Hirata M, Ishihama S, Sanjo K, Idezuki Y
Second Department of Surgery, Faculty of Medicine, University of Tokyo, Japan.
Surgery. 1994 Jul;116(1):8-16.
We studied new prognostic factors of esophageal variceal rupture by analyzing electronic video images of esophageal varices in 30 patients with portal hypertension.
Fifteen of the patients were emergency or elective cases (bleeders), and the remaining 15 were prophylactic cases (nonbleeders). A comparison was made between the bleeders and nonbleeders in terms of endoscopic findings and the image processing data, especially variceal color tone and red color sign.
Endoscopic findings based on the general rules prepared by the Japanese Research Society for Portal Hypertension showed no significant difference between the two groups. However, with regard to the image processing data, both the ratio of red signal and the ratio of value were significantly lower in bleeders than in nonbleeders. In addition, the area ratio of red color sign was significantly higher in the former than in the latter. A follow-up study of nonbleeders also indicated that image processing data were more reliable than traditional endoscopic rules.
By adding these image processing data to the traditional general rules for recording endoscopic findings, it is possible to select patients with varices that have a higher risk of rupture.
