Diagnosing, Treating and Preventing Food Allergy
WSAVA 2002 Congress
Richard EW Halliwell, MA VetMB PhD MRCVS DECVD
Professor, University of Edinburgh, Royal (Dick) School of Veterinary Studies, Summerhall
Edinburgh, UK


Adverse reactions to foods can have an immunological basis (food hypersensitivity) or can develop without involvement of the immune system (food intolerance). The latter can develop form a number of different reaction types, namely direct toxicity, idiosyncratic reactions, pharmacological reactions and metabolic reactions. In some instances, intolerance and hypersensitivity may coexist in the same patient.


Large quantities of foods are presented to the gastrointestinal tract over the lifetime of an animal, and it is known that significant amounts are absorbed in an antigenically recognisable state. This is limited by,

1.  A vigorous local immune response, especially involving IgA, and

2.  A backup mechanism of immune elimination via the liver.

Despite these damage limitation exercises, appreciable amounts of food antigens are still absorbed. In many situations, systemic tolerance to the food allergens results. Nonetheless, IgG antibodies to many food allergens are detectable in the serum of both dogs and man. Generally, they are non-pathogenic.


Although the gastrointestinal tract has been shown capable of supporting all recognised types of hypersensitivity, it seems likely that IgE is predominantly involved, although some believe that high levels of IgG antibody may be pathogenic. Where local hypersensitivity results, GI disease follows. Systemic hypersensitivity results in the involvement of other systems, and especially the skin.


The incidence of food hypersensitivity as a cause of allergic dermatitis is quite controversial, and in many cases food allergy may co-exist with atopy, and a partial improvement only results from feeding a hypoallergenic diet. In general, most clinicians agree that food allergy is much less frequent than is atopy-probably occurring with around 10% of the frequency of the latter.

Food allergy can start at any age, with approximately one third of cases commencing clinical signs prior to one year of age. Thus commencement of clinical signs at a very young age, or after 7-8 years of age will raise the index of suspicion for food allergy as compared with atopy.

Clinical signs

Dermatological signs

Pruritus is present in almost all cases, with a small proportion presenting as severe seborrhoea. Many feel that a primary eruption is more likely to be seen than in atopy. It is a common cause of recurrent pyoderma, and especially folliculitis which relapses very soon after discontinuing the antibiotic therapy. It has also been shown to be a predisposing cause for German shepherd dog pyoderma (Rosser, 1997). Major clinical features of 3 reported case studies are noted in Table 1

Table 1. Incidence of various clinical signs of food allergy

White (1986)

Rosser (1993)

Carlotti (1990)
















Otitis externa




Distribution of dermatological signs

The distribution of signs in cases of food allergy is highly variable, and is generally not characteristic. They are unlikely to mimic flea allergy, but sometimes they may mimic atopy.

Gastrointestinal signs

These range from the trivial to the severe, and may co-exist with the dermatological signs. Although most authors state that cases involving both systems are in the minority (at most 10-15%), others state that very minor signs, such as increased frequency of defecation-sometimes with soft stools, excessive borborygmi and wind, frequently accompany cases of dermatological disease resulting from food allergy. In a recent study, every one of 20 cases diagnosed as suffering from adverse food reactions, and presented primarily for dermatological signs, showed some manifestation of gastrointestinal disease (Paterson, 1995)

Adverse reactions to foods resulting in gastrointestinal disease are likely to be far more common that reported in the literature, in that in many cases of diarrhoea and/or vomiting, the owner makes an association with the feeding of different foods, without ever consulting the veterinarian. Chronic gastrointestinal diseases for which food hypersensitivity should be on the top of the list of differentials are canine lymphocytic plasmacytic enteritis, canine eosinophilic gastroenteritis and canine colitis. In all of these, an inflammatory infiltrate with cells known to be involved in immunological responses is seen, and the triggering factor could well be dietary antigen.


Clinical signs

Clinical signs suggestive of food allergy include:

1.  Perennial pruritic skin disease in which atopic dermatitis from house dust mite allergy has been ruled out.

2.  Intermittent skin disease in animals fed a varied diet.

3.  Pruritic disease starting below one year of age, and particularly less than 6 months of age.

4.  Distribution of clinical signs that are inconsistent with flea allergy and atopic dermatitis.

5.  Co-existence with gastrointestinal signs, which may range from the subtle to the severe.

6.  Cases of recurrent pyoderma which relapse quickly after the cessation of antibiotic therapy.

7.  Cases in which recurrent otitis externa is a major feature, and where no other cause for the latter can be identified.

Intradermal skin tests

It is generally believed that skin tests are of little or no value in the diagnosis of food allergy. In a study in Florida, the incidence of positive reactions in suspected food-allergic dogs was 48%. However, when 30 of these dogs were fed a hypoallergenic diet, only 3 improved. In contrast, when 52 of those without reactions were fed a hypoallergenic diet, 6 improved (Kunkle and Horner, 1992).

In vitro tests

Published data on the use of in vitro tests assaying allergen-specific IgE is again not supportive of their value (Jeffers et. al. 1991). However, more recent data in the author's laboratory which will be presented at the meeting, suggests that a further investigation of ELISA is warranted. In contrast to reported experience with commercial ELISAs, the incidence of positive reactions is very low, and will certainly be of value in aiding the selection of a hypoallergenic diet for clinical use, if not truly diagnostic.

Hypoallergenic diet trial

The definitive way to confirm the diagnosis is by observing remission of clinical signs upon feeding a hypoallergenic diet, and noting a relapse upon reintroduction of the original diet. However, of course, this will not differentiate hypersensitivity from intolerance. Although there are now many commercial diets on the market, the possibility of reactions to non-nutrient constituents means that there is no substitute for the feeding of a home-prepared diet. The following are the guidelines:

1.  A single protein source is selected, based upon the dietary history, or results of in vitro tests.

2.  A single carbohydrate source is also chosen-ordinarily potato, which is preferred to rice.

3.  This is fed exclusively for a minimum of 6 weeks, taking care to exclude treats, rawhide bones etc. If there is any improvement at all, it is continued for a further 4 weeks.

4.  The initial diet (or components of it) is then introduced to observe for recrudescence of the clinical signs.


Commercial diets

Having made the diagnosis using a home-prepared diet, the goal is to introduce a commercial hypoallergenic diet for long-term control. Many are now available, including those containing digested protein sources, which are supposedly no longer allergenic.

Home-prepared diets

If it proves necessary to use a home-prepared diet, care must be taken to ensure that it is adequately balanced with minerals and vitamins-especially in growing dogs and the larger breeds.

Anti-allergic therapy

In rare instances, it proves impossible to find a diet that the patient can tolerate. Also, on occasions, the animal becomes asymptomatic for a while, and then becomes sensitised to dietary antigens in a sequential manner. The use of corticosteroids, or even more potent immunosuppressive/antiinflammatory agents such as azathioprine or cyclosporine A becomes necessary.


1.  Carlotti, D-N, Vet. Dermatol. 1:55, 1990

2.  Jeffers, J et al. J. Am. Vet. Med. Assoc. 198:677, 1991

3.  Kunkle G et al. J. Am. Vet. Med. Assoc. 200:245, 1992

4.  Patterson, S, J small Anim. Pract. 36:529, 1995

5.  Rosser, EJ, J. Am. Vet. Med. Assoc. 203:259, 1993

6.  Rosser, EJ, J. Am. Anim. Hosp Assoc 33:355, 1997

7.  White, S. J. Am. Vet. Med. Assoc. 188:695, 1986

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Richard EW Halliwell, MA, VetMB, PhD, MRCVS, DECVD
Professor, University of Edinburgh, Royal (Dick) School of Veterinary Studies, Summerhall
Edinburgh, UK

MAIN : : Food Allergy
Powered By VIN