基于核磁共振氢谱建立多发伤患者微循环损伤程度的预测模型

Establishment of a Prediction Model of Microcirculation Injury Degree in Patients with Multiple Injuries Based on Nuclear Magnetic Resonance Hydrogen Spectroscopy

目的应用以核磁共振氢谱(HNMR)技术为基础的代谢组学方法,结合机器学习,建立多发伤患者微循环损伤程度的预测模型,从而达到对多发伤患者进行及早精准治疗的目的。方法本研究纳入2021年6月至2022年5月四川省人民医院急诊科收治的30例多发伤合并失血性休克患者,在患者进入急诊抢救室后2 h内采集血清样本,经预处理后采集HNMR数据。根据休克指数(SI)将多发伤患者分为A组(SI≥1.5)和B组(1.5>SI≥1),利用MestReNova软件提取两组的HNMR数据,再利用MetaboAnalyst 5.0进行偏最小二乘判别分析(PLS-DA),结合人类代谢组数据库(HMDB)和京都基因和基因组百科全书(KEGG),找出能够区分两组的代谢物及相关的代谢途径。结果A组共纳入20例,B组共纳入10例。与B组比较,A组的烟酰胺、L-5-羟色氨酸、3-羟基苯乙酸、4-羟基苯甲醛、烟尿酸、L-犬尿氨酸、N-甲基烟酰胺、烟酸、5-羟基吲哚、伞形花序酮、N-乙酰血清素、N-氧代烟酰胺、肌苷水平升高;L-乙酰肉碱水平降低(P<0.05)。通路富集分析得出色氨酸代谢及烟酸和烟碱胺代谢途径在两组间的差异均有统计学意义(P<0.05),其中色氨酸代谢途径的影响最大(影响因子=0.29)。结论本研究首次通过检测不同休克指数多发伤患者血清,找到了组间的代谢差异,为今后使用其他更方便的样本建立多发伤患者微循环损伤程度的预测模型提供了理论支持,但本研究样本量偏小,未来还需要对更大的样本量进行定量实验。

Objective The metabolomics method based on nuclear magnetic resonance hydrogen spectroscopy(HNMR)technology and machine learning were used to establish a prediction model for the degree of microcirculation injury in patients with multiple injuries,so as to achieve the purpose of early and precise treatment for patients with multiple injuries.Methods A total of 30 patients with multiple injuries complicated with hemorrhagic shock who were admitted to the Emergency Department of Sichuan Provincial People s Hospital from June 2021 to May 2022 were included in this study.Serum samples were collected within 2 hours after admission to the emergency room,and nuclear magnetic resonance hydrogen spectrum were collected after pretreatment.According to the shock index(SI),patients with multiple injuries were divided into group A(SI≥1.5)and group B(1.5>SI≥1).MestReNova software was used to extract the nuclear magnetic resonance hydrogen spectrum data of the two groups,and MetaboAnalyst 5.0 was used to perform partial least squares discriminant analysis(PLS-DA).The Human Metabolome Database(HMDB)and Kyoto Encyclopedia of Genes and Genomes(KEGG)were combined to identify the metabolites and related metabolic pathways that could distinguish the two groups.Results A total of 20 cases were included in group A and 10 cases in group B.Compared with group B,the levels of Niacinamide,5-Hydroxy-L-tryptophan,3-Hydroxyphenylacetic acid,4-Hydroxybenzaldehyde,Nicotinuric acid,L-Kynurenine,N-Methylnicotinamide,nicotinic acid,5-Hydroxyindole,Umbelliferone,N-Acetylserotonin sulfate,Nicotinamide N-oxide,and inosine in group A were increased,while the level of L-Acetylcarnitine was decreased(all P<0.05).Pathway enrichment analysis showed that there were statistically significant differences in the tryptophan,niacin and nicotinamide metabolism pathways between the two groups(P<0.05),among which,the tryptophan metabolism pathway had the greatest impact(Impact=0.29).Conclusion This study,for the first time,demonstrates the metabolic differences between groups by detecting the serum of patients with multiple injuries with different shock indices,providing theoretical support for establishing prediction models of microcirculation damage degree of patients with multiple injuries with other more convenient samples in the future.However,the sample size of this study is small quantitative experiments with larger sample sizes are needed in the future.