PURPOSE: The red color sign observed by endoscopic examination is a reliable predictive factor for variceal bleeding. The aim of this study was to calculate the incidence of the appearance of the red color sign and to evaluate its predictive factors. METHODS: Endoscopic examination was repeatedly performed in 359 consecutive patients diagnosed as having liver cirrhosis with or without esophageal varices, during a median follow-up period of 2651 days. RESULTS: The incidence of the appearance of the red color sign on esophageal varices at the end of the tenth year was compared among patients without varices (11.4%), those with small varices (45.4%), and those with mid-size varices (65.0%). The difference was significant (P < 0.0001). The number of varices (P = 0.0010), size of varices (P = 0.0064), platelet count (P = 0.0168), and alpha-fetoprotein level (P = 0.0207) were significantly correlated with the appearance of the red color sign, as estimated by the multivariate Cox hazard model. To exclude the influence of carcinogenesis, observation was stopped when hepatocellular carcinoma was discovered. Additive predictive factors with significance were: number of varices (P = 0.001), size of varices (P = 0.027), and platelet count (P = 0.0315). CONCLUSIONS: Endoscopic signs of esophageal varices and platelet count were significant predictors for the appearance of the red color sign.
PURPOSE: The red color sign observed by endoscopic examination is a reliable predictive factor for variceal bleeding. The aim of this study was to calculate the incidence of the appearance of the red color sign and to evaluate its predictive factors. METHODS: Endoscopic examination was repeatedly performed in 359 consecutive patients diagnosed as having liver cirrhosis with or without esophageal varices, during a median follow-up period of 2651 days. RESULTS: The incidence of the appearance of the red color sign on esophageal varices at the end of the tenth year was compared among patients without varices (11.4%), those with small varices (45.4%), and those with mid-size varices (65.0%). The difference was significant (P < 0.0001). The number of varices (P = 0.0010), size of varices (P = 0.0064), platelet count (P = 0.0168), and alpha-fetoprotein level (P = 0.0207) were significantly correlated with the appearance of the red color sign, as estimated by the multivariate Cox hazard model. To exclude the influence of carcinogenesis, observation was stopped when hepatocellular carcinoma was discovered. Additive predictive factors with significance were: number of varices (P = 0.001), size of varices (P = 0.027), and platelet count (P = 0.0315). CONCLUSIONS: Endoscopic signs of esophageal varices and platelet count were significant predictors for the appearance of the red color sign.