Population Pharmacokinetics And Pharmacodynamics Research Of Piperacillin/Tazobactam In Patients In Nephrology Department

ObjectivePiperacillin/tazobactam has been widely used in infectious diseases of the nephrology department. However, the renal function impairment is very common in the patients of nephrology department. As a special group, these patients’ drug disposition progress is not only different from other groups, but also shows a wide variance within the group, which makes it necessary to adjust the individualized dosage regimen of antibiotics based on the pharmacokinetics (PK) and pharmaco-dynamics (PD) properties. In this study, an HPLC method was established for the determination of the piperacillin and tazobactam in human plasma of patients with different renal function; and nonlinear mixed effect model(NONMEM) method was used to establish the population pharmacokinetics (PPK) model of piperacillin/ tazobactam in the specific group of nephrology department.This PPK model makes it highly possible to design rational dosage regimens for individuals during anti-infective therapy.MethodsDetermination of the piperacillin/tazobactam in plasma by HPLC:The mobile phase for piperacillin was 0.01 mol·L-1 sodium dihydrogen phosphate (pH6.5)-acetonitrile (75:25, v/v), and was 0.01 mol·L-1 sodium dihydrogen phosphate (pH6.5)-acetonitrile (90:10, v/v) for tazobactam. Other chromatographic conditions included:flow rate (1.0 ml·min-1),column temperature(30℃), detection wavelength (220 nm) and injection volume(20 μL). The specificity, linear correlation, recovery, precision and stability were evaluated.Establishment and authentication of population pharmacokinetic model in the patients of nephrology department:Patients were enrolled according to the inclusion and exclusion criteria and all the participants signed the informed consent before treatment. Different dosage regimens were given according to the severity of infection, the results of bacterial culture and the renal function. When steady state was reached in patients with intravenous injection after three days, venous blood samples were collected before the injection,0.5 h and 1-6 h after injection. Blood samples of each patient should be more than two. NONMEM was applied to obtain the final PPK model and pharmacokinetic parameters of piperacillin/tazobactam in patients with infections in nephrology department by the following steps:basic structure model, statistical model, screening of covariants and model validation. Evaluate the fixed effects and the random effects on the clearance rate and the apparent volumn of distribution of piperacillin/tazobactam and establish the full amount of regression models of piperacillin/tazobactam by stepwise regression, and then build the final model via the backward regression.Bacterial resistance monitoring and strategy of optimizing dosage regimen by Monte Carlo simulation:Retrospective statistical method was performed to analyze the data of bacteria culture and drug resistance in nephrology department of our hospital from 2011 to 2013 through the WHONET5.4 software. Monte Carlo simulation was conducted by Crystal Ball software7.2.2 to determine the probability of target attainment for different dosage regimens. Pharmacodynamic target of piperacillin/tazobactam is %fT>MIC≥50%, the PTA values of higher than 90% were considered to achieve satisfactory therapeutic effect. Simulated regimens included 2.25g q8h,2.25g q6h,3.375g q12h,3.375g q8h,4.5g q12h and 4.5g q8h with traditional intravenous infusion (30min) and prolonged intravenous infusion (3h).ResultsAccording to FDA’s guideline for bioanalytical method validation, all the criterias for validation have been met with this method. The retention time for piperacillin and tazobactam was 4.2 and 3.7 min, respectively. The endogenous substances had no influence on the determination. The calibration range was linear from 6.25 to 400 mg·L-1 for piperacillin and from 1 to 100 mg·L-1 for tazobactam. Intra-and inter-day precision for piperacilln was less than 6.35% and 8.06%, respectively. Intra- and inter-day precision for tazobactam was less than 5.72% and 4.44%, respectively.The accuracy for piperacillin and tazobactam was within the range of 92.83～102.93%. Extraction recoveries for piperacillin and tazobactam in the plasma were 84.29±4.59±%、85.66±2.80%, respectively. The results of stability test also met all requirements for determination.One-compartment model has been applied for estimating values of population pharmacokinetic parameters. The established PPK final regression model is as follow: Piperacillin CL=9.26+4.54(CLCR/687) (L·h-1), V=12.30+9.54(WT/61.1) (L); tazobactam CL=5.84+3.52(CLCR/68.7) (L·h-1), V=7.97+8.05(WT/61.1)(L). PPK parameters are depicted as the population value and between subject variability(BSV):CL=13.80 L·h-1, BSV=10.7%; V=21.84 L, BSV=7.6% for piperacillin, and CL=9.36 L·h-1, BSV=8.9%; V=16.02L, BSV=5.2% for tazobactam.1230 strains of pathogens were isolated from 2011 to 2013. Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa were the top three among the Gram-negative bacteria and their detection rates were 46.50%,7.07% and 4.15%, respectively.The drug resistance rates of three bacteria to amikacin, piperacillin/ tazobactam and carbapenems were low, and the average drug resistance rates to piperacillin/tazobactam were 3.84%,11.24% and 11.76%. In Monte Carlo simulation study, for the infections caused by E.coli, the following regimens reached the PTA values of higher than 90%:2.25g q8h PIT,2.25g q6h TIT/PIT,3.375g q8h TIT/PIT, 4.5g q12h PIT,4.5g q8h TIT/PIT. For K.pneumonia, the PTA values remained higher than 90% when the regimens of 2.25g q8h PIT,2.25g q6h TIT/PIT,3.375g q8h PIT, 4.5g q8h PIT were conducted. Only three regimens of 2.25g q6h,3.375g q8h and 4.5g q8h in PIT met requirements as to P.aeruginosa.Conclusions1. The HPLC method has been developed and validated for the determination of piperacillin/tazobactam in human plasma. The validated method was proved to be simple,sensitive and rapid.2. The PKK model of piperacillin/tazobactam in patients with infectious diseases of Nephrology Department has been established by NONMEM. Creatinine clearance and weight had significant effects on pharmacokinetic parameters.The model was authenticated to have satisfied stability and predictive efficacy, and is reliable to provide basis on the individualized dosage regimens.3. Pharmacodynamic evaluation was performed for the traditional and prolonged intravenous infusion by the Monte Carlo simulation. The results showed that compared to the traditional regimens, the optimal PTAs can be obtained through increasing administration frequency and prolonged intravenous infusion, which could be considered as the empiric antibiotic therapy for severe infections.