Temporal trends in new and recurrent esophageal strictures in Department of Veterans Affairs

Abstract

Background: Anecdotal experience suggests that gastroenterologists are encountering fewer patients with esophageal strictures; however, the temporal trends of esophageal strictures have not been well examined in large studies.

Methods: We examined the incidence rates of new esophageal strictures as well as esophageal dilations among patients presenting to VA facilities between 1998 and 2003 as a proportion of the annual frequency of all endoscopic procedures and/or imaging procedures. Recurrent strictures were defined as those recorded at least 3 months after the index date of a new stricture or the date of the preceding recurrent stricture. Stricture was defined only in the presence of an upper endoscopy or upper gastrointestinal (UGI) contrast imaging within 30 days before or after the date of diagnosis. This algorithm was validated by manually reviewing the medical records of 180 patients; the positive predictive value was approximately 84%. The risk of recurrent strictures was examined using Cox proportional hazards model for multiple failures.

Results: There were 19,157 patients identified between fiscal year (FY) 1998 and FY 2003 with new esophageal strictures. New strictures comprised a constant annual proportion of total EGD and UGI imaging studies. New strictures declined by 11.6% as a proportion of all upper endoscopies (not including UGI imaging). However, esophageal dilations associated with new stricture diagnoses declined among patients with new esophageal strictures, 16.8% had at least one episode of recurrence. The 1-yr incidence rate of recurrent strictures declined significantly (-36.0%) from 13.9 per 100 person-years for patients diagnosed with new strictures in 1998 to 8.9 per 100 person-years in patients diagnosed with new strictures in 2003. In the full multivariable model, new strictures diagnoses during 2001-2003 were associated with a reduced risk of having recurrent strictures (19% in the first year and 13% throughout follow-up) independent of age, gender, race, geographic region, or the presence of Barrett's esophagus.

Conclusion: The incidence of recurrent strictures has declined between 1998 and 2003 by approximately one-third. Although there has been only a small decline in the incidence of new esophageal strictures among patients undergoing endoscopy, the use of esophageal dilations has declined considerably among these patients. The role of proton pump inhibitors needs to be examined.