The Clinical Significance of Ultrashort Barrett’s: Persistence and Progression
Samuel Schueler, MD1 & Romy Chamoun, MD2
1Assistant Professor of Medicine, Gastroenterology & Liver Diseases; 2GI Fellow, George Washington University School of Medicine & Health Sciences, Washington, DC
This article reviews Skef W, Haydel J, Rao A, et al. High risk of persistence and risk of dysplasia after diagnosis of ultrashort Barrett’s esophagus. Am J Gastroenterol. 2025; 120(11): 2520-2528.
Access the article through The American Journal of Gastroenterology
Correspondence to Romy Chamoun, MD. Associate Editor. Email:EBGI@gi.org
Keywords: ultrashort Barrett’s esophagus, esophagus, dysplasia
STRUCTURED ABSTRACT
Question: What is the prevalence of ultrashort Barrett’s esophagus (USBE; <1 cm), how often does it persist on follow-up, and what is the associated risk of dysplasia or neoplasia compared with Barrett’s esophagus ≥1 cm?
Design: Retrospective cohort study evaluating long-term endoscopic and histologic outcomes over 32 years.
Setting: Michael E. DeBakey Veterans Affairs Medical Center (Houston, Texas), with follow-up through April 2023.
Patients: Adults diagnosed with Barrett’s esophagus between 1990 and 2022 with biopsy-confirmed intestinal metaplasia and at least 1 follow-up endoscopy.
• USBE: <1 cm segment with intestinal metaplasia.
• Comparison group: BE ≥1 cm by Prague criteria.
Intervention / Exposure: Diagnosis of ultrashort Barrett’s esophagus (<1 cm) with intestinal metaplasia on index endoscopy.
Outcome Measures: Primary—Persistence of BE or intestinal metaplasia on follow-up endoscopy. Secondary—Development of definite dysplasia or neoplasia.
Data Analysis: Descriptive statistics were used for baseline characteristics, comparing continuous variables with Student’s t-tests and categorical variables with Pearson’s χ² or Fisher’s exact tests. Multivariable logistic regression—including variables significant on univariate analysis and known BE risk factors—identified predictors of persistent BE after USBE. Cumulative incidence, 95% confidence intervals (CIs), and incidence rates per 1,000 person-years were calculated for any and definite dysplasia. Follow-up time extended from index endoscopy to dysplasia diagnosis, last endoscopy, or death. Differences in incidence rates were evaluated with log-rank tests. Cox proportional hazards models estimated hazard ratios comparing persistent USBE to BE ≥1 cm. Analyses were performed using Stata 15.1, with P< 0.05 considered statistically significant.
Funding: Supported by U.S. Veterans Affairs institutional resources; no external funding.
Results: Of 739 patients with Barrett’s esophagus, 167 (22.6%) had USBE on index endoscopy; 86 patients (11.6% of the total cohort) had persistent BE on follow-up. Clinical characteristics—including age, sex, BMI, reflux risk factors, hiatal hernia, medication use, and biopsy volume—did not differ between persistent USBE and BE ≥1 cm. Persistence was significantly associated with race/ethnicity: non-Hispanic White (adjusted odds ratio [aOR] 3.85; 95% CI 1.37–10.8) and Hispanic patients (aOR 4.78; 95% CI 1.18–19.4) had higher odds of persistent BE compared with non-Hispanic Black patients. Among those with persistent USBE, 41.9% remained <1 cm and the remainder developed short-segment BE. Cumulative incidence of any dysplasia (indefinite, LGD, HGD, or EAC) was 17.4% in persistent USBE vs 31.2% in BE ≥1 cm; cumulative incidence of definite dysplasia/neoplasia was 11.6% and 17.8%, respectively. Incidence rates of all dysplasia/neoplasia were lower in persistent USBE (30.2 vs 66.2 per 1,000 person-years; HR 0.54; 95% CI 0.32–0.91), while incidence rates of definite dysplasia were not significantly different (19.5 vs 33.8 per 1,000 person-years; HR 0.67; 95% CI 0.35–1.29). No demographic or clinical factors predicted dysplasia among patients with persistent USBE.
COMMENTARY
Why Is This Important?
Ultrashort Barrett’s esophagus (USBE), defined as <1 cm of columnar-lined mucosa with intestinal metaplasia, is a common but controversial finding in clinical practice. Many gastroenterologists encounter patients with subtle tongues of salmon-colored mucosa or an irregular Z-line that is biopsied, but guidelines disagree on whether segments <1 cm should be labeled as BE or followed at all. Some societies exclude USBE due to measurement variability and presumed minimal cancer risk. This study provides long-term outcome data to help clinicians understand whether USBE is clinically meaningful, challenging the assumption that these short segments are benign and transient.
Key Study Findings
Persistence was notable—more than half of USBE cases had ongoing BE or intestinal metaplasia on follow-up, suggesting that many of these segments represent true metaplasia rather than sampling artifact or inflammation.
Race and ethnicity significantly influenced outcomes, with White and Hispanic patients having a markedly higher risk of persistence compared with Black patients, underscoring potential biological or environmental influences. Importantly, persistent USBE was associated with an 11.6% cumulative dysplasia risk, and the dysplasia incidence rate did not differ significantly from that of ≥1 cm BE. These findings collectively suggest that USBE may carry more prognostic weight than previously believed.
Caution
Several limitations must temper the interpretation of these findings. The study population consisted primarily of older male veterans, which may restrict applicability to women or broader community populations. Measurement of <1 cm mucosa is inherently subjective and susceptible to interobserver variability, particularly across decades of endoscopy. Although the dysplasia risk appears meaningful, the absolute number of dysplasia events in the USBE subgroup remains modest, resulting in wide confidence intervals and limiting precision. Retrospective data collection may also introduce bias, particularly regarding biopsy practices and follow-up intervals. Therefore, while USBE should not be dismissed, these findings should be applied thoughtfully and individualized.
My Practice
In my practice, I do not aim to biopsy an irregular Z line. Biopsies are obtained when there is at least 1 centimeter of proximal displacement of columnar mucosa from the top of the gastric folds. This approach is consistent with guidelines from the American College of Gastroenterology and most other national and international guidelines. Studies show that patients with an irregular Z line rarely progress to high-grade dysplasia (HGD) or esophageal adenocarcinoma (EAC) during surveillance. These potentially low-yield endoscopic surveillance studies may lead to unnecessary burden for patients and increased health care cost. However, the findings of this study highlight that ultrashort Barrett’s esophagus may carry a measurable risk of dysplasia, particularly in certain demographic groups. At present, in borderline cases of an irregular Z line vs > 1 cm of abnormal appearing columnar mucosa in the distal esophagus, I tend to err on the side of caution and obtain a biopsy or multiple biopsies encompassing the proximal tip of columnar mucosa and normal appearing esophageal mucosa in an effort to only capture esophageal (and not gastric) mucosa. The findings of this study may support this approach, as there is not only benefit in capturing subtle > 1 centimeter tongues of columnar mucosa, but also potential benefit in capturing columnar mucosa that is borderline in length of proximal extent given the rate of definite dysplasia or neoplasia (11.6%) found in surveillance of patients with irregular Z lines in the study. That said, more studies are needed to demonstrate the yield of endoscopic surveillance in patients with irregular Z lines before I routinely biopsy this finding to assess for possible USBE.
Future Research
Future work should include large prospective multicenter cohorts that include women and more racially diverse populations to validate these findings beyond the VA system. Studies are needed to identify endoscopic factors that predict persistence, such as subtle mucosal features or advanced imaging patterns. Research should clarify optimal surveillance intervals for persistent USBE and determine whether molecular, genomic, or biomarker-based tools can better stratify dysplasia risk. Ultimately, comparative effectiveness trials evaluating selective surveillance versus no surveillance strategies in USBE could help refine guideline recommendations and determine the true clinical significance of these ultrashort segments.