This issue was sponsored by Unilever Bestfoods Robertsons (Pty) Ltd http://www.unilever.co.za/

Contents A. Case study B. More information C. Editors' comments D. References E. CPD questions (South Africa, Australia)

Index

A. Case study
A 5-month-old girl was taken to the doctor because she had developed generalized eczema. She was on a cow's milk-based formula exclusively. After taking a brief history, the doctor performed a skin prick test for cow's milk on the girl. It was positive (10mm). The mother was asked to change the formula to soy formula. The child's symptoms improved dramatically. She still had very slight eczema at times, but it was manageable with steroid creams. At the age of 13 months the eczema relapsed. It seemed to be a constant rash (with no specific flare-ups). By this age she was already eating many different foods, but she was still on a soy formula (the follow-on formula).

THOUGHT PROCESS:
What could be the cause of the reaction re-occurring?
a. She was being exposed to a hidden source of cow's milk.
b. Something had been added to her environment (e.g., exposure to house dust mite, pets, stress, smoke, perfumes or fabric softeners) that she could be reacting to or that could aggravate the reaction.
c. She was reacting to a specific food or foods in her diet.

DISCUSSION:
a. A careful diet history was taken. No hidden sources of cow's milk could be identified.
b. The mother said that the house was a bit dusty at times; they did have a dog, but she had not seen the child having a reaction after contact with the dog. The child was not under any type of stress. No one in the house smoked. And she had changed the laundry detergent and related items a couple of months before (to brands that would more certainly not affect her child). House dust mite and the dog might therefore play a role. However, there were no recent changes that would bring the child into contact with these more regularly.
c. No one specific food could be identified as the cause, as the child had a constant rash (i.e., there were no flare-ups that could be directly associated with a specific food). The mother had, however, only recently introduced fish into the child's diet.

THOUGHT PROCESS:
What could be done at this point?
a. Consider other allergens as a possible cause by doing more skin prick tests, or do serum-specific IgE tests.
b. Ask the mother to keep a food-symptom diary for the child, to determine which food could be causing the problem.
c. Put the child on an elimination diet until the symptoms have disappeared and then challenge the child to determine the culprit food.

DISCUSSION:
a. This could be helpful.
b. The diary would probably not be helpful in the case of a child experiencing constant eczema. One would not be able to pinpoint a specific food as the cause.
c. This is also a possibility, but only after skin prick tests are done.

A range of skin prick tests were performed: house dust mite, dog (as she was exposed to these in her environment), cow's milk, soya, wheat, egg and fish (as these are 5 of the 6 main allergens; she had not been exposed to peanut yet). The results were:
Histamine control: 8mm
House dust mite: 6mm
Dog: 0mm
Soya: 0mm
Wheat: 0mm
Cow's milk: 0mm
Egg: 15mm
Fish: 12mm

DISCUSSION:
What do the results of the tests mean?
The results of SPTs cannot be interpreted on their own. One has to consider the history, the symptoms, and the results of the other allergen SPTs.
a. House dust mite:
The SPT wheal was smaller than that of the histamine control, which could have indicated a number of things (see below). The reaction could explain the slight eczema that she continued to experience after the exclusion of milk. She probably had been exposed to house dust mite since birth and therefore could have been reacting to it since then.

b. Dog, soya and wheat:
Comparing these negative results with the histamine control and the positive results of the other allergens, it is very unlikely that the child was reacting to any of these. (SPTs have a >95% negative predictive value.)

c. Cow's milk:
Why was the cow's milk test negative? Was the previous test done incorrectly, or did the patient outgrow the allergy? Did this mean that milk could be reintroduced into the child's diet?
The first test could have been done incorrectly, but the child's symptoms did improve following the exclusion of cow's milk. A more likely reason is that, because the child was not exposed to the allergen for the previous 8 months, the body had not produced any IgE to cow's milk during that period and IgE would therefore not show up on the skin prick test, whether the allergic tendency persisted or not. To outgrow a sensitivity to an allergen requires a period of time that varies between individuals. The longer the allergen is avoided, the more likely it is that the sensitivity has been outgrown. This child may not have outgrown the allergy and could still react to cow's milk if she ingested it. It was suggested that the child be challenged with milk (to see whether she was still sensitive to it and would react), but only once the cause of the current eczema was established, the culprit food(s) excluded, and the symptoms brought under control. Only then would it be possible to see the effects of a challenge.

d. Egg and fish:
The child had started eating egg and fish only in the last couple of months and therefore had probably developed a sensitivity to them.

To confirm egg and fish allergy, the mother was asked to exclude all sources of these from her daughter's diet until the symptoms disappeared. Thereafter, the child could be challenged with each allergen to see whether she in fact did react to each. The mother was also given guidelines on how to minimise exposure of the child to house dust mite.

TIP for Allergy Advisor users: Information about the diagnostic tests that are useful or applicable for assessing a specific allergen can be found in the main menu for that allergen, under "Adverse Reactions", "Diagnostics". This function lists the tests that are available as well as the diagnostic value of each test.

B. More information:
The primary tools to diagnose food allergy are the clinical history, physical examination, food-symptom diaries, elimination diets, oral food challenges, and allergy tests.^1,2 Appropriately designed challenge testing remains the gold standard, although it has many limitations.³

The clinician should use the findings from the history and physical examination to establish whether the patient experiences a food-induced allergic disorder and whether an IgE-mediated or non-IgE-mediated mechanism is most likely responsible. If an IgE-mediated mechanism is suspected, laboratory tests can assist in the diagnosis, but there are no rapid, specific diagnostic tests for non-IgE-mediated disorders (except in the research setting).^4,5

Numerous biological tests have been developed to assist in the diagnosis of food sensitization. These include multi-detection assays, immunoblotting, screening of basophil activation (BAT or FAST), assays for leukotriene LTC4 release (CAST), measurement of plasma histamine, serum tryptase, serum ECP (eosinophil cationic protein), urinary EDN (eosinophil-derived neurotoxin), assay of fecal IgEs, and RAST for specific IgG.^3,6 As a group, these tests are not very widespread or accessible. They have either been limited to academic and research settings for various technical reasons, or it is simply too early to evaluate their clinical usefulness.¹ Most of these are in vitro methods of diagnosis. The in vivo tests, on the other hand, include skin testing and challenge procedures.¹

The tests most often used to diagnose IgE-mediated allergy are the skin prick test (SPT) and the serum IgE test (total and/or serum-specific IgE). However, patch tests have recently been used more often in the diagnosis of food allergy as well. Numerous additional tests may be needed in various clinical scenarios to assist in determining whether an adverse reaction to food is the cause of a clinical problem. Among available resources are stool culture, endoscopy with biopsy, total eosinophilic count, pH probe, breath hydrogen, and reducing substances in the stools.¹

The SPT and the serum IgE tests indicate the presence of allergen-specific IgE, but they do not necessarily indicate that an allergic reaction would occur upon ingestion of the tested food.^2,4 The presence of IgE antibodies to a specific food merely indicates a certain probability of a clinical reaction to that food; the risk levels are unique in each patient.¹ This is explained in more detail below.

This Educational Review will focus on the use, benefits and limitations of skin prick tests. Serum IgE (RAST) tests, atopy patch tests, how to take a thorough clinical history, elimination diets, oral challenges, and controversial allergy tests will be discussed in future issues.

THE SKIN PRICK TEST
Several techniques are used for allergy skin testing (e.g., scratch test and intradermal test), but studies indicate that the SPT is the most useful technique with the most predictable results.⁵ Variations exist on the name and method of the SPT, including the prick-puncture skin test, prick skin test and prick-to-prick method. This Review will, however, focus on the most commonly used method.

With the scratch test, a scratch is made on the skin and a drop of allergen is deposited onto the site.5,7 Intradermal skin testing (where the allergen is injected into the skin dermis from a syringe) is not recommended for the evaluation of food allergy. This method gives an unacceptably high false-positive rate, and has been associated with systemic reactions, including fatal anaphylactic reactions.1,3,4,5,7,8

Skin testing is simple, quick to perform with immediate results, relatively safe, cost-effective, highly reproducible and, when properly performed and interpreted, very informative as a diagnostic tool.^1,9,10 Another benefit is that the patient sees the reaction occurring on his or her arm (compared to the blood test, during which the patient is not there); there tends to be a better buy-in to the diagnosis. The SPT does, however, have limitations, which will be discussed below.

Method
The SPT involves placing a liquid extract of the allergen (1:10 or 1:20 dilution) on the skin and pricking the skin through it (with a device such as a needle, bifurcated needle, probe, or lancet) so as to just puncture the skin.^1,11 It is unnecessary to scratch or lift the skin, and no blood should be drawn. The test is best performed on the inner aspect of the forearms, avoiding the flexures and the wrist area. It may also be performed on the back.⁹ It must be performed on intact skin; no active eczema may be present.

Any number from 1 to 25 allergens can be tested at a time, although it is more usual to test only 3 to 6. Although the amount of allergen introduced into the skin is very little, there have been reports of anaphylactic reactions following skin prick testing, especially with allergens such as peanut and bee sting.¹¹

Up until recently, it was believed that SPT could not be performed on children under the age of 4 years, but there is no evidence to substantiate this. Although a IgE response is not always present in young children, the test has been anecdotally shown to be useful in detecting elevated IgE levels to allergens such as soya and milk in breastfeeding infants. It may therefore be of use, especially where taking blood from an infant is difficult.

The skin test detects any IgE that has been produced by the immune system through previous contact to the allergen. The allergen-specific IgE is attached to the surface of mast cells in the skin. When the test allergen comes into contact with the IgE, inflammatory mediators are released into the surrounding skin, resulting in a small swollen area at the site of application (the wheal) and the spreading of histamine into the area surrounding it (flare; the flat, reddened area). This is the wheal-and-flare reaction that is the cornerstone of skin testing for allergy.7,12

The skin is also punctured through positive (histamine) and negative (saline-glycerin, the diluent used to preserve the allergen extract) controls.^1,4,9 The test site is examined 10 to 20 minutes later.^1,9,11

Interpretation
A positive skin test is indicated by the development of a wheal-and-flare reaction. There are different ways to grade the extent of the reaction that is experienced.⁷ The most often used is the measurement of the wheal diameter. If the diameter is 3mm or greater than a negative control (which should be 0), it is generally considered positive.^1,4 If the wheal is <3mm, it can be a sign of sensitisation or of the beginning of the development of allergy. Increasing skin test wheal size correlates directly with increasing IgE antibody to the allergen and increasing probability of clinical reactions.¹ The grading of 0 to 4+ is subjective and no longer recommended.⁹

The negative control: No response is expected to the negative control. If there is a response, it may be that the skin is oversensitive to pressure (as with dermatographism) and that the response to the test is a 'false positive'.¹¹ That is why a positive test is generally regarded as one with a mean wheal diameter at least 3 mm greater than the positive control.¹

The positive control: Everyone is expected to react to the positive control, as it is a solution of histamine. If there is no response, it is possible that something is preventing the skin from developing an allergic reaction. A number of drugs can do this, including all antihistamine preparations, some antidepressants, some cough mixtures and, under certain conditions, preparations that contain steroids. These medications should, if possible, be stopped for an appropriate period of time (e.g., 1 to 2 days with a short-acting and up to 1 week with a long-acting antihistamine) before the skin testing.^9,11

Sensitivity and specificity
The sensitivity and specificity of a test comprise its ability to identify a known condition. Sensitivity is defined as the proportion of test-positive patients in relation to the total number of patients who have the disease, whereas specificity is the proportion of test-negative persons in relation to the total number of patients who do not have the disease. Sensitivity and specificity are affected by intrinsic properties of the test, but the clinically important issue for the health professional is the probability that a patient has food allergy if the test is positive (positive predictive value, PPV) or does not have food allergy if the test is negative (negative predictive value, NPV).¹

The sensitivity of the SPT is usually high (>80%); the specificity is usually lower than the sensitivity but usually better than 50% if good-quality food extracts are utilized.^1,10 The intrinsic properties of the test that affect the sensitivity and specificity are factors such as the extract concentration, the device used, the pressure applied, and the location where the test is performed.¹ More details are set out below.

Negative skin tests for IgE-mediated allergy to the highly allergenic foods such as egg, peanut, wheat, milk, fish, and tree nuts are considered to be accurate about 95% of the time (i.e., NPV of >95%). However, negative reactions to other food allergens, such as soy, have a significantly reduced rate of accuracy, and some practitioners rate them no higher than 30 to 50% (i.e., only 30 to 50% of the reactions are true). (Commercial extracts from fruit and vegetables have an even lower accuracy, depending on the speed at which the allergen denatures.) Positive skin tests have a lower predictive value: less than 50% accuracy (i.e., PPV of <50%).^5,7,11

SPT's are, therefore, most valuable when they are negative; they are an excellent means of excluding IgE-mediated food allergies. Positive SPT responses only "suggest" the presence of symptomatic allergy.^4,10 However, the larger the wheal, the higher probability that it is a truly positive indicator. It follows that a positive skin test in isolation cannot be considered proof of clinically relevant hypersensitivity, whereas a negative test virtually rules out IgE-mediated food allergy to the food in question. Oral food challenge may be necessary to confirm food allergy, except where the history is overwhelmingly convincing.¹³

It must be remembered that, crucially, a negative result does not exclude the possibility of cell-mediated allergic reactions or intolerance.¹

In some patients a delayed skin reaction can occur 3 to 5 hours after the skin test has been performed, and it is important to remind all patients to look out for these reactions. Delayed reactions are often associated with more severe and significant allergies.⁹

The selection of which skin test extract to use should be based on the individual patient history. The indiscriminate use of large numbers of different allergen tests often leads to more confusion than clarification. If the history does not highlight possible culprits, but the symptoms suggests a food sensitivity, it may be useful to perform a limited number of skin tests to establish the probability that food allergy is present. The major foods for which skin testing has been found to be most helpful (as they have a high negative predictive accuracy, as discussed above), include egg, milk, wheat, peanut, tree nuts, fish, and shellfish.5

Virtually every step in performing and interpreting skin tests may affect the final interpretation. For example:

It has been suggested that neo-allergens are created during the processing and digestion of foods. For example, the allergenicity of peanuts is enhanced and/or formed by the roasting process. In some cases, food allergens are destroyed during processing, which is important in the case of patients who tolerate the cooked but not the raw food.^1,7 This variability may naturally influence the effective diagnostic contribution of extracts, according to their preparation. Also, the allergen stability of commercial extracts over time varies from food to food. Extracts of the major allergens retain their allergenicity for long periods, whereas most of the other extracts that are commercially available, especially the fruits, denature fairly quickly. (This may be the reason there is only a limited number of extracts available for skin prick testing, compared to the hundreds of RAST tests available.) The extract that is used for the SPT may therefore no longer be in its natural form and would not elicit an immune response.⁷

It is evident that the selection of source material, the extraction process, and the preparation of stable skin test solutions are all important factors influencing the suitability of extracts.^1,3,14,15 The reliability of the extract may, however, be enhanced by using fresh or native foods or using the "prick-by-prick" technique (pricking the fruit and then the patient with the same needle, thereby transferring the juice containing the allergen).³ Reagents sourced in-house are more concentrated and may intensify reactions (causing irritation), a possible drawback of this method, as it may increase the risk of a false-positive reaction.^1,8 The patient may be reacting to natural non-allergenic constituents (or even the acidity) in the food, which will not be present in commercial extracts. To obviate confusion, negative SPT responses with commercially prepared extracts that contradict convincing histories of allergic reactions to a particular allergen should be replicated with the fresh food before any conclusion that the allergen-specific IgE is absent.

Below is a summary of the imperfections of SPT, including some factors not covered, or not covered in detail, above.

Reasons for false positive tests include the following:

One always needs to consider that a person may have specific IgE antibodies to a certain food, and therefore be sensitized, but experience no symptoms to the food when it is ingested.^7,11

Reasons for false negative tests include the following:

If an SPT is done and the serum IgE to the same allergen is measured, the results may differ. The proposed reason for this is that IgE is attached to the mast cells (which are evaluated with a SPT), but occurs freely in the serum (which is evaluated when serum IgE is tested for). The concentration of the IgE in different locations is an important issue. This is also the reason antihistamines do not have to be avoided before the serum IgE is measured. In patients with severe eczema, it is often very difficult to stop using medication for a period. They are also more likely to have active eczema at the sites where the SPTs would normally be performed.

For more information on this subject and other allergy and intolerance related topics, visit:
http://www.allallergy.net
http://www.allergyadvisor.com
http://users.bigpond.net.au/allergydietitian

D. References
1. Metcalfe DD, Sampson HS, Simon RA. Food allergy: adverse reactions to foods and food additives 3rd edition. Blackwell Publishing, 2003.
2. Sicherer SH. Beyond oral food challenges: improved modalities to diagnose food hypersensitivity disorders. Current opinion in allergy and clinical immunology 2003;3:185-8.
3. Bahna SL. Diagnosis of food allergy. Ann Allergy Asthma Immunol. 2003 Jun;90(6 Suppl 3):77-80.
4. Sampson HA. Food allergy. Part 2: diagnosis and management. J Allergy Clin Immunol. 1999 Jun;103(6):981-9.
5. Bock SA. Diagnostic evaluation. Pediatrics 2003;111(6):1638-44.
6. Moneret-Vautrin DA, Kanny G, Fremont S. Laboratory tests for diagnosis of food allergy: advantages, disadvantages and future perspectives. Allerg Immunol (Paris). 2003 Apr;35(4):113-9.
7. Joneja JV. Dealing with Food Allergies. A practical guide to detecting culprit foods and eating a healthy, enjoyable diet. Bull Publishing Company, USA, 2003.
8. Williams LW. Skin testing and food challenges for the evaluation of food allergy. Curr Allergy Rep. 2001 Jan;1(1):61-6.
9. Toerien A, Potter PC, Buys C. Appendix IX. The skin prick test. The ALLSA handbook of practical allergy, 2nd Edition. Ince (Pty) Ltd., South Africa. 2001.
10. Sampson HA. Improving in-vitro tests for the diagnosis of food hypersensitivity. Current opinion in allergy and clinical immunology 2002;2:257-61.
11. Wright T. Food allergies: enjoying life with a severe food allergy. Class Publishing, London, 2001.
12. Poulsen LK. In vivo and in vitro techniques to determine the biological activity of food allergens. J Chromatogr B 2001;756:41-55.
13. Williams LW, Bock SA. Skin testing and food challenges in allergy and immunology practice. Clin Rev Allergy Immunol. 1999 Fall;17(3):323-38.
14. Roberts G, Lack G. Food allergy - getting more out of your skin prick tests. Clinical and experimental allergy 2000;30:1495-8.
15. Vieth S, et al. Factors influencing the quality of food extracts for in vitro and in vivo diagnosis. Allergy 1998;(Suppl 46)53:65-71.
16. Host A, et al. Allergy testing in children: why, who, when and how? Allergy 2003;58:559-69.

2. Which of the following is not true regarding the benefits of skin prick tests?
(a.) Quick to perform, with immediate results
(b.) Cost-effective
(c.) Safe, even for persons who have experienced anaphylactic reactions to the allergen
(d.) Highly reproducible

3. True or false: Skin prick tests cannot be performed on children under the age of 4 years.
(a.) True
(b.) False

4. Which of the following is not true regarding the interpretation of skin prick tests?
(a.) The wheal size should always be compared to the positive and negative controls.
(b.) The negative predictive value of skin tests for the highly allergenic foods such as egg, peanut, wheat, milk, fish, and tree nuts is >95%.
(c.) The positive predictive value of skin tests is >50%.
(d.) The larger the wheal size, the higher the probability that the skin test is truly positive.

5. True or false: A positive skin test in isolation cannot be considered proof of clinically relevant hypersensitivity, whereas a negative test virtually rules out IgE-mediated food allergy to the food in question.
(a.) True
(b.) False

6. True or false: In some instances, a skin prick test performed using the prick-to-prick technique will be more accurate than one using commercial extracts.
(a.) True
(b.) False

7. True or false: The materials used for pricking the skin and the technique used (the area, depth and pressure at which the allergen is introduced) may negatively influence the results.
(a.) True
(b.) False

8. Which of the following is not a reason for a false-positive skin prick test?
(a.) The food can cause a nonspecific ("irritant") positive skin reaction.
(b.) The skin on which the test is performed was not clean and free of active eczema or dermatographism.
(c.) The person has been sensitized to a cross-reactive allergen, without having a clinical reaction to it.
(d.) The person being tested is very young or very old.

This issue was sponsored by Unilever Bestfoods Robertsons (Pty) Ltd http://www.unilever.co.za/