Subscribe
Share
Share
Embed
Eosinophilic esophagitis is a food-driven Th2 allergic disease that scars the esophagus if it is left untreated, so the episode is organized around telling inflammation from fibrosis. It starts with a young man's food impaction and the rule to biopsy every adult impaction. It then covers the mechanism, the diagnosis at fifteen eosinophils per high-power field with no PPI trial, and EREFS scoring with its inflammatory and fibrostenotic split. First-line options are a high-dose PPI, a swallowed steroid, or a step-up diet, with biopsy-confirmed remission and ongoing maintenance. A second patient who is refractory leads to adequacy checks, dupilumab, and dilation.
Topics covered
Key decisions
This is an AI-generated podcast, and some pronunciations may be imperfect. Thank you for your understanding, and we hope you enjoyed this content.
For the full chapter with MCQs, tables, and primary-guideline references, visit www.boardpearls.com.
Questions or feedback: hello@boardpearls.com.
The most complete gastroenterology board-review podcast on the web, and it is entirely free. This is the GI series from Board Pearls: board review built around clinical reasoning, not recall. Every episode takes one high-yield topic and works it the way you would on rounds: a case to anchor it, the framework that sorts the differential, and the specific decisions the exam rewards. No trivia, no textbook dictation, just the thinking that earns the point.
Over 100 episodes span the full blueprint across nine modules (esophagus, stomach and duodenum, small bowel, colon, pelvic floor, liver, pancreas and biliary, endoscopy, and the cross-cutting topics), grouped by chapter and built from the guidelines and pivotal trials the boards are written from: ACG, AGA, AASLD, and ASGE.
Pair it with the written curriculum, question bank, and AI tutor at boardpearls.com. Questions or feedback: hello@boardpearls.com.
Welcome to Board Pearls. This is episode one of two of the Eosinophilic and Infectious Esophagitis chapter, in the Esophageal Disorders module. This episode is eosinophilic esophagitis, from the first food impaction through refractory disease, through two patients. A twenty-eight-year-old man with steak stuck in his esophagus teaches why every adult food impaction is biopsied before the scope comes out, what the disease is and why it ends in rings and strictures, how the diagnosis is made without a proton pump inhibitor, PPI, trial, how the endoscopy is scored, and how first-line treatment is chosen, checked, and maintained. A forty-one-year-old woman whose disease hasn't responded to first-line treatment teaches how to confirm she's truly refractory, how to separate inflammation from scar, how dupilumab works and is dosed, and why dilation is now a routine and safe part of care.
Here's the first patient. A twenty-eight-year-old man comes to the emergency department at midnight because a bite of steak has been stuck in his chest for four hours. He can't swallow his own saliva. He has asthma and seasonal allergies. When asked, he admits that for years he's eaten slowly, cut his meat into small pieces, and drunk water with every bite. Emergency endoscopy removes the meat. The esophagus shows faint rings, vertical furrows, and a few white specks.
He has a food impaction, and the likeliest cause in a young atopic man is eosinophilic esophagitis, an allergic inflammatory disease of the esophageal lining that ends in fibrosis, rings, and impactions if it's left untreated. That's why the one thing his endoscopist must do before finishing is take biopsies. Any adult with a food impaction should have esophageal biopsies at that endoscopy, because eosinophilic esophagitis is the leading cause of food impaction in adults, and the esophagus looks normal in five to ten percent of patients who have it. Removing the meat without biopsies sends him home with the cause undiagnosed. His endoscopist takes biopsies from the proximal and distal esophagus before the scope comes out.
His history fits the typical patient. Eosinophilic esophagitis is a disease of atopic men in their twenties and thirties, with a male predominance of about three to one, and the typical patient has a personal or family history of asthma, allergic rhinitis, atopic dermatitis, or food allergy. It's most common in cold, dry climates and in white patients, but it occurs across all demographics. Adults present with years of solid-food dysphagia, often with repeated impactions, from meat or bread at home or a meal in a restaurant. They adapt by eating slowly, washing every bite down with water, and avoiding the textures that stick, so the disease can go unrecognized for years. That's why eating habits are asked about directly: a patient who says he swallows fine may simply have learned to eat around the problem. Children present differently, with feeding intolerance, abdominal pain, vomiting, and failure to thrive rather than dysphagia, because the fibrosis that causes adult dysphagia hasn't developed yet. Peripheral blood eosinophilia is sometimes present but is neither sensitive nor specific, so a normal blood count doesn't argue against the diagnosis.
The mechanism explains both why his disease scars and what each treatment targets. Eosinophilic esophagitis is a type two helper T cell, or Th2, allergic disease, triggered by food antigens in a genetically susceptible person. The Th2 response in the esophageal lining produces interleukins four, five, and thirteen. Interleukin five makes eosinophils in the marrow and recruits them to tissue. Interleukins four and thirteen act on the esophageal lining itself, making it produce eotaxin three, the signal that pulls activated eosinophils across the basement membrane into the squamous epithelium. Interleukin thirteen also weakens the lining's barrier directly, by lowering filaggrin and the tight-junction proteins, which lets more antigen reach the tissue beneath and amplifies the response. Mast cells contribute in parallel. Staining for tryptase and for CD117, cluster of differentiation one seventeen, shows more mast cells, actively degranulating, and they can persist after eosinophils have fallen below the threshold with treatment, which is part of why some treated patients still have symptoms.
Years of that inflammation lay down fibrosis under the lining, driven by transforming growth factor beta, and the fibrosis produces the rings, the strictures, and impactions like his. Left untreated, eosinophilic esophagitis becomes a scarring disease, and scar doesn't reverse with drugs. That's why the emphasis is on diagnosing it early and bringing it into histologic remission.
His biopsies decide the diagnosis. It requires at least fifteen eosinophils per high-power field on esophageal biopsy, in a patient with symptoms of esophageal dysfunction, after other causes of esophageal eosinophils are excluded. Because the disease is patchy, at least six biopsies are taken from at least two levels, proximal and distal, and most endoscopists sample the middle too. A single level misses disease in about a third of patients when it's segmental.
The cause that most needs excluding is reflux, because reflux can also put eosinophils in the lining. Reflux usually produces fewer than five to ten eosinophils per high-power field, and the inflammation is mostly neutrophils with reactive epithelial changes. Eosinophilic esophagitis shows dense eosinophils, eosinophil microabscesses, thickening of the basal layer, and swelling between cells. His biopsies show up to sixty eosinophils per high-power field at both levels, with microabscesses. He has eosinophilic esophagitis.
He doesn't need a PPI trial to make the diagnosis. Until the late twenty-tens, an eight-week PPI trial was used to separate eosinophilic esophagitis from a condition labeled PPI-responsive esophageal eosinophilia. But that condition turned out to be identical to eosinophilic esophagitis clinically, endoscopically, under the microscope, and in gene expression. And the PPI doesn't work by lowering acid; it blocks the Th2 cytokine-driven release of eotaxin three by the lining cells, an anti-inflammatory effect. Current guidelines treat PPI-responsive disease as eosinophilic esophagitis and count the PPI as a treatment, not a diagnostic test. So a patient with at least fifteen eosinophils per high-power field and the right clinical picture has the diagnosis at the first endoscopy.
His endoscopy, described with a standard score, shows how far the disease has progressed. The score is the Endoscopic Reference Score, EREFS, for edema, rings, exudates, furrows, and strictures. Edema is pale lining that has lost its normal vessel pattern, scored absent or present. Rings are stacked circular folds that make the esophagus look like a trachea, sometimes called a feline esophagus or trachealization; they're scored mild, moderate when distinct but still passable with the scope, or severe when they block a standard adult scope. Exudates are white specks or plaques, which are eosinophil abscesses on the surface, scored mild when they cover less than ten percent of the surface and severe above that. Furrows are vertical grooves along the length of the esophagus, scored mild without visible depth or severe with visible depth. Strictures are scored absent or present and reported with the diameter at the narrowest point.
Those five findings split into two groups, and the split changes management. Edema, exudates, and furrows form the inflammatory subscore, and they respond to anti-inflammatory treatment. Rings and strictures form the fibrostenotic subscore, and they need dilation as well as drugs. The fibrostenotic findings predict long-term dysphagia, the need for dilation, and future impactions. His furrows and white specks are inflammation. His faint rings, and the impaction itself, show that fibrosis has already begun.
His inflammation is the treatment target, and there are three first-line options: a high-dose PPI, a swallowed topical steroid, or an empiric elimination diet. A fourth drug, the biologic dupilumab, is kept for refractory disease. None of the first-line options is clearly superior, so the choice is driven by adherence, allergic history, and the patient's preference.
Whichever he chooses, the response is judged by biopsy eight to twelve weeks later, with remission defined as fewer than fifteen eosinophils per high-power field. Symptoms aren't enough, and they lag behind histologic remission, for two reasons. Adaptive eating hides active disease. And fibrosis can cause dysphagia after the inflammation is gone. So a patient who says he's better may still be inflamed, and a patient who still has dysphagia may already be in remission.
A PPI is the most accessible option, because it's oral, cheap, and familiar, and it's a sensible choice for a patient who also has typical reflux, because it treats both. The dose for eosinophilic esophagitis is high: omeprazole twenty to forty milligrams twice daily, or an equivalent such as esomeprazole twenty to forty milligrams twice daily or lansoprazole thirty milligrams twice daily, for at least eight weeks before reassessment. It produces histologic remission in about thirty to fifty percent of patients on that first course. Loss of response during long-term maintenance is more common in patients who are rapid metabolizers by the liver enzyme cytochrome P450 two C nineteen, CYP2C19, and raising the dose often brings the response back, which explains why some patients need twice-daily dosing after once daily first worked.
A swallowed topical steroid is the most reliably effective drug class, inducing remission in roughly sixty to ninety percent depending on dose and formulation. It works by local anti-inflammatory action on the esophageal lining, with little systemic absorption. Fluticasone from a metered-dose inhaler is sprayed into the mouth and swallowed, eight hundred eighty micrograms twice daily in adults, given as four puffs of two hundred twenty micrograms, for a total of one thousand seven hundred sixty micrograms a day. The patient must not breathe in while spraying, because the goal is to coat the esophagus, not the lungs. Budesonide can be made into a thick slurry by mixing the one-milligram respiratory suspension with sucralose or honey, given at one milligram twice daily in adults. An approved budesonide oral suspension, two milligrams twice daily, is licensed by the Food and Drug Administration, FDA, for adults and adolescents eleven and older and avoids compounding.
The way the steroid is taken matters as much as the dose. With any formulation, the patient rinses his mouth and takes nothing by mouth for thirty minutes after each dose, so the drug stays in contact with the esophagus. The main side effect is candidiasis of the mouth or esophagus, which is why he rinses; it's treated with a short topical or systemic antifungal course while the steroid continues. Adrenal suppression, and in children growth suppression, are uncommon at standard doses, but they prompt consideration of a lower dose for maintenance.
An elimination diet works because the disease is driven by food antigens, and removing the trigger reverses the Th2 response. The usual triggers are milk, wheat, eggs, soy, nuts, and seafood. The step-up approach removes just milk and wheat for six to eight weeks, then repeats endoscopy. If that fails, it expands to four foods, milk, wheat, eggs, and soy, then to all six. Removing milk and wheat alone brings about forty percent into remission, and each wider step brings in more responders, with the six-food diet the most effective of the three. Starting small means fewer endoscopies overall than the older step-down approach, which began with all six. Once he's in remission, foods are added back one at a time, each followed by endoscopy and biopsies, so he ends up avoiding only his specific trigger. Milk is the dominant trigger, found in roughly half to sixty percent of responders, followed by wheat, with soy, eggs, and nuts accounting for a smaller share.
Allergy testing doesn't shorten that process. Skin prick tests and blood immunoglobulin E, IgE, panels don't reliably identify the triggers, because the disease is largely mediated by immunoglobulin G four, IgG4, rather than IgE. Diets chosen by allergy testing reach remission in only about forty-five percent, and that's why empiric elimination is the standard.
An elemental diet, an amino acid formula with no whole food, achieves remission in over ninety percent of patients. But it's rarely tolerated outside children or refractory adults, because of taste and cost, so it's kept for severely symptomatic or polysensitized patients who can't find a diet they can maintain.
The choice follows the patient's life. A patient with a strongly atopic history and known food allergy may prefer a diet, because it removes the antigen directly. A patient with typical reflux may prefer a PPI as a drug that treats both. A patient who travels often, or can't control where his food comes from, may prefer a swallowed steroid. This patient travels for work and eats at restaurants most nights, which makes a diet hard to keep, so he chooses budesonide suspension.
His repeat endoscopy at twelve weeks shows only a few eosinophils per high-power field, and the furrows and exudates are gone. The faint rings remain. He's in histologic remission, and he's eating normally.
Remission doesn't mean he can stop, because relapse after stopping is the rule. After steroids are stopped, about ninety percent of patients relapse within a year, with an average time to recurrence of about nine months, and the same is true of the other treatments. So responders continue maintenance therapy at the lowest effective dose. Stopping lets the inflammation and the fibrosis start again, and the next impaction may come with a stricture.
Not every patient responds like him, and the second patient shows what happens next. A forty-one-year-old woman was diagnosed with eosinophilic esophagitis three years ago. She has asthma and eczema. Twice-daily omeprazole for twelve weeks didn't bring her into remission, and neither did budesonide suspension for twelve weeks. She tried a two-food elimination diet but couldn't keep to it. She still has solid-food dysphagia several times a week and has had two more impactions. Her latest endoscopy shows furrows, exudates, and moderate rings, with a fixed narrowing of about thirteen millimeters that the standard scope passes only with resistance. Biopsies show forty eosinophils per high-power field.
Before escalating, the first step is to confirm she's truly refractory. Refractory means disease that hasn't reached histologic or clinical remission after an adequate trial of first-line treatment, so each trial is checked for adequacy. The PPI counts only if the dose was high enough for long enough: omeprazole twenty to forty milligrams twice daily for at least eight weeks qualifies, and once daily doesn't. A patient who responded once and then lost it may be a rapid metabolizer who needs a higher dose.
The steroid counts only if it was taken correctly, so she's asked to describe exactly how she takes it. Budesonide is swallowed, then nothing by mouth for thirty minutes. If she's been taking it with breakfast, or washing it down with coffee, it hasn't had a fair trial. With fluticasone, the common error is inhaling the spray into the lungs instead of swallowing it. And a diet she couldn't keep to wasn't a failure of the diet.
She took both drugs correctly, at full doses, for long enough, so she's refractory. The modern definition of refractory also includes a second group: patients whose biopsies have cleared but who still have dysphagia from fibrosis that's already set in. That group is easy to mistreat, because their problem is no longer inflammation.
Her plan depends on which of her problems is inflammation and which is scar, read from the same EREFS split. Forty eosinophils per high-power field, furrows, and exudates mean active inflammation. Moderate rings and a thirteen-millimeter narrowing mean fibrosis. She has both, so she needs a treatment that controls the inflammation and dilation for the narrowing. That combined approach is the standard for refractory disease, and either one alone is incomplete.
For the inflammation, the next step is dupilumab, a fully human monoclonal antibody against the alpha chain of the interleukin four receptor. That chain is part of two receptors. The type one receptor pairs it with the common gamma chain and binds interleukin four. The type two receptor pairs it with the interleukin thirteen receptor alpha one chain and binds both interleukin four and interleukin thirteen. So blocking the alpha chain shuts off both cytokines at once. That stops the lining from producing eotaxin three and reverses the barrier damage interleukin thirteen causes. It acts upstream of the eosinophils and mast cells, at the cytokine signal that drives the disease.
Dupilumab is the only biologic approved for eosinophilic esophagitis. The FDA approved it in May twenty twenty-two for patients twelve and older weighing at least forty kilograms, at three hundred milligrams subcutaneously every week. In January twenty twenty-four the approval was extended to children one year and older weighing at least fifteen kilograms, dosed by weight. The weekly schedule comes from the registration trial, which randomized patients with PPI-refractory disease and at least fifteen eosinophils per high-power field to weekly dosing, dosing every two weeks, or placebo for twenty-four weeks. Histologic remission occurred in roughly sixty percent with either dose. But only weekly dosing improved dysphagia compared with placebo. So in eosinophilic esophagitis it's given weekly, even though other dupilumab indications use every two weeks.
Her other atopic diseases make dupilumab especially attractive, because it's also approved for asthma, atopic dermatitis, and chronic rhinosinusitis with nasal polyps, so one drug can treat conditions that would otherwise need separate biologics. Its main side effects are injection site reactions and conjunctivitis, so she's warned about red, irritated eyes. Other biologics are being studied but aren't approved for eosinophilic esophagitis: cendakimab against interleukin thirteen, benralizumab against the interleukin five receptor alpha chain, which depletes eosinophils through antibody-dependent cellular cytotoxicity, and mepolizumab against interleukin five.
Her response to dupilumab is judged by biopsy, not symptoms, with remission still defined as fewer than fifteen eosinophils per high-power field. That matters even more in her, because her narrowing will keep causing dysphagia whether or not the inflammation clears. By symptoms alone, a drug that failed can't be told apart from a drug that worked on an esophagus that's still scarred.
An elemental formula is another option for inflammation that won't settle, but she couldn't keep to a two-food diet, and a formula-only diet is far harder, so it isn't right for her.
For the narrowing, the treatment is dilation, because no drug will open a thirteen-millimeter fibrotic narrowing; scar doesn't reverse with anti-inflammatory treatment. Dilation in eosinophilic esophagitis was once feared, because the strictures were thought to be unusually fragile and prone to extensive tears and perforation. That concern has been substantially revised. Across more than thirteen hundred dilations, the perforation rate is about two tenths of a percent, within the usual range for dilating any benign stricture. So dilation is a routine, safe part of care. It's also safe and effective in patients who respond to medical therapy but have residual fibrotic narrowing, and it shouldn't be put off indefinitely because of an outdated idea of risk.
The technique is gradual. It starts at a small diameter and advances by about three millimeters per session, toward a goal of fifteen to eighteen millimeters over several sessions, which usually controls symptoms. Through-the-scope balloons and wire-guided Savary bougies both work, with neither clearly better. The older rule of three, no more than three successive dilator sizes in one session once moderate resistance is felt, is no longer strict dogma in eosinophilic esophagitis, but graduated dilation remains the conservative norm.
She's told beforehand about two expected findings that are not complications. Mucosal tears, called rents, appear in roughly seventy-five to eighty-five percent of dilations in eosinophilic esophagitis; they show that the fibrotic ring is being broken open, which is the goal. And chest pain afterward is common, with reported rates from a few percent up to seventy percent depending on the series and the dilator, and it usually settles within hours to days.
The real contraindications are few: deep active ulceration, suspected perforation, and a high-grade stricture in a patient who hasn't had medical treatment first. That last one is about order. When the stricture is very tight and the esophagus has never been treated, medical therapy calms the inflammation first, so the dilation works on the fibrosis that's left rather than on inflamed tissue. She's had months of treatment, and her stricture isn't extremely tight, so she can be dilated now while dupilumab starts.
Dilation treats the narrowing, not the disease. Uncontrolled inflammation will rebuild the fibrosis, so dilation always goes together with maintenance medical therapy: dupilumab, a swallowed steroid, a PPI, or a diet. She starts dupilumab three hundred milligrams weekly and has serial dilations from thirteen millimeters toward eighteen over three sessions. Dupilumab alone would leave a fixed narrowing that keeps catching food; dilation alone would leave the inflammation that caused the narrowing. Six months later, her biopsies show two eosinophils per high-power field, the furrows and exudates are gone, and so is her dysphagia.
Her follow-up shows why symptoms and biopsies have to be read together. If she returns a year later in histologic remission, with no furrows or exudates but with dysphagia again and a fixed fourteen-millimeter narrowing, the drug hasn't failed and she doesn't need a different biologic. She needs another dilation. When the inflammation is controlled and dysphagia persists, the problem is fibrosis, and fibrosis is dilated. Persistent mast cells after eosinophils clear can also keep some patients symptomatic, one more reason not to read symptoms alone as treatment failure. She stays on dupilumab, because relapse after stopping is the rule.
Here are the traps the boards set in these patients.
At diagnosis, removing a food impaction without taking biopsies misses the leading cause, even when the lining looks normal. Taking too few biopsies, or sampling only one level, misses patchy disease. And requiring a PPI trial before making the diagnosis is outdated, because PPI-responsive eosinophilia is eosinophilic esophagitis.
In first-line treatment, judging the response by symptoms instead of biopsies is wrong in both directions. Letting the patient inhale the fluticasone, or eat right after a dose, wastes the steroid. Stopping a steroid because of thrush is unnecessary, because a short antifungal course treats it. Choosing diet triggers from allergy tests fails, because the disease isn't IgE-driven. And stopping treatment once the patient is in remission invites relapse.
In refractory disease, calling a patient refractory before checking that each trial was adequate leads to needless escalation. Dosing dupilumab every two weeks clears eosinophils without improving dysphagia. Avoiding dilation for fear of perforation leaves fibrosis untreated, and calling rents or chest pain after dilation a complication misreads the expected result. Dilating without continuing medical therapy lets the narrowing rebuild. And switching biologics in a patient whose biopsies have cleared but whose esophagus is still narrow treats the wrong problem.
So, start to finish.
Eosinophilic esophagitis is a food-driven Th2 allergic disease of the esophagus that ends in fibrosis, rings, and impactions. Biopsy every adult food impaction. The diagnosis is at least fifteen eosinophils per high-power field on at least six biopsies from two levels, in a patient with esophageal symptoms, with no PPI trial needed. Describe the endoscopy with EREFS, separating the inflammatory findings from the fibrostenotic ones.
First-line treatment is a high-dose twice-daily PPI, a swallowed topical steroid taken with nothing by mouth for thirty minutes, or a step-up elimination diet starting with milk and wheat, matched to the patient. Remission is confirmed by biopsy at eight to twelve weeks, and maintenance continues, because relapse after stopping is the rule.
Before calling the disease refractory, check that each treatment was taken correctly, at full dose, for long enough. Dupilumab blocks the interleukin four receptor alpha chain, shutting off interleukins four and thirteen, and is dosed three hundred milligrams weekly. Fibrotic narrowing is dilated gradually toward fifteen to eighteen millimeters, with a perforation risk no higher than other benign strictures; rents and chest pain are expected. Dysphagia with clean biopsies means fibrosis, which needs dilation, not another drug.
Episode two turns to infections and chemical injuries of the esophagus: Candida, herpes simplex, and cytomegalovirus, then caustic and pill injury.