Introduction: We recently demonstrated (Kusano M, et al. Neurogastroent Motil, 2012) that both Japanese men and women with cascade stomach (CS) on barium studies have a markedly elevated incidence of upper gastrointestinal symptoms, especially reflux symptoms. We hypothesized that CS might not only be associated with symptoms, but also with esophageal mucosal injury. Methods: Study 1: We developed a new endoscopic classification of CS based on the fundal incisura angularis (ridge), which is observed immediately after the endoscope passes through the cardia. The classification comprised four grades: grade 0, no ridge between fundus and corpus; grade 1, a ridge between fundus and corpus at endoscope insertion; grade 2, the ridge is also seen by retroflexed observation; and grade 3, the ridge is still observed after full gastric inflation. To assess inter-observer variation of this classification by calculating kappa statistics, 60 endoscopic photos were selected, and grading of CS was done three times at intervals > 1 week by 10 experienced endoscopists. Study 2: In 500 consecutive men and 189 women undergoing endoscopy for health screening, the incidence of reflux esophagitis (RE) and the grade of CS were compared. RE was defined as changes exceeding grade A of the LA classification, while hiatus hernia (HH) was protrusion > 2 cm. Assessment was performed by the three endoscopists with the highest kappa values (kv) in Study 1. Multivariate analysis was also performed to identify factors (age, BMI, cigarette smoking, and alcohol intake) related to CS. Results: In study 1, the kv for inter-observer agreement at the 3rd assessment were 0.85, 0.58, 0.50, and 0.78 for CS grades 0, 1, 2, and 3, respectively, while the kv for overall agreement was 0.68. Kv increased significantly after repeated assessment. In study 2, CS was classified as grade 0, 1, 2, and 3 in 40%, 28%, 17%, and 13% of men, respectively. The incidence of RE was 20% in men with grade 0 CS, 17% in grade 1, 25% in grade 2, and 30% in grade 3, showing significant differences. In addition, the incidence of RE was 10% in women with grade 0 CS, 3% in grade 1, 9% in grade 2, and 29% in grade 3 (P=0.09). Multivariate analysis showed that CS, BMI, and HH were independent risk factors for RE. Conclusion: Our new endoscopic classification of CS was shown to be both usable and meaningful. CS was identified as an independent risk factor for RE, suggesting that it could have a potential role in various functional gastrointestinal disorders.
Read full abstract