Subscribe
Share
Share
Embed
Board Pearls is a gastroenterology board review built around clinical reasoning, not recall. Each 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.
The gastroenterology series covers 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. Episodes are grouped by chapter and built from the primary guidelines and pivotal trials the boards draw from (ACG, AGA, AASLD, ASGE), not from textbook summaries.
Use it as an audio companion to the written curriculum, MCQs, and AI tutor at boardpearls.com. Questions or feedback: hello@boardpearls.com.
Welcome to Board Pearls. This is episode two of two of the GERD and Refractory Reflux chapter, in the Esophageal Disorders module. In this episode we cover refractory reflux: the Lyon Consensus framework that splits proven reflux from reflux hypersensitivity and functional heartburn, the antireflux surgery that follows only after objective reflux is documented, and the workup of the patient whose PPI failed, where the diagnosis is often not GERD at all.
The patient who anchors this episode is on twice-daily PPI, has been on twice-daily PPI for months, and still has heartburn. The question is not how to escalate. The question is what they have. Most of the work in refractory reflux is sorting that vignette into the right diagnostic bucket, because the treatments diverge sharply once the sort is right. The temptation in any persistent-symptoms vignette is to add more acid suppression or to refer for surgery. Both moves fail more often than they help, because most PPI-refractory patients do not have refractory acid reflux. They have something else, and the something else has its own treatment.
Start with reflux monitoring, because reflux monitoring is the test that decides which bucket the patient is in. The framework is the Lyon Consensus, updated to version two in twenty twenty-four, and it organizes evidence for GERD into three tiers: definitive, supportive, and against. Definitive evidence is anatomic or physiologic proof that the antireflux barrier has failed. Long-segment Barrett esophagus is definitive. LA grade B, C, or D esophagitis is definitive. Peptic stricture is definitive. An acid exposure time above six percent on a wireless pH study is definitive. Acid exposure time below four percent on every day of the study is evidence against pathologic reflux. The borderline zone is four to six percent, and that zone is where the supportive metrics earn their keep.
The supportive metrics are what Lyon two added to the framework, and they exist to rescue interpretation in the borderline zone. The reflux episode count above roughly eighty events in twenty-four hours suggests pathologic reflux. The mean nocturnal baseline impedance, which measures mucosal integrity during a long stable period when no one is swallowing, runs low when the mucosa has been chronically inflamed. Values under fifteen hundred ohms support GERD. Values over twenty-five hundred ohms argue against it. The post-reflux swallow-induced peristaltic wave index measures how often a reflux event triggers a saliva-clearing swallow, and a low index reflects impaired clearance. Symptom-reflux association adds a separate layer on top of the burden metrics, and there are two ways to measure it. The symptom index counts the percentage of symptoms preceded by a reflux event, with fifty percent or higher considered positive. The symptom association probability is a statistical test on a two-by-two table of symptom and reflux time intervals, with ninety-five percent or higher considered positive. The symptom association probability is the more rigorous metric and is the one the boards usually invoke.
The on-PPI versus off-PPI decision follows from a simple question. The question is what you are trying to prove. If you do not yet know whether the patient has GERD, you are trying to prove that reflux exists, and PPI on board would mask the answer. Test off PPI, with the drug held for seven days. If you already know the patient has GERD, because of LA grade C or D esophagitis or long-segment Barrett or a prior abnormal pH study, the question changes. Now you are trying to characterize what is breaking through despite acid suppression. Test on PPI. The on-PPI study has to be impedance-capable, because the residual events on suppression include weakly acidic and non-acid reflux, and pure pH does not see them. The off-PPI study is usually the Bravo wireless capsule, clipped six centimeters above the squamocolumnar junction, recording for forty-eight to ninety-six hours. The longer window catches a typical day and improves the chance of capturing a symptomatic episode. The on-PPI study is the twenty-four-hour catheter pH-impedance.
Run the Lyon framework on a refractory patient and they sort into four buckets. Confirmed GERD is acid exposure time above six percent, regardless of symptom association. The patient has pathologic reflux and the symptoms can be attributed to it. Borderline GERD is acid exposure time in the four to six percent zone with supportive evidence, where the diagnosis hangs on the adjunctive metrics. Reflux hypersensitivity is acid exposure time below four percent with a positive symptom association. The patient genuinely associates symptoms with reflux events, but the events are physiologic in volume. The esophagus is hypersensitive rather than overexposed. Treatment combines a neuromodulator with acid suppression. Functional heartburn is acid exposure time below four percent with a negative symptom association. The patient has heartburn that is not driven by reflux at all, and the treatment is a neuromodulator alone. More PPI does not help functional heartburn. Fundoplication does not help functional heartburn. The single most common reason a fundoplication fails is that the patient referred for it had functional heartburn, not reflux.
That sort is the central reason the boards reward objective testing before antireflux surgery, and it sets up the second piece of the episode. Antireflux surgery is matching the right operation to the right patient, and the matching breaks down completely if the indication is wrong. The first principle is that no patient should be sent for a wrap on PPI failure alone. Objective evidence of reflux is mandatory. LA grade B or higher esophagitis, long-segment Barrett, peptic stricture, or an abnormal pH-impedance study off PPI. PPI failure is a workup trigger, not an operative indication.
Every fundoplication shares the same mechanism. The surgeon reduces the hiatal hernia, approximates the diaphragmatic crura, brings a segment of distal esophagus into the abdomen, and wraps a portion of the gastric fundus around it. The wrap reproduces the angle of His and its flap-valve effect, restores the abdominal-pressure environment of the distal esophagus, and buttresses the LES from outside. It also blunts transient LES relaxations by preventing the fundus from distending freely with each meal. The choice of wrap geometry is then a single question. Can the esophagus generate enough peristaltic force to push a bolus past the wrap.
The Nissen is the three-hundred-and-sixty-degree complete posterior wrap. It maximizes reflux control because the full wrap closes the distal esophagus most effectively against the gastric pressure gradient. The trade-off is postoperative dysphagia and gas-bloat syndrome, because a tight full wrap cannot vent during a normal transient relaxation. The Nissen is the operation of choice in the patient with normal esophageal motility. It is contraindicated in scleroderma esophagus and other absent-contractility states, because peristalsis cannot drive a bolus past a tight full wrap. The LOTUS trial randomized just over five hundred patients with PPI-responsive GERD to esomeprazole versus laparoscopic Nissen and showed five-year remission rates of around ninety-two percent on PPI and around eighty-five percent after surgery. The lesson is that high-quality PPI therapy is competitive with surgery in selected responders, and surgery's main advantage is in regurgitation control and in patients who cannot or will not stay on long-term acid suppression.
The Toupet is the two-hundred-and-seventy-degree partial posterior wrap. It leaves the anterior third of the distal esophagus uncovered, which lowers the resistance a peristaltic wave has to overcome. That is the operation of choice in hypomotility, because a Nissen on a weak esophagus produces dysphagia that the peristaltic pump cannot overcome. Scleroderma and ineffective esophageal motility get a Toupet if they get any wrap at all. Reflux control with the Toupet is slightly less complete than with the Nissen in normal-motility patients, but the difference matters less than the consequences of mismatching wrap to motility. The Dor is the one-hundred-and-eighty-degree anterior wrap, and it is most often paired with a Heller myotomy in achalasia, where the surgical goal is partial reflux protection without compromising the deliberately disrupted LES. Dor has a smaller role as a primary GERD operation.
Magnetic sphincter augmentation, the LINX device, is a ring of titanium beads with magnetic cores placed around the distal esophagus laparoscopically. The magnets close the LES against the small pressure differential of resting respiration and the abdominothoracic pressure swings that drive reflux. A swallowed bolus generates enough force to separate the beads, and the device reapproximates after the bolus passes. The principal advantage over a Nissen is preservation of belching and vomiting, which means much less gas-bloat syndrome. The device is also potentially reversible. The exclusion criteria are essential. Hiatal hernia larger than three centimeters excludes the device. BMI over thirty-five excludes it. Barrett esophagus, LA grade C or D esophagitis, absent or severely impaired peristalsis, titanium or nickel allergy, and prior major upper GI surgery all exclude it. The current ACG guideline endorses LINX specifically for troublesome regurgitation that fails medical therapy, on the basis of randomized data showing LINX superior to medical therapy for that endpoint.
Transoral incisionless fundoplication, the EsophyX-based TIF procedure, creates a partial fundoplication endoscopically by plicating the gastric fundus onto the distal esophagus through stapled fasteners. It avoids skin incisions and is less invasive than any laparoscopic option. The selection criteria are restrictive. The hiatal hernia must be under two centimeters. Severe esophagitis at LA grade C or D excludes the procedure. The best candidate has PPI-responsive symptoms with troublesome regurgitation as the dominant complaint. Symptom improvement at one to two years is good. Durability past five years is debated, with reoperation and PPI resumption rates rising over time. The combined version, sometimes called concomitant TIF, pairs a laparoscopic crural repair with a subsequent endoscopic plication, and extends candidacy to patients with larger hiatal hernias.
The morbidly obese patient with reflux is a separate algorithm. Roux-en-Y gastric bypass is the antireflux operation of choice in that population. The Roux limb diverts bile and acid away from the small gastric pouch. The pouch itself contains much less parietal cell mass. And the procedure addresses the obesity that is driving the reflux in the first place. Sleeve gastrectomy causes or worsens reflux in roughly a third of patients, because it removes the gastric fundus and converts the stomach into a high-pressure tubular reservoir. Sleeve should generally be avoided in obese patients with established GERD. When a patient with prior sleeve develops refractory reflux with erosive disease or Barrett, the conversion to Roux-en-Y bypass is the supported operation. BMI over thirty-five is also an independent predictor of fundoplication failure, which is the other reason that obese refluxers go to bariatric surgery rather than to a wrap.
The historical endoscopic procedures deserve a brief note because they show up as distractors. Stretta delivers radiofrequency energy to the gastroesophageal junction with a proposed mechanism of mucosal remodeling and reduced transient relaxations. Sham-controlled trials have produced modest results and current ACG guidance does not routinely recommend endoscopic antireflux procedures outside of TIF. Bulking-agent injections such as Enteryx and Gatekeeper are no longer marketed. Anti-reflux mucosectomy and the RefluxStop device are under investigation but have not displaced the standard options.
The pre-surgical workup is the same regardless of which procedure is selected, and it is non-negotiable. Endoscopy is required to document esophagitis grade and hiatal hernia size, and to exclude Barrett, EoE, and malignancy. High-resolution manometry is required to confirm normal peristalsis or to document the hypomotility that mandates a partial wrap, and to exclude achalasia masquerading as reflux. Multiple rapid swallow provocation on the manometry, where five rapid swallows are followed by a contractile response, is the test of contractile reserve. Augmented post-swallow contraction predicts tolerance of a full wrap. Absent augmentation biases the decision toward a partial wrap. Reflux monitoring off PPI is required when the diagnosis of GERD has not already been objectively established. Postoperative complications are the slipped wrap, the tight wrap, gas-bloat syndrome, and telescoping. Suspected wrap failure is evaluated first with a barium swallow rather than EGD, because barium shows wrap geometry that endoscopy does not.
The third piece of the episode is the refractory workup itself, the algorithm that runs when a patient on optimized PPI still has symptoms. The framework is mechanism-driven and the candidate runs it in order. Each step asks whether the diagnosis is what we think it is before escalating treatment.
The first step is to confirm that the PPI is being taken correctly. PPIs are prodrugs activated in the parietal cell canaliculus, and they bind only proton pumps that are actively pumping. Pumps are activated by meal-stimulated gastrin and acid secretion. The drug has to be in the bloodstream when the pumps light up. That means thirty to sixty minutes before breakfast for once-daily dosing, and thirty to sixty minutes before breakfast and before dinner for twice-daily. A patient taking the PPI at bedtime, with food, or whenever they remember is functionally untreated. The Spechler trial in twenty nineteen showed that roughly twelve percent of patients referred for PPI-refractory heartburn responded to a structured trial that fixed the timing. That correction is the cheapest intervention in the algorithm and is never skipped.
The second step is to escalate acid suppression. Once-daily PPI moves to twice-daily, before breakfast and before dinner. The agent can be switched to a higher-potency PPI, with esomeprazole or rabeprazole substituting for omeprazole or pantoprazole. The switch matters most in suspected CYP2C19 ultrarapid metabolizers, in whom omeprazole and pantoprazole and lansoprazole clear too fast to maintain pump-binding levels. Vonoprazan, the potassium-competitive acid blocker, is the modern answer for genuine breakthrough acid in the patient who has already failed twice-daily PPI. Vonoprazan does not require meal-stimulated pump activation. It blocks the pump in a meal-independent, sustained, more complete way than any PPI dose. A bedtime H2 blocker can be added for documented nocturnal acid breakthrough, but tachyphylaxis erodes the H2 effect within two to four weeks of continuous use, so the H2 add-on is a short-term measure.
The third step is endoscopy with biopsies, even when a prior endoscopy was normal. The biopsies are non-negotiable, and they are taken at two levels, proximal and distal, two to four samples per level. The reason is eosinophilic esophagitis. EoE causes PPI failure, is histologic by definition at fifteen or more eosinophils per high-power field, and can look entirely normal endoscopically in a meaningful minority of cases. A negative-appearing endoscopy without biopsies misses EoE in exactly the patients who most often have it. The full EoE framework, including the diet, the topical steroids, and the biologic option, lives in the eosinophilic chapter. The point in the refractory algorithm is that the biopsy is mandatory. Endoscopy at this step also documents the hiatal hernia size, looks for Barrett, and rules out structural disease.
The fourth step is reflux monitoring, on or off PPI as dictated by whether GERD has been objectively established. The framework is the Lyon Consensus. The cutoffs are acid exposure time over six percent, in the borderline four to six zone, or under four. The symptom association metrics are the symptom association probability at ninety-five and the symptom index at fifty. The sort produces the four buckets: confirmed GERD, borderline GERD, reflux hypersensitivity, and functional heartburn.
The fifth step is where most of the diagnostic work actually lives. It is the step that asks whether the patient has GERD at all, or whether they have one of the diagnoses that mimic it. Functional heartburn is heartburn with normal acid exposure and negative symptom association. The treatment is a neuromodulator. A low-dose tricyclic, an SSRI, or hypnotherapy, and not more PPI and not surgery. Reflux hypersensitivity is heartburn with normal acid exposure but positive symptom association. The treatment combines acid suppression with a neuromodulator, because the patient is responding to physiologic reflux that the esophagus is reading as injurious. Rumination syndrome is effortless regurgitation of recently swallowed food, beginning within ten to fifteen minutes of starting a meal, with the patient re-swallowing or spitting the bolus. There is no preceding nausea and no retching. On high-resolution manometry, the diagnostic finding is an abdominal-strain pressure wave with simultaneous gastric and intra-esophageal pressurization and concurrent LES relaxation. The treatment is diaphragmatic breathing taught by a behavioral therapist, not acid suppression and not a wrap.
Supragastric belching is the rapid suction of air into the esophagus followed by immediate expulsion before the air ever reaches the stomach. It is often misread as reflux because the patient describes belching and regurgitation. On impedance the recognition pattern is unmistakable. Air enters the esophagus from above and exits from above, never crossing the LES. The treatment is speech therapy and behavioral retraining. Achalasia can present with regurgitation and a sensation that looks like reflux, particularly when retained food and saliva ferment in a non-relaxing LES. High-resolution manometry with the Chicago Classification framework is the test that excludes it. The full achalasia algorithm lives in the motility chapter. Eosinophilic esophagitis is the structural mimic that the biopsies in step three were designed to catch.
Adjunctive medications are useful in selected confirmed-reflux patients with symptoms despite optimized acid suppression. Baclofen ten milligrams three times daily is a GABA-B agonist that suppresses transient LES relaxations and reduces both acid and non-acid reflux. It helps the patient with troublesome regurgitation and the patient with persistent non-acid reflux on impedance. Drowsiness and fatigue limit dose escalation and long-term use. Alginate, the Gaviscon-family preparations, precipitates into a gel raft on contact with gastric acid. The raft floats on top of the gastric contents and displaces the postprandial acid pocket distally, away from the gastroesophageal junction. That displaces the substrate that the next transient relaxation would otherwise carry up. Alginate is particularly useful for postprandial breakthrough symptoms and as an add-on in non-erosive disease. Sucralfate is a sulfated sucrose-aluminum complex that coats ulcerated mucosa. Evidence in non-pregnant GERD is limited. Sucralfate is the preferred adjunct in pregnancy, where the systemic agents are second-line. Antacids provide on-demand relief and do not heal erosive disease.
Surgery is the final step in the algorithm and is reserved for the small subset with objectively confirmed reflux-related heartburn after the mimics have been excluded. The Spechler trial in twenty nineteen screened three hundred and sixty-six PPI-refractory referrals and randomized seventy-eight carefully selected patients with truly refractory reflux-related heartburn. Arms were laparoscopic Nissen, active medical therapy with omeprazole and baclofen and desipramine, or placebo medical therapy. Surgery was superior at around sixty-seven percent treatment success at one year, with active medical therapy around twenty-eight and placebo around twelve. The two lessons of the trial are inseparable. Surgery wins in the small carefully selected subgroup, and the rest of the original three hundred and sixty-six referrals should never have reached surgery, because most did not have refractory reflux. They had PPI-timing failures, functional heartburn, reflux hypersensitivity, EoE, motility disorders, or rumination. The Spechler trial is the single best argument for the structured workup that precedes any surgical referral.
The atypical and extraesophageal manifestations of GERD deserve a separate caution before we close. Chronic cough, laryngitis, asthma, and dental erosion show up as test items, and their relationship to reflux is much weaker than typical heartburn. Empiric PPI response rates in extraesophageal symptoms are low. Roughly half of patients with confirmed reflux-associated cough respond to PPI, and less than a fifth of patients with cough and no confirmed reflux respond. Laryngopharyngeal reflux is over-diagnosed by laryngoscopy, because most healthy volunteers have at least one laryngeal finding traditionally attributed to reflux. Objective reflux testing is required before committing a patient with isolated cough or laryngitis to long-term PPI. Fundoplication has poorer outcomes in atypical reflux than in typical heartburn, with significantly higher hazard for symptom recurrence.
So the way of thinking the episode taught is this. PPI failure is not refractory GERD. It is a workup trigger that mostly turns up something other than reflux. The Lyon Consensus framework, with acid exposure time at six and four percent and the symptom association probability at ninety-five, sorts the persistent-symptoms patient into confirmed GERD, borderline GERD, reflux hypersensitivity, or functional heartburn. The treatments for those four buckets diverge, and the divergence is what makes the sort load-bearing. Antireflux surgery is matched to the patient on objective reflux evidence and on motility. Nissen for normal motility, Toupet for hypomotility, LINX for selected anatomies with troublesome regurgitation, TIF for small hernias without severe esophagitis, and Roux-en-Y for the obese refluxer. And the refractory workup runs in order: timing, escalation, endoscopy with biopsies, reflux monitoring on or off, and the explicit search for the mimics that look like reflux but are not. Each step asks whether the diagnosis is what we think it is, because most of the time the answer is no.
That closes the GERD chapter. In the next chapter we move from the inflamed esophagus to the columnar one, the conjunction of visible salmon-colored mucosa and intestinal metaplasia that defines Barrett esophagus. The dysplasia ladder drives surveillance intervals, and endoscopic eradication therapy with resect-then-ablate handles visible disease, while CROSS and FLOT are the locally advanced regimens for the cancers that come out of it.