FOOD ALLERGY - FACT OR FICTION?

by Dr David W Hide FRCP, DCH, Honorary Consultant in Clinical Allergy, St Mary's Hospital NHS Trust, Newport, Isle of Wight

Food allergy resulting from the presence of IgE antibodies is common in childhood and also limits the diet of many adults. The author explains how to identify the allergen so that it can be avoided, and describes emergency treatment for accidental ingestion in severe cases.

A substantial number of adults restrict their diet in the belief that particular items of food upset them. Possibly even more children have their diet limited in some way because they are perceived to be intolerant to one or more items of food. Yet when the problem is discussed with the general practitioner, the possibility of food allergy may be greeted with scepticism or even with a total denial that such problems exist.

The cynical attitude many doctors adopt to food allergy is readily explained. The term has been grossly overused as an explanation for somatic and psychological ill-health. Any symptom or idiosyncrasy of behaviour not easily explained is inevitably attributed to "food allergy".

Much quasi-medical literature has reinforced the belief that identification of offending foods will relieve arthritis, skin disease, bowel disorders, migraine, enuresis and even cancer. Such claims are often associated with unproven diagnostic methods. So it is no surprise that many doctors, trained in the scientific method, are critical of the exaggerated claims of food allergy.

However, the doctor who takes an extremely sceptical view is doing a disservice to him/herself and to his/her patients. Food allergy does exist. It can be validated and its recognition will resolve ill-health and even, on rare occasions, be life-saving.

Definition of food allergy

The definition of food allergy adopted by the Royal College of Physicians and the British Nutrition Foundation is a form of food intolerance in which there is evidence of an abnormal immunological reaction to the food. This excludes other perceived or real reactions to foods. Thus, the idiosyncratic reaction to monosodium glutamate, or milk intolerance from primary or secondary lactase deficiency, should not be called "allergy". Neither is the dislike of any particular food which is better termed "food aversion". There are few long-term population studies to enable a cumulative prevalence to be given.

Foods that commonly cause allergic reactions.

Children:

• milk
• egg
• wheat
• peanuts
• nuts
• soya

Adults:

• fish
• shellfish
• nuts
• peanuts
• citrus fruit
• wheat
• milk
• egg

Allergy to cows' milk

Cows' milk allergy is most frequently encountered in infancy and early childhood. Prevalence rates vary from 0.3 to 7.5%. (1) A whole-population study of infants born on the Isle of Wight showed that 3.8% of infants had some manifestation of milk intolerance, and in 2.5% this could be regarded as allergy. (2)

Allergy to egg

Allergy to egg affects 0.5% of children, again most often in infancy.(3) It is often associated with atopic eczema. Many children, even some who have shown a profound hyper-sensitivity to milk or egg, lose this clinical intolerance as they grow older. In some, skin tests remain positive, although others become skin-test negative as the allergy disappears.

Allergy to egg should not be regarded as a contra-indication to the measles/mumps/rubella vaccination. The Department of Health advises that only generalised anaphylaxis to egg is a contra-indication to immunisation. In borderline patients it is prudent to administer the vaccine on a day-case basis in hospital.

Allergy to nuts

Allergy to nuts and to the pulse-related peanut is more persistent. There is concern, particularly in the USA, about the apparent increase in prevalence of peanut allergy. Three of six fatal cases of anaphylaxis to food reported recently were attributed to peanuts. (4) All had developed their allergy by the age of 18 months. Deaths from this cause are also being reported in the UK, but precise statistics do not exist.

Two deaths were reported from nuts in the same US report. Both were reactions to cashew nuts, one boy developing his allergy by the age of 18 months, the other at 5 years.

I have encountered allergy to Brazil nuts quite frequently, and have reported 12 cases, ranging in age from 1 to 39 years. (5) The onset of symptoms was rapid in all, occurring within 3 minutes of ingestion of the nut. Tingling of lips and tongue gave the first warning of the often severe anaphylaxis that was to follow. No patients died, but several were unconscious when seen and required immediate resuscitation. Ten of the 12 had other features of atopy. All nine in whom skin-prick tests were performed reacted positively. Specific IgE was identified by radio-allergo-sorbent testing (RAST) in eight out of eleven. Several individuals have now been followed for up to 10 years, and all appear to retain their allergy to Brazil nuts. Several are allergic to other nuts or peanuts as well, but in some the allergy appears to be exclusively to Brazil nuts.

Allergy to soya

Soya allergy is encountered more frequently in communities where soya milks are used as an alternative to cows' milk formula for infants. Initial reports suggested that soya-based formula may be less antigenic than that produced from cows' milk. They have therefore been used both to treat cows' milk allergy, and as a "hypo-allergenic" formula to prevent allergy in infants considered at high risk of developing such problems.

However, neither indication is appropriate. Soya milks are potentially as antigenic as cows' milk formulae. As many as 20% of infants with cows' milk allergy will become allergic to soya. (6) Other investigators have confirmed soya sensitivity in 4-5% of children with eczema.(7,8)

Allergy to wheat

Many other foods may cause allergy. In childhood, allergy to wheat may present as a transient event, causing gastro-intestinal symptoms and even a failure to thrive. This is associated with a partial villous atrophy, and should be differentiated from the persistent gluten intolerance that is coeliac disease.

Allergy to fish

Allergy to fish is encountered most frequently in communities where large amounts of fish are eaten. It occurs in 1 in 1,000 Norwegians, with a higher prevalence in children. In a series from Sweden, 39% of all children with food allergy were allergic to fish. (9) Although some children lose this allergy, for many it continues into adult life. There is cross-sensitivity between many species of fresh- and salt-water fish.

Food allergy, as defined here, is unlikely to affect more than 4% of children. The prevalence in adults is almost certainly lower than this.

Presenting signs

Food allergy can involve any system of the body, although it most frequently presents with gastro-intestinal symptoms. The immediate oral sensation, possibly including swelling of the tongue, alerts the older patient to his intolerance. Younger patients may react to the food by increased salivation, retching or explosive vomiting that can mimic the projectile vomiting of pyloric stenosis. Colic and abdominal distension may be manifestations of food allergy. Recurrent bleeding from the gut mucosa is now recognised as a sign of allergy to cows' milk.

Symptoms remote from the gastro-intestinal tract are not always associated with foods. Nevertheless, urticaria is a feature of anaphylaxis with immediate onset. The exacerbation of eczema by food may be rapid, or may follow many hours or even days after the offending food has been ingested. Presumably, the underlying immunological process is more complicated than the immediate IgE-mediated reaction that causes the rapidly occurring symptoms.

Respiratory symptoms

The role of food allergy in causing respiratory symptoms has yet to be confirmed. Some children with recurring rhinitis may be food allergic. Undoubtedly, a small percentage of both child and adult asthmatics are made worse by foods. Whether this is a true allergy, or an idiosyncratic reaction to a colorant or additive, should be determined.

Behavioural symptoms

The possible role of allergy in disturbances of the central nervous system, or of behaviour, is even more difficult to establish. Extravagant claims have been made that food-triggered mechanisms are responsible for intractable migraine, or for behaviour regarded as outside normal limits of acceptability. Possibly, in the highly selected series reported, food has been causal, but often food allergy is difficult to confirm and symptoms are not reproduced by blind challenge. It is unhelpful to dismiss the possibility of food provoking the symptoms, but it is equally inappropriate to interpret a presentation as food allergy when there is no indication of any immunological disorder.

Anaphylaxis

The most dramatic consequence of food allergy is immediate anaphylaxis. This occurs more often than is realised and is a cause of sudden death as frequently as the much better understood reaction to bees and wasps. The sudden onset of extreme distress, perhaps associated with swelling of the tongue, glottic spasm or wheezing, with a generalised urticarial rash, and accompanying hypotension and sensation of imminent demise, is extremely frightening. The patient should be educated to anticipate and manage any recurrence of the life-threatening experience. This alone makes knowledge of the identification and treatment of food allergy important to all practising physicians.

Diagnosis of food allergy

The history will often identify the offending food. If symptoms occur after all formula feeds, cows' milk is the obvious suspect. In older children or adults, the possibilities are many. The patient may be unaware that the food contained particular items. I recently encountered two cases where peanuts in pretzels and pine kernels in pesto sauce caused anaphylaxis. A very careful dietary analysis may reveal an unsuspected food.

When food allergy is strictly defined, the mechanism is usually IgE mediated. Therefore, the affected individual will have an increase in specific IgE to the offending foods. This may be revealed by skin-prick tests. Indeed, many parents of children with milk, egg or fish allergy will volunteer that skin contact with these foods produces an immediate wheal and flare reaction.

Alternatively, specific IgE may be measured by the more protracted and less sensitive method of RAST. This measures specific IgE antibody in serum, and the correlation with skin-prick testing is close. The procedure is more expensive and time-consuming, but in some circumstances, such as the patient with severe eczema, or who is taking antihistamines, may be preferable.

When a positive skin test, or RAST, occurs without supporting history, it should be interpreted cautiously. Many individuals have IgE antibodies without symptoms of intolerance to the food. In these circumstances, a trial of elimination of the food may be justified by the nature of the presenting complaint, such as eczema.

Food challenge testing

The definitive test of food allergy is claimed to be the double-blind, placebo-controlled, food challenge. The food is usually administered concealed in capsules and the patient is carefully observed. This so-called "gold standard" can itself be criticised. It is tedious, time-consuming and unphysiological, as not many people take food in capsular form. If the reaction to the food had been severe anaphylaxis, then food challenge would be hazardous and is not indicated. In practice, a probability of food allergy is often accepted if a suggestive history is accompanied by supporting skin tests or RAST. In some circumstances, an open challenge may be performed.

Treatment of food allergy

Once identified, the treatment of food allergy is avoidance. The mother of the milk-allergic child requires guidance as to which foods contain milk and milk proteins (see next section). There is rarely a problem in finding replacement sources of protein from fish, meat or pulses. Particular care must be taken to ensure that calcium intake is adequate, and calcium supplements may be necessary. The importance of a dietician in the management of food allergy cannot be overemphasised.

Foods that may contain cows' milk protein

• Butter
• Cheese
• Yoghurt
• Margarine
• Biscuits
• Savoury snacks
• Cakes and scones
• Chocolate
• Toffee
• Instant desserts
• Battered products
• Muesli
• Savoury meals
• Sauces
• Malted drinks
• Coffee whiteners

Many food allergies, particularly those occurring early in infancy, are of limited duration. The physician and dietician can advise how and when previously offending foods can be introduced into the diet. Sometimes, such food challenge is best carried out in hospital. It always should be done with extreme caution - starting with the smallest possible amount of milk, egg or other allergen. A fall in specific antibody level on the RAST or a negative skin test may suggest that such a challenge is appropriate.

Problems continue even when food allergy has been diagnosed. Many individuals are unaware of the ubiquity of cows' milk protein. Labelling of foodstuffs is not sufficiently reliable to prevent susceptible individuals from accidentally eating foods that should be avoided. A recent paper from the USA showed that 50% of individuals with known food allergy had, in the preceding 12 months, suffered reactions to food which they had inadvertently taken.(10)

Where food produces a major reaction or anaphylaxis, the individual, parent or other carer should have available, and know how to administer, adrenaline 1:1,000, 0.25-0.5 ml subcutaneously, repeated if necessary after 10-15 minutes. Some individuals with multiple and ill-defined food allergy may need to use their adrenaline on many occasions. The majority will never require it, and their greatest problem will be to remember to keep the adrenaline in date, as it has a shelf-life of only 9-18 months.

At present, immunotherapy, or desensitising, has not been established as a safe or effective option for the treatment of food allergy.

Conclusion

In this article, I have discussed only those food allergies resulting from the presence of IgE antibody. Different immunological mechanisms, perhaps utilising IgG or immune complexes, may explain other forms of food allergy. These tend to occur rather more slowly, hours or even days after ingesting the food. Some forms of cows' milk allergy fall into this group and these mechanisms probably occur in food-sensitive eczema. (11)

Food allergy, as a consequence of IgE-mediated reaction, is common in childhood and limits the diet of significant numbers of adults. Deaths from food-related anaphylaxis occur approximately as frequently as deaths from bee and wasp stings, yet they receive far less attention. Food allergy should be identified, and appropriate advice given on avoiding offending foods and the immediate action to take if ingested by accident.

Practical points

• Food allergy is a form of food intolerance where there is evidence of an abnormal immunological reaction to the food
• In infancy and childhood, cows' milk allergy has a prevalence of from 0.3 to 7.5%; allergy to egg affects 0.5% of children and is often associated with atopic eczema
• Many infants with cows' milk allergy will become allergic to soya; other common food allergies include nuts, wheat and fish
• Presenting symptoms of food allergy include angioedema, vomiting, abdominal colic and distension, recurrent bleeding from the gut, urticaria and exacerbation of eczema
• Food allergy may lead to anaphylaxis and sudden death; this occurs as frequently as sudden death caused by bee and wasp stings
• Diagnosis of food allergy is confirmed by skin tests, a radio-allergo-sorbent test or a food challenge
• Adrenaline 1:1,000 should be available for subcutaneous injection in the event of a food reaction in those known to be susceptible

References

1. Bahna SL, Heiner DC. Cows' milk allergy: pathogenesis, manifestations, diagnosis and management. In: Barness LA et al, eds. Advances in pediatrics, vol. 25. Chicago Year Book, 1978, pp. 1-37.
2. Hide DW, Guyer BM. Cows milk intolerance in Isle of Wight infants. Br J Clin Pract 1983; 37: 285-7.
3. Ford RPK, Taylor B. Natural history of egg hypersensitivity. Arch Dis Child 1982; 57: 649-52.
4. Sampson HA, Mendelson L, Rosen JP. Fatal and near-fatal anaphylactic reactions to foods in children and adolescents. N Engl J Med 1992; 327: 380-4.
5. Arshad SH, Malmberg E, Krapf K, Hide DW. Clinical and immunological characteristics of Brazil nut allergy. Clin Exp Allergy 1991; 21: 373-6.
6. Gerrard JW, Mackenzie JWA, Goluboff N, Garson JZ, Maningis CS. Cow's milk allergy: prevalence and manifestations in an unselected series of newborn. Acta Paediatr Scand 1973; 234 (Suppl.): 1-21.
7. Sampson HA. The role of food sensitivity and mediator release in atopic dermatitis. J Allergy Clin Immunol 1988; 81: 635-45.
8. Businco L, Bruno G, Giampetro PG, Cantani A. Allergenicity and nutritional adequacy of soy protein formulas. J Pediatr 1992; 121 (Suppl.): S21-8.
9. Dannaeus A, Ingunas M. A follow-up study of children with food allergy: clinical course in relation to serum IgE and IgG antibody to milk, egg and fish. Clin Allergy 1981; 11: 533-9.
10. Bock SA, Atkins FM. The natural history of peanut allergy. J Allergy Clin Immunol 1989; 83: 900-4.
11. Hill DJ, Firer MA, Ball G, Hosking CS. Natural history of cows' milk allergy in children: immunological outcome over 2 years. Clin Exp Allergy 1993; 23: 124-31.