Analysis Of Peripheral Blood Nontargeted Lipidomics In Patients With Sepsis

Objective By using liquid chromatography-mass spectrometer(lc-ms),the lipids and their metabolites in the plasma of sepsis patients and normal people are identified and analyzed,and the main lipid metabolites are screened out to analyze the changes in the process of sepsis;The diagnostic accuracy of PCT,il-13 and TNFαwas compared,so as to explore the influence of changes in lipid molecules and their metabolites in the pathogenesis of sepsis,and to find more effective biomarkers for early diagnosis of sepsis.Methods From September 2017 to August 2018,peripheral blood samples of sepsis patients diagnosed in the department of respiratory and critical illness medicine(MICU)of baotou iron and steel hospital and healthy patients who underwent physical examination were selected,The plasma was analyzed by mass spectrometry using UHPLC Nexera lc-30 a ultra-high performance liquid chromatograph,and the contents of lipid molecules and metabolites were detected.Concentrations of tumor necrosis factor(TNFα),interleukin-13(il-13)and procalcitonin(PCT)were determined by ELISA and electrochemical luminescence immunoassay.T test,spearman correlation analysis,regression analysis,principal component analysis and neural network diagram analysis were performed on the corresponding data to find out the components of lipid metabolites that have important influence on sepsis.Linear correlation analysis and ROC curve were conducted to evaluate the accuracy of lipid molecular changes in the diagnosis of sepsis.Results Mass spectrometry was used to screen lipid small molecule metabolites between the sepsis group and the normal group,and 17 different metabolites including 16:0 LPC were found between the sepsis group and the normal group(P<0.05).Among them,the plasma levels of sepsis patients were lower than those in thenormal group: 18:2LPA,20:4LPA,S1 P,16:0 LPC,18:2 LPC,18:1 LPC,18:0 LPC,16:0 LPE,18:2 LPE,18:1 LPE,and 18:0 LPE.The 16:0 cers,18:0 cers and 20:0 cers in the plasma of sepsis patients were highly expressed compared with the normal group(P<0.05).In Spearman correlation analysis,all the different lipid metabolite components shown by t test were correlated with sepsis(P < 0.05),and the correlation coefficient between 13 metabolites such as 18:2LPA,S1 P,16:0 LPC and sepsis fluctuated between 0.856-0.655,suggesting that the above lipid components were strongly correlated with sepsis.According to the principal component analysis,sepsis can be divided into four major components that have a great impact on sepsis,among which18:2 LPC,16:0LPE,18:2LPE,16:0LPC,S1 P and 20:0Cer have a high contribution value in the principal component(> 0.8,P<0.05).The Logistic regression model was-5.482 18:1 lpc-17.598 18:1LPE,indicating that there was a correlation between18:1LPC and 18:1LPE and sepsis(P<0.05).The analysis of neural network shows that the importance percentages of 20:4LPA,16:0 LPC,18:2LPC and 18:2LPA are100%,88.7%,53.4% and 37.3% respectively.PCT,TNFα,and il-13 were highly expressed in peripheral blood of patients with sepsis compared with the normal group(t=3.93,4.26,2.26,P < 0.05).Linear correlation analysis showed that il-13 and TNFα were positively correlated with PCT changes in the sepsis group.Four lipid metabolites,namely,0 LPC,18:1LPC,18:2LPA and 20:4LPA,were positively correlated with PCT content changes in the sepsis group(P<0.05).The aucs of PCT,TNF-,il-13,18:1LPC,16:0 LPC,18:2LPA and 20:4LPA were 1,0.871,0.999,0.918,0.913,0.857 and 0.679.Conclusion 1.16:0 LPC and 17 other differential lipid metabolites have an influence in sepsis.The four lipid metabolic components,namely 16:0 LPC,18:1LPC,18:2LPA and 20:4LPA,were strongly correlated with sepsis.2.Low expression of 18:2LPA,20:4LPA,S1 P,16:0 LPC,18:2 LPC,18:1 LPC,18:0LPC,16:0 LPE,18:2 LPE,18:1 LPE and 18:0 LPE in sepsis patients.The 16:0 cers,18:0 cers and 20:0 cers were highly expressed.The changes in the contents of varioussmall lipid molecules suggest that the differences in lipid metabolism during sepsis may be related to the different lipid metabolism processes in inflammatory reactions.3.Four lipid metabolic components,namely 16:0 LPC,18:1LPC,18:2LPA and20:4LPA,are involved in the cascade inflammatory response of sepsis.The AUC area was used to evaluate the accuracy of 18:1LPC and 16:0 LPC in the diagnosis of sepsis.Note: 18:1LPC and 16:0 LPC have similar diagnostic sensitivity to sepsis compared with PCT,which can be used as molecular markers for early screening of sepsis.