药物非房室模型分析.ppt

《药物非房室模型分析.ppt》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。

1、NONCOMPARTMENTALANALYSISDeficienciesofcompartmentalanalysis:Lackofmeaningfulphysiologicalbasisforderivedparameters.Lackofrigorouscriteriatodetermine#ofcompartmentsnecessarytodescribedisposition.Lackofabilitytoelucidateorganspecificelimination.Inabilitytorelatederivedparameterst

2、oquantifiablephysiologicalparameters.Inabilitytopredictimpactofpathophysiology.Inabilitytoprovideinsightintomechanismofdrug-druganddrug-nutrientinteractions.Highlysensitivetosamplingfrequency.1GENERALPRINCIPLESOFSTATISTICALMOMENTSMOMENT:Amathematicaldescriptionofadiscretedistri

3、bution.STATISTICALMOMENTS:UtilizedinchemicalengineeringtodescribeflowdataFirstappliedtobiologicalsystemsbyPerlandSamuelin1969todescribethekineticsofcholesterol2ExamplesofStatisticalMomentUsageInstatisticsInphysicsM0NweightM1(mean)CenterofmassM2(variance)MomentofinertiaM3(skewne

4、ss)M4(kurtosis)3Instatistics,themeanisameasureofasamplemeanandisactuallyanestimateofthetruepopulationmean.Inpharmacokinetics,wecancalculatethemomentofthetheoreticalprobabilitydensityfunction(i.e.,thesolutionofadifferentialequationdescribingtheplasmaconcentrationtimedata),orweca

5、ncalculatemomentsfrommeasuredplasmaconcentration-timedata.Thesecurvesarereferredtoassamplemomentsandareestimatesofthetruecurves.4AssumeatheoreticalrelationshipofC(t)asafunctionoftime.Thenon-normalizedmoments,Sr,abouttheoriginarecalculatedas:5Non-normalizedmomentsKineticparamete

6、rAUCAreaunderthecurveAUMCAreaunderthemomentcurve6From:RowlandM,TozerTN.ClinicalPharmacokinetics–ConceptsandApplications,3rdedition,WilliamsandWilkins,1995,p.487.7NormalizedmomentsKineticparameterFirstmoment:MRTMeanresidencetime8AREADETERMINATIONA.IntegrationofSpecificFunctionMu

7、stelucidatethespecificfunctionInfluencedbythequalityofthefit9B.NumericalIntegrationLineartrapezoidalLogtrapezoidal10B.NumericalIntegrationLineartrapezoidal11B.NumericalIntegrationLineartrapezoidalAdvantages:Simple(cancalculatebyhand)Disadvantages:Assumesstraightlinebtwndatapoin

8、tsIfcurveissteep,errormaybelargeUnderoroverestimatedep