When the presentations were finished, we had a panel discussion and took questions from the audience. This 20 minute Q & A went close to 45 minutes.

I have linked to Google Documents this presentation. It is entitled ‘Allergy Testing and Referral to the Allergist’.  This presentation was completed in April. You will notice slides with red titles. It has been a rule for speakers, especially in Continuing Medical Education (CME) offerings to not make changes. An article in JAMA was published the week prior to this presentation. In an attempt to provide the 50 learners at the seminar with the most up-to-date information, I quickly added these slides.

The reference list for the presentation (Food Allergy Testing Reference List)  is also available via a link to google documents.

FEL

This week I am presenting at a seminar at the Riley Child Care Conference. The seminar was the idea of Dr. Sandeep Gupta (pediatric gastroenterology). The title is “Puzzling, Perplexing, Problematic Allergies in Children”. The third lecturer is Dr. Jeff Travers (dermatology). I have the task of talking about allergy testing and referral to the allergist. Food and specifically food allergy is the common ground for the three specialties. My focus will be on food allergy evaluations and management.

I have been putting this presentation together for a number of weeks. I was ahead of deadline and sent my slide copy and handouts to the organizers for inclusion in the syllabus. For those involved in continuing medical education (CME) programs, that is what is called being a good citizen. However, I just changed major parts of my talk due to the appearance of an article in the Journal of the American Medical Association (JAMA) this past week. The article is entitled ‘Diagnosing and Managing Common Food Allergies: A systematic review’ by J Schneider Chafen and colleagues (JAMA, May 12, 2010- Vol 303, No 18, pages 1848-1856). I know that I am in trouble for doing this because showing slides that are not part of the syllabus has always been a most dangerous behavior for a CME speaker. The audience tends to yell at the speaker for this violation, however I will take the hits in favor of providing the most up to date information.

The idea of a ‘systematic review’ is a very specific and intense look at the literature on a specific topic. First a few basic questions are established. This is followed by an extensive review of everything in the literature on the topic/question. There are strict inclusion criteria. The data from the studies is abstracted, the quality of the study is assessed, and the data is then synthesized. The hope is that those studies which have substantial numbers of cases in randomized controlled studies were evaluated and included in the review. This is done to present the best, the most valid, and most convincing work.

This systematic review caused me to revise my talk and add seven slides that will truly reflects the most current information.

This review was sponsored by the National Institute for Allergic and Infectious Diseases (NIAID). It is the prelude to the establishment of National Food Allergy Diagnosis and Management Guidelines, a topic I have talked about previously. There were 12,378 literature citations on food allergy found between the January 1988 and September 2009. From this, only 72 articles qualified for this review. That represents approximately 0.05% of the starting total. This is important to note. Almost all of the articles pulled did not fit the purpose of this review. The specific topics that were sorted inclluded; food allergy prevalence, studies of diagnostic tests, and studies on management and food allergy prevention. Further restriction involved looking at studies that dealt with specific food allergies; milk, egg, peanut, tree nut, fish, and shellfish (50% of all food allergy).

The overall summary was that the literature/evidence regarding food allergy prevalence, diagnosis, and management is voluminous, diffuse, and according to this review is also critically limited by the lack of uniformity on what food allergy is (lacking uniformity for criteria for the diagnosis of food allergy). The point is that when looking at an article on food allergy we have to be sure what is being talked about. All too often the diagnosis is based on laboratory study results alone. This lack of defining food allergy has severely limited making conclusions regarding the best practices for managing and preventing food allergy.

It may come as a surprise, but food allergy has no universally accepted definition. The NIAID suggested definition is “an adverse immune response that occurs reproducibly on exposure to a given food and is distinct from other adverse responses to food, such as food intolerances, pharmacologic reactions, and toxin-mediated reactions.

The results were as follows;

• Prevalence- food allergy affects more than 1-2% of the population but less than 10%.
• Diagnosing- food challenges, skin prick tests (SPT), & serum food-specific IgE (blood tests for food allergy) all have a role in making the diagnosis, but no one test has sufficient ease of use or sensitivity or specificity to be recommended over the other tests. The food challenge suffers from not being easy to use in general clinical practice.
• Management (elimination diets)- only 1 randomized controlled trial (RCT), established as the more scientifically rigorous test,  was identified for the effect of elimination diets. RCT are generally lacking for atopic dermatitis and eosinophilic espophagitis. The benefits for elimination diets are uncertain based on published evidence, and potential benefits need to be weighed against the potential nutritional risks especially in children. It is important to point out that this is not referring to trials for serious life-threatening food allergy reactions, such a trial would be unnecessary and unethical.
• Immunotherapy- not a currently licensed method for treating food allergy. May be effective in generating desensitization. The effect on long-term tolerance needs to be determined.
• Prevention- In high-risk infants hydrolyzed formula may prevent against cow’s milk allergy, but standard definitions of high risk and hydrolyzed formula do not exist.

There were a few general comments made that are worth noting.

• There is the potential for the over-diagnosis of food allergy
• Consequences

                Dietary restriction

                Nutritional problems

                Anxiety/worry

                Social challenges due to food allergy

There were a few final comments in this paper that are worth consideration.

• Proper interpretation of SPTs and serum food-specific IgE results requires evaluation of the data within the context of the clinical history and physician understanding of symptoms consistent with clinical food allergy to separate true positives for food allergy.
• The over-diagnosis or misdiagnosis of food allergy by medical practitioners obscures the substantial morbidity caused by patients truly affected by immune-mediated food allergy and serves to perpetuate some public misperceptions that food allergy is a trivial medical condition.

We all have a significant amount of work to sort this all out. The first steps are coming to some consensus as to what a food allergy is and what it is not. We then need to perform a detailed medical history to tease out a reproducible immune response with exposure to a food. Next we need a diagnostic tool or tools to be used to confirm our impression. The food challenge has been the gold standard for this, however it is not easy to perform food challenges. The next need is a plan of management. There has always been avoidance. We can add ‘Father Time’ as some food allergies can be outgrown. We eagerly look forward to immunotherapy that not only provide desensitization but will lead to tolerance.

FEL

Last week I was interviewed by a reporter from the Wall Street Journal. The topic was food allergy. The reporter came across this website and thought that I be a good resource for her article. We had a delightful talk that went on for 45 minutes. Questions were asked about the increase in food allergy; is it real or is it possibly due to the over use of diagnostics (allergy testing).

Needless to say I was excited about the prospect of being quoted in the Journal.

My hopes were dashed. The reporter had to cutback on material. My name did not appear in the article. The article was very well done and did quote a number of outstanding leaders in the field of food allergy (Drs. Hugh Sampson and Robert Wood).

I do encourage you to read the article written by Melinda Beck.

FEL

Authors: Amy Branum and Susan Lukacs

Reference: Pediatrics Volume 124 (6) December 2009

This title caught my eye. The impression in clinical practice is that more and more children have food allergy. This article looks at the prevalence of food allergy in children. I wanted to get this review posted this week. I am off to Santa Fe to moderate an AAP Practical Pediatrics Course. This AAP meeting is similar to the one I reported on earlier on this home page (Rhode Island). This meeting has an excellent cast of presenters. I plan to take notes and post a few updates upon my return.

Purpose of the article: To describe trends in the prevalence of food allergy and food allergy-related health care utilization in children in the United States.

Methods (how was this study conducted?): Data from a number of national health surveys were reviewed.

• Food allergy prevalence was evaluated in children 0-17 years of age from surveys conducted over the years 1997-2007. The question asked about food allergy was “During the past 12 months has the child had any kind of food or digestive allergy?”
• Blood tests for IgE antibodies to foods were taken from the National Health and Nutrition Examination Survey (NHANES) 2005-2006. Specific IgE antibodies to peanut, egg, and milk were measured using the Pharmacia ImmunoCap 1000 System. Specific IgE to shrimp was measured only in children over the age of 6 years. The range of specific IgE values was 0.35 to 1000 kU/L.
• Information regarding food allergy-related visits to physician offices and hospital facilities was taken from two additional surveys.
• The results were analyzed using rather sophisticated statistical tools that included weighing the data for the analysis of trends.

Results (what the study found):

• The prevalence of reports of food allergy in children has increased from 3.3% in 1997 to 3.9% in 2007.
• Peanut IgE antibodies were found in 9.3%, egg IgE antibodies were found in 6.7%, milk IgE antibodies in 12.2%, and shrimp specific IgE was found in 5.2% of children.
• Ambulatory care visits for food allergies tripled between 1993 and 2006. Between the years 2003 and 2006 there were 317,000 visits/years to emergency departments and outpatient offices. Hospitalizations with a recorded diagnosis related to food allergy increased from 2600 to 9500 discharges/year.

Conclusions:

                These national surveys show that food allergy prevalence and/or food allergy awareness has increased in recent years.

Commentary:

                The authors point out a number of limitations in the study, however the major contribution here is reporting on what these surveys reveal about the parent’s report regarding food allergy. Food allergy may be rising however it is possible that the results may be due to increased food allergy awareness which is also a very good thing. This is a report of prevalence and does not go into the possible reasons for the increases.

                It is important to note that this was a survey. A simple question was asked. These were not absolutely proven cases of food allergy. The question included digestive allergy which has the potential to include a number of clinical conditions that are more common and may or may not be allergy; lactose intolerance, eosinophilic esophagitis, and celiac disease for example. This was a report on what a parent thought about food allergy in their child.

                The report has a few ‘between the lines’ issues as well. The conclusion is that food allergy and digestive tract allergy has a prevalence of 3.9%. The study also included a survey in which a blood test for allergy was performed. Using the blood test the prevalence of peanut, egg, milk, and shrimp ‘allergy’ exceeds the overall food allergy prevalence. The authors do point out this difference and are very careful about what is allergy and what sensitization to food is.  “Although serum IgE measurements cannot be used alone to determine the prevalence of food-specific allergies or to predict reactions to certain foods, they give an indication of increased atopy and risk for allergic reactions to food.” I define allergy and atopy on my allergy testing page.

                We also need to be a bit careful on the hospital data. The information on health care utilization included children who had a diagnosis of a food allergy. This did not necessarily mean that they were in the health care facility for a food allergy issue. There is a tendency in coding encounters to include as many codes as possible and to include codes that will help with health care utilization reimbursements.

                The statistical analyses on papers like this always fascinate me. During my MPH training I had a number of biostatistics courses. The weighing of the data is frequently done and when it is done, differences can be found. Sometimes it is interesting to see what the results were before any weighing. I have also wondered what went into the ‘weighing’ of the data. What elements of the data were assigned a ‘weight’ to make them work into the analysis?

                This was a nicely done paper and does answer some questions however as many quality studies also do it has us asking many more questions about food allergy in children.

Fred Leickly

 The article did contain something that I strongly feel needs to be critiqued and clarified. The content expert for the quote was an associate professor of family practice at Michigan State University. Dr. Reinhardt is quoted as saying that the National Heart, Lung, and Blood Institutes (NHLBI) Guidelines for the Diagnosis and Management of Asthma 2007 advocates the use of the Phadia test to determine whether a person has allergies to common indoor allergens. He also states that too few clinicians are using this test.

 The first comment is out of context with other considerations that need to be addressed prior to ordering an expensive and perhaps unreliable laboratory test. The comment supports one commercially available product (fair balance). It assumes that this commercial product is the one the Guidelines recommend. As to the last comment regarding the use of the Phadia test for inhalant allergens, this is most probably because clinicians may be aware of the problems with this approach to allergy.

 The 2007 asthma guidelines do not specific the Phadia test for determining allergy. On page 168 of the NHLBI Asthma Guidelines Key Points clearly indicate that there are more steps involved- a relevant history, skin testing or in vitro testing, and an assessment of the significance of the test results. The guidelines do not mention specifically the Phadia test in the Key Points. This test has value, but only in the context of a patient’s history. The Phadia test for inhalant allergens is frequently falsely positive. In a recent national survey (JACI 2009;123:1163-9) 20% had symptoms of allergic disease. The Phadia test for inhalants demonstrated that 50% of the population had a positive test (for more on this article see ‘Incidence of Allergy in Children: Using Allergy Testing Panels (Pharmacia ImmunoCap) or Symptoms?’ one of my earlier posts on this topic). It is very important to point out that any test used in allergy only tells us that antibodies are produced. It is the history of symptoms plus the results of the test together that fulfill the criteria for allergy. A test result alone makes no one allergic and is only as useful as the history that supports it.

 In my opinion the reason why the Phadia test is not being used is because clinicians are aware of the need of taking a history, they dislike the obligatory panel of analysis offered by the Phadia panel, there is worry about the cost, and the test declares many more positive than actually have symptoms.

 Do we let the laboratory test dictate the diagnosis/treatment or the patient’s history matched to relevant laboratory studies to decide how to manage asthma?

I am old school- I make sure that I understand the history of the condition and the environmental exposures. From this I decide what tests for allergy are relevant and use testing to verify my clinical impression. Sometimes we don’t need to test. Sometimes we have to take the test results and re-evaluate the history and exposures to make sense of it all. Health care is expensive, laboratory tests for specific IgE is expensive. The consequences of questionable results taken out of context can have a cost as well. Allergen-environmental control is a big business. Consider the hidden costs of extra work, alteration of living environments, and the status of pets. Seek out consultation of experts who can sort through both a medical and environmental history and make sense of what to test for. You also need to consider what the gold standard is for the assessment of allergic sensitization and that is the allergen skin prick test. Look also to the credentials of the providers that you are working with.

Fred Leickly

A young African-American girl of 11 months presented upon the recommendation of her pulmonologist. The pulmonologist saw the need for a flu shot however there was some concern about a positive allergy test to egg. This little girl had a ‘Phadia Specific IgE’ set of food allergy panels performed for about 31 foods (the bill to the mother was >$800). The test was performed because of a concern for a constant runny nose. The child was tested to the usual suspects for food allergy in young children. The test array went significantly further. As I have talked about previously, because of the marketing of these allergy blood tests to include extra items (at more expense) these are in most cases irrelevent. Many of the foods tested are not part of a young child’s diet. The test included foods that this young lady of 11 months has never ingested; lobster, shrimp, tree nuts, shrimp, and clams to name a few. So in the end mother was at risk to pay for information that was not relevent to the child’s situation or exposures. There is also the argument within this case regarding the pursuit of a food allergy for the complaint of a runny nose in an 11 month old child.

The test for egg was 0.51 kU/L. The cutoff for a negative response is <0.35 kU/L. When you look at the literature for critical cut-off levels for doing an egg challenge, this value would indicate that the child will have a high probability of a negative and successful egg challenge.

Now the real kicker. As we all know there is no better test for a food allergy than the challenge; give the food and see what happens. We do have to be careful depending upon the history of previous reactions for those challenges. In this baby’s history she had been eating scrambled eggs without any problems. There was no history of a cause/effect relationship with egg exposure.

Now let us add in the variable of the flu shot. The flu shot contains egg protein. The recommendations from the CDC and the AAP are to not give the flu shot if there is a history of a severe reaction. This pertains to what happened when egg was ingested. It does not pertain to the presence or size of an allery test for egg. In this case, cooked eggs have been ingested without any reactions. So in my opinion, the history of egg exposure and having no serious reactions with that exposure should clear the way for the flu immunization. For those children who have serious reactions to egg but can eat heated egg products, there is a need for considering a desensitization for the fluogen. The egg in the vaccine is more like cooked egg (scrambled, hard boiled, over easy etc) than heated egg (cakes, cookies, muffins, and waffles). The history of eating products with heated egg should not be considered a ‘safe’ for those with serious egg reactions.

The children need to be immunized for this upcoming flu season.

My humble opinion,

Fred Leickly

The study is very interesting and very well done. I was fascinated by not only their conclusions, but their inclusion of additional information on this large representative group of children. It is this ‘between the lines’ information that I want to go over in this report. I think that there are a few more points that can be made from the NHANES aside from the obesity/allergy issue.

Information, data for those scientifically oriented, is shown on 4,111 children. Each of these children underwent a blood study that looked for specific IgE antibodies to a variety of allergens. For the children over age 2 years but less than 5 years of age they were tested to a panel of allergens that included the following; house dust mites, cat, dog, cockroach, Alternaria (mold), peanut, egg, and milk. The children who were 6 years and older were also tested for IgE antibody to ragweed, ryegrass, Bermuda grass, white oak, birch, shrimp, Aspergillus (mold), thistle, mouse, and rat. Nine allergens were looked at in the younger group and 19 were evaluated in the older age group.

In addition to the blood tests questions were asked about previous diagnoses of allergic conditions such as hay fever, eczema, and allergies. Atopy (the condition in which someone makes an allergy antibody- IgE antibody) was defined as having any measurable IgE to any of the allergens tested for. An atopic child had at least one positive blood test.

In my opinion the NHANES methodology was very appropriate in being selective as to what allergens to test for. Note that the younger age group was not tested to tree, grass, or ragweed pollen. Also note that for both groups, the selection of foods was limited to a only a few.

This table is redrawn from Table 1.

What caught my eye was the difference between those who were given a diagnosis of allergy and/or who stated that they had symptoms of allergy versus the percentage of children who had one or more measure of atopy (allergic sensitization) by a blood test.  Only 18.7% of these children (overall number) had symptoms of allergy yet 46.4% (overall percentage) had one or more blood test that was positive thus fulfilled the criteria for atopy. The key question is why almost half of the population shows an allergy antibody yet only 20% have symptoms? The difference is not small between the two groups. Should we diagnose allergy by a set of symptoms due to exposures of a relevant allergen or dash the history and let the blood tests dictate the diagnosis and treatment. Almost 50% of the time that blood test panel will show a positive response.

What has happened here? Who is the allergic child? Does this blood test reveal too many false positive responses? Is it predicting an allergic future? Is it reflecting an allergic past? The result is not giving us a meaningful look at the current situation. I think this shows a significantly high level of falsely positive results.

My interpretation and what this side bar of the study has clearly shown is that the positive blood tests when done as a panel and done outside the context of a good history will yield clinically irrelevant information for a large number of children.

In the medical community of central Indiana there has been subject to very active marketing of allergy test panels (blood test, RAST). One of my referring pediatrician friends told me last week that an allergist from California came to town and visited primary care offices talking about allergy test panels on behalf of a sponsoring company. That reminds me of an old western- Paladin. The cowboy’s business card read ‘Have gun, will travel’.

The original purpose of the article is worthy of a future report and a watchful eye on how this association.

No pun intended, but food for thought?

Fred Leickly