Abstract
Nonalcoholic fatty liver disease (NAFLD) is common among Saudi patients with type 2 diabetes (T2DM). However, recommended clinical procedures to detect it are unavailable in many locations. Therefore, better and more available diagnostic biomarkers for NAFLD are needed. Various serum parameters were suggested, and algorithms that employ routine measurements in clinical practice have been developed for the prediction of fat stores in the liver in different populations. However, no such studies have been conducted on Saudis. We aimed to compare selected biochemical markers and calculated indices in T2DM patients diagnosed with NAFLD and patients without NAFLD to find the best markers associated with NAFLD. A cross-sectional study was employed to recruit 67 people with T2DM from endocrine outpatient clinics at King Abdul-Aziz University Hospital. NAFLD was detected by ultrasonography in 28 patients. Demographic information, anthropometric, and blood pressure (BP) measurements were taken. Fasting blood samples were obtained to measure glucose, glycated haemoglobin, lipid profile, liver function tests, and highly sensitive C-reactive protein. Fatty liver index, hepatic steatosis index, NAFLD-liver fat score, and triglyceride and glucose index were calculated. Following stepwise forward likelihood ratio regression with independent variables included in one model using binary logistic regression with age and waist circumference (WC) entered as covariates, elevated diastolic BP and low high-density lipoprotein- cholesterol remained significantly associated with NAFLD (
p
= 0.002 and 0.03, respectively). However, none of the investigated indices could be used to diagnose the disease adequately due to low specificity, even after calculating new cut-off values. Investigating novel markers and adjusting existing equations used to calculate indices to improve sensitivity and specificity in our population is needed.