Daming Kong (1), Jeroen V. Koomen (1), Hong Su (1), Douglas J. Eleveld (1), Michel M.R.F. Struys (1, 2), Pieter J. Colin (1)
(1) Department of Anesthesiology, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands, (2) Department of Basic and Applied Medical Sciences, Ghent University, Ghent, Belgium
Objectives: Piperacillin-tazobactam is a β-lactam/β-lactamase inhibitor combination product used against severe infections in the intensive care unit setting[1]. However, pharmacokinetics (PK) of piperacillin typically depends on the type of patient population and disease progression[2,3]. This study aimed to describe the PK of piperacillin in different patient populations by a single model instead of multiple subgroup-specific ones.
Methods: Studies on piperacillin-tazobactam pharmacokinetics were identified through a PubMed search. Individual patient data was provided by the corresponding authors for 13 out of 58 identified studies[4-16]. Based on these data, a pooled population PK model was developed. Different patient characteristics were tested in the model as covariates to explain the observed variability in PK parameters across patients and studies.
Results: A two compartment model with covariates including weight, postmenstrual age and serum creatinine was found to best describe the data. V1, V2, CL and Q were 9.12 L kg-1, 12.8 L kg-1, 10.9 L (kg h)-1 and 18.7 L (kg h)-1 respectively for a 35-year-old, 70-kg patient with a serum creatinine level of 0.83 mg dL-1. With postmenstrual age, CL increases, reaching 50% of the maximum value at 1.12 years. Later in life, CL decreases with postmenstrual age, to 50% of maximum values at 83.9 years. Based on that, for every 1 mg dL-1 rise in serum creatinine, CL decreases 28.5%. Lonsdale et al.[3] also estimated CL of piperacillin with weight-based allometry and age-based maturation-decline with pooled data. In their study, CL for a 35-year-old, 70-kg patient was 12.7 L (kg h)-1, which was slightly larger than our estimate. The cause of the difference might be the absence of creatinine-based correction in their analysis. Besides, we found that the estimates of V1, V2 and CL for the study by Sime et al.[4] were different from other included studies[5-16]. Considering that the patients in the study by Sime et al. suffered from hematological malignancies, we hypothesized that they received high fluid volumes during the treatment, which resulted in higher CL estimate and lower estimates of V1 and V2 in the study by Sime et al. compared with other included studies. Given that the nonlinear elimination of piperacillin was reported in previous studies[17], single saturable elimination pathway was tested by a sigmoidal function in this work. However, our estimate of C50 (12836 mg L-1), describing the total plasma concentration reducing CL to half of its maximum value, was much greater than our observations (maximum = 813 mg L-1). With objective function value increasing by 0.704, no improvement was showed in fitting the data after using the nonlinear elimination. These results indicated that our data did not support the nonlinear PK of piperacillin.
Conclusions: A pooled population PK model was established, which can be used to predict PK exposure of piperacillin in a variety of patients. Based on this work, the dose-exposure-response model can be further optimized, which will help to further individualize piperacillin dosing regimens[18].
References:
[1] Hayashi Y, et al. Expert Opin Drug Metab Toxicol, 2010, 6(8): 1017-31.
[2] El-Haffaf I, et al. Clin Pharmacokinet, 2021, 60(7): 855-875.
[3] Lonsdale D O, et al. Br J Clin Pharmacol, 2019, 85(2): 316-346.
[4] Sime F B, et al. Antimicrob Agents Chemother, 2014, 58(6): 3533-7.
[5] Felton T W, et al. Clin Pharmacol Ther, 2014, 96(4): 438-48.
[6] Cohen-Wolkowiez M, et al. Antimicrob Agents Chemother, 2014, 58(5): 2856-65.
[7] Udy A A, et al. Crit Care, 2015, 19: 28.
[8] Taccone F S, et al. Crit Care, 2010, 14(4): R126.
[9] Sukarnjanaset W, et al. J Pharmacokinet Pharmacodyn, 2019, 46(3): 251-261.
[10] Roberts J A, et al. Int J Antimicrob Agents, 2010, 35(2): 156-63.
[11] De Cock P, et al. J Antimicrob Chemother, 2017, 72(7): 2002-2011.
[12] Tsai D, et al. Antimicrob Agents Chemother, 2016, 60(12): 7402-7406.
[13] Thibault C, et al. Antimicrob Agents Chemother, 2019, 63(11).
[14] Alobaid A S, et al. Antimicrob Agents Chemother, 2017, 61(3).
[15] Dhaese S a M, et al. Int J Antimicrob Agents, 2019, 54(6): 741-749.
[16] Weinelt F A, et al. PAGE 28 (2019) Abstr 9093 [www.page-meeting.org/?abstract=9093]
[17] Vinks A A, et al. Antimicrob Agents Chemother, 2003, 47(2): 541-7.
[18] Colin P J, et al. Clin Pharmacokinet, 2019, 58(6): 767-780.
Reference: PAGE 30 (2022) Abstr 10193 [www.page-meeting.org/?abstract=10193]
Poster: Drug/Disease Modelling - Infection