Open AccessDetermination of the predictive factors for significant liver fibrosis assessed through transient elastography
Author(s) -
Kao WeiYu,
Su ChienWei,
Fang ShengUei,
Tang JuiHsiang,
Chang ChunChao,
Liu JeanDean
Publication year - 2019
Publication title -
advances in digestive medicine
Language(s) - English
Resource type - Journals
ISSN - 2351-9800
DOI - 10.1002/aid2.13102
Subject(s) - transient elastography , medicine , gastroenterology , fibrosis , nonalcoholic fatty liver disease , univariate analysis , gamma glutamyltransferase , body mass index , diabetes mellitus , bilirubin , cirrhosis , liver fibrosis , fatty liver , endocrinology , disease , multivariate analysis , biochemistry , chemistry , enzyme
Several studies have reported the accuracy of transient elastography (TE), a noninvasive, reproducible, and reliable method for predicting liver fibrosis. This study evaluated the predictive factors associated with significant liver fibrosis assessed through TE. This study enrolled 1799 consecutive patients who underwent TE at Taipei Medical University Hospital between February 2014 and June 2015, including 810 patients with normal serum alanine aminotransferase (ALT) levels (<40 U/L). We evaluated the risk factors for significant liver fibrosis (liver stiffness measurement [LSM] > 7 kPa) assessed through TE and compared the results with those of the aspartate aminotransferase (AST)/platelet ratio index (APRI), a validated serum marker of liver fibrosis, in 672 patients with detailed laboratory data. Significant liver fibrosis was defined as an APRI score of >0.7. Significant liver fibrosis was detected through TE in 103 (12.7%) of 810 patients with normal serum ALT levels and in 91 (19.4%), 17 (18.7%), and 13 (5.8%) patients with chronic hepatitis B, chronic hepatitis C, and nonalcoholic fatty liver disease, respectively. Univariate analysis of patients with detailed biochemical test data showed that compared with without significant fibrosis, patients with significant fibrosis in the LSM > 7 kPa group were older; had higher body mass index (BMI) values; had higher fasting glucose, ALT, AST, total bilirubin, gamma‐glutamyltransferase, and ferritin levels; had higher APRI and fibrosis index based on four factors scores; had lower serum albumin levels and platelet counts; and higher rates of hypertension, diabetes, hepatitis C virus (HCV) positivity, splenomegaly, and hepatocellular carcinoma than those without significant fibrosis. Multivariate analysis disclosed that high BMI ( P < 0.001), high AST levels ( P = 0.006), low platelet counts ( P < 0.001), and low albumin levels ( P = 0.008) were the independent factors associated with significant liver fibrosis. In patients with normal serum ALT levels, the factors determining significant liver fibrosis were high BMI ( P < 0.001), high AST levels ( P < 0.001), low platelet counts ( P = 0.001), and anti‐HCV positive ( P = 0.006). LSM exhibited a moderate correlation with APRI scores ( R 2 = 0.325, P < 0.001). Among patients with normal serum ALT levels, the percentage of significant liver fibrosis defined as LSM > 7 kPa (11.2%) was higher than that of significant liver fibrosis defined as APRI > 0.7 (2.6%) ( P < 0.001). Significant liver fibrosis is not uncommon in patients with chronic hepatitis and normal serum ALT levels. TE is a reliable method for assessing the degree of liver fibrosis. High BMI values, high AST levels, low platelet counts, and low albumin levels are correlated with the diagnosis of significant liver fibrosis through TE.