Readily available tools that identify diabetic patients with clinically significant liver fibrosis secondary to NAFLD are needed. The NFS is designed to categorise NAFLD fibrosis risk and it uses the following variables: age, BMI, presence of diabetes, AST, ALT, platelets and serum albumin. A high NFS is a predictor of advanced NAFLD and also mortality.
Method: Patients were randomly recruited from a tertiary diabetes centre and excluded if they reported significant current alcohol intake or liver disease other than NAFLD. Each patient underwent a clinical assessment, fasting blood tests, liver ultrasound and assessment of liver fibrosis by transient elastography (FibroScan®). XL probe readings were used unless unavailable. A FibroScan® measure of ≥10.3kPa was considered consistent with cirrhosis.
Results: A total of 102 serial patients completed all assessments (61% male and 97% type 2 diabetes). Results are presented as either mean ± SD or median and IQR: age 62 ± 10 years, duration of diabetes 11 (IQR: 5 - 15 years), BMI 30.9 ± 5.8 kg/m2 and HbA1c 7.6 ± 1.2 %. Steatosis was present on ultrasound in 88% of patients and FibroScan® ≥10.3kPa in 19%. NFS classified 28 patients ‘low risk’, of whom one had a FibroScan® ≥10.3kPa (history of heavy alcohol consumption >10 years previously). Of the 65 patients classified as ‘intermediate risk’, 13 had a FibroScan® ≥10.3kPa. Of the 9 patients classified ‘high risk’, 6 had a FibroScan® ≥10.3kPa; one patient had a FibroScan® of 9.2kPa and cirrhosis could not be excluded. The other 2 of 9 patients were either morbidly obese or elderly.
Conclusions: Severe liver fibrosis secondary to NAFLD is prevalent in patients with type 2 diabetes. When applied in a correct clinical setting, a ‘low risk’ NFS adds value as it helps to exclude cirrhosis. NFS accuracy in ‘intermediate risk’ would likely be improved in combination with other biomarkers or FibroScan®.