BACKGROUND & AIMS: Narrow band imaging (NBI) facilitates mucosal surface evaluation and may improve the endoscopic diagnosis of gastroesophageal reflux disease (GERD). We investigated the utility of NBI in improving the endoscopic diagnosis of GERD when compared with conventional endoscopy. METHODS: A total of 107 subjects (nonerosive reflux disease [NERD], 36; erosive reflux disease [ERD], 41; controls, 30) were recruited prospectively. The mucosal morphology at the squamocolumnar junction (SCJ) in GERD and controls was visualized using conventional endoscopy and NBI. The main outcome measurements were as follows: (1) The differences in mucosal morphology at the SCJ between conventional endoscopy and NBI; and (2) the differences in mucosal morphology at the SCJ between GERD and controls with NBI. RESULTS: Micro-erosions, increased vascularity, and pit patterns at the SCJ not seen on conventional endoscopy were well seen with NBI. Compared with controls, ERD and NERD had a significantly higher prevalence of micro-erosions (ERD, 100%; NERD, 52.8%; controls, 23.3%), and increased vascularity (ERD, 95.1%; NERD, 91.7%; controls, 36.7%), but a lower prevalence of round pit pattern (ERD, 4.9%; NERD, 5.6%; controls, 70%). ERD and NERD were similar in terms of increased vascularity and pit patterns. Increased vascularity with absence of round pit pattern was useful to distinguish NERD from controls (sensitivity, 86.1%; specificity, 83.3%). Interobserver agreement was good for increased vascularity (kappa = 0.95) and micro-erosions (kappa = 0.89), but lower for pit pattern (kappa = 0.59). CONCLUSIONS: NBI enhanced mucosal morphology at the SCJ and appeared useful for improving the endoscopic diagnosis of GERD.
BACKGROUND & AIMS: Narrow band imaging (NBI) facilitates mucosal surface evaluation and may improve the endoscopic diagnosis of gastroesophageal reflux disease (GERD). We investigated the utility of NBI in improving the endoscopic diagnosis of GERD when compared with conventional endoscopy. METHODS: A total of 107 subjects (nonerosive reflux disease [NERD], 36; erosive reflux disease [ERD], 41; controls, 30) were recruited prospectively. The mucosal morphology at the squamocolumnar junction (SCJ) in GERD and controls was visualized using conventional endoscopy and NBI. The main outcome measurements were as follows: (1) The differences in mucosal morphology at the SCJ between conventional endoscopy and NBI; and (2) the differences in mucosal morphology at the SCJ between GERD and controls with NBI. RESULTS: Micro-erosions, increased vascularity, and pit patterns at the SCJ not seen on conventional endoscopy were well seen with NBI. Compared with controls, ERD and NERD had a significantly higher prevalence of micro-erosions (ERD, 100%; NERD, 52.8%; controls, 23.3%), and increased vascularity (ERD, 95.1%; NERD, 91.7%; controls, 36.7%), but a lower prevalence of round pit pattern (ERD, 4.9%; NERD, 5.6%; controls, 70%). ERD and NERD were similar in terms of increased vascularity and pit patterns. Increased vascularity with absence of round pit pattern was useful to distinguish NERD from controls (sensitivity, 86.1%; specificity, 83.3%). Interobserver agreement was good for increased vascularity (kappa = 0.95) and micro-erosions (kappa = 0.89), but lower for pit pattern (kappa = 0.59). CONCLUSIONS: NBI enhanced mucosal morphology at the SCJ and appeared useful for improving the endoscopic diagnosis of GERD.