Barrett esophagus occurs when the squamocolumnar junction is displaced ≥1 cm proximal to the gastroesophageal junction and the squamous lining of the lower esophagus is replaced by metaplastic columnar epithelium, which predisposes to the development of esophageal adenocarcinoma. Although cardia-type epithelium has been shown to predispose to esophageal cancer, the presence of intestinalized epithelium is still considered essential for the diagnosis. Recent data show that the absolute annual risk for esophageal carcinoma in Barrett esophagus is 0.12%, which is much lower than the assumed risk of 0.5% that is the basis for current surveillance guidelines.^1-3

Esophagus, Barrett

Esophagus, columnar-lined

Ulcer, Barrett

• Male:female ratio of 4:1.
• Mean age of onset is 40 yr, with a mean age range of diagnosis of 55 to 60 yr.
• Occurs more frequently in white and Hispanic individuals than in African American individuals, with a ratio of 10 to 20:1.
• Mean prevalence of 5% to 15% in patients undergoing endoscopy (esophagogastroduodenoscopy [EGD]) for symptoms of gastroesophageal reflux disease (GERD). It is estimated that 5.6% of adults in the U.S. have Barrett esophagus. The prevalence rate may be as low as ∼1% in patients without risk factors and varies based on patient location (Western vs. non-Western countries) and number of risk factors.
• Independent risk factors include chronic reflux (>5 yr), hiatal hernia, age >50 yr, male sex, white ethnicity, smoking history, and intraabdominal obesity. A family history of at least one first-degree relative with Barrett esophagus or adenocarcinoma of the esophagus is associated with a threefold to fivefold increased risk of Barrett esophagus and esophageal adenocarcinoma.^4,5

• Metaplasia is thought to result from re-epithelialization of esophageal tissue injured as a result of chronic GERD. Activation of proliferative and inflammatory pathways likely induces a molecular phenotypic reprogramming of progenitor cells to form intestinal epithelium. Several candidate cells have been identified, including migrating gastric epithelial cells, residual embryonic cells, transitional columnar epithelium, or submucosal glands or ducts.^2,3,4,6
• Patients with Barrett esophagus tend to have more severe esophageal motility disturbances (decreased lower esophageal sphincter pressure, ineffective peristalsis) and greater esophageal acid exposure on 24-h pH monitoring. Table 1 lists some physiologic abnormalities that have been reported in Barrett patients and suggests how these abnormalities may contribute to GERD severity.
• Bile salts in the presence of acid may produce oxidative stress, induce reactive oxygen species, and alter transcriptional factor activity, all of which may induce the formation of Barrett epithelium, DNA damage, and dysplastic progression.
• Familial clustering of GERD and Barrett esophagus suggests a genetic predisposition. Early data suggest that patients who develop Barrett are genetically predisposed to a severe inflammatory response to GERD.
• One meta-analysis of genome-wide association studies identified eight risk loci associated with Barrett esophagus and esophageal adenocarcinoma. Candidate susceptibility loci include CRTC1, BARX1, and FOXP1.^7,8
• Progression from metaplasia to dysplasia to carcinoma is associated with accumulating changes in gene structure and expression, including the caudal-related homeobox family of transcription factors (CDX1 and CDX2), the embryonic transcription factor SOX2, and the tumor suppressors p16 (CDKN2A) and TP53. Biomarkers and developing technology like TissueCypher, an image analysis platform, may help better predict cancer progression compared to histology alone, though current studies show low sensitivity and specificity.
• Chromosomal deletions, duplications, and translocation are all potential early events in progression of Barrett esophagus to malignancy.⁹ A recent study suggests c-MYC and HNF4A as transcriptional drivers of metaplasia.¹⁰
• A recent Australian study suggests the presence of four pathologic criteria (loss of surface maturation, mucin depletion, nuclear enlargement, and increase of mitosis), together, was associated with greater progression from low-grade dysplasia to high-grade dysplasia or adenocarcinoma.¹¹
• Transcriptionally active high-risk human papillomavirus has been described in cohorts of patients with Barrett esophagus with dysplasia and esophageal adenocarcinoma.
• Patients with Barrett esophagus and esophageal adenocarcinoma may have a shift in the esophageal microbiome toward a gram-negative population, which may play a role in oncogenesis through the release of proinflammatory cytokines and activation of oncogenic transcriptional regulators such as nuclear factor-κB (NF-κB).^12,13

• GERD, uncomplicated
• Erosive esophagitis
• Gastritis
• Peptic ulcer disease
• Angina
• Malignancy
• Stricture or Schatzki ring

• The American Gastroenterological Association Medical Position Panel gave a weak recommendation for screening patients with multiple risk factors, including age over 50, male sex, white race, chronic GERD, hiatal hernia, elevated body mass index, and intraabdominal distribution of body fat.^14-16 An international consensus group suggested screening men over 60 with GERD symptoms for more than 10 yr. An American College of Gastroenterology (ACG) Clinical Guideline suggests a single screening endoscopy for patients with chronic GERD symptoms and ≥3 of the following risk factors: male sex, age >50, white race, tobacco use, obesity, first-degree relative with Barrett esophagus or esophageal adenocarcinoma.² An American Society for Gastrointestinal Endoscopy (ASGE) guideline stated that although there was insufficient evidence on the effectiveness of screening; if done, it should be performed for high-risk individuals, defined as patients with a family history of esophageal adenocarcinoma or Barrett esophagus, or moderate-risk patients with GERD and one of the earlier-defined risk factors.^4,17 General population screening is not currently recommended as screening using current techniques may not improve mortality rates from adenocarcinoma.^14,15
• EGD (Fig. E1, Fig. E2, Fig. E3) with biopsy is necessary for diagnosis. Ideally, this should be done via high-resolution white-light endoscopy using the Seattle protocol with biopsies of any lesions followed by four-quadrant biopsies for every 2-cm segment. If erosive esophagitis is present, patients should be treated to heal esophagitis with repeat endoscopy and biopsies 8 to 12 wk later to determine if underlying Barrett esophagus is present. ASGE and ACG recommend concurrent use of chromoendoscopy or virtual chromoendoscopy if available. Repeat endoscopy following initial normal screening endoscopy has low utility.
• Unsedated transnasal endoscopy and esophageal cytology devices (Cytosponge, EsophaCap, or Esocheck) may be acceptable alternatives to conventional endoscopy.¹⁸ Wireless esophageal capsule endoscopy may detect Barrett esophagus but with too low a sensitivity and specificity to be recommended.¹⁰ Imaging studies are not useful.
• Diagnosis requires the presence of metaplastic columnar epithelium at least 1 cm proximal to the gastroesophageal junction (Fig. E4). Longer-segment (≥3 cm) Barrett esophagus is more readily diagnosed (Fig. E5). Barrett esophagus may be described using the Prague criteria, documenting the circumferential and maximal length of involvement via a C and M score, though the Prague criteria are unreliable when Z line displacement is <1 cm.¹⁹ Biopsy of a normal Z line or a Z line with <1 cm displacement is not recommended given the potential high false-positive rate of Barrett diagnosis, which can lead to unnecessary surveillance endoscopy.²⁰ The Paris classification may be used to report associated visible lesions.
• At least two expert GI pathologists should concur if any grade of dysplasia is diagnosed.^2,14 A recent meta-analysis demonstrated a change in diagnosis in 55% of cases when reviewed by an expert pathologist, most commonly with indeterminate or low-grade dysplasia.²⁰
• Intestinal metaplasia of the gastric cardia is not Barrett esophagus and does not have the same risk for malignancy.
• Biomarkers using tissue samples and, more recently, blood tests involving whole-genome sequencing and even breath tests for volatile organic compounds, and advanced imaging techniques, such as mutational load, fluorescence in situ hybridization, autofluorescence imaging, white light postprocessing algorithms, confocal laser endomicroscopy, and optical coherence tomography (recently incorporated into volumetric laser endomicroscopy), are being evaluated to assist with diagnosis and to better understand progression of disease, prediction of response to therapy, or prognosis.
• Recent advancements also have been made in artificial intelligence. One example from the United Kingdom involves 3D reconstruction of endoscopic images with which Prague criteria C and M scores have been determined with >97% accuracy.²¹
• Screening for Helicobacter pylori infection in patients with GERD and Barrett esophagus is not recommended.

Nonpharmacologic therapy includes lifestyle modifications; elevating head of bed; and avoiding chocolate, tobacco, caffeine, mints, and certain drugs (see “Gastroesophageal Reflux Disease”).

• Proton pump inhibitors are the most effective treatment for GERD. Therapy should be titrated to control symptoms and/or to promote healing of endoscopic signs of disease.²²
• If patient is asymptomatic and incidentally found to have Barrett esophagus, once-daily proton pump inhibitors should be prescribed, as they may reduce the risk of neoplastic progression. Increasing dosage to twice daily is only recommended if breakthrough symptoms arise.¹⁴

• Chronic acid suppression is recommended to control symptoms, maintain healing, and reduce neoplastic progression. For patients with Barrett esophagus, the benefits of the use of proton-pump inhibitors are thought to outweigh the potential risks.^14,21,23,24
• Antireflux surgery may be considered for management of GERD and associated sequelae, but it has not been proven to be superior to medical therapy and has no clear role in the management of Barrett esophagus. Barrett patients continue to require endoscopic surveillance of their esophagus.
• When GERD is controlled by either medical or surgical therapy, ablation of metaplastic epithelium usually leads to replacement by normal squamous epithelium. Because only a minority of patients with Barrett esophagus progress to high-grade dysplasia or carcinoma, endoscopic eradication therapy (EET) is not recommended for the general population of patients with nondysplastic Barrett esophagus.^16,24,25
• EET is becoming the treatment of choice for low-grade dysplasia and is the treatment of choice for high-grade dysplasia and intramucosal cancer.^2,11,25-27 Radiofrequency ablation or photodynamic therapy, combined with endoscopic mucosal resection (EMR) or endoscopic submucosal dissection (ESD) of visible mucosal irregularities, should be performed in conjunction with aggressive surveillance and eradication of all remaining Barrett epithelium. Endoscopic therapy is preferred over surgical treatment in properly staged individuals, given lower morbidity and mortality rates and similar long-term survival rates. These therapies may even be considered for patients with focal intramucosal carcinoma, if properly staged (T1 SM1 or lower). Radiofrequency ablation is currently first-line, while cryotherapy (carbon dioxide, liquid nitrogen, or nitric oxide) and argon plasma coagulation are being evaluated for the complete eradication of both dysplasia and intestinal metaplasia and reduced risk for disease progression. All these options run the risk for residual or buried metaplasia. Ultimate goals of therapy include both complete eradication of dysplasia (CE-D) and complete eradication of intestinal metaplasia (CE-IM). Even with CE-IM, recurrence has been documented in 5% to 39.5% of patients. Modalities for endoscopic treatment of Barrett esophagus are summarized in Table 2. Given the high rates of recurrence of Barrett esophagus, those who have undergone successful EET should then continue with endoscopic surveillance.²
• Surgical resection is definitive therapy and may be offered for multifocal high-grade dysplasia, carcinoma that has extended into the submucosa (T1 SM2 or 3), or patients with poorly differentiated carcinomas or with evidence of lymphovascular invasion. Surgery may also be considered in patients with resistant, progressive, or recurrent disease in the face of endoscopic therapy. Mortality appears to be lower with experienced surgeons operating in high-volume centers.
• Patients with cardiovascular risk factors may be considered for low-dose aspirin therapy for chemoprevention of esophageal adenocarcinoma. NSAIDs or NSAIDs combined with statins are potentially effective at reducing the risk of esophageal adenocarcinoma but are not currently recommended for use because of the risk of adverse effects.

• The relative risk of developing esophageal adenocarcinoma for a patient with Barrett esophagus, as compared with the general population, is 11.3, a substantial drop from the relative risk of 30 or 40 estimated in earlier reports. The risk is greater in men and in patients with longer (≥3 cm) columnar-lined segments.
• The risk of progression to esophageal adenocarcinoma from untreated Barrett with low-grade dysplasia is ∼0.5% per yr, and with high-grade dysplasia ranges from 6% to 19% per yr.
• In the United States and Europe, the Prague criteria are recommended as a means of risk stratification and determination of the frequency surveillance endoscopy. However, adherence is extremely variable.^1,28,29
• Prasad et al. developed a model to predict progression of Barrett esophagus to neoplasia (“Progression in Barrett’s oesophagus score”) based on sex, smoking history, length of Barrett’s involvement, and histology.³⁰
• Frequency of monitoring is controversial. No prospective studies have proved that endoscopic surveillance is cost effective or increases life expectancy. Similarly, there are no clear guidelines for when surveillance should be stopped, although the ACG suggests stopping with less than an expected 5-yr survival or when a patient is no longer fit enough to undergo endoscopy or EET.²
• Patients with Barrett esophagus currently undergo EGD and systematic four-quadrant biopsy (“Seattle protocol”) using white light and chromoendoscopy at intervals determined by the presence and grade of dysplasia. Wide-area transepithelial sampling with computer-assisted 3D analysis (WATS-3D) may increase the detection of dysplasia. All visible mucosal abnormalities should undergo biopsy.^14,28
  1. Patients without dysplasia should have follow-up every 3 to 5 yr (≥3 cm or <3 cm Barrett involvement, respectively).²
  2. Patients indefinite for dysplasia require aggressive, intensified medical therapy and close follow-up and resampling within 6 mo.
  3. Patients with low-grade dysplasia should have aggressive antisecretory (proton pump inhibitor) therapy, repeat endoscopy within 2 to 6 mo, and either endoscopic ablation therapy or extensive mucosal sampling at 6 and then every 12 mo thereafter. Patients with high-grade dysplasia should have expert confirmation and extensive mucosal sampling. Advanced imaging techniques should be used to localize all neoplastic lesions. This includes virtual chromoendoscopy, narrow-band imaging (NBI), blue laser imaging (BLI), and i-scan. A simple and inexpensive technique is the use of acetic acid spray, which turns Barrett epithelium white, before turning neoplastic lesions red after 30 to 40 sec. High-grade dysplasia with visible mucosal irregularities should be removed by EMR or ESD, followed by mucosal ablation. Consider intensive surveillance every 3 mo for at least 1 yr and then annually thereafter. Endoscopic treatment is preferred over intensive surveillance in patients with high-grade dysplasia.³¹
  4. The ACG recommends after complete eradication that surveillance endoscopy be performed at 1 yr and 2 yr for low-grade dysplasia, and 3, 6, 12 mo and annually for high-grade dysplasia.² This is less frequent than previously recommended.
• Patients should be treated aggressively for GERD before surveillance.

• Consider EGD with biopsy in male or selected female patients with multiple risk factors who have not had previous EGD.
• Refer patients with GERD for evaluation if “red flag” symptoms are present (dysphagia, odynophagia, weight loss, vomiting, early satiety, GI bleeding, iron deficiency).
• Refer patients with biopsy-proved Barrett esophagus for surveillance.
• For those with low-grade or high-grade dysplasia, refer for ablative therapy with EMR or ESD if appropriate, followed by intensive surveillance. Esophageal resection may be considered.

Barrett Esophagus (Patient Information)

Esophageal Tumors (Related Key Topic)

Gastroesophageal Reflux Disease (Related Key Topic)