Article Text


Gluten exposure and risk of autoimmune disorders
  1. F Biagi,
  2. D Pezzimenti,
  3. J Campanella,
  4. G R Corazza
  1. IRCCS Policlinico S Matteo, Università Degli Studi di Pavia, V le Golgi, 19, 27100 Pavia, Italy
  1. Correspondence to:
    Professor GR Corazza;

Statistics from

We applaud the paper from Sategna Guidetti et al (Gut 2001;49:502–5) showing that, contrary to what was described in a paediatric population,1 there is no clear relationship between gluten exposure and risk of autoimmune disease (AID) in adult patients affected by coeliac disease (CD).

Although in our clinical experience we have seen CD patients developing AID after many years on a strict gluten free diet, we thought that the data published by Ventura and colleagues1 should have been easily confirmed in an adult population. However, in common with Sategna Guidetti et al, this did not appear to be the case in the coeliac patients under our care. Moreover, although Sategna Guidetti et al could at least confirm the relationship between age at diagnosis of CD and risk of developing AID found by Ventura et al, we could not confirm this finding.

We retrospectively studied the notes of 462 adult patients affected by CD attending our outpatient clinic. All had been diagnosed on the basis of villous atrophy in jejunum biopsy specimens which improved after withdrawal of dietary gluten. There were 327 females and mean age at diagnosis of CD was 33.3 ± 15.6 years (range 1–79). Ninety six patients were affected by at least one AID. Age at diagnosis of CD did not differ between CD patients with and without AID (32.6 v 33.5 years; p=0.6). The prevalence of AID was not related to age at diagnosis of CD (κ2, p=0.7). However, an upward trend, similar to that shown by Ventura et al, was noted in the first three decades (fig 1).

Since we had previously shown that diagnostic delay is a very important feature in the history of CD,2 we hypothesised that the longer the diagnostic delay the greater was the risk of developing AID. However, the prevalence of AID was not related to duration of diagnostic delay. Moreover, remarkably, diagnostic delay in patients with CD and AID was significantly shorter than that of patients with CD but not AID (9.1 v 13.2 years; p=0.02). We feel that this is a sensible result. The more reasons a patient has to be diagnosed, the more likely it is that he/she will be diagnosed sooner. However, this strongly argues against a relationship between the risk of AID and exposure to gluten.

Finally, as we agree with Sategna-Guidetti et al on the fact that dermatitis herpetiformis should not have been considered, we analysed our data both including and excluding this skin condition. The final outcome was unaffected.

In conclusion, our data in adult patients with CD confirm those of Sategna-Guidetti et al. Our only finding that was similar to that of Ventura et al was an upward trend in the first three decades. As patients affected by latent CD clearly prove that CD can start in adult life and not exclusively during childhood,3 this upward trend could mean that the findings of Ventura et al are valid only for the paediatric age. However, as Londei points out, the only way to answer this question would involve an unacceptable prospective study.4

Figure 1

Prevalence of coeliac disease (CD) patients affected by autoimmune disease (AID) according to age at diagnosis of CD.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.