Usefulness of the Glasgow–Blatchford score to predict 1-week mortality in patients with esophageal variceal bleeding

OBJECTIVESEsophageal variceal bleeding is one of the most severe complications of liver cirrhosis, with high mortality. However, there is no established scoring system for short-term mortality in patients with esophageal variceal bleeding. The aim of this study was to evaluate the usefulness of the Glasgow–Blatchford score (GBS), the Model for End-Stage Liver Disease (MELD) score, and the Child–Pugh score for predicting short-term and hospital mortality in patients with esophageal variceal bleeding. METHODSA total of 47 patients with esophageal variceal bleeding were studied between September 2009 and March 2015. The GBS, the MELD score, and the Child–Pugh score were assessed for their ability to predict 1- and 6-week mortality rates using a receiver operating characteristic curve. RESULTSThe 1- and 6-week mortality rates were 17.0 and 31.9%, respectively. The median GBS, MELD, and Child–Pugh scores were 13 (range4–19), 10 (range0–34), and 9 (range5–13), respectively. The GBS was superior to both the MELD and the Child–Pugh scores for prediction of 1-week mortality [area under the curve=0.82 (95% confidence interval0.66–0.98) vs. 0.71 (0.47–0.96) and 0.72 (0.53–0.91)]. The MELD score was superior to both the Child–Pugh score and the GBS for prediction of 6-week mortality [area under the curve=0.83 (95% confidence interval0.69–0.97) vs. 0.69 (0.52–0.85) and 0.67 (0.50–0.83)]. CONCLUSIONFor 1-week mortality, the GBS was superior to the Child–Pugh and the MELD scores in patients with esophageal variceal bleeding. However, for 6-week mortality, the MELD score was superior in patients with esophageal variceal bleeding.