Dear Editor

We are concerned about the conclusions of the recent publication by RJ Lock et al [1] on the subject of gluten ataxia and food antibodies.

A number of previous editorials written by the same authors express doubts on the existence of gluten ataxia as a disease entity as well as the whole notion of neurological involvement in coeliac disease and gluten sensitivity. They have proposed large studies looking at the prevalence of these antibodies in neurological populations. Within this context the current study by the same authors lacks the power to prove or disprove the existence of gluten ataxia. Screening only 20 patients with idiopathic ataxia and 7 patients with familial ataxia and lacking data on the prevalence of food antibodies in the healthy control population renders this study statistically uninformative.

Despite the small patient numbers, their findings are broadly in keeping with our previously published data. We reported that the prevalence of antigliadin antibodies in sporadic ataxia was 41% (54 out of 132 patients)[2], compared with their value of 40%. This compares with a prevalence in the healthy population of Sheffield of 12%.

Unfortunately, their conclusions however are misleading and not justified by the data presented:

The authors acknowledge in their abstract that the group of 7 patients with familial ataxia that they screened is very small and statistical analysis was considered inappropriate. It therefore follows that it is unsound to conclude, as they do, that ..food antibodies were equally common in hereditary ataxia.

The presence of enteropathy (which defines CD) is independent of the diagnosis of gluten ataxia. Gluten ataxia is a disease entity discovered and defined by our group as sporadic idiopathic ataxia with positive antigliadin antibodies in the absence of alternative aetiology for the ataxia. It has an immunological basis the evidence for which includes the reactivity between antigliadin antibodies and cerebellar Purkinje cells, the presence of oligoclonal bands in up to 50% of patients, the inflammatory pathology in cerebellar tissue, and the higher prevalence of additional autoimmune diseases seen in patients with gluten ataxia.[3] A systematic study has demonstrated improvement of the ataxia associated with adherence to a strict gluten-free diet, even in those patients with gluten ataxia without enteropathy.[4]

It is true that a subgroup of patients with gluten ataxia (12%, as per the prevalence in the healthy population) may have antigliadin antibodies that are coincidental rather that aetiologically linked to the ataxia. At present there is no better serological marker to distinguish these patients. We have recently demonstrated the presence in vivo of IgA antibodies against TG2 in the small bowel biopsy of patients with gluten ataxia, a finding typical of patients with CD, dermatitis herpetiformis and patients with potential CD.3 This finding may help in distinguishing those patients with gluten ataxia from those with coincidental positivity for antigliadin antibodies.

Interpretation of the authors' finding of DQ2 in 50% of only 8 patients with idiopathic ataxia and positive antigliadin antibodies contrast with our finding of 70% DQ2 in 100 patients with gluten ataxia, but is problematic to interpret. The authors would need to study 180 patients to be sure that this prevalence was significantly different from that which we have previously reported, making any conclusion from their study of only 8 patients unsound. Furthermore, the authors seem not to appreciate that the presence of the DQ2 haplotype is not essential to the diagnosis of gluten ataxia.

The authors conclude that there is a striking absence of serological evidence for occult CD in their patients with sporadic ataxia, despite the presence of circulating antigliadin antibodies of 40%. Serological evidence of CD based solely on anti-tTG positivity (which is what these authors suggest) does not equate to the presence of CD (defined by the demonstration of an enteropathy). We have several patients with gluten ataxia and gluten enteropathy who had circulating IgG antigliadin antibodies and with neither circulating antiendomysium nor transglutaminase antibodies. Without biopsies no conclusions can be drawn as to the presence of enteropathy particularly in this population that differs from those patients presenting to gastroenterologists. As to the significance of the presence of circulating antigliadin antibodies the authors should perhaps be reminded that the landmark studies that first highlighted the fact that CD was much commoner in the so-called healthy population used antigliadin antibodies as their screening tool. [5,6]

The vast majority of studies in the literature on the prevalence of antigliadin antibodies/gluten sensitivity with or without CD (including this one) confirm a higher prevalence in sporadic ataxias. All studies have been summarised in detail in a previous publication by our group.[7] As the prevalence of CD in the healthy population is now known to be 1%, in order to disprove a higher prevalence of CD in patients with idiopathic ataxia (say 12% as we have reported[7]) the authors would have to screen 120 patients, assuming a power of 80% and an 1 value of 0.05.

The authors' statement HLA DQ2 expressing, antigliadin positive cases were rare in our clinics (four cases in 2 years from a population of 2 million) is misleading as it implies that the authors have screened all patients with idiopathic ataxia out of 2 million population. The authors recognise the fact that they have not done so as a major weakness of their study yet still include this comment in their abstract.

Although the definition of gluten sensitivity remains controversial for some, published literature supports the contention that gluten sensitivity represents a diverse spectrum of which coeliac disease is just one part. Mike Marsh has shown that the mucosal abnormalities in gluten sensitivity range from normal (Marsh grade 0) to severe enteropathy (Marsh grade 3) [8]. He went on to demonstrate enteropathy by increasing the gluten load in those patients with grade 0. This is what lead him to define gluten sensitivity as a state of heightened immunological responsiveness to ingested gluten in genetically susceptible individuals. This widely adopted definition makes no reference to the presence or absence of an enteropathy. How then does one define gluten sensitivity in a patient with Marsh grade 0 (i.e. normal mucosa on biopsy)? Given that antiendomysium and to a lesser extend tissue transglutaminase antibodies are not present in the absence of enteropathy then currently only antigliadin antibodies (ie the antibodies against the aetiological agent of the disease) can be used in defining the spectrum. The finding of circulating food antibodies in this cohort of patients is irrelevant to the issue of gluten sensitivity. The presence of food antibodies might reflect altered permeability of the bowel as a result of gluten sensitivity, but in the absence of a healthy control group to compare it is difficult to conclude anything about the significance of such antibodies. In contrast, gluten is an antigen which can trigger a spectrum of immune disorders.

References

1 Lock RJ, Pengiran Tengah DSNA, Unsworth DJ, Ward JJ, Wills AJ. Ataxia, peripheral neuropathy, and amtigliadin antibody. Guilt by association? J Neurol Neurosurg Psychiatry 2005;76:1601- 1603

2 Hadjivassiliou M, Gr|newald RA, Sharrack B et al. Gluten ataxia in perspective: epidemiology, genetic susceptibility and clinical characteristics. Brain 2003;126:685-691

3 Hadjivassiliou M, Mdki M, Sanders DS, Williamson CA , Gr|newald RA, Woodroofe N, Korponay-Szabs IR. Autoantibody targeting of brain and intestinal transglutaminase in gluten ataxia. (Neurology 2005, in press)

4 Hadjivassiliou M, Davies-Jones GAB, Sanders DS, Gr|newald RAG. Dietary treatment of gluten ataxia. J Neurol Neurosurg Psychiatry 2003;74(9):1221-1224.

5 Grodzinsky E, Franzen L, Hed J, Strom M. High prevalence of celiac disease in healthy adults revealed by antigliadin antibodies. Annals of Allergy 1992; 69:66-69.

6 Catassi C, Ratsch IM, Fabiani E, Rossini M, Bordiccia F, Candela F, Coppa GV, Giorgi PL. Coeliac disease in the year 2000: exploring the iceberg. The Lancet 1994;343:200-3.

7 Hadjivassiliou M, Williamson CA, Woodroofe N. The immunology of gluten sensitivity: beyond the gut. Trends in immunology 2004;25: 278-282.

8 Marsh M.N. Gluten, Major Histocompatibility Complex, and the small intestine. Gastroenterol 1992;102:330-354

9 Wills A, Dale R, Giovannoni G. Gluten Ataxia and Post- streptococcal Central Nervous System Syndromes: Emerging Immune-mediated disorders of the Central Nervous System? Curr Treat Options Neurol 2005;7:183-189