Non-Food-Dependent Exercise-Induced Anaphylaxis: Prophylactic Therapy Versus Exercise Restriction
A first-time marathoner has had three anaphylactic episodes on long runs with no food, drug, or weather pattern behind any of them. The real disagreement isn't whether to try an antihistamine — it's what actually keeps him safe if it doesn't work.
T.O., a 27-year-old man training for his first marathon, has had three anaphylactic episodes during long runs this year — hives, wheezing, and once a blood pressure low enough to require paramedics on scene — each during runs of eight miles or more, with no consistent pattern he or his doctors have been able to pin to food, medication, or the specific weather that day. He has had episodes on an empty stomach and after eating, with and without a pre-run ibuprofen for an old knee injury, ruling out the kind of food-cofactor pattern that would point toward wheat-dependent exercise anaphylaxis instead. A cholinergic-urticaria challenge was negative, his baseline tryptase is normal, and nothing about his exam or history suggests mastocytosis — which leaves non-food-dependent exercise-induced anaphylaxis, a diagnosis defined largely by exclusion of everything else, as the working label. He has no cardiac history, a normal resting ECG obtained after his second episode specifically to exclude an exertional arrhythmia mimicking anaphylaxis, and no family history of anaphylaxis or mast cell disease of any kind. He has no interest in giving up running altogether and has already told his coach he intends to run the marathon regardless of what today's visit recommends.
The honest limitation in non-food-dependent exercise-induced anaphylaxis is that, forty-odd years after Sheffer and Austen first described the syndrome, its evidence base is still almost entirely case reports and small series, and the antihistamine premedication that reliably helps cholinergic urticaria on its own has a far less consistent record here — some patients respond, a meaningful fraction don't, and nothing currently predicts in advance which group T.O. would fall into. What the literature is more consistent about is the actual mechanism of harm: these episodes are dangerous less because of any single mediator and more because they happen mid-exertion, often alone, on a route where reaching help takes time — a structural risk that a pill either controls or doesn't, but that changing how and where he runs addresses directly regardless.
At the bedside
I'd start him on a non-sedating antihistamine today, and I'd be willing to push the dose up toward what we use for refractory chronic urticaria if a standard dose doesn't move the needle. It's a low-risk trial, some patients in Sheffer and Austen's original series and the ones that followed did respond, and there's no real reason to withhold something this benign while he's actively training and at real risk of another episode.
My concern isn't the antihistamine's safety, it's what we're actually protecting him from if it doesn't work — and the response rate for this specific non-food-dependent variant is genuinely inconsistent, not the reliable response cholinergic urticaria alone tends to show. What the case reports agree on far more clearly is that the real danger here is collapsing mid-run with nobody around. No solo long runs, a partner or populated route, an intensity and distance cap below whatever's triggered episodes so far, an epinephrine autoinjector on him every time — that protects him regardless of whether the antihistamine turns out to help.
I'm not against trying the medication. I just don't want the running plan built as if it's already working.
I don't think this needs to be either-or. Start the antihistamine, but treat it explicitly as unproven for him specifically — a series of monitored, moderate-distance runs first, with someone present and epinephrine on hand, before he goes back to eight-plus-mile long runs at all, medicated or not. The failure mode I'd actually worry about is the quiet one: three clean runs on the antihistamine start feeling like evidence it's working, and the partner requirement or the distance cap gets relaxed before we've actually confirmed anything. Keep the behavioral safeguards in place on their own merits regardless of how the drug trial goes, and let the medication earn any relaxation of them later, not in advance.
Agreed: start cetirizine 10mg daily, up-titrate if tolerated and needed; resume training with a hard rule of no solo runs over the prior trigger distance, partner or populated route required, epinephrine autoinjector carried on every run regardless of distance.