Barrett Esophagus and Esophageal Cancer: Barrett: Diagnosis, Pathogenesis, and Surveillance

0:00Welcome to Board Pearls. This is episode one of three of the Barrett Esophagus and Esophageal Cancer chapter, in the Esophageal Disorders module. This episode is diagnosing Barrett's, understanding why it turns into cancer, and deciding how often to look. The single organizing idea is that the worse the cells look under the microscope, the faster they progress toward cancer, and that speed is what sets how often you scope, so the whole surveillance scheme is really just that one relationship turned into intervals.

0:30Start with the definition, because everything else depends on getting it exactly right, and it has two parts you have to hold together. First, on the scope, you have to see salmon-colored columnar lining extending at least a centimeter above the junction of esophagus and stomach. Second, on biopsy, that lining has to show intestinal metaplasia, meaning columnar cells with goblet cells in them. Either part alone is not Barrett's. You need the visible segment and the goblet cells together.

0:59That junction of esophagus and stomach is where the diagnosis is made or lost, because Barrett's is defined relative to it, so you need to know where it is. Two landmarks mark it: the top of the gastric folds, and the bottom end of the palisade vessels running in the esophageal lining. Above that junction is the squamous-columnar junction, the Z line, which is where the columnar lining starts and what the measurement system tracks. And the one-centimeter floor isn't arbitrary. Salmon lining with goblet cells less than a centimeter above the junction is just intestinal metaplasia of the cardia, which is common in ordinary people and carries no measurable cancer risk. That has a direct practical consequence: you don't biopsy an irregular Z line with little tongues under a centimeter, because there's no Barrett's down there to find, and calling those goblet cells Barrett's is a mistake.

1:51The goblet-cell requirement is the piece that gets tested for precision, because the definitions differ across the Atlantic. The British framework accepts non-goblet columnar lining above the junction as Barrett's, arguing that it's also a reflux-driven metaplasia. The American framework requires goblet cells, and the reason is concrete: the cancer-risk numbers the whole surveillance scheme relies on came from patients defined by goblet-cell metaplasia. For exams, use the American definition: salmon lining over a centimeter above the junction plus intestinal metaplasia with goblet cells.

2:26Once you have the diagnosis, you measure the segment two ways: the circumferential extent, how tall the fully wrapping-around portion is, and the maximum extent, the longest tongue reaching up above that. So a segment described as C-three M-six wraps all the way around for three centimeters with a tongue reaching six. Long-segment is three centimeters or more, short-segment is under three, and ultra-short is under one, and that last group is genuinely hard to measure reliably, which is part of why it's treated separately.

3:00You sample the segment with a defined protocol, and the logic behind it is that dysplasia is patchy and doesn't announce itself visually, so you can't just biopsy what looks abnormal. You start by inspecting the whole segment carefully under good lighting with dye or its electronic equivalent, and here's a rule worth its own emphasis: any visible nodule or irregularity gets resected, not forceps-biopsied, because the resection is also how you stage it, which is an episode-two point. Then the flat rest of the segment gets four-quadrant biopsies every two centimeters when there's no dysplasia, tightened to every centimeter once dysplasia is known to be present. The protocol still only samples a fraction of the surface, which is why targeting visible lesions and using dye add yield, and why a careful exam takes real time, more than a minute per centimeter of segment; a rushed look measurably misses more cancer.

3:52A couple of adjuncts show up by name. One is a wide-area brushing that abrades the whole segment and uses computer analysis to flag possible metaplasia or dysplasia, which adds to the yield of standard biopsies and is supported as a supplement, not a replacement. The others are the non-endoscopic capsule devices, a swallowed sponge on a string or a small balloon dragged through the lower esophagus, which collect cells for molecular markers of metaplasia. Their accuracy is reasonable but not perfect, and the thing to recognize is the setting: a question about screening a whole population or a primary-care setting is pointing at one of these devices rather than scoping everyone.

4:33Which raises who to screen at all, because the general population's risk is too low to justify scoping everybody. The approach is a single screening endoscopy for someone with chronic reflux, meaning weekly heartburn or regurgitation for at least five years, plus several additional risk factors: male sex, age fifty or older, white race, central obesity, smoking, and a first-degree relative with Barrett's or esophageal adenocarcinoma. Central obesity is judged by waist size rather than body mass index, because belly fat drives reflux and Barrett's independent of overall weight. And it's worth knowing the limits, because they're increasingly tested: a large share of people who present with esophageal adenocarcinoma never had heartburn and wouldn't have qualified for screening at all, and screening hasn't shown a survival benefit in randomized data, which is why the field is drifting toward broader, cheaper, non-endoscopic tools. One pointer from the reflux chapter folds in here: severe erosive esophagitis needs a repeat scope after healing precisely because the inflammation can hide underlying Barrett's.

5:37Now why this tissue becomes cancer, because the risk is graded by where the cells sit on a sequence. Barrett's is the price the lower esophagus pays for chronic acid and bile injury: the reflux destroys the squamous lining, and the wound heals back not as squamous tissue but as intestinal-type lining with goblet cells, because the ongoing reflux reprograms how the healing cells develop. From there the cells can move stepwise, from metaplasia, to low-grade dysplasia, to high-grade dysplasia, to cancer confined to the mucosa, to invasive cancer, driven by acid and bile causing DNA damage and the loss of tumor-suppressor genes. Worth knowing conceptually: a large share of these cancers take a fast route where an early damaged clone doubles its whole genome and then tolerates rapid further mutation, and that fast route can outrun the usual multi-year surveillance interval, which is part of why surveillance sometimes fails.

6:32Dysplasia is the microscope read that grades the risk, and the key clinical fact about it is how unreliable it is. Low-grade dysplasia keeps some surface maturation while high-grade loses it, but pathologists disagree with each other constantly, especially on low-grade, and the same biopsy can be read three different ways by three pathologists. When community low-grade calls are re-reviewed by an expert, most don't survive, with the majority downgraded to no dysplasia at all. So the rule, which is tested relentlessly, is absolute: any dysplasia call has to be confirmed by a second pathologist with gastrointestinal expertise before you change management, because ordering ablation on an unconfirmed community low-grade read is acting on a diagnosis that's wrong more often than not.

7:19The progression risk climbs with the grade, and that gradient is the whole point, even without memorizing the exact figures. Non-dysplastic Barrett rarely progresses. Confirmed low-grade dysplasia progresses meaningfully faster, and the better the confirmation, the more its behavior approaches high-grade. Untreated high-grade dysplasia progresses to cancer quickly, and a real fraction of high-grade biopsies already hide an occult cancer at diagnosis. That same gradient continues into early cancer, and the principle to carry is depth: a cancer confined to the mucosa carries very little risk of lymph node spread and stays an endoscopic problem, while once it invades into the submucosa the node risk climbs steeply and it becomes a surgical one. That mucosa-versus-submucosa line is the hinge we use in the next episodes to decide who stays endoscopic and who goes to surgery.

8:11A few things push a given patient's risk up: male sex, smoking, each added centimeter of segment length, and confirmed low-grade dysplasia. Tissue markers are increasingly used, and abnormal p53 staining is the single most validated one, consistently linked to progression, with a commercial multi-marker score available to risk-stratify non-dysplastic patients. And one epidemiologic reality explains why screening alone won't fix this: roughly half of people diagnosed with esophageal adenocarcinoma have no detectable Barrett's left at diagnosis, fitting the idea that the cancer arose from a metaplastic field that it then overgrew, so the visible Barrett's is really a marker of the field and the window of opportunity.

8:54On prevention, the calibrated answer is once-daily acid suppression for every Barrett patient. Higher-dose suppression and aspirin showed a protective signal in a large trial but not strong enough to make them mandatory for everyone, so they're individualized considerations. And antireflux surgery is not done for cancer prevention, because it doesn't beat medication for that purpose.

9:17Now the surveillance itself, which is that risk gradient turned into intervals, higher grade meaning shorter interval. Non-dysplastic Barrett gets scoped every five years for short segments and every three years for long ones, the split reflecting the higher risk of the longer segment. But the first surveillance after a new diagnosis is usually at one year regardless, and the reasoning is that the highest-yield moment for catching dysplasia or cancer that was missed on the first exam is that first follow-up, since a striking share of cancers turn up within the first year. Once that first pass is clean, the patient settles into the three- or five-year rhythm.

9:57Indefinite for dysplasia is a real category, and its management follows from what it means: the pathologist can't tell reactive change from acid injury apart from true dysplasia, because inflamed mucosa looks atypical. So you optimize acid suppression with twice-daily dosing for three to six months and repeat the biopsies, and most indefinite reads resolve to non-dysplastic under good acid control. If it persists, you tighten to a twelve-month interval. So a vignette with mild esophagitis and an indefinite biopsy is asking for twice-daily suppression and a repeat scope, not ablation.

10:31Low-grade dysplasia is the real decision point and the most-tested one. The first move after a community low-grade call is expert confirmation, because most don't survive it. Once an expert confirms it, there are two acceptable paths: endoscopic eradication, meaning resect any visible lesion and then ablate the flat field, or continued surveillance every twelve months. The data favor eradication in patients with a reasonable life expectancy, because ablation lowers progression to high-grade and cancer, while surveillance is reasonable for older patients, those with limited life expectancy, or those who decline. And the first repeat after a low-grade call is usually at six months on twice-daily suppression, because low-grade that vanishes on a clean acid-controlled exam was probably overcalled; it's the low-grade that persists across exams that earns ablation.

11:22High-grade dysplasia gets endoscopic eradication in essentially everyone with a reasonable life expectancy, for two reasons: it progresses to invasive cancer quickly, and a meaningful fraction already harbor an occult cancer. The sequence is a careful re-exam, resection of any visible nodule, both to treat and to stage it, and then ablation of the remaining flat high-grade tissue. Esophagectomy, the old answer, is now reserved for disease that can't be cleanly resected endoscopically or for endoscopic failures. So an eighty-year-old with a six-centimeter high-grade segment confirmed by an expert and no visible nodule does not need an esophagectomy; the answer is endoscopic eradication. Cancer confined to the mucosa follows the same endoscopic-first logic, resect the visible lesion and ablate the flat segment, and it's the move into the submucosa that brings surgery into the conversation, which is an episode-two thread.

12:16Two final ideas at the edges. First, when to stop: surveillance only helps someone who'll live long enough to develop and treat a cancer caught on a future scope, so you stop in the elderly and the seriously comorbid, somewhere in the seventies to eighties for healthy patients and earlier with significant illness. Second, the humility point that's increasingly rewarded: a large trial comparing regular surveillance against scoping only for symptoms found no improvement in survival or in the stage at which cancers were caught, so the surveillance dogma has real limits.

12:50So the whole arc: Barrett's is a visible columnar segment at least a centimeter above the junction plus intestinal metaplasia with goblet cells on biopsy. You measure it by circumferential and maximum extent, and you sample it with a careful, unhurried protocol, resecting visible nodules rather than biopsying them. It's reflux-driven metaplasia that can climb a dysplasia sequence toward cancer, and the grade sets the interval: non-dysplastic short segments every five years, long ones every three, indefinite and surveillance-path low-grade every twelve months after confirming with acid suppression, and confirmed low-grade in fit patients and essentially all high-grade going to endoscopic eradication. And every dysplasia call needs a second gastrointestinal pathologist before anything changes.

13:40The next two episodes pick up from here. Episode two takes the patient who needs endoscopic eradication through the techniques in their fixed order: resect the visible disease first, because that specimen is the staging test, then ablate the flat field. Episode three takes the cancer side, starting right at that mucosa-versus-submucosa line, where the node risk climbs and the patient stops being an endoscopic problem and starts being a surgical one.

14:07For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode one of three of chapter four, and I'll see you in the next one.