| Summary: | Type 2 diabetes is an established risk factor for non-alcoholic fatty liver disease (NAFLD). Despite this, there are few data investigating NAFLD within large populations of people with Type 2 diabetes. In the initial chapters of this thesis NAFLD is defined and the pathological processes linking it to Type 2 diabetes are described. The epidemiological evidence that underpins current understanding of prevalence, risk factors and progression of the condition is reviewed in detail. Subsequent chapters describe research undertaken within the Edinburgh Type 2 Diabetes Study on the epidemiology of NAFLD in this population. The prevalence of NAFLD was determined within 939 subjects, aged 61 - 76 years, with well-characterised Type 2 diabetes. NAFLD was defined by the presence of ultrasounddiagnosed steatosis with detailed exclusion of secondary causes for liver disease. Ultrasound gradings were validated in a subgroup of 58 participants using 1Fi magnetic resonance spectroscopy, the non-invasive gold-standard method for hepatic lipid quantification, and intra- and inter-observer variability in ultrasound grading was calculated. The final prevalence of NAFLD, 42.6%, was higher than in the general population, but lower than in the few studies that have been performed in populations with Type 2 diabetes. It is likely that this was due to comprehensive screening for secondary causes of liver disease and validation of the ultrasound measure which resulted in re-categorisation as "normal" subjects initially graded as having mild steatosis. Independent predictors of NAFLD were diabetes variables or components of the metabolic syndrome. The utility of conventional "liver function tests" in detecting hepatic steatosis and NAFLD was examined. Although liver enzyme levels (alanine aminotransferase, aspartate aminotransferase and gamma-glutamyltransferase) were significantly higher in participants with hepatic steatosis compared with those with normal liver on ultrasound, values remained within the normal range in the majority of cases. The negative predictive values of normal levels were therefore low, calling into question their utility as screening tests for liver disease in the diabetic population. The prevalence of advanced liver disease had not previously been studied in a population with Type 2 diabetes, and was investigated in our population using surrogate markers of hepatic fibrosis. Hyaluronic acid (HA) levels were significantly elevated in 5.7% of subjects and 0.4% had ultrasound-diagnosed cirrhosis. Hepatocellular carcinoma was detected in 0.2% of participants. Liver enzyme markers were poorly predictive of high HA levels. Clinical risk scores identified a large proportion of subjects as potentially having severe fibrosis, but these scores have not previously been validated in populations with Type 2 diabetes and therefore have to be interpreted with caution. NAFLD has previously been shown to be predictive of cardiovascular disease in the context of Type 2 diabetes, but mechanisms underlying this have not been fully elucidated. In the current study the association of hepatic steatosis and NAFLD with non-classical cardiovascular risk factors - clot formation and lysis dynamics, prothrombotic mediators and markers of inflammation - was investigated. NAFLD was significantly associated with a longer clot lysis time and higher levels of complement C3 and plasminogen activator inhibitor type 1. Following adjustment for these prothrombotic compounds, the association of NAFLD and clot lysis time was lost, and it is therefore hypothesised that NAFLD may influence clot lysis time via increased production of these mediators.
|
|---|
