This study aimed to analyze the diagnostic accuracy of liver stiffness for predicting esophageal variceal grading and the risk of esophageal variceal bleeding (EVB) in cases of cirrhosis. Hematological and biochemical parameters were measured and transient elastography was performed in 88 patients with hepatitis B–related cirrhosis undergoing endoscopy for esophageal varices. Esophageal varices grade was highly correlated with liver stiffness measurement (LSM) and the liver stiffness spleen diameter-to-platelet score in cirrhosis. Compared with those from endoscopy, the LSM and the liver stiffness spleen diameter-to-platelet score for the absence of esophageal varices were as follows: area under the receiver operating characteristic curve (AUROC), 0.894/0.926; sensitivity, 0.836/0.818; and specificity, 0.875/1.000, respectively. The AUROC and the sensitivity and specificity of LSM and the liver stiffness spleen diameter-to-platelet score for predicting grade III esophageal varices were 0.954 and 0.901, respectively. The AUROCs of LSM and the liver stiffness spleen diameter-to-platelet score for discriminating grades II and III from grade I or the absence of esophageal varices were 0.958 and 0.941, respectively. We also found that EVB was closely associated with LSM and spleen diameter. The AUROC, sensitivity, and specificity were 0.855/0.819, 0.857/0.875, and 0.747/0.780, respectively. Meanwhile, LSM and spleen diameter were 2 independent factors for predicting EVB. These data suggest that LSM and the liver stiffness spleen diameter-to-platelet score could accurately rule out cirrhosis without esophageal varices and differentiate high- and low-risk patients. Furthermore, LSM and spleen diameter had excellent abilities to predict EVB.