Optimal liver stiffness measurement values for the diagnosis of significant fibrosis and cirrhosis in chronic liver disease in Singapore.

INTRODUCTION Despite the widespread use of transient elastography for non-invasive assessment of liver fibrosis, the optimal cut-off liver stiffness measurement (LSM) values remain unclear. This study aimed to validate the optimal cut-off LSM values for significant fibrosis and cirrhosis in patients with chronic liver disease (CLD). METHODS Prospective multicentre data of CLD patients who underwent paired liver biopsy and LSM was analysed to determine the optimal cut-off LSM values for predicting significant fibrosis (METAVIR F >= 2) and cirrhosis (METAVIR F4). A high-quality cohort was selected by excluding those with failed LSM and invalid LSM readings. RESULTS Of the 481 patients recruited, 322 fulfilled the pre-defined quality criteria. CLD aetiology was chronic hepatitis B (CHB) in 49%, non-alcoholic steatohepatitis (NASH) in 16% and chronic hepatitis C (CHC) in 12%. Area under the receiver operating characteristic curve for LSM was 0.775 (95% confidence interval [CI] 0.724-0.826) for significant fibrosis and 0.810 (95% CI 0.738-0.882) for cirrhosis. Optimal cut-off LSM values were 9 kPa for significant fibrosis and 13 kPa for cirrhosis in the general cohort. Optimal cut-off LSM values were 9 kPa for significant fibrosis and 12 kPa for cirrhosis for both CHB and CHC, while the corresponding values for NASH were 11 kPa and 15 kPa. CONCLUSION Optimal cut-off LSM values should be selected based on disease aetiology. In Singapore, the optimal cut-off LSM values for CHB and CHC are 9 kPa for significant fibrosis and 12 kPa for cirrhosis. Optimal cut-off values for NASH require further validation.