Salmon-colored mucosa in the esophagus is not always Barrett’s esophagus. The pinkish-orange tissue that catches a gastroenterologist’s eye during an endoscopy can have several explanations, and only a fraction of cases turn out to be true Barrett’s when biopsies come back from the lab. In one study tracking over 7,300 patients who underwent upper endoscopy, salmon-colored mucosa was flagged in about 5% of them, yet histology confirmed Barrett’s esophagus in fewer than a third of those flagged cases. The gap between what looks suspicious on camera and what the microscope reveals is wider than most people realize, and it matters because a Barrett’s diagnosis triggers years of surveillance, anxiety, and cost.
What the Salmon Color Actually Means
The normal lining of the esophagus is made of squamous epithelium, which looks pale and whitish during endoscopy. The stomach, by contrast, is lined with columnar epithelium, which appears reddish-orange or salmon-colored. When tissue that looks like stomach lining shows up where esophageal lining should be, it raises the possibility that the normal squamous cells have been replaced by columnar cells. Barrett’s esophagus is one version of this replacement, specifically a metaplastic change characterized by its salmon-colored lining and the presence of columnar epithelium with goblet cells.1PubMed Central. Diagnosis and Management of Barrett’s Esophagus But columnar tissue in the esophagus is not automatically Barrett’s. It can be gastric-type columnar tissue without any of the intestinal features that define the condition, or it can be something else entirely.
The process behind true Barrett’s involves chronic exposure to stomach acid and bile, which damages the squamous lining. When the body repairs the damage, stem or progenitor cells sometimes differentiate into columnar cell types instead of restoring the original squamous cells.2PubMed Central. Origins of Metaplasia in Barrett’s Esophagus: Is this an Esophageal Stem or Progenitor Cell Disease? This columnar replacement can take several forms, and only the form containing goblet cells (intestinal metaplasia) is considered Barrett’s under American guidelines. That distinction is invisible during the endoscopy itself. The tissue looks salmon-colored regardless of which type of columnar cell has taken up residence, so the camera alone cannot make the diagnosis.
The Gap Between What Is Seen and What Is Confirmed
The numbers are striking. In a decade-long review at a single university medical center, roughly a quarter of patients suspected of having Barrett’s during endoscopy were confirmed by biopsy.3PubMed. Increased Barrett’s esophagus for the decade between 1991 and 2000 at a single university medical center A separate study looking at ethnic and gender differences found salmon-colored mucosa in 391 patients out of more than 7,300 endoscopies, but only 111 of those 391 had Barrett’s confirmed on histology.4PubMed Central. The frequency of histologically confirmed Barrett’s esophagus varies by the combination of ethnicity and gender That is a confirmation rate of about 28%, meaning roughly seven out of ten patients whose esophagus looked suspicious turned out not to have Barrett’s.
The confirmation rate also varied dramatically by demographics. Non-Hispanic white men had the highest rate of histologic confirmation at about 42%, while African American women had the lowest at around 12%.4PubMed Central. The frequency of histologically confirmed Barrett’s esophagus varies by the combination of ethnicity and gender This disparity means that when an endoscopist sees salmon-colored mucosa, the prior probability of true Barrett’s depends in part on who the patient is. A finding that would be confirmed histologically almost half the time in one group might be confirmed barely one time in ten in another.
The Irregular Z-Line Problem
One of the most common reasons salmon-colored mucosa gets flagged incorrectly is the irregular Z-line. The Z-line is the visible border where the pale esophageal lining meets the pinkish stomach lining. In many people, this junction is not a tidy, straight ring; it zigzags, sending little tongues of salmon-colored tissue up into the esophageal zone. These short tongues look a lot like the earliest stages of Barrett’s, and patients who have them are often mislabeled, with real downstream consequences including higher healthcare costs and reduced quality of life.5PubMed Central. Irregular Z-Line: To Biopsy or Not to Biopsy?
A systematic review and meta-analysis pooling data from multiple studies found that among patients with an irregular Z-line, roughly 29% had intestinal metaplasia on biopsy. Among those with intestinal metaplasia, dysplasia was present in about 6%, and esophageal adenocarcinoma in about 1.5%.6Clinical Endoscopy. Prevalence of intestinal metaplasia, dysplasia, and esophageal adenocarcinoma in patients with irregular Z-line: a systematic review and meta-analysis Those numbers are not trivial, but they also mean that more than 70% of irregular Z-line patients biopsied for intestinal metaplasia did not have it. The question of whether to routinely biopsy an irregular Z-line is still debated, because the risk of missing something real has to be balanced against the cost and consequences of mislabeling a normal anatomic variant.
Inlet Patches, the Other Salmon-Colored Mimic
There is another condition that produces salmon-colored patches in the esophagus and has nothing to do with Barrett’s: heterotopic gastric mucosa, commonly called an inlet patch. These are islands of stomach-type tissue that appear in the upper esophagus, usually just below the upper esophageal sphincter, rather than in the lower esophagus where Barrett’s occurs.7PubMed Central. Clinical significance of heterotopic gastric mucosal patch of the proximal esophagus They are thought to be congenital, meaning the tissue was there from birth, rather than being a response to acid reflux damage.
Inlet patches are described as salmon-colored and velvety, which makes them visually similar to Barrett’s mucosa.8PubMed Central. Esophageal Stricture: An Uncommon Complication of Cervical Inlet Patch Their location in the proximal (upper) esophagus is the main clue that they are not Barrett’s, since Barrett’s occurs in the distal (lower) esophagus near the stomach. But in practice, during a busy endoscopy session, a flash of salmon-colored tissue can raise alarm before the endoscopist has mentally registered the exact location. Inlet patches are considered rare but are probably underdiagnosed because many endoscopists do not specifically look for them when withdrawing the scope through the upper esophagus. Most inlet patches are harmless, though in uncommon cases they can cause local complications like stricturing.
Why Short Segments Are So Hard to Read
Barrett’s esophagus comes in different lengths. Long-segment Barrett’s, where multiple centimeters of the esophageal lining have been replaced, is relatively easy to spot and measure consistently. Short-segment Barrett’s, typically under a centimeter, is where diagnostic reliability falls apart. A multinational study measured how well endoscopists agreed on what they were seeing when the Barrett’s segment was short versus long. For segments of one centimeter or more, agreement between endoscopists was excellent. For segments under one centimeter, agreement was poor to the point of being nearly random.9PubMed Central. Interobserver reliability in the endoscopic diagnosis and grading of Barrett’s esophagus: an Asian multinational study In a separate validation study, shorter Barrett’s segments of five centimeters or less also showed lower agreement compared with longer segments.10PubMed. Validation of the Prague C&M classification of Barrett’s esophagus in clinical practice
This means the cases that are hardest to call endoscopically are precisely the ones most likely to be confused with an irregular Z-line or other benign variation. The margin between “a tiny tongue of Barrett’s” and “a normal anatomic variant” is often a matter of millimeters, and even experienced endoscopists cannot agree on which side of the line it falls on. When two experts looking at the same images cannot agree, it is not surprising that overdiagnosis occurs in everyday clinical practice.
Pathologists Disagree Too
Even after biopsies are taken, the story does not become straightforward. There is worldwide controversy over the definition of Barrett’s esophagus at the microscopic level, particularly around whether goblet cells must be present for the diagnosis.11PubMed Central. Barrett’s Esophagus: A Comprehensive and Contemporary Review for Pathologists American guidelines require goblet cells. British and Japanese guidelines do not, accepting any columnar metaplasia as Barrett’s. This means the same biopsy slide could be read as Barrett’s in London and as not-Barrett’s in New York.
Beyond the basic diagnosis, pathologists struggle to agree on the grading of dysplasia, which is the cellular abnormality that signals cancer risk. A study comparing pathologists in the United States and Europe found only slight agreement on low-grade dysplasia, with a kappa value of 0.11, essentially no better than chance. Agreement improved for high-grade dysplasia but was still only moderate overall.12Gastroenterology. Discordance Among Pathologists in the United States and Europe in Diagnosis of Low-Grade Dysplasia for Patients With Barrett’s Esophagus When pathologists were confident in their calls, agreement jumped substantially, but confidence was not the norm. The practical takeaway is that a single pathologist’s reading of a Barrett’s biopsy, especially for low-grade dysplasia, should be treated with caution. Many centers now require a second expert pathologist to confirm any dysplasia diagnosis before treatment decisions are made.
Gastric Metaplasia Versus Intestinal Metaplasia
When columnar cells replace the normal esophageal lining, the replacement can take different forms. Gastric metaplasia means the new cells resemble stomach lining. Intestinal metaplasia means the new cells have features of intestinal tissue, including the goblet cells that define Barrett’s under American criteria. The distinction matters for cancer risk. Chronic acid and bile exposure to the lower esophagus can lead to intestinal metaplasia, which is closely linked to the development of esophageal adenocarcinoma.13PubMed Central. From reflux esophagitis to Barrett’s esophagus and esophageal adenocarcinoma
A study analyzing individual-level data on disease progression found that patients with only gastric metaplasia progressed to cancer at significantly lower rates than patients with intestinal metaplasia.14PubMed. Understanding the malignant potential of gastric metaplasia of the oesophagus and its relevance to Barrett’s oesophagus surveillance: individual-level data analysis Patients who had gastric metaplasia on some biopsies and intestinal metaplasia on others fell somewhere in between, but still progressed less than those with consistent intestinal metaplasia. This has real implications: if you are told you have salmon-colored mucosa and the biopsy shows gastric-type columnar cells but no goblet cells, the risk profile is not the same as confirmed Barrett’s with intestinal metaplasia.
Biomarkers That Help Resolve Ambiguous Cases
When the microscope alone leaves room for doubt, special stains and molecular markers can sharpen the picture. One protein that has proven useful is CDX2, a transcription factor normally active in intestinal cells. In one study, CDX2 staining was present in about 88% of Barrett’s biopsies, while being completely absent in normal esophageal and gastric cardia tissue.15PubMed. Cytokeratin and CDX-2 expression in Barrett’s esophagus Another study confirmed that CDX2 and a protein called villin were positive in all cases that had definite goblet cells on standard staining, making them sensitive markers for catching intestinal metaplasia early. Both markers were negative in control samples of normal gastric mucosa.16American Journal of Clinical Pathology. CDX2 and Villin Are Useful Markers of Intestinal Metaplasia in the Diagnosis of Barrett Esophagus
Cytokeratin patterns also help. CK7, when it stains fully through the tissue thickness, was present in 96% of biopsies with intestinal metaplasia but only 58% of adjacent cardia-type tissue and none of fundic-type tissue.16American Journal of Clinical Pathology. CDX2 and Villin Are Useful Markers of Intestinal Metaplasia in the Diagnosis of Barrett Esophagus These markers are not used in every clinical lab for every biopsy, but when the standard staining leaves a pathologist uncertain about whether goblet cells are present, immunohistochemistry can break the tie. The research trajectory here suggests that molecular tools will play a growing role in reducing the ambiguity that salmon-colored mucosa creates.
The Real Costs of Getting It Wrong
Overdiagnosis of Barrett’s esophagus is not just an academic problem. It carries concrete consequences for patients. A diagnosis means enrollment in an endoscopic surveillance program, with repeat procedures every few years for the rest of a patient’s life. It can affect insurance eligibility and premiums. And the psychological weight of being told your esophagus harbors a pre-cancerous condition is significant, even when the actual cancer risk is low. Barrett’s esophagus is frequently overdiagnosed in clinical practice, with implications for costs, insurability, and psychological stress.17Gastrointestinal Endoscopy. Barrett’s esophagus is frequently overdiagnosed in clinical practice: results of the Barrett’s Esophagus Endoscopic Revision (BEER) study
The financial impact has been modeled for cases where low-grade dysplasia is over-called. The estimated excess cost for every patient in the United States who is incorrectly diagnosed with low-grade dysplasia in Barrett’s runs between roughly $3,100 and $8,000, driven mostly by repeat endoscopies, pathology review, and ablation procedures that would not have been needed.18PubMed. Healthcare Cost of Over-Diagnosis of Low-Grade Dysplasia in Barrett’s Esophagus Multiply that across the number of over-diagnosed patients nationally, and the aggregate cost becomes substantial. This is part of why the push for expert second opinions on dysplasia readings has become standard practice at most academic centers.
A Name That Does Not Quite Fit
The condition’s eponym has its own peculiar history. Norman Barrett, the British surgeon whose name attaches to the diagnosis, was far from the first to describe columnar-lined esophagus. Descriptions go back to at least 1906, and the key conceptual work was done by other investigators, particularly Philip Allison and his colleague Johnstone, who preceded Barrett’s widely credited 1957 presentation. A historical review argued that Barrett contributed relatively little to the core understanding of the condition compared with his predecessors and proposed abandoning the eponym in favor of the descriptive term “columnar-lined esophagus.”19PubMed. Columnar-lined esophagus: time to drop the eponym of “Barrett”: Historical review The suggestion has not caught on clinically, and “Barrett’s esophagus” remains the universal shorthand. But it is worth knowing that the name carries more historical accident than scientific precision, which is oddly fitting for a diagnosis where the visual appearance and the pathologic reality so often fail to line up.
What to Do If You Are Told You Have Salmon-Colored Mucosa
If an endoscopy report mentions salmon-colored mucosa, the first thing to understand is that this finding alone does not mean you have Barrett’s. Biopsies are necessary, and the pathology results determine the actual diagnosis. If biopsies show columnar tissue without goblet cells, you may have gastric-type metaplasia, which carries a lower cancer risk than intestinal metaplasia. If biopsies show intestinal metaplasia with goblet cells, that is Barrett’s under American criteria and warrants a surveillance plan. If biopsies show no metaplasia at all, you may simply have an irregular Z-line, an anatomic variant that is not a disease.
For patients who receive a Barrett’s diagnosis, the length of the affected segment and the presence or absence of dysplasia are the two factors that drive management decisions. Longer segments carry more risk than shorter ones. Any dysplasia reading, especially low-grade, deserves confirmation by a second pathologist with expertise in gastrointestinal pathology, given how poor inter-observer agreement is at that end of the spectrum. Being proactive about understanding exactly what was seen, what the biopsies showed, and what grade of cellular change was found puts you in a much better position to have an informed conversation about whether surveillance, treatment, or reassurance is the right next step.