Abstract

Introduction: Reflux esophagitis (RE) traditionally has been assumed to start when acid and pepsin destroy esophageal squamous cells at the luminal surface, inducing an acute, granulocyte-predominant inflammatory response. However, recent studies in animal RE models have challenged this traditional notion, suggesting that refluxed gastric juice does not destroy squamous cells directly, but instead stimulates them to secrete chemokines, which attract lymphocytes that mediate tissue damage. Patients typically have months to years of GERD symptoms before seeking medical attention, and so the early histologic changes of RE have not been evaluated prospectively in humans. To elucidate those changes, we induced acute RE by discontinuing PPIs in patients with healed RE. Methods: Patients with a prior endoscopy showing LA grade C RE were treated with PPIs BID for at least 1 month before a baseline endoscopy to document healing. PPIs then were stopped and endoscopy repeated 1 and 2 weeks later. At each endoscopy, 4 biopsies were taken from nonulcerated mucosa in the distal esophagus for histology and immunohistochemical staining for lymphocyte markers (CD3, CD20). Esophageal pH monitoring and GERD-HRQL symptom scoring were performed at baseline and 2 weeks. Biopsies were scored for type and degree of inflammation, for basal and papillary hyperplasia, and for spongiosis (dilated intercellular spaces) on a scale of 0 to 3. Results: 9 patients (8 men; mean age 67 years) had mean esophageal acid exposure (total % time pH<4) that increased from 6.6% at baseline to 28.4% at 2 weeks (p=.002), and mean GERD-HRQL scores that increased from 8.3 to 13.4 (p= .03). All patients developed endoscopic RE by 2 weeks (2 LA-A, 3 LA-B, 4 LA-C). Lymphocytes were by far the predominant type of inflammatory cell. Lymphocytes, basal and papillary hyperplasia, and spongiosis increased significantly from baseline to week 1, but not from week 1 to 2 (see Table). Lymphocytes were almost exclusively CD3+T cells, and found predominantly in peripapillary areas. There were few eosinophils in any biopsy at any time point. Neutrophils were also few in number and generally found only in association with micro-erosions at weeks 1 and 2; 4 of the 9 patients had no neutrophils in any biopsy at any time. Conclusions: In patients with severe RE healed by PPI therapy, stopping PPIs for 2 weeks causes a profound increase in acid reflux with acute RE that is characterized histologically by infiltration of the esophageal mucosa with T lymphocytes. Neutrophils and eosinophils do not appear to play an important role. The finding that acute RE in humans is a form of lymphocytic esophagitis supports the new paradigm for RE pathogenesis suggested by animal studies in which reflux does not destroy esophageal squamous cells directly, but rather stimulates them to secrete chemokines that attract lymphocytes.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call