Study design and participants
A total of 69 subjects including 37 NAFLD and 32 control subjects were selected from Golestan Cohort Study (24). This study was approved by the medical ethics committee of Semnan University of Medical Sciences, and the written informed consent was obtained from all participants according to the guidelines of Declaration of Helsinki. The diagnosis of NAFLD was established using ultasonography and fibroscan. All participants were male, aged 50-81 years old. The subjects were excluded if they had excessive alcohol consumption (> 30 g/d), diabetes, viral hepatitis, autoimmune liver disease, hemochromatosis, Wilson’s disease. None of the patients were taking medication that has been reported to cause steatosis.
Anthropometric and laboratory evaluation
Anthropometric parameters including age, height, weight, blood pressure (BP), waist circumference (WC) were measured in accordance with the standardized protocols. Body mass index (BMI) was calculated as body weight (kg) divided by the square of height (m2). WC was measured at the midpoint between the lowest rib and the iliac crest.
Plasma samples were collected from participants following an overnight fasting and stored at −80°C until analysis. Fasting blood glucose (FBG), serum total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and levels of alanine amino transferase (ALT), aspartate amino transferase (AST), gamma glutamyl transferase (GGT) were determined by automated enzymatic methods and commercial kits (Pars Azmoon, Iran).
Circulating level of FAM19A5 was measured with an enzyme linked immunosorbent assay (ELISA) with a minimum detectable concentration of 15 pg/ml, Intra assay CV of <10% and Inter assay CV of <12% (RayBiotech Inc.).
Ultrasonography and elastography
Ultrasound assessment was performed using Accuvix XQ ultrasound unit (Medison, South Korea) equipped with a 3–7 MHz curved-array and a 5–12 MHz linear-array transducer for the evaluation of liver, abdominal fat and carotid arteries as previously described (24). Ultrasonographic scoring was used to determine fatty liver with specificity of 100% and sensitivity of 91.7%. In this protocol, ultrasonography results included vascular blurring (score 0 to 1), hepatorenal echo contrast and/or liver brightness (score 0 to 3) and deep attenuation (score 0 to 2). A total score of at least 2 was needed for the diagnosis of NAFLD (24).
Visceral Adipose Tissue thickness (VAT) was measured between the anterior wall of the aorta and the internal face of the rectus abdominis muscle perpendicular to the aorta. Ultrasonographic measurements have been shown to have strong correlations with the visceral fat area measured by the computed tomography (24).
Carotid Intima-Media Thickness (cIMT) was assessed as the distance between the lumen–intima interface and the media– adventitia interface, measured at its thickest point on the distal (far) wall of the common carotid arteries, 1.5–2 cm proximal to the carotid bulb. The average of right and left sides was used for cIMT analysis (24).
LS was measured by transient elastography using the FibroScan 502 machine (EchoSense, Paris, France, 5MHz). According to the manufacturer’s guidelines the M probe was used for the subjects with thoracic perimeter less than 110 cm and the XL probe for 110 cm and above. At least 10 measurements were done for each patient and the median value was recorded. Values were considered valid if the inter-quartile range (IQR) was less than 30% of the median reading (24).
Statistical analyses
The sample size was calculated based on data from a previous study about circulating levels of FAM19A5 in the subjects with and without diabetes (9). We calculated that 32 subjects in each group would provide 90% power to achieve a difference of 80% in the circulating levels of FAM19A5 between the studied groups assuming a 2-sided t-test with alpha of 0.05.
The Shapiro-Wilk test was applied to test the normal distribution of values. Non-pairwise comparisons of the concentrations of normally and non-normally distributed variables were performed between the two groups with the students t test and nonparametric Mann–Whitney test respectively. Binary logistic regression models were calculated to identify independent risk factors. Associations between the concentrations of FAM19A5 and other variables of interest were tested with the use of Pearson correlation coefficients after appropriate log-normalizations of the concentration values. Additionally, multivariate linear regressions with the stepwise variable selection were used to test for significant relations in continuous data with adjustment for possible confounders. We used two-tailed hypothesis tests and P-values < 0.05 were considered significant.