Laryngopharyngeal Reflux: Whether an Empiric PPI Trial Still Makes Sense
Chronic throat-clearing and hoarseness with a laryngoscopy read as suggestive of reflux sit at the center of a real, still-unsettled controversy: multiple placebo-controlled trials have failed to show PPI therapy clearly outperforms placebo for these symptoms, yet empiric trials remain common clinical practice.
Harold M., 61, retired from thirty years as a high-school choir director last spring and still volunteers directing the community chorus twice a week, which is how he first noticed his voice giving out partway through rehearsal — something that had never happened across three decades of daily singing. He's had a persistent urge to clear his throat and a sensation of something “stuck” in his throat for about four months, with intermittent mild hoarseness that worsens by evening. He denies heartburn or regurgitation entirely. An ENT evaluation found mild arytenoid erythema and edema on laryngoscopy — findings often read as suggestive of reflux, though neither specific nor sensitive for it, since the same appearance shows up in a meaningful fraction of people with no reflux at all.
His Reflux Symptom Index scored 19, above the commonly used cutoff of 13. That's where the actual controversy lives: several placebo-controlled trials, including one of the largest and most frequently cited, have failed to show twice-daily PPI therapy clearly outperforming placebo for laryngeal symptoms like his, even in patients selected using laryngoscopy findings and symptom scores much like these. A substantial fraction of patients empirically trialed on PPIs for LPR-type symptoms improve on placebo alone, which makes distinguishing a real drug effect from natural symptom fluctuation and placebo response genuinely difficult in an individual case rather than a settled empiric-trial decision. Those trials enrolled patients selected exactly the way he was — by RSI score and laryngoscopic appearance — which is what makes the negative result hard to set aside as the wrong population. The ACG reflux guideline (Katz and colleagues) reflects that, declining to endorse empiric PPI therapy for laryngeal symptoms in patients without typical reflux symptoms, and Harold denies them entirely. He was also asked directly whether his voice fatigue tracked more closely with rehearsal length than with meal timing, and his answer — rehearsal length, regardless of what or when he'd eaten — was itself a data point the team weighed, since a cleaner meal-timing association would have argued more strongly for reflux specifically.
An empiric trial with genuinely mixed evidence behind it
I want to name the evidence honestly before we start him on months of a drug. Several placebo-controlled trials — including some of the larger ones specifically designed around patients selected the way he was, by RSI and laryngoscopy findings — have failed to show twice-daily PPI therapy clearly outperforming placebo for these symptoms. That's not a fringe finding; it's one of the more consistent negative results in reflux medicine, and it should genuinely inform whether an empiric trial is the right first move here.
I've seen real patients who looked exactly like Harold respond meaningfully once acid suppression actually controlled their reflux, and I don't want to withhold a low-risk trial from someone who might be a true responder just because the population-level trial data is mixed.
I take the trial data seriously, and I'm not arguing it's wrong — I'm arguing that population-level negative trials don't rule out real individual responders, and an 8- to 12-week trial is a reasonable, low-risk way to find out which he is.
I'd support a trial, but only with real structure around it, because the actual harm here isn't the drug's side-effect profile at this dose — it's what happens when an unproven empiric trial has no defined endpoint. Twelve weeks, a repeat RSI at the end, and a genuine conversation about stopping if it hasn't moved meaningfully, rather than continuing by inertia because he's already on it. Given the negative trial data, I'd also want confirmatory pH-impedance testing offered now rather than only after a failed empiric trial, so we're not defaulting to months of medication as the only diagnostic tool available to us.
Agreed: a defined 12-week pantoprazole trial with reflux-precaution counseling started concurrently, and a repeat Reflux Symptom Index at the endpoint rather than an open-ended continuation.
Continue at the lowest effective dose with a planned taper attempt, since sustained response in an individual patient is still real evidence even against mixed population-level trial data.
Stop the PPI and proceed to ambulatory pH-impedance testing rather than extending the trial further on the chance it eventually works.