Cookie Notice

This site uses cookies. By continuing to browse this site, you are agreeing to our use of cookies. Review our cookies information for more details.

skip to main content

Specific milk IgE with contact urticaria but without anaphylaxis


A 13 month-old breast fed infant was given cow's milk few weeks ago. He ingested 3-4 ounces and developed urticarial lesions where ever the milk came in contact with the skin. There were no systemic symptoms. He was introduced to cow's milk 4-5 times and every time it is contact urticaria with no systemic symptoms. He was given yogurt at 7 months and soon developed generalized urticaria which resolved with Benadryl. He is tolerating pancake and cooked foods with milk as an ingredient.

A skin prick test is strongly positive for milk. A positive reaction was seen in less than 5 minutes. Immuno cap RAST for milk is class 5 (very high). Also, component milk testing shows high antibody level to casein, alpha lactalbumin and beta lactoglobulin. How do you explain the disparity between the skin reaction, positive test results and no systemic symptoms?


Let’s tackle the multiple issues here. The question I have is if you are asking if the child is actually milk allergic because they had urticaria with yogurt but is able to tolerate milk in pan cooked (not extensively heated) pancake and other non-extensively heated but not liquid milk forms?

My questions to clarify this would be
1) If the kid ingested cow milk formula, are you certain that the urticaria observed was entirely related to skin contact and wasn’t generalized (it sounds as if you are)
2) Are you sure when the child ate yogurt that they didn’t get this all over them, causing the hives from contact
3) Are you asking if urticaria from ingestion is not considered systemic
4) Are you asking how a milk allergic child could tolerate milk in a cooked but not extensively heated form (wondering specifically about pan heated pancakes made with milk, and other cooked but not “baked” products”)

Contact urticaria is not uncommon and does not necessarily indicate that there is an IgE mediated allergy to the food. Foods can irritate skin, and even in non-allergic children, contact-related rashes including urticaria can occur. For contact rashes, these generally resolve once you wash the skin area, but some may be better managed with antihistamine. There is no evidence these generalize to systemic symptoms, or they sensitize the child to develop future IgE mediated symptoms on ingestion.

In this case, the child has booth contact urticaria from milk, and appears to have had a possible/likely IgE mediated reaction with ingestion independent from contact with the skin. Yogurt is a cow milk product, and contains cow milk protein. Generalized urticaria from ingestion, in the setting of a positive skin test, and a positive ImmunoCAP test indicate sensitization. In general, sensitization in the setting of a strongly suspicious history indicates a likely allergy to milk. Based on what you describe, on face, this is an infant that ingests yogurt, and has a generalized urticarial rash. If the child has not had milk products previously, with this sensitization pattern, this is a highly probable allergy. If they have tolerated milk before, and this was the only time they had symptoms, then something else is in play, and we can come back to that at the end.

The contact issue is probably unrelated but does overlap in some kids also allergic to milk. It is not a marker of anything in and of itself, however. The milk components are also positive, but there is far less clinical utility for these. Some studies have hinted that stronger sensitization of certain casein subtypes may be associated with more persistent milk allergy, but there is no hard and fast evidence this is a useful marker of anything. Some studies have also suggested that higher casein may indicate a lower likelihood of tolerating baked milk, but these are somewhat narrow and confounded studies that lack the ability to generalize to a population outside the sample in which the study was performed. However, as you can see in your patient, the casein level has no bearing on the child’s ability to tolerate milk cooked into things. The bigger question is if this child is actually milk allergic since cooked milk is not different as an allergen from yogurt, whereas extensively heated milk (e.g., “baked” milk) is allergenically distinct.

Extensively heated is a loosely defined term, but generally refers to something heated to at least 350 degrees Fahrenheit for a minimum of 30 minutes. Higher temps and shorter times have been reported as well, and the term is a “catch all” that lacks a precise scientific definition. The theory here is that milk and egg have a predominance of heat labile epitopes, which get denatured in the heating process, thereby making the form less allergenic to some milk or egg allergic individuals (~75-80% of all milk/egg allergic individuals). Pancake is generally prepared at pan cooked temperature, which is not extensively heated, so that is a little harder to explain. However a commercially prepared product, such as an Ego pancake (a breakfast staple for my children growing up!) may actually be cooked at an extensively heated temperature, whereas a pancake made from a mix from scratch is cooked at a standard pan temperature, that would not reach an extensively heated level. Again, these are loose criteria, and some children may demonstrate some variability (or there may be unawareness of the degree of cooking/baking that indicates it was more cooked than reported). There also may be issues of threshold and quantity of allergen and/or how completely denatured the epitopes must be to elicit symptoms in a patient that can explain some variation. But, a good rule of thumb is to stick with the aforementioned 350-F x 30 min as a working definition of “baked” or extensively heated.

Offhand, I would suspect one of two things:
a) The milk exposure in the cooked items is either at a very low threshold or is more extensively heated than realized
b) The child may have had urticaria with yogurt exposure, but this is being misinterpreted as an IgE mediated allergy, which explains why he tolerated several exposures (contact issues notwithstanding) and the non-baked milk is tolerated. To determine this, you’d need to do a milk challenge to disprove the urticaria was an allergy.

Hard to say which is the more likely story without a bit more background. Conservatively, I’d probably favor option a, presume allergy, watch this for a year, try to get the child on truly extensively heated milk, and have them only very cautiously continue with the products that have not yet provoked symptoms, but not encourage anything else at that unlikely extensively heated level. After a year, I’d retest (skin and serum), look to see if there was change, and consider a challenge. I would unequivocally not wait for the skin test or blood test to become “negative”. I’d personally challenge when I see close to a 50% drop in a sensitization level, or I see ample evidence that shows perhaps the story is wrong (like continued ingestion of truly non baked milk in a decent quantity making me believe there may be some tolerance).

If you are more aggressive or inclined, then option b will provide a definitive outcome. I don’t think the risk is higher per se—this is a standard OFC, and can be managed as such if there is a reaction. I’d not avoid this just because of fear of possibly provoking a reaction—you are in a position where you’d need to, potentially. The parents may not favor this though.

One last note on reporting serologic results. Technically, ImmunoCAP is a fluorescent enzyme immunoassay, using a 3D binding to a surface. It is distinct from RAST, a separate technology, and ImmunoCAP and RAST are not interchangeable terms (or numbers) per se. As for the class levels, the integer levels of reporting have been split into sextiles by reference labs to establish these classes, but this is not the preferred nomenclature for reference—the specific kilounits of antibody per liter (KUa/L) are the more traditional way to refer to these, given different labs may use different breakpoints for their classes and could refer to entirely different degrees of sensitization.

Matthew Greenhawt, MD, MBA, MSc