Does duodenal juice reflux into the esophagus of patients with complicated GERD? evaluation of a fiberoptic sensor for bilirubin

Background It is controversial whether duodenal juice can damage esophageal mucosa in patients with gastroesophageal reflux disease (GERD). The issue remains unresolved partly because of difficulties in detecting the presence of duodenal juice in the lower esophagus. Objectives and methods This study utilized an in vitro portable spectrophotometer with a fiberoptic probe capable of detecting bile as a marker of duodenal juice. Absorbance/concentration curves were established with known bilirubin concentrations at pH 1.4 and pH 7.7. Esophageal pH and bilirubin absorbance were monitored in vivo over a 24-hour period in 20 healthy volunteers to determine the absorbance threshold for clinical use. The study population consisted of 21 patients with GERD. Four had no mucosal injury, 5 erosive esophagitis, and 12 Barrettʹs esophagus. Results The correlation between absorbance and bilirubin concentration was 0.98 and 0.99 for acid and alkaline environments, but bilirubin absorbance was 35% less in an acid environment. Using an absorbance threshold of 0.14, patients with GERD taken in toto had elevated esophageal exposure to bilirubin. No difference was observed in bilirubin exposure between reflux patients without mucosal injury and controls. Highest exposure occurred in patients with Barrettʹs esophagus. An important observation was that esophageal bilirubin exposure frequently occurred during periods when the esophageal pH was in the normal range. Conclusions The fiberoptic probe accurately detects esophageal bilirubin as a marker of duodenal juice. Esophageal exposure to bilirubin is uncommon in normal subjects. Patients with erosive esophagitis and Barrettʹs metaplasia have increased esophageal exposure to duodenal juice compared to normal subjects. Reflux of duodenal juice into the esophagus can occur when the esophageal pH is within its normal range and is undetectable by pH monitoring alone.