PDB1 COMPARISON OF NONINVASIVE SCORES FOR THE PREDICTION OF NON-ALCOHOLIC FATTY LIVER DISEASE AMONG METABOLIC SYNDROME PATIENTS

Abstract

Patients with metabolic syndrome (MetS) have a higher risk of non-alcoholic fatty liver disease (NAFLD). The comparison of predictive performance of NAFLD noninvasive predicting score among MetS patients has never been reported. The aim of this study was to compare the accuracy and performance of 4 noninvasive scores (i.e. fatty liver index, FLI; lipid accumulation product, LAP; hepatic steatosis index, HSI, and NAFLD-MS score) for the prediction of NAFLD among MetS patients. FLI, LAP, HSI, and NAFLD-MS score were determined among a total of 499 MetS patients and 249 patients in type 2 diabetes subgroup. Ultrasonography, a widely accepted method for screening NAFLD, was used to assess the disease. The accuracy and performance of the scores were analyzed by using previously published cutoff values and presented as area under the receiver operating characteristic (AuROC) curve, sensitivity, specificity, positive predictive value, negative predictive value, and likelihood ratio. NAFLD was detected in about 68% of total patients and 77% in T2DM subgroup. According to the AuROC curves, HSI and FLI provide good performance for predicting the presence of NAFLD among MetS patients and T2DM subgroup. NAFLD-MS score provided the highest specificity of 99.4% and the highest positive likelihood ratio of 21.4 among MetS patients. All scores provide better accuracy when apply in T2DM subgroup. FLI, HSI, LAP, and NAFLD-MS score were capable to predict NAFLD among MetS patients. Routine ultrasonographic screening in high-risk group is still not recommended due to the lack of long-term benefit data and higher costs. Noninvasive scores for the prediction of NAFLD would help to confirm the need for further investigation. In terms of simplicity and ease of calculation, LAP and NAFLD-MS are recommended.