Endoscopic screening for esophageal varices in cirrhotic patients

Therefore, our results do not appear to confirm the validity of the model proposed by Schepis et al. It should be noted that our cohort included less Child-Pugh class B patients than the population reported by Schepis et al. (19% vs. 41%). However, a model to predict the presence of varices should be especially useful in a population of compensated cirrhotic patients such as our cohort. In conclusion, the available data are scarce and insufficient to de- HEPATOLOGY, Vol. 35, No. 2, 2002 CORRESPONDENCE 501 termine a reliable noninvasive predictive tool to identify and discriminate cirrhotic patients with and without esophageal varices. Thus, a larger prospective study including patients with cirrhosis of all origins is required to determine the factors predicting the presence of varices (if they exist) (Letter)