Background: Endoscopic surveillance of Barrett's esophagus is recommended to detect dysplasia or cancer at an early and potentially treatable stage. However, little is known about the clinical practice patterns for endoscopic surveillance in the United States. Methods: A questionnaire regarding surveillance intervals, techniques and management approaches for patients with Barrett's esophagus was mailed to 1000 randomly selected members of the Clinical Practice Section of the American Gastroenterological Association. Results: The response rate was 455 of 1000 (45%). Not all respondents answered all questions. Seventy-nine percent of respondents were in community practices, and 21% were in academic practices. Nearly all (96%) performed endoscopic surveillance, but it was practiced more commonly in the community (334 of 341 [98%]) than in the academic setting (83 of 93 [89%], p < 0.001). For patients without dysplasia, endoscopic surveillance was most commonly performed every 2 years (264 of 415 [64%]). Patients with low-grade dysplasia usually had surveillance endoscopy at B-month intervals (215 of 413 [52%]), whereas those with high-grade dysplasia most commonly had endoscopy every 3 months (201 of 404 [50%]). These surveillance patterns did not differ between the academic and community groups. Random biopsies were performed by 93 of 403 (23%), 4-quadrant biopsies by 310 (77%). Most physicians (83%) used standard capacity forceps. Brush cytology was done uncommonly (69 of 414 [17%]). The most common indications for esophagectomy were high-grade dysplasia by 82% and cancer by 83%. Ablation therapy was performed for Barrett's esophagus without dysplasia by 3.5%, Barrett's with dysplasia by 20%, and cancer by 8%. Conclusions: Surveillance for Barrett's esophagus is widely practiced in the United States but there is considerable variation in interval and technique. A clearer consensus on endoscopic surveillance is warranted to optimize care of patients with Barrett's esophagus.
