Intended for healthcare professionals


Lack of oats toxicity in coeliac disease

BMJ 1997; 314 doi: (Published 18 January 1997) Cite this as: BMJ 1997;314:159

Toxic fraction makes up less of total protein than in other cereals

  1. Jacques Schmitz, Professora
  1. a Gastro-Entérologie Pédiatrique, Hôpital des Enfants Malades, 149 rue de Sèvres, 75743 Paris Cedex 15, France

    In his pioneering study of “the harmful effects of certain types of cereal on patients suffering from coeliac disease,” Dicke showed that wheat and rye could reproducibly trigger anorexia, diarrhoea, and steatorrhoea in these patients.1 Soon after, using the same prolonged fecal balance studies, Dicke found that oats were also noxious whereas corn, rice, and potatoes were not.2 3 Reports suggesting that barley was toxic came later.4 5 Simultaneously, the “injurious constituent of wheat” was found to be its prolamin (or alcohol soluble protein), gliadin.3 Secalin, hordein, and avenin, the prolamins of rye, barley, and oats respectively, were thus considered as the toxic fractions of these cereals.

    However, whereas the noxious effects of wheat, barley, and rye could be reproduced, the harmful effects of oats remained controversial–observed by some workers,2 3 denied by others,6 7 and variable for still others.5 8 This uncertainty stems from several factors. Firstly, the early studies included small numbers of patients (from two to 12) and followed them for short periods (from several weeks to less than three months). Secondly, the methods used to determine the harmful effects varied from insensitive functional tests–balance studies2 3 8 and xylose tests5 –to histological4 and biochemical studies7 of small intestinal biopsies that were sometimes difficult to interpret. These methodological limitations are relevant when considering the variability of gluten sensitivity from patient to patient; several years of a gluten containing diet are sometimes necessary before a patient will relapse.9 However, despite the variability of the clinical and histological responses observed after these early challenges with oats, it seems clear that taking small amounts of oats (about 50 g) for short periods (less than one month) is not generally noxious whereas more than 100 g for longer than a month leads to recurrence of steatorrhoea, a strong sign of serious mucosal damage.2 3 8

    A recent Finnish study avoided the pitfalls of variability by following a large number of patients (92) for one year and using stringent histological criteria of mucosal damage. Patients recently diagnosed or in remission were given a gluten free diet and randomised to receive either no oats or 50 g of oats a day. Severe cases were excluded. After one year, the two groups showed no significant difference in clinical symptoms, laboratory measures, or histological criteria. All the newly diagnosed patients were in clinical and histological remission. The conclusion is straightforward: moderate amounts of oats (40-60 g/day) are not toxic in most patients with coeliac disease.10

    The recent study by Srinivasan et al points in the same direction.11 Ten patients, including two who were particularly sensitive to gluten, consumed 50 g of oats as porridge daily for three months while maintaining a strict gluten free diet. During the challenge the patients remained symptom free and maintained low titres of antiendomysium and antiangliadin antibodies. Quantitative histological evaluation at the end of the observation period showed no change in mucosal appearance, in particular no increase in intraepithelial lymphocyte count. Thus, as in the Finnish study, although with a smaller number of patients and shorter duration of challenge, moderate amounts of oats proved non-toxic. These findings have recently been confirmed in patients with dermatitis herpetiformis,12 and in the case of the Finnish study, by a five year follow up.13

    Although concordant, these studies do not show that larger daily amounts of oats (100-160 g) would be equally non-toxic in these patients. Indeed, as suggested by the early experiments,2 3 8 large amounts of oats are theoretically likely to be toxic. The oat is a member of the Avena tribe, whereas wheat, rye, and barley are members of the neighbouring Triticeae tribe, both tribes being part of the Pooideae subfamily. Thus, avenin, the prolamin of oats, is genetically less like gliadin than are secalin and hordein. Despite this greater difference, sequence homologies (and weak immunological cross reactivity) have been found between avenin and the prolamins from barley, wheat, and rye.14 15 16 Moreover, avenin accounts for only 5-15% of the total protein in oats compared with the 40% contribution from gliadin in wheat and the prolamins in rye and barley.14 Thus, taking into account a smaller number of toxic sequences per unit weight of avenin and the smaller amounts of avenin as a proportion of the oat seed proteins, it seems likely that only considerable amounts of oats consumed over long periods will be toxic for patients with coeliac disease.

    However, until randomised studies are performed on large enough numbers of patients consuming large amounts of oats for long periods, it seems reasonable to assume that moderate amounts of oats may be consumed by most patients without risk.10 10 As noted by Watson,17 in Scotland “it would have been obvious many years ago if coeliac children and adults who are taking porridge relapsed.”


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    View Abstract