infection in the preceeding month. Acute gastroenteritis was defined as a brief illness of 1–6 days with increased frequency of stools (>3/day) of soft consistency with or without blood. Acute respiratory tract infection was defined as an acute illness characterized by fever (100–103 °F) with cough and malaise with or without expectoration. Electrophysiological studies were done to classify the pattern of GBS.
   Age- and sex-matched (within 2 years) inpatients suffering from a non-GBS neurological illness with no antecedent history of gastroenteritis or respiratory tract infections were included as controls (hospital control). None of the hospital controls suffered from peripheral neuropathies. Blood was also collected from a member of the patient’s household (household control). An informed consent was obtained from all the patients and controls after explaining to them the study protocol.

Microbiological investigations
Two stool specimens were collected from each patient within 48 hours of admission. Stool specimens were cultured on a medium selective for C. jejuni, i.e. Campylobacter blood agar base with Butzler supplement (Hi-media Laboratories, Mumbai, India). Specimens were also cultured by the filtration technique on 5% sheep blood agar as previously described.8 The latter method was used to detect strains of Campylobacter that may have been sensitive to the antibiotics contained in the selective media. The plates were incubated at 42 °C in a microaerophilic atmosphere for 3–5 days. The Gram-negative, oxidase-positive, catalase-positive isolates were identified by conventional biochemical methods.8
   Serum samples were obtained from the patients at the time of admission, and serum samples from cases and controls were stored at –20 °C till tested. The investigator who analysed the specimens for various immunological studies was blinded to the origin of the specimens (whether from a patient, hospital control or household control). The presence of serum IgA, IgG and IgM antibodies specific for C. jejuni was detected by enzyme-linked immunosorbent assay (ELISA) at dilutions of 1:50, 1:200 and 1:400 as described by Blaser and Duncan.9 A serum sample was considered positive when two or more immunoglobulin classes showed a value greater than or equal to the mean+2SD optical density of 40 uninfected children (OD ratio value) in ELISA. M. pneumoniae antibodies were detected by the gelatin particle agglutination test. For this, Serodia Myco II test kit (Fuji Rebio Inc., Tokyo, Japan) was used according to the manufacturer’s instructions. A titre of 1:80 was taken as indicative of recent infection. 10

Statistical analysis
The association of seropositivity in the patient and control groups was determined using the chi-square test. When one of the values was <5, the Fischer exact test was used.

RESULTS
Twenty patients with a clinical diagnosis of GBS were included in the study. Of these, 18 were men. The mean age of the patients was 31 (range 13–72) years. Eighty-five per cent of the patients were in the age group of 13–40 years. Most of the cases presented during the summer and rainy season. Three patients gave a history of diarrhoea, 8 of acute respiratory tract infection and 3 of fever in the preceding 3 weeks (Table I).

Table I. Frequency of preceding infections in patients with Guillain–Barré syndrome
Type of infection	n	C. jejuni serology	M. pneumoniae
Diarrhoea	3	2	1
Respiratory tract	8	0	5
Fever	3	0	1
None	6	5	4

Microbiological investigations
Stool cultures of all cases and controls were negative for C. jejuni. Of the 7 patients with serology positive for recent C. jejuni infection, 2 gave a history of gastroenteritis. One patient with a history of gastroenteritis was negative for C. jejuni serology. Of the 11 patients with serology positive for M. pneumoniae, 5 gave a history of acute respiratory tract illness. One patient with a history of fever was positive for M. pneumoniae serology and of the 6 cases with no preceding infection 3 were positive for M. pneumoniae and C. jejuni, 2 for C. jejuni and 1 for M. pneumoniae serology. The association between patients and household controls for both infections was not significant (Tables I and II).

Table II. Comparison of M. pneumoniae and C. jejuni positive serology in patients and controls
Organism	Patients n (%)	Hospital controls n (%)	p value	OR (95% CI)	Household controls n (%)	p value	OR (95% CI)
C. jejuni	7 (35)	0	0.008	Infinite	5 (25)	0.49	1.62 (0.34–7.88)
M. pneumoniae	11 (55)	3 (15)	0.008	6.93 (1.29–46.3)	5 (25)	0.49	3.67(0.81–17.74)
OR odds ratio CI confidence interval

Clinical data
There was no significant age or sex difference between those with and without evidence of recent infection. The mean (SD) duration of symptoms at the time of presentation was 4 (3.2) days (range 0.5–14 days). Three of the patients died during hospitalization. Of these, 2 were positive for M. pneumoniae serology and 1 for M. pneumoniae and C. jejuni serology. One patient had a variant of GBS with pure motor weakness and bilateral ptosis. Autonomic disturbances in the form of rhythm disturbances and fluctuating blood pressure were seen in 4 patients and cranial nerve involvement in 11.
   Electrophysiological studies revealed demyelinating neuropathy in 1 patient, 6 had totally unresponsive nerves and 13 could not be classified into any of the patterns of GBS. There was a statistical significance between the patients with recent c. jejuni infection and the hospital controls (p=0.008) but not with the household controls (p=0.49) (Table II). The association of recent M. pneumoniae infection was found to be highly significant when compared with hospital controls (p=0.008) and nearly significant with household controls (p=0.05) (Table II).

DISCUSSION
We used serology to detect recent M. pneumoniae and C. jejuni infections. The diagnosis of recent M. pneumoniae infections is usually based on serology, as culture is time-consuming and not readily available. 11 Although isolation from the faeces is the ideal method for the diagnosis of C. jejuni infection, 1 we were unable to isolate C. jejuni in our patients. This could be because the median duration of excretion of Campylobacter in the stool is only 16 days and because of the 1–3-week lag time between the infection and the onset of GBS, many patients with GBS and preceding Campylobacter infection might have had falsely negative stool cultures. 2 These limitations make serology the mainstay of diagnosis of antecedent C. jejuni infections in patients with GBS. The availability in India of over-the-counter antibiotics could have countributed to the negative culture results. A variety of antibody assays for detecting specific antibodies to C. jejuni have been reported. However, there are no standards for testing with regard to the antigen used or end-points for positivity. We used stringent serological criteria for the diagnosis of recent C. jejuni infection (positive ELISA for at least 2 classes of antibodies) to exclude false-positive results. A diagnosis based on positive ELISA for a single class of antibody may result in false-positive results. Rees et al. have found that an isolated IgM immune response to C. jejuni may also occur after Salmonella enteritis.12
   In our study, 35% of the patients with GBS had evidence of recent C. jejuni infection. This is higher than that reported by Rees et al. (26%), Winer et al. (14%) and in the Emilia–Romagna study group (15%). 12–14 Two retrospective studies that used an ELISA to detect C. jejuni infection, reported comparable seropositivity rates of 36% and 38%.1,2 These 2 studies were done at tertiary care referral hospital and proposed that the high C. jejuni seropositivity could be due to a selection bias favouring severe cases, as GBS associated with precedent C. jejuni infection has been reported to be more severe.1,12 Our study was also done at a tertiary care hospital and this could explain the high C. jejuni seropositivity in our patients with GBS. The incidence of C. jejuni diarrhoea reported from India is 4%–5%. 15,16 In our study, 35% of patients with GBS, 30% of household controls and none of the hospital controls were seropositive for C. jejuni. The case–control study by Rees et al. reported 26% seropositivity in patients, 2% in household controls and 1% in hospital controls.12 In the present study, high rates of seropositivity among household controls could be due to high rates of C. jejuni transmission and infection with suboptimal hygienic conditions in the respective geographical areas.
   M. pneumoniae seropositivity was found in 55% of patients with GBS as compared with 25% in household controls and 15% in hospital controls. These figures are much higher than those reported by others. Jacobs et al.3 reported M. pneumoniae seropositivity in 5% of patients with GBS compared with 1% of controls, and Hao et al. 5 reported M. pneumoniae seropositivity in 2% of patients with GBS as compared with 0.5% of healthy controls. The high frequency of M. pneumoniae infections in our patients and controls could reflect the high frequency of M. pneumoniae in the Indian population, which has been described previously.17,18 Six patients in our study had a positive serology for both infectious agents. As suggested by Jacobs et al.3 this may indicate that a dual antigen-induced immune response has a role in a subgroup of patients with GBS. We found that 5 patients did not recall a preceding episode of gastroenteritis but were C. jejuni serology positive. This might be due to asymptomatic infection as described earlier.5 Porter and Reid. reported 57 asymptomatic cases among 205 patients in an outbreak of C. jejuni infection.19 Similarly, only 5 of 11 patients with M. pneumoniae seropositivity gave a history of respiratory tract infection in the previous 4 weeks.
   This preliminary investigation suggests that there may be an association between recent C. jejuni and M. pneumoniae infections in patients with GBS in India.

REFERENCES