A Long Skin-Test Panel and a Question of What It's Actually Worth in EoE
A single patient with eight positive skin tests and a diagnosis those tests weren't built to guide. The disagreement is whether a panel that looks informative should be trusted to shrink his diet, or whether the honest answer is that his own skin doesn't know what's hurting his esophagus.
Marcus T., a 34-year-old man who works installing home theater systems, has been the slowest eater at every family dinner for as long as anyone can remember — a habit his siblings used to tease him about before it stopped being funny and started being dysphagia. Eighteen months of progressively worse trouble swallowing solids finally brought him to a gastroenterologist, and an endoscopy showed the rings and linear furrows now familiar to anyone who reads for eosinophilic esophagitis, confirmed on biopsy at 38 eosinophils per high-power field. His allergy history is dense: seasonal rhinitis since childhood, and a skin prick panel ordered at the same visit came back positive to eight separate foods — peanut, tree nuts, wheat, dairy, egg, soy, shellfish, and dust mite cross-reacting with several of them.
That panel is exactly what makes his case, rather than a routine one. Eight positive results reads, at first glance, like a map straight to his trigger foods — eliminate what he's sensitized to, skip the guesswork. But skin-prick and patch testing were built to identify IgE-mediated reactions, and EoE is driven by a different, largely non-IgE mechanism; Molina-Infante and colleagues found a multimodal allergy-testing-directed diet left ten of fifteen adult patients short of remission, while an empiric six-food elimination diet with no testing at all cleared roughly 70% in separate series. The number of positive tests Marcus is carrying doesn't tell the team which mechanism is actually driving his esophagus, and treating a long panel as more informative simply because it's longer is the exact substitution the literature argues against. There's a genuine irony worth naming underneath the length of his panel: a shorter, more targeted list of positives tends to read as more convincing precisely because it looks more specific, but a separate comparison of serum-IgE-directed elimination against empiric six-food elimination found the two performed comparably — not because targeted testing is worthless in principle, but because the testing modality matters more than how long the resulting list is. Marcus's own panel used skin-prick and patch testing rather than serum IgE, the modality that comparison didn't use, which narrows rather than strengthens the case for building his diet around it.
Eight positive tests, and what they're actually worth here
Marcus has eight positive results in front of him and a strong instinct to act on them. I'd rather build his diet around foods he's already been told he's sensitized to than ask him to eliminate six food groups where five might have nothing to do with his esophagus at all.
I recognize this is an adherence argument more than an evidence one — I’m naming that up front rather than dressing it up as a clinical judgment it isn’t.I'd push back on treating his adherence preference as a reason to override what's actually been tested here. Molina-Infante's group looked at exactly this question in adults — a multimodal allergy-testing-directed diet using skin prick, prick-prick, and atopy patch testing together — and it cleared five of the fifteen adults enrolled, against SFED's roughly 70% in separate series.
You’re right that a shorter, targeted list is easier to follow, but easier to follow isn’t the same as more likely to work. Even taking that small a series at face value, most of the people who complete his preferred diet will still be eosinophilic at the end of it.
There's a version of this that takes both of you seriously without picking one extreme. The 2-4-6 step-up data show a four-food elimination diet — milk, wheat, egg, and soy, none of it drawn from his skin tests — achieving remission in a clear majority of adult patients, at real burden savings over the full six.
That's not a compromise between his preference and the evidence — it's a different empiric starting point that happens to ask less of him. If it doesn't work, stepping up to the full six is still available, and nothing about starting at four forecloses that.
Agreed: start a four-food elimination diet (milk, wheat, egg, soy) with repeat endoscopy at eight weeks, and step up to the full six-food protocol if remission isn't achieved.
Not agreed, and named directly rather than papered over:
The primary care physician still wants his skin-test panel revisited before stepping up further, on the reasoning that eight positive results are too specific to set aside permanently, even granting the comparative data.
The allergist would step directly to the full six-food diet, arguing the panel has already been weighed against the literature and shouldn't be reintroduced as a tiebreaker each time a step fails.
The plan for now proceeds on the evidence rather than the panel. Whether the panel gets a second look if this step doesn't work is left open, not settled by today's decision.