Supplement Abstracts Submitted for the 70th Annual Scientific Meeting of the American College of Gastroenterology: ESOPHAGUS
Purpose: The sensitivity of EGD for detecting relevant esophageal disease in patients with refractory reflux symptoms has been reported to be low. It remains unclear whether the incidental discovery of gastric pathology increases the overall yield of EGD in this population. The aim of this study was to evaluate the prevalence and characteristics of gastric pathology detected by endoscopy in a multi-ethnic population with refractory reflux symptoms.
Methods: Patients evaluated with EGD for heartburn between July 2002 and December 2004 in a large municipal hospital were reviewed. The study population consisted only of patients with persistent heartburn despite at least 6 weeks of PPI therapy. Pre-procedure demographic and clinical data was collected and patients were grouped according to the presence of alarm symptoms (dysphagia, odynophagia, GI bleeding, weight loss, or iron deficiency anemia). Reflux complications were defined as: grade 2 or above esophagitis, Barrett's esophagus >3cm, or esophageal stricture. Significant gastroduodenal findings were defined as: gastric ulceration, gastric cancer, duodenal ulceration, duodenal cancer, or H. pylori positivity.
Results: A total of 263 patients (mean age 53 ± 12 years) met inclusion criteria of which 33 (13.6%) were Caucasian, 21 (8.6%) were African American, 149 (61.3%) were Hispanic, and 40 (16.5%) were Asian or of Middle Eastern descent. Reflux complications were present in 14% of individuals, while 35% had gastroduodenal pathology, and 3% of the group had both. Alarm symptoms as defined did not predict gastric findings, although the 38 patients with nausea & vomiting had a higher prevalence of gastric pathology compared to the 154 without nausea & vomiting (55%vs. 32%, p = 0.006). The most frequent gastric finding was H. pylori colonization in 39% of individuals. Patients with reflux complications had a significantly lower prevalence of H. pylori colonization compared to the group without reflux-related findings (20%vs. 43% p = 0.017). This negative association remained significant in multivariate logistic regression analysis after controlling for age, gender, and ethnicity (OR 0.37; 95% CI 0.14 to 0.96; p = 0.043).
Conclusions: In this diverse population with refractory GERD, the overall yield of EGD was increased more than twofold by the incidental detection of gastric pathology. H. pylori colonization was negatively associated with reflux complications, while the presence of nausea & vomiting was predictive for gastric pathology.