Anaphylaxis During a Supervised Food Challenge: Push Through or Abort
A 9-year-old boy develops hives and vomiting on dose four of a baked-milk challenge. He already meets the formal criteria for anaphylaxis — the real question is what a fully monitored room earns him before epinephrine is given.
C.D., a 9-year-old boy with a well-documented cow's milk allergy diagnosed as a toddler, has spent the morning in the allergy clinic's observation room working through a baked-milk oral food challenge — five carefully measured, progressively larger doses of a baked-milk muffin, the standard way allergists test whether a child has outgrown enough of a milk allergy to tolerate baked forms even if he still reacts to fresh milk. He tolerated the first three doses without incident, chatting with the nurse about a video game between doses. Fifteen minutes after the fourth, he developed hives around his mouth and vomited twice in quick succession — objective findings, not just an anxious child's report of feeling funny, and enough on their own, involving both his skin and his gastrointestinal tract after a known allergen exposure, to meet the formal definition of anaphylaxis even though nothing about him right now looks dramatically sick. His mother, watching from the chair beside his bed, asks the obvious question: does this mean he failed, and does failing mean giving him the epinephrine she has always been told is for emergencies, right now, for what still looks like hives and an upset stomach.
The uncomfortable fact behind her question is that ‘looks mild’ and ‘meets criteria for anaphylaxis’ are not the same test, and the literature on anaphylaxis fatalities returns to the same finding often enough to be uncomfortable — Pumphrey's fatal-reaction series, and the food-fatality registry work that followed it, both landed on delayed epinephrine, rather than the severity of the initial presentation, as among the most consistent factors separating reactions that resolve from ones that don't. Two organ systems, skin and gastrointestinal, reacting together after a known allergen exposure meets the current diagnostic threshold for anaphylaxis regardless of how contained it looks in this exact minute, in this exact monitored room, with staff and equipment already at the bedside. Whether that technical threshold should translate immediately into epinephrine, or whether a supervised setting this controlled earns a brief window to treat symptomatically and watch first, is the actual disagreement unfolding at his bedside, not whether his reaction is real.
At the bedside
He meets the formal criteria for anaphylaxis right now — hives and vomiting, two organ systems, right after a known allergen exposure — and I don't think how mild it currently looks should change what we do next. Pumphrey's fatality series keeps getting reproduced on this point: it's delayed epinephrine, far more than initial severity, that predicts a bad outcome. I'd give it now.
I agree he meets the technical criteria. What I'd weigh differently is that this is exactly the setting built to tolerate a short, closely watched interval that a reaction at home never could — full monitoring, epinephrine already drawn up, staff at the bedside. Give him cetirizine, watch him for the next fifteen minutes with everything ready to go, and if anything progresses — any respiratory or cardiovascular finding, or the GI and skin symptoms worsening rather than plateauing — give epinephrine immediately. I don't think ‘meets criteria’ has to mean ‘must be treated with epinephrine in this exact minute,’ specifically because of how controlled this room actually is.
I take the fatality-literature point seriously. I don't think it was measured in a room where epinephrine is already in someone's hand.
I don't think the two of you actually need to resolve the epinephrine-timing question to answer his mother's real question, which is whether he failed. He didn't. Reacting at dose four of five is a complete, informative result — it tells us his current baked-milk tolerance threshold precisely, which is exactly what this challenge was designed to find. Whatever you decide about epinephrine right now, I'd make sure she leaves today understanding this challenge worked, not that it went wrong.
Agreed: cetirizine given immediately; a strict, explicit fifteen-minute observation window with epinephrine already drawn up at the bedside and a hard rule that any new or worsening finding in any organ system triggers immediate epinephrine without further discussion; his symptoms plateau and begin resolving within that window without requiring epinephrine, and the challenge is documented as a clear, valid positive result at dose four rather than an inconclusive or aborted attempt.
Not agreed, and left open rather than smoothed over:
The allergist wants the strict epinephrine-at-criteria default reinstated for the next similar patient.
The pediatrician thinks today's clean resolution is real evidence the brief observation window is defensible in a fully-monitored setting.