Screening of lipid biomarkers in septic patients with different survival outcome
10.3760/cma.j.cn101721-20220315-000067
- VernacularTitle:不同生存结局脓毒症患者脂质生物标志物的筛选
- Author:
Jifang LIANG
1
;
Shan WANG
;
Xiuzhe WANG
;
Haipeng SHI
;
Meini JIANG
;
Jing LI
;
Wenjing WU
;
Caixia ZHAO
;
Weidong WU
Author Information
1. 山西白求恩医院 山西医学科学院 同济山西医院 山西医科大学第三医院重症医学科，太原 030032
- Keywords:
Sepsis;
Lipidomics;
Survival outcome
- From:
Clinical Medicine of China
2022;38(5):414-419
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To screen lipid biomarker in sepsis patients with different survival outcome based on ultra high performance liquid chromatography-mass spectrometry(UHPLC-MS/MS) technique.Methods:From September 2019 to April 2020, 30 septic patients admitted in Department of Intensive Care Unit and 30 cases of physical examination at the same time in Shanxi Bethune Hospital were studied. Lipid metabolite in serum were detected by UHPLC-MS/MS technique. According to the 28 day survival outcome of sepsis patients, they were divided into survival group (21 cases) and death group (9 cases). The baseline data of case group and control group, survival group and death group were compared respectively. Independent sample t-test and orthogonal partial least squares discriminant analysis (OPLS-DA) were further performed to identify lipid biomarkers related to sepsis survival outcome. Receiver operating characteristic (ROC) curve to evaluate the predictive efficacy of differential lipids on the survival outcome of biomarker sepsis patients. Results:There were 32 lipid subclasses and 1 437 differential lipid molecules in the sepsis group compared with the control group. 196 differential lipid molecules in the sepsis survival group and the death group were screened according to the OPLS-DA model (variable weight of projection (VIP)>1), which were glycerophosphingolipids (129), sphingolipids (52), glycerides (14), and sterols (1).All the original data were statistically analyzed by univariate independent sample t-test. There were statistically significant differences in 15 lipid molecules between the two groups. Combined with VIP > 1 and P < 0.01, three lipid molecules were finally screened, which were sphingomyelin (SM) lipid molecules, SM (d30∶1), SM (d32∶2), SM (d32∶1). ROC curve analysis showed that the areas under curves of the above three lipid molecular were 0.915, 0.892, 0.898, respectively. The sensitivity was 77.27%, 95.45%,72.73%. The specificity was 100.0%, 87.5%,100.0%. Further Z-test showed that there was no significant difference in the area under the ROC curve ( Z(SM (d30∶1) and SM (d32∶1)) =0.36, P=0.722; Z(SM (d30∶1) and SM (d32∶2))=0.34, P=0.732; Z(SM (d32∶1) and SM (d32∶1))=0.07, P=0.942). Conclusions:Sphingomyelin may be involved in the formation of different clinical outcomes of sepsis, and has a good predictive effect on the survival outcome of sepsis.
