One of the major complications of cirrhosis is the occurrence of portal hypertension and esophageal varices. At present, universal endoscopic screening of esophageal varices is recommended in association to primary prophylaxis in patients at high risk of variceal bleeding. But this screening is invasive and could be not cost-effective. Besides, pre-primary phrophylaxis is not effective and hampared by side effects. So, non invasive diagnosis of portal hypertension might be useful. This one could depend on non invasive measurement of hepatic venous pressure gradient, but its application to screening is not well-documented and its use in treatment monitoring is debated. A second way could be non invasive diagnosis of large esophageal varices because of prognostic and economic issues. Indirect echographic markers of portal hypertension and esophageal varices (ascites, portal vein diameter > or = 13 mm, spleen length, maximal and mean velocimetry of portal vein flow, respectively < 20 cm/s and < 12 cm/s) could be useful. Among this parameters, spleen length is an independent predictive marker of esophageal varices. Besides, several direct or indirect blood markers of fibrosis have been tested. Platelet count is repeatedly a predictive marker of esophageal varices in multivariate analysis. The other predictive factors of esophageal varices could be: prothrombin time, splenomegaly, spider naevi, Child-Pugh class, bilirubinemia, platelet count/spleen diameter ratio and Fibrotest, but these data require validation. In summary, in regard to actual results, non invasive diagnosis of portal hypertension might be useful in esophageal varices screening, but the substitutes to endoscopy have limited place actually in clinical practice, and exclusive non invasive diagnosis of portal hypertension is not applicable; the only test that seems to be useful in clinical practice is conventional endoscopy awaiting the results of videocapsule.