中华胰腺病杂志
中華胰腺病雜誌
중화이선병잡지
CHINESE JOURNAL OF PANCREATOLOGY
2014年
5期
293-298
,共6页
许宏敏%张磊%康华%王宇凡%吴静%刘怀平%周淑芬%张凤美%刘树业
許宏敏%張磊%康華%王宇凡%吳靜%劉懷平%週淑芬%張鳳美%劉樹業
허굉민%장뢰%강화%왕우범%오정%류부평%주숙분%장봉미%류수업
胰腺炎%代谢组学%胆石%色谱法,液相%质谱法
胰腺炎%代謝組學%膽石%色譜法,液相%質譜法
이선염%대사조학%담석%색보법,액상%질보법
Pancreatitis%Metabolomics%Gallstones%Chromatography,liquid%Mass spectrometry
目的 分析轻症急性胰腺炎(MAP)患者血清代谢组学,筛选出具有潜在临床诊断价值的代谢标志物.方法 应用超高效液相色谱与高分辨质谱联用(UPLC-HRMS)分析平台对38例MAP患者、26例胆石症患者及36例健康志愿者的血清进行分析,构建主成分分析(PCA)及正交偏最小二乘(OPLS-DA)模型,并进行模型验证,筛选差异代谢物.应用支持向量机(SVM)分类模型及受试者工作特征曲线下面积(AUC)判断差异代谢物对MAP的诊断价值.同时对其中11例MAP患者入院时、治疗中期(治疗后第6天)、治愈出院前的血清代谢组学进行动态观察,筛选与MAP病程监测相关的差异代谢物.结果 从总共122个血清样本中提取出内源性代谢物432个.构建了“MAP-胆石症-健康对照”的PCA(R2X =77.1％,Q2=23.5％)及OPLS-DA模型(R2Y =84.0％,Q2=62.8％),通过验证筛选出49个差异代谢物.经非参数检验确定了在3组之间均存在着显著性差异的12个差异代谢物.由这12个差异代谢物分别构建MAP组与胆石症组、MAP组与对照组的二级SVM分类模型,两个模型的敏感度、准确度均为100％,均无假阳性.12个差异代谢物的AUC均大于实验室常用的淀粉酶、脂肪酶的AUC.经过物质鉴定,结果可靠的有鞘氨醇、辛酰胆碱、甘氨胆酸、十四烷酸、癸酰基胆碱、十二烷醇、2-十四酮、L-甲状腺原氨酸8个差异代谢物,它们在MAP时的血清含量均较健康者升高.与胆石症患者比较,MAP患者血清鞘氨醇、辛酰胆碱、十四烷酸、癸酰基胆碱、十二烷醇、2-十四酮含量显著升高,而甘氨胆酸、L-甲状腺原氨酸含量显著下降.经动态观察,鞘氨酸、L-甲状腺原氨酸、甘氨胆酸及2-十四酮4种代谢物随着治疗进程均呈逐渐下降趋势,出院前的水平均较入院时水平显著下降,差异均有统计学意义(P值均＜0.05).结论 利用代谢组学分析方法发现8个对MAP具有潜在诊断价值的代谢物,其中4个可用于病程监测.
目的 分析輕癥急性胰腺炎(MAP)患者血清代謝組學,篩選齣具有潛在臨床診斷價值的代謝標誌物.方法 應用超高效液相色譜與高分辨質譜聯用(UPLC-HRMS)分析平檯對38例MAP患者、26例膽石癥患者及36例健康誌願者的血清進行分析,構建主成分分析(PCA)及正交偏最小二乘(OPLS-DA)模型,併進行模型驗證,篩選差異代謝物.應用支持嚮量機(SVM)分類模型及受試者工作特徵麯線下麵積(AUC)判斷差異代謝物對MAP的診斷價值.同時對其中11例MAP患者入院時、治療中期(治療後第6天)、治愈齣院前的血清代謝組學進行動態觀察,篩選與MAP病程鑑測相關的差異代謝物.結果 從總共122箇血清樣本中提取齣內源性代謝物432箇.構建瞭“MAP-膽石癥-健康對照”的PCA(R2X =77.1％,Q2=23.5％)及OPLS-DA模型(R2Y =84.0％,Q2=62.8％),通過驗證篩選齣49箇差異代謝物.經非參數檢驗確定瞭在3組之間均存在著顯著性差異的12箇差異代謝物.由這12箇差異代謝物分彆構建MAP組與膽石癥組、MAP組與對照組的二級SVM分類模型,兩箇模型的敏感度、準確度均為100％,均無假暘性.12箇差異代謝物的AUC均大于實驗室常用的澱粉酶、脂肪酶的AUC.經過物質鑒定,結果可靠的有鞘氨醇、辛酰膽堿、甘氨膽痠、十四烷痠、癸酰基膽堿、十二烷醇、2-十四酮、L-甲狀腺原氨痠8箇差異代謝物,它們在MAP時的血清含量均較健康者升高.與膽石癥患者比較,MAP患者血清鞘氨醇、辛酰膽堿、十四烷痠、癸酰基膽堿、十二烷醇、2-十四酮含量顯著升高,而甘氨膽痠、L-甲狀腺原氨痠含量顯著下降.經動態觀察,鞘氨痠、L-甲狀腺原氨痠、甘氨膽痠及2-十四酮4種代謝物隨著治療進程均呈逐漸下降趨勢,齣院前的水平均較入院時水平顯著下降,差異均有統計學意義(P值均＜0.05).結論 利用代謝組學分析方法髮現8箇對MAP具有潛在診斷價值的代謝物,其中4箇可用于病程鑑測.
목적 분석경증급성이선염(MAP)환자혈청대사조학,사선출구유잠재림상진단개치적대사표지물.방법 응용초고효액상색보여고분변질보련용(UPLC-HRMS)분석평태대38례MAP환자、26례담석증환자급36례건강지원자적혈청진행분석,구건주성분분석(PCA)급정교편최소이승(OPLS-DA)모형,병진행모형험증,사선차이대사물.응용지지향량궤(SVM)분류모형급수시자공작특정곡선하면적(AUC)판단차이대사물대MAP적진단개치.동시대기중11례MAP환자입원시、치료중기(치료후제6천)、치유출원전적혈청대사조학진행동태관찰,사선여MAP병정감측상관적차이대사물.결과 종총공122개혈청양본중제취출내원성대사물432개.구건료“MAP-담석증-건강대조”적PCA(R2X =77.1％,Q2=23.5％)급OPLS-DA모형(R2Y =84.0％,Q2=62.8％),통과험증사선출49개차이대사물.경비삼수검험학정료재3조지간균존재착현저성차이적12개차이대사물.유저12개차이대사물분별구건MAP조여담석증조、MAP조여대조조적이급SVM분류모형,량개모형적민감도、준학도균위100％,균무가양성.12개차이대사물적AUC균대우실험실상용적정분매、지방매적AUC.경과물질감정,결과가고적유초안순、신선담감、감안담산、십사완산、계선기담감、십이완순、2-십사동、L-갑상선원안산8개차이대사물,타문재MAP시적혈청함량균교건강자승고.여담석증환자비교,MAP환자혈청초안순、신선담감、십사완산、계선기담감、십이완순、2-십사동함량현저승고,이감안담산、L-갑상선원안산함량현저하강.경동태관찰,초안산、L-갑상선원안산、감안담산급2-십사동4충대사물수착치료진정균정축점하강추세,출원전적수평균교입원시수평현저하강,차이균유통계학의의(P치균＜0.05).결론 이용대사조학분석방법발현8개대MAP구유잠재진단개치적대사물,기중4개가용우병정감측.
Objective A serum metabolic profiling analysis was performed on mild acute pancreatitis (MAP) patients to select potential biomarker with clinical diagnostic value.Methods A UPLC-MS platform was introduced for serum metabolic profiling analysis on 38 mild acute pancreatitis (MAP) patients,26 cholelithiasis patients and 36 healthy volunteers.Principal component analysis (PCA) model and orthogonal partial least squares discriminate analysis (OPLS-DA) model was established and validated to select characteristic metabolites.Support vector machine (SVM) classification model and area of curve (AUC) was used to determine the diagnostic value of the metabolites.At the same time,for 11 patients with MAP,the serum metabolic condition was dynamically observed,and potential biomarkers related to disease course monitoring were selected.Result Among the 122 serum samples,432 endogenous metabolites were selected,then a PCA model (R2X =77.1％,Q2 =23.5％) and an OPLS-DA model (R2Y =84.0％,Q2 =62.8％) of "MAP patients-cholelithiasis patients-health control" was established,49 different metabolites were validated.After non-parametric test,12 characteristic metabolites detected with significant differences in serum level in three groups were found.The secondary SVM classification model of MAP and cholelithiasis,MAP and control group consisted of these 12 characteristic metabolites had sensitivity,specificity of 100％,and no false positivity was detected.The AUC of these 12 characteristic metabolites was higher than amylase and lipase.After material identification,the levels of sphingosine,symplectic phosphatidyl choline,tetradecanoic acid,decyl acyl choline,dodecanol,2-tetradecanone were significantly increased,while glycocholic acid,L-thyronine were significantly decreased.Dynamic observation showed sphingosine,L-thyronine,glycocholic acid and 2-tetradecanone gradually decreased with the treatment course,and the levels before discharge were significantly decreased when compared with those at admission (P ＜ 0.05).Conclusions After metabolic profiling analysis on MAP patients,there are 8 potential biomarkers detected with outstanding differential diagnoses ability for MAP and 4 biomarkers can be used for disease course monitoring.
