【作者】 张勇；

【导师】 彭曦；

【作者基本信息】 第三军医大学 ， 外科学， 2014， 博士

【摘要】 背景顽固而持久的高代谢反应是重症烧伤患者最显著的代谢特征，其主要的病理表现为能量消耗和分解代谢显著增强，并伴有营养底物利用障碍。目前临床常用的营养治疗方案并未充分考虑患者对营养底物的耐受能力，大量补充的营养物质不仅不能被很好利用，还可加重代谢紊乱。在这种营养支持模式下，患者自噬代谢依旧，机体不断消耗，预后不良。目前烧伤营养治疗最大的难点在于对患者代谢规律的认识不够深入，营养治疗的针对性和有效性还有待进一步提高。准确把握烧伤患者代谢规律是进行合理营养治疗的前提，现有的检测手段难以满足快速获取患者完整代谢信息的需求。代谢组学研究为全面系统地了解烧伤后机体代谢变化规律提供了有效手段。目的利用1H-NMR波谱技术绘制重症烧伤患者代谢指纹图谱，了解烧伤代谢基本规律，通过数据建模和判别分析筛选出烧伤患者特异性代谢靶标，并希望建立起能预判烧伤患者预后的代谢组学公式。在此基础上，通过对烧伤大鼠代谢指纹的分析，并结合多项生理生化检测指标，以期找到最佳的烧伤动物营养支持方式，为优化临床营养治疗方案奠定基础。方法1.重症烧伤患者代谢指纹图谱的绘制：按照纳入和剔除标准选取21例重症烧伤患者为病例组和10例正常健康人为对照组，患者在烧伤后第1、3、7、14、21、28和35天留取血浆标本，进行核磁共振检测(NMR)，获得患者各时相的所有核磁共振波谱数据，采用组成分分析（PCA）、偏最小二乘法（PLS）以及偏最小二乘法-判别分析（PLS-DA）法对数据进行模式识别，绘制出重症烧伤患者代谢指纹图谱。2.在获得烧伤代谢指纹图谱的基础上，对不同烧伤病程、不同损伤程度及不同临床结局的患者的数据进行挖掘分析，在人类代谢组学数据库（HMDB）中筛选出能体现烧伤代谢特征的标志物，并通过数学模型处理建立起能预判患者预后的代谢组学公式。3.建立重度烧伤大鼠模型，采用代谢组学分析判别技术和静息能量消耗(REE)、血糖、糖耐量、创面细菌定量等多种检测方法，观察不同能量供给和不同营养素配比以及谷氨酰胺强化后对机体代谢和预后的影响，筛选出最佳的烧伤动物营养支持方案。结果1.成功绘制出重症烧伤患者代谢指纹图谱，发现重症烧伤患者与正常对照、烧伤休克期与其他时期、死亡和存活患者之间的在代谢模式上存在显著差异；2.发现烧伤休克期和健康人血浆中有12种重要代谢标志物存在明显差异，烧伤休克期和休克后各时期的差异性代谢标志物有15种，死亡和存活患者之间存在明显差异的代谢靶标有25种；3.通过代谢组学研究方法对不同结局患者的代谢特征进行分析，以及采用模式识别技术，建立了烧伤患者死亡风险预测的代谢组学公式：Y=-1.1568X-5.4133；4.烧伤大鼠代谢组学数据建模及代谢模式分析显示，不同能量供给、非蛋白热卡不同配比以及是否使用谷氨酰胺在代谢指纹图谱上存在显著差异。5.对烧伤大鼠多种营养支持方案进行比较，发现最佳的能量供给方案为：在伤后第1周能量供给应达到能量消耗的75%，第2周能量供给与消耗应一致。最佳的糖，脂肪和蛋白比例为70%：10%：20%。此外，添加谷氨酰胺的肠内营养能减轻烧伤后脏器损伤，降低代谢率，有利于机体修复。结论1.基于核磁共振波谱分析的代谢组学技术为准确了解烧伤后代谢变化提供了有效手段，通过该方法能将复杂的代谢网络转变为直观的代谢指纹图谱，并有可能为烧伤临床治疗提供能反映烧伤损伤程度的定量判断工具；2.明确了正常与烧伤、休克期与休克后期、死亡与存活患者之间的代谢差异，找到了一些重要的代谢标志物；3.成功建立了能初步判断烧伤患者是否具有死亡风险的代谢组学公式：Y=-1.1568X-5.4133。4.烧伤大鼠最佳营养治疗模式为：伤后第1周给予75%REE，第二周给予100%REE的能量供给，糖、脂肪和蛋白供热比为：70%：10%：20%，尽早使用谷氨酰胺强化的肠内营养。更多还原

【Abstract】 Background:Persistent and irreversible hypermetabolic response is the main metaboliccharacteristics in severe burned patients. After burn injury, the primary pathologicalphenomena are energy consumption and catabolism increased significantly, and nutrientsutilize barriers. Current clinical nutritional therapy not fully consider the tolerance degree ofnutrition in severe burned patients. A lot of exogenous nutrients can not be utilized well,and may aggravate metabolic disorder. The result is the autophagy metabolism, continuousconsumption, and progressive emaciation continuity. Now the major obstacle in burnnutrition therapy is the understanding of metabolic change characteristics postburn notdeep enough. The pertinence and effectiveness of nutrition therapy remains to be furtherimproved. An accurate knowledge of metabolic pattern change characteristics postburnis a prerequisite for rational nutrition therapy. The current testing methods can not meetrapid access to complete metabolic information. Metabolomics provide an effective tool forcomprehensive and systematic understanding of metabolic pattern changes after burn injury.Objective:To establish plasma metabolomics fingerprint spectrum in severe burn patients using1H-NMR spectroscopy. To understand the basic rules of severe burn metabolism, throughdata mining, data modeling and pattern recognition analysis to screen a set of biomarkersthat could be used for prognostic prediction in severe burned patients，and establish adiscriminant formula for predicting prognosis. Using the advantages of metabonomics tojudge the rationality of the new nutritional therapy protocol by severe burned rats model,and optimize current burn nutritional therapy.Methods:1. Establish a plasma metabolomics fingerprint spectrum in severe burned patients.According to including and excluding standard, twenty-one severe burned patients and tenhealthy control individuals were enrolled in this study. The plasma samples from patients and healthy individuals were collected for NMR measurements. The NMR spectra data waspattern recognized by using principal component analysis and partial least squares fordrawing burn patients metabolomics fingerprint.2. Base on establishing severe burn patients metabolic fingerprint, further mining andanalyze the data of different disease course, damage degree, and clinical prognosis.Screening metabolic markers and establish death prediction formula for judging prognosisthrough The Human Metabolome Database.3. The effects of different energy supplement, variety of nutrients ratio and glutamineadminstration on metabolism and prognosis were observed and the best nutritional supportscheme was screened by using metabonomics analysis technology and resting energyexpenditure (REE), blood glucose, glucose tolerance, bacterial quantitative detectionmethods on burned rats model.Results1. Successfully established the severe burn patients metabolic fingerprint. Theresults showed that the metabolic patterns were significant difference among burnedpatients, healthy individuals, burn shock stage and burn other stage, as well as death andsurvival patients.2. Using metabolomics techniques, we found an eigen-metabolome that consists of12metabolites to distinguish burned patients and healthy individuals, and15metabolites todifferentiate burn shock stage and other stage, as well as25metabolites to identify deathand survival patients.3. Analyze the data of different outcomes by metabolomics techniques, establishedmetabonomics formula for predicting death in severe burned patients: Y=-1.1568x-5.4133.4. NMR spectra of the plasma samples showed that there were significant differenceamong different dose of nutrition support, different nutrients ratio and glutamine treatment.5. Animal experimental results showed that the best nutritional support scheme wasenergy supplement should reach75%REE in the first week after burn, and reach100%REEin the second week. The best proportion of sugar, fat and protein was70%：10%：20%.More over, enteral nutrition with glutamine administration was beneficial to body recoveryafter burn injury. Conclusion1. Metabolomics techniques based on NMR can be used to monitor metabolism insevere burn patients. The integrated1H-NMR metabolome and metabolic network analysisis useful for visualizing complex metabolic disturbances after severe burn injury and mayprovide a new quantitative injury severity evaluation for future clinical use.2. NMR spectra of the plasma samples showed significant differences between burnedpatients and healthy individuals, burn shock stage and other stage. Using metabolomicresearch methods find some metabolic markers for predicting death.3. Successfully established prognostic prediction formula in severe burned patients:Y=-1.1568x-5.4133.4. Create the best nutritional support scheme base on burn rat model: energysupplement should reach75%REE in the first week after burn, and reach100%REE in thesecond week, and the best proportion of sugar, fat and protein was70%：10%：20%. Moreover, give early enteral nutrition supplemented with glutamine as soon as possible.更多还原