Osteonecrosis of the femoral head (ONFH), also known as vascular necrosis of the femoral head, is combined with lipid metabolism disorders in most patients. This study aims to explore the lipid metabolism profiles in different subtypes of ONFH. The subjects were divided into an alcohol-induced osteonecrosis of the femoral head (AONFH) group, a steroid-induced osteonecrosis of the femoral head (SONFH) group, and a normal control (NC) group (n=16, 29, and 32, respectively). Ultra-performance liquid chromatography-mass spectrometry/mass spectrometry (UPLC-MS/MS) was used to detect the lipidomics analysis in the peripheral blood samples of subjects and identify the underlying biomarkers. The samples were preprocessed, the partial least squares discriminant analysis (PLS-DA) was adopted, and the variable importance for the projection (VIP) values were calculated to measure the expression pattern of each lipid metabolite and observe the influence and explanatory power of the expression pattern of each lipid metabolite on the classification and discrimination between the different groups. The lipid metabolites with fold change (FC)>2, P<0.05 and VIP>1 in the different groups were screened as differential lipids. Among them, the differential lipids co-existing in the AONFH group and the SONFH group were regarded as common differential lipids for ONFH, and the differential lipids that exist separately were regarded as specific differential lipids in the AONFH group or the SONFH group. Binary logistic regression was used to evaluate the diagnostic value of differential lipid metabolites on the basis of the receiver operator characteristic (ROC) curve analysis. Based on the disease stage information, the correlation between the differential lipids and the disease stage was analyzed in the AONFH group and the SONFH group. In this study, 1 358 lipid metabolites were detected in each plasma sample. Compared with the NC group, there were significant difference in the expression patterns of lipid metabolism profiles in the AONFH group and the SONFH group. A total of 62 and 64 differential lipid metabolites were screened in the AONFH and SONFH patients (FC>2, P<0.05, VIP>1) respectively, and these differential lipids were mainly up-regulated in the disease samples. Nine differential lipid metabolites were further identified, which were shared by the AONFH group and the SONFH group; the area under the curve (AUC) in 6 kinds of lipid components was greater than 0.7, including 1-myristoyl-2-docosahexaenoyl-sn-glycero-3-phosphocholine, hypoxanthin, serotonin, PE (19:0/22:5), PE (19:0/22:5), and cholest-5-en-3-yl beta-D-glucopyranosiduronic acid. Fifty-three specific differential lipid metabolites were identified in the AONFH group, and 55 specific differential lipid metabolites were identified in the SONFH group. The AUC in 6 kinds of lipid components was greater than 0.9, including 1D-myo-Inositol 1,2-cyclic phosphate, L-pyroglutamic acid, DL-carnitine, 8-amino-7-oxononanoic acid, Clobetasol, and presqualene diphosphate. In the AONFH group, there were 9 differential lipid metabolites related to the disease stages, including LPG 18:1, serotonin, PC (22:4e/23:0), PC (19:2/18:5), hypoxanthin, PE (18:1/20:3), LPE 18:1, 1-stearoyl-2-arachidonoyl-sn-glycerol, and PE (16:0/18:1); with AONFH disease progresses from I/II stages to III/IV stages, the relative content of these 9 differential lipid metabolites was increased. In the SONFH group, 8 differential lipid metabolites were found to be related to the stage of the disease, including TM6076000, 4-(1,1-dimethylpropyl)phenol, D-617, asarone, phenylac-gln-OH, creatine, leu-pro, and 8-amino-7-oxononanoic acid; and with the SONFH progressed from stage I/II to stage III/IV, the content of these 8 differential lipid metabolites were gradually increased. This study analyzes the characteristics of the plasma lipid metabolism profile in the AONFH and SONFH patients, and which identifies the differential lipid metabolites related to disease diagnosis and evaluation. These results provide evidence for exploring lipid metabolism alterations and the mining of novel lipid biomarkers for the ONFH.
Read full abstract