Utility of serum Campylobacter specific antibodies in determining prior Campylobacter infection in neurological disease

Abstract

Campylobacter jejuni has been implicated in the pathogenesis of Guillain-Barre syndrome (GBS); however, little information exists on the utility of Campylobacter serology in determining recent infection in the patient population. C. jejuni specific antibodies (CAs) were measured in 420 blood donors (controls), 99 patients with recent C. jejuni infection, 34 patients with central nervous system disorders (neurology controls), and 44 patients with peripheral nervous system (PNS) disorders: 18 with GBS, 12 with MND and 14 with chronic inflammatory neuropathies. Elevated CA titres consistent with recent C. jejuni infection were found in six of the 44 patients with PNS disorders (three with GBS, two with neuropathy, and one with MND, only one of whom had a history of recent C. jejuni infection), compared with two of 454 controls (p = 0.00001). Therefore, we conclude that CAs are often raised in patients with PNS disorders who do not have a history of recent C. jejuni gastroenteritis, so Campylobacter serology may be an unreliable marker of recent infection in this patient group.

Introduction

There is a clear association between prior infection with the Gram-negative bacteria Campylobacter jejuni and the development of immune-mediated inflammatory peripheral neuropathies, particularly the acute motor axonal neuropathy (AMAN) form of Guillain-Barre syndrome (GBS).[1], [2], [3] This association is based on the development of GBS following acute gastrointestinal infection with C. jejuni, first reported in 1982.⁴ An association with particular subtypes of C. jejuni has also been noted: Penner type HS:19 in Japan,[5], [6] and HS:41 in South Africa.⁷

Whether C. jejuni infection is the antecedent trigger for development of the demyelinating form of GBS, acute inflammatory demyelinating polyneuropathy (AIDP), is less clear, with some doubting such a link.³Campylobacter jejuni infection has been implicated in the aetiopathogenesis of AIDP solely based on the finding of elevated titres of C. jejuni specific serum antibodies,[8], [9] often without a recorded antecedent gastrointestinal infection typical of C. jejuni enteritis. In the Western world, C. jejuni infections are almost always associated with acute and often severe gastroenteritis; asymptomatic infections are rare, and chronic carriage of the organism in an immune competent host occurs very infrequently, if at all.[10], [11] In a study of 1049 stool samples from Japanese patients with GBS and the Miller-Fisher syndrome (MFS), Takahashi et al. noted that antecedent enteritis symptoms were reported in 88% of C. jejuni culture-positive patients (n = 112; 11% of all samples), but upper respiratory tract symptoms in only 5%, fever in only 3% and no symptoms in 4%.⁶

Following acute enteritic infection with C. jejuni, the mean duration of excretion of the organism in the stool is 2–3 weeks.[12], [13] Various studies have isolated C. jejuni from the stools of patients with GBS in 4–64% of cases.[6], [14], [15], [16], [17] Given that GBS occurs a mean of 9 days postinfection, it is likely that a significant proportion of patients will have cleared C. jejuni from their stools prior to the development of neurological symptoms. As a consequence, a number of investigators have attempted to use serology as a means of determining prior C. jejuni infection in patients presenting with GBS. The findings of these studies are summarized in Table 1. Together, these findings suggest that the rate of prior C. jejuni infection is in the region of 12–15% in an unselected GBS population, with the vast majority having the AMAN variant. The figure of 15% of GBS patients is consistent with the figure arrived at by Tam et al. in 2003 from a large study of infectious intestinal diseases in the UK.¹⁸

The utility of Campylobacter specific antibodies (CAs) in the diagnosis of acute C. jejuni infection has been controversial. We have recently described and validated our C. jejuni antibody assay.¹⁹ The enzyme-linked immunosorbent assay (ELISA) method used was highly specific for CAs, but not particularly sensitive, with only two-thirds of acute C. jejuni infections associated with significant rises in at least two classes of CAs as per criteria established by Kaldor et al.²⁰ CAs can remain elevated following an acute infection, in some cases for years.¹⁹ Similar levels of sensitivity were reported by Strid et al. in a Danish study.²¹

It is possible that CAs in the sera of patients with GBS represent antibodies cross-reacting with some components of peripheral nerves damaged during the immune attack rather than recent C. jejuni infection. There exists significant evidence for molecular mimicry between Campylobacter epitopes and ganglioside epitopes, this being one of the favoured mechanisms of Campylobacter-induced AMAN, MFS and AIDP.[22], [23], [24], [25], [26], [27], [28]

Guillain-Barre syndrome is associated with the development of a non-uniform anti-ganglioside antibody response, with antibodies to gangliosides GM1, GM2, GD1b, GD1a and others recorded in varying numbers of cases. For a review of this issue see Willison and Yuki.²⁹

Therefore, we hypothesized that elevated titres of CAs in patients without clinical evidence of a recent acute C. jejuni infection may represent an epiphenomenon of cross-reactivity to antibody epitopes expressed by peripheral nerves undergoing immune attack or degeneration. If this were the case, we would expect to see elevated titres of CAs in active peripheral neuropathies such as GBS/MFS/multifocal motor neuropathy(MMN)/AMAN and in active neuronopathies such as motor neuron disease (MND), but not in neurological diseases without peripheral nervous system (PNS) involvement.