By Dr. Paul Ehrlich
The Journal of Allergy and Clinical Immunology, pronounced by allergists everywhere as “Jackie”, has a terrific new offshoot called In Practice. As the name implies, it is targeted at issues facing allergists in day-to-day practice, as opposed to researchers. Thus, the current issue of In Practice contains an article entitled “Asthma Adherence Management for the Clinician”, while mother JACI has one called “Examination of the relationship between variation at 17q21 and childhood wheeze phenotypes.” Both are welcome in my office, but one has more immediate application for me than the other.
One article in the inaugural issue of In Practice that ought to be required reading not just by allergists but by all doctors in general practice, particularly pediatricians, is “Advances in Diagnosing Peanut Allergy” by Dr. Scott H. Sicherer of Mount Sinai and Dr. Robert A. Wood of Johns Hopkins. If there were an allergist all-star team, both of them would be in the starting line up.
Acknowledging that, “the diagnosis of food allergy is far from a perfect science” they proceed to analyze the procedures and probabilities that ought to be part of the doctor’s thought process, and evaluate the “diagnostic toolbox”.
I won’t recapitulate the article point by point, but I would like to highlight their section called “History, the key diagnostic test.”
They write:
Consider two extreme examples of patients with PN[peanut]-IgE levels of 20 kUA/L, a level often considered diagnostic of PNA [peanut allergy]. The first is a patient who was tested to peanut on a “panel” of tests performed as part of an evaluation for respiratory allergies. This patient routinely eats peanuts and peanut butter in full serving sizes and has never had a complaint. The correct diagnosis, based on the history, is that this patient does not have a PNA (prior probability is zero). The patient may be particularly sensitized to proteins in peanut that are labile, pollen-related ones, but this is a moot point (in this situation, although it may be crucially relevant in other circumstances) that needs no further evaluation for this patient. Testing to a tolerated food is poor practice because it is not only a waste of money, but it can also cause unnecessary avoidance and anxiety.
On the other extreme is a patient who has experienced 3 isolated ingestions of peanut and on each occasion experienced immediate symptoms of urticaria, angioedema, and wheezing. The most recent episode was 5 days before testing. One could argue that testing is unnecessary in this case, but a PN-IgE concentration of 20 kUA/L, or for that matter any evidence of sensitization, would simply be confirmatory given the history, as the prior probability of PNA was 100%.
You may think that these two extremes are “for discussion purposes only”, that in effect most cases fall in the middle—fifty shades of peanut allergy grey. Well, yes, but every day we see patients who have been on the one hand over-tested and over-treated because they have eliminated peanuts and other foods from their diets unnecessarily and on the other hand courted disaster by not understanding their allergies and taking the requisite precautions.
The authors conclude: “Until better characterization of the diagnostic utility of current tests is undertaken, and until new tests and novel uses of current tests arrive, the current findings can provide excellent diagnostic information. Results suggest, however, that OFCs (oral food challenges) will likely remain a mainstay of diagnosis for many patients presenting with possible PNA and that the history is still the most important test for all patients who have previously consumed peanut.”