Chronic liver disease causes 1.75 million deaths globally and is within the top 10 leading causes of death in middle income countries. Chronic liver injury occurs via a process of inflammation and fibrosis formation. Patients often do not present to healthcare until advanced stages of disease and when there is already decompensated cirrhosis. Liver biopsy has been used to identify earlier stages of fibrosis, but it is poorly accepted by patients and has limitations. Transient Elastography (TE) using Fibroscan ® is a non-invasive tool for the diagnosis liver fibrosis. The clinical application of Fibroscan in non-alcoholic fatty disease (NAFLD), chronic hepatitis B (CHB), and methotrexate induced liver fibrosis were examined. Clinical Utility of Transient Elastography in non-alcoholic fatty liver disease: Chapters 2, 3 & 4 Non-alcoholic fatty liver disease affects 20-35% of the global population, but only a small subset develop the histological subtype of non-alcoholic steatohepatitis (NASH), which can lead to progressive liver disease by causing fibrosis and eventually cirrhosis. Fibroscan can potentially identify those patients who have fibrosis and are at increased risk of further progression. Patients with type 2 diabetes, who are at high risk of NASH, were assessed. A liver stiffness measurement (LSM) ≥9.8 kPa, used as a cut-off for advanced fibrosis (1), was found in 12% (10/77) of subjects. Higher LSM readings correlated with higher BMI and the use of insulin therapy. Patients on insulin had LSM ≥9.8 kPa with likelihood ratio (LR): 12.3, p=0.002 (Chapter 2). The study was limited by a small sample size, and a high failure rate as the medium (M) probe was only available. A systemic review evaluating all non-invasive methods for diagnosing NASH and NAFLD fibrosis was undertaken. This included a meta-analysis that focused on what was found to be the most widely studied markers of NASH and NAFLD fibrosis: cytokeratin-18 (CK-18) fragments and transient elastography respectively (Chapter 3). Not only was TE found to be the most extensively studied, it had also one of the highest diagnostic accuracies with pooled sensitivities and specificities to diagnose F≥2, 3 and 4 to be: 79% and 75%, 85% and 85%, and 92% and 92% respectively. We then proceeded to perform a much larger study in diabetic subjects using the latest generation of Fibroscan ® 502 touch model (Chapter 4). This included the extra-large (XL) probe for obese subjects and also featured the novel Controlled Attenuation Parameter (CAP), which assesses liver steatosis. A total of 1918 diabetes patients at Prince of Wales Hospital, Hong Kong were recruited. Each had a TE and CAP to assess liver stiffness and steatosis. Reliable scans were achieved in 98.2% of patients using the M or XL probes. The proportion of patients with increased CAP (suggestive of steatosis) and increased LSM (suggestive of advanced fibrosis) were 72.8% and 17.7% respectively. By multivariate analysis, female gender, higher body mass index, triglycerides, fasting plasma glucose and alanine aminotransferase, and non-insulin use were associated with increased CAP. Longer duration of diabetes, higher body mass index, alanine aminotransferase, spot urine albumin-creatinine ratio, and lower high-density lipoprotein-cholesterol were associated with increased LSM. The17.7% prevalence of advanced fibrosis suggests type 2 diabetic patients would benefit from routine screening for liver disease. Clinical Utility of Transient Elastography in chronic hepatitis B: Chapters 5 and 6 Transient elastography was initially applied for staging patients with chronic hepatitis C (CHC) with data rapidly growing on its utility for the assessment in patients with CHB infection. Our study contributes to this by further evaluating the diagnostic accuracy and usefulness of TE, and also comparing its performance against the FIB4 index, Aspartate Platelet Ratio Index (APRI), Aspartate Alanine Aminotransferase Ratio (AAR), Age Platelet Index (API), Fibrosis Index (FI) and Caffeine Breath Test (CBT) (Chapter 5). In 71 CHB patients, the diagnostic performance of the LSM for Metavir fibrosis stage F≥1, 2, 3 and 4 were: Area under Receiver Operator Characteristic (AUROC) = 0.825, 0.792, 0.874 and 0.945 respectively. Patients with high ALT required higher LSM cut-offs. Dual cut-offs are needed to “rule in” and to “rule out” stage of fibrosis with a high level of certainty. Using normal vs high ALT specific cut-offs, F≥2 and F≥3 can be “ruled in” or “ruled out” with certainty in 49.3% and 57.7% of CHB patients respectively. TE was the superior non-invasive test when compared with FIB4-I, APRI, API, AAR and FI. Caffeine breath test compared well against TE in a small cohort, but is not as practical. Liver histology is limited by interobserver variability, with 44% of liver biopsies being classified a different stage on second evaluation, and the intraclass correlation coefficient showing moderate agreement (K =0.457). Although routinely compared, this highlights the limitations of assessing the accuracy of TE and other non-invasive tests against a reference standard that has such a degree of variation. The use of TE in the longitudinal monitoring of fibrosis is important in the follow up of patients with CHB (Chapter 6). Current literature was conflicting and seemed to suggest that decline in LSM was influenced more by the fall in ALT with decline in necroinflammatory activity, rather than fibrosis regression. We sought to evaluate the factors that affected LSM change and assess which clinical subgroups experienced an LSM decline. In 124 CHB patients who were followed for 31.2 months (SD 13.1), LSM decline was greatest in those who had active disease and were subsequently treated with antivirals. This is associated with ALT normalization, HBeAg seroconversion and viral suppression. In CHB patients with quiescent disease - ie did not require antiviral treatment, or who had persistently normal ALT irrespective of treatment - only a small or non-significant decline in LSM was observed. The change in LSM was strongly correlated with length of time and may suggest fibrosis regression. Further studies are required, as our findings are limited by a lack of correlation with liver biopsy, and the low baseline levels of liver stiffness in those with inactive CHB. Clinical Utility of Transient Elastography in methotrexate induced liver fibrosis: Chapter 7 Long term use of methotrexate has been associated with risk of liver fibrosis and the role of TE in this cohort was evaluated. The relationship between liver fibrosis and methotrexate dose, and other factors associated with moderate fibrosis (F2) using an LSM cut-off of ≥7.1 kPa were examined. In 39 patients with a mean intake dose of 5.3g of methotrexate, no correlation was found between the LSM and the cumulative dose or duration of treatment. Of the 7/39 cases of LSM≥7.1 kPa (17.9%), BMI≥30 was the only risk factor with a likelihood ratio (LR) of 4.442, p=0.029. One patient had cirrhosis (2.6%). This is much lower than rates reported from early studies [26% (2, 3)], and more in line with recent data [around 2% (4)], and lends support to the suggestion that early studies overestimated the risk of methotrexate induced fibrosis due to lack of controls for pre-existing liver disease (5). There was also no difference in the LSM of methotrexate subjects and matched population controls. Conclusion Our studies lend further support to the utility of LSM on identifying those at increased risk of liver fibrosis progression, which will continue to remain a significant clinical challenge in both individuals and as a public health burden. In particular we feel that major contributions have been made on the subject of screening for advanced fibrosis in a high-risk population of type II diabetic patients. Our longitudinal studies on the role of using TE in follow up and comparing its performance in CHB patients are also significant. Despite the small cohort of methotrexate users, this further supports the utility of TE in a wide range of liver diseases that manifest with progressive fibrosis. The next area of further development in the clinical use of TE is as a stand-alone marker that has prognostic significance.