Pill-Induced Esophagitis When the Causative Drug Can't Be Stopped
Confirmed doxycycline-induced esophageal injury in a patient who has a real, ongoing clinical need for the drug turns the usual first instinct — stop the offending agent — into a genuine choice between administration modification and drug substitution.
Priya N., 24, moved into her first apartment alone six months ago and still texts her mother a photo of dinner most nights, half joke and half genuine reassurance-seeking as she settles into living by herself for the first time. She's been on doxycycline 100mg twice daily for eight months for moderate inflammatory acne that had failed topical therapy and had a real, documented psychosocial impact on her before treatment started — two dermatologists' notes describe it as significantly affecting her confidence at a stressful, high-visibility point in a new job. Three days ago she developed sharp retrosternal chest pain and odynophagia, worse a few minutes after taking her evening dose while lying on the couch scrolling her phone rather than sitting upright. Endoscopy showed a discrete, well-demarcated ulcer in the mid-esophagus, at the level of aortic-arch compression where esophageal transit is known to slow — the classic location and appearance of pill-induced esophageal injury, and doxycycline is among the most commonly implicated drugs.
The reflexive move in pill-induced esophagitis is usually to stop the causative drug, but that's genuinely harder to justify here than it sounds: her acne was significant enough that two separate dermatology notes documented real psychosocial impact before doxycycline controlled it, and eight months of meaningful improvement is not a trivial thing to abandon for a drug-administration problem rather than a drug-tolerability one. The injury itself is a well-described, largely preventable complication — caused by the tablet's acidic dissolution sitting directly against esophageal mucosa when it doesn't clear promptly into the stomach, not by any systemic toxicity of the drug — which makes fixing how she takes it a genuinely reasonable alternative to switching agents entirely. The location is itself diagnostic. Pill esophagitis clusters at points of anatomic compression — the aortic arch, the left atrium, and the gastroesophageal junction — where a swallowed tablet is most likely to pause in transit even in a patient with entirely normal underlying motility, which is part of why correcting administration technique, rather than assuming an underlying motility problem, is the evidence-consistent first move for an otherwise healthy 24-year-old. Priya's dermatologist was looped in directly rather than informed after the fact, specifically so the administration-correction plan and the acne-treatment plan were coordinated as one conversation rather than two clinicians independently telling her slightly different things about the same medication.
Fixing how the drug is taken, not necessarily what drug it is
This is a textbook pill-esophagitis location and appearance, and the mechanism matters here: it's direct local mucosal contact from the tablet sitting in place too long, not a systemic reaction to doxycycline itself. If she takes it upright, with a full glass of water, and stays upright afterward, that mechanism largely goes away. I'd correct the administration and continue the drug rather than switch.
I'd genuinely prefer to keep her on doxycycline if it's reasonably safe to do so. She's had eight months of real improvement on a regimen that took real trial and error to get to, and her acne had documented psychosocial impact before it worked. I don't want to switch her off something that's working for a problem that has a described fix.
My hesitation is entirely about whether the fix actually holds — she was taking it lying down at bedtime, which is exactly the pattern that produces this injury, and I don't have a clean way to confirm that habit has genuinely changed once she leaves today.
I agree with continuing doxycycline given how meaningful her response has been, but I don't think 'we told her to sit up and drink water' is a complete plan on its own. I'd add a short PPI course to support healing regardless, and set an actual follow-up call in two weeks specifically to confirm the administration change is happening, not just to check on her symptoms. If it turns out the habit genuinely isn't sticking, that's the point to revisit switching agents — not before we've given the fix a real, verified chance.
You're right that the mechanism has a described fix, but "we told her to sit up and drink water" isn't the same as confirming she actually will — the plan needs a way to check that, not just state it.
Agreed: doxycycline continued with explicit administration counseling (upright, full glass of water, remain upright 30 minutes, never at bedtime), a 4-week PPI course to support healing, and a two-week follow-up call specifically to confirm the administration change is happening, not just to check symptoms.