Background & aims
Alcohol abuse causes half of all deaths from cirrhosis in the West, but few tools are available for non-invasive diagnosis of alcoholic liver disease. We evaluated 2 elastography techniques for diagnosis of alcoholic fibrosis and cirrhosis; liver biopsy with Ishak score and collagen-proportionate area were used as reference.
We performed a prospective study of 199 consecutive patients with ongoing or prior alcohol abuse, but without known liver disease. One group of patients had a high pretest probability of cirrhosis, in that they were identified at hospital liver clinics (in Southern Denmark). The second, lower-risk group, was recruited from municipal alcohol rehabilitation centers and the Danish national public health portal. All subjects underwent same-day transient elastography (FibroScan), 2-dimensional shear wave elastography (2D-SWE, Supersonic Aixplorer), and liver biopsy after an overnight fast.
Transient elastography and 2D-SWE identified subjects in each group with significant fibrosis (Ishak score ≥3) and cirrhosis (Ishak score ≥5) with high accuracy (area under the curve ≥0.92). There was no difference in diagnostic accuracy between techniques. The cut-off values for optimal identification of significant fibrosis by transient elastography and 2D-SWE were 9.6 kPa and 10.2 kPa, and for cirrhosis 19.7 kPa and 16.4 kPa. Negative predictive values were high for both groups, but the positive predictive value for cirrhosis was >66% in the high-risk group vs approximately 50% in the low-risk group. Evidence of alcohol-induced damage to cholangiocytes, but not ongoing alcohol abuse, affected liver stiffness. The collagen-proportionate area correlated with Ishak grades and accurately identified individuals with significant fibrosis and cirrhosis.
In a prospective study of individuals at risk for liver fibrosis due to alcohol consumption, we found elastography to be an excellent tool for diagnosing liver fibrosis and for excluding (ruling-out rather than ruling-in) cirrhosis.