English Audio Request

Out study covered 260 patients with chronic liver diseases (120 female and 140 male patients) who underwent both abdominal ultrasonography and upper gastrointestinal endoscopy at the National Hospital Organization (NHO) Nagasaki Medical Center from January through December.
We included 260 patients (120 females and 140 males) with chronic liver disease who underwent both abdominal ultrasonography and upper gastrointestinal endoscopy between January and December 2006 at the National Nagasaki Medical Centre, Japan.
The etiologies of their liver diseases were hepatitis C in 153 patients (accounting for 59% of all the patients), hepatitis B in 56 (22%), alcoholic hepatitis in 23 (9%), autoimmune hepatitis (AIH) with 6 (2%), primary biliary cirrhosis (PBC) with 9 (3%) and others with 13 (5%).
The etiologies of their chronic liver disease were hepatitis C in 153 patients (59%), hepatitis B in 56 (22%), alcoholic hepatitis in 23 (9%), autoimmune hepatitis (AIH) in 6 (2%), primary biliary cirrhosis (PBC) in 9 (3%), and other in 13 (5%).
In all patients, we calculated liver US scores and its 3 components (i.e., edge blunting, texture roughening and surface unevenness in ultrasonographic images) and the spleen index.
In all patients we calculated the hepatic US score and its 3 components (bluntness of the edge, coarseness of the parenchymal texture, and irregularity of the surface on the liver ultrasonogram), as well as the spleen index.
We set appropriate threshold values for results of these tests and various factors, such as ages and genders, and calculated sensitivity, specificity and the positive likelihood ratio.
Appropriate thresholds of variables, including these results, age, and sex were determined for presence of varices in order to calculate sensitivity, specificity, and positive likelihood ratio.
In studying significant differences, we used the Mann-Whitney U test for univariate analysis and logistic regression for multivariate analysis.
To determine statistically significant differences, the Mann-Whitney U test was used for the univariate analysis and logistic regression analysis was used for multivariate analysis.
We plotted a Receiver Operating Characteristic (ROC) curve and compared Az values. For the ROC curve, we used the continuous confident scale, and performed analysis using ROCKIT software (The University of Chicago).
Receiver operating characteristic (ROC) curves of the investigated variables were drawn and the area under the ROC curve (Az) compared. ROC curves were analyzed according to the maximum-likelihood estimation of ROC curves from continuously distributed data using ROCKIT (The University of Chicago).
Findings of gastrointestinal endoscopy were obtained by a specialist certified by the Japan Gastroenterological Endoscopy Society (JGES) and were classified based on the general rules for recording endoscopic findings in the General Rules for Study of Portal Hypertension 2004.
Gastrointestinal endoscopy findings were read by a physician accredited by the Japan Gastroenterological Endoscopy Society to classify them according to the general rules for recording endoscopic findings in the General Rules for Study of Portal Hypertension 2004.