| Dextran40and Mannitol Injection was studied as the object in this paper. Thepharmacokinetics and pharmacodynamics characteristics of the compound injectionwas explored in rabbits, and the PK-PD modeling of the injection was established forthe first time,which provided a reliable theoretical basis for the clinical application.The concentration of dextran40could not be measured in plasma as it is apolymer. So a derivatization-gas chromatography （Derivatization-GC） was first setup to determinate mannitol’s concentration in rabbit’s plasma. The samplepre-treatment method,the choice of internal standard substance and the choice of thederivatization method were discussed. Ultimately protein precipitation method waschosen as the sample pre-treatment method, and xylitol was selected as an internalstandard substance. Specimens were then derivated by adding pyridine/aceticanhydride （1:1, v/v）100μL, heating at80°C for40min. The validation of themethod was tested by a series projects including specificity, sensitivity, precision,accuracy, stability. The mannitol peak and impurity peaks could be completelyseparated through this method, and impurities in the plasma did not interfere with thesample determination, which indicated the method had a high specificity. Theminimum detectable concentration was0.02mg mL-1（signal to noise ratio>3）;thelowest detection limit was0.05mg mL–1; the linear range was from0.05to10.00mg mL–1; the recovery was98.13%; the intraday and interday RSD were less than15.0%. In the room temperature texts, the freeze-thaw tests and the long-term stabilitytests, the determination of relative error （RE%）between the value and add value wereless than15%, in line with the biological sample analysis requirements. Dextran40inthe compound injection does not affect the determination of the results of themannitol.In the second part of this paper, the pharmacokinetics of mannitol of thecompound injection in rabbits was studied,as well as the dextran40’s influence onmannitol in vivo metabolic processes.12rabbits were randomly divided into twogroups, with crossover study of dual-cycle, dextran40and mannitol injection (according to rabbit weight inject mannitol2g kg-1, dextran0.6g kg-1) and mannitolinjection(mannitol2g kg-1) was injected into ear vein of each rabbit in10mL kg-1respectively. Then the concentration of mannitol in plasma was determinated and theplasma concentration-time curve was drawn. Total blood collection time was12h.Plasma concentration data were fitted by the Kinetica4.4software and results showedthat it was fitting the two-compartment model. After injected mannitol injection andthe compound injection, mannitol’s pharmacokinetic parameters in rabbits wererespectively: A5.53±2.09mg mL-1,5.11±1.20mg mL-1;B1.63±1.44mg mL-1,1.83±1.13mg mL-1;α1.21±0.60h-1,1.41±0.44h-1;β0.39±0.35h-1，0.36±0.08h-1;Cmax7.25±1.10mg mL-1,6.94±0.68mg mL-1;t1/20.87±0.15h,0.87±0.13h; AUC9.09±1.79mg mL-1 h,8.78±1.74mg mL-1 h.Each rabbit took itself as reference, the twogroups of pharmacokinetic parameters were processed by SPSS13.0statisticalsoftware.Bilateral t-test analysis results showed that the main pharmacokineticparameters of the two groups had no significant difference （P>0.05）. Thus, in thedextran40mannitol injection, dextran40does not affect the pharmacokinetic behaviorof mannitol in rabbits.In the third part of this paper, PK-PD models were built in mannitol injection anddextran40mannitol injection. The pharmacokinetic model was compartment model.The rabbit urine changes were treated as pharmacodynamic index in pharmacodynamicmodels. Combined with the the Sheiner effect compartment theory, Linear Model,Log-Linear Model, Emax Model and Sigmoid-Emax Model were used to fit the datarespectively.Then the parameters of each pharmacodynamic model were calculated andcompared to choose the best form of pharmacodynamic model. The best model of thetime course of drug effect was distinguished by using AIC, the residual sum of squares（SUM） and fitting degree （r2） theories, as well as the statistical techniques.The resultsshowed that mannitol plasma concentration-time curve was fitting atwo-compartment model; the diuretic effect of mannitol fell behind the concentration.There was an obvious counter-clockwise hysteresis loop between the drug effect anddrug concentration. Efficacy and effect-site concentration was most in line with theSigmoid-the Emax model. Diuretic pharmacodynamic parameters for the mannitolinjection were as follows: Emax17.7±3.2drop min-1, EC501.17±0.16mg mL-1,Ke02.82±0.43h-1. Pharmacodynamic parameters for dextran40mannitol injection were asfollows: Emax24.2±4.8drop min-1, EC501.21±0.32mg mL-1,Ke02.41±0.52h-1. Afterthe model fitting, in both groups the lag relationship of the effect-plasma drugconcentration（E-Cp） can be transformed into positive correlation relations of theeffect-plasma concentration in effect compartment （E-Ce）. Diuretic effect andplasma concentration in effect compartment could peaked of sync.Takingpharmacokinetic and pharmacodynamic parameters into the effect-time equations, effect-time process of prediction was close to the measured process.Through thecomparison between the two groups, the maximum diuretic effect(24.2±4.8drop min-1)of the compound group was significantly higher than that of mannitol group(17.7±3.2drop min-1), which indicated that efficacy enhanced by using Dextran40and mannitolin combination compared with using mannitol alone. Ke0in mannitol group (2.41±0.52h-1) was small than that of the compound group (2.82±0.43h-1), prompting thatthe lag time had been shortened and the drug worked in advance when using mannitoland dextran40used in combination.Methods for the determination of mannitol in rabbit plasma was successfullyestablished in this experiment,as well as pharmacokinetic-pharmacodynamic models.Dynamic variation of mannitol diuresis was studied quantitatively, which couldprovide valuable pharmacokinetic and pharmacodynamic parameters and provide areference for the development of rational clinical regimen. |
PDF Full Text Request