Diagnostic value analysis of different new inflammatory markers in MAFLD/NAFLD
Objective At present,there are no effective clinical biomarkers to diagnose NAFLD and metabolic related fatty liver (MAFLD).The purpose of this study is to analyze several potential new inflammatory biomark-ers,such as MHR,NLR,PLR,LMR,SⅡ and NHR,and evaluate their predictive value in the diagnosis of NAFLD/MAFLD.Methods The clinical data of patients diagnosed as NAFLD/MAFLD in the Affiliated Hospital of Nanjing University of Traditional Chinese Medicine in 2022 were collected.Mann-Whitney U test and binary logistic regression analysis were used to evaluate the correlation between these six new inflammatory biomarkers and NAFLD/MAFLD,and the predictive efficacy of these markers for NAFLD/MAFLD was measured according to the AUC.Results There were 805 cases of NAFLD (400 cases)and MAFLD (405 cases).The age,AST,GGT,ALT,creatinine,urea,TG,TC and fibrosis index (FIB-4)in MAFLD group were significantly different from those in NAFLD group (P<0.001),while ALP,uric acid,and the ratio of AST/ALT was not significantly different between the two groups (P>0.05 ).There were significant differences in six inflammatory indexes be-tween the two groups (P<0.001 ).Regression analysis showed that age,MHR,FPG,TG,LDL-c and FIB-4 were independent risk factors for MAFLD.The ROC curve analysis based on these six risk factors showed that the AUC of MHR is 0.816,the best cutoff value is 0.451,the sensitivity is 0.956,the specificity is 0.65,and the Jordan index is 0.606,indicating a good prediction efficiency.After combining the above six risk factors,the AUC,sensitivity and specificity of the joint forecasting model are 0.941,0.862 and 0.885,respectively,and the Jordan index is 0.747,which indicates that combining these six factors can significantly improve the forecasting efficiency of MAFLD.Conclusion Compared with NAFLD patients,patients with MAFLD have more serious risk of metabolic abnormalities such as liver function,blood lipid,and liver fibrosis.MHR combined with AGE,FPG,TG,LDL-c and FIB-4 as a joint prediction model has higher prediction value and is worthy of clinical application and popularization.
万方数据
维普
