Stefanie Hennig(1,2), Joseph Standing(2,3), Christine Staatz(1), Alison H Thomson(4, 5)
1. School of Pharmacy, University of Queensland, Brisbane, Australia; 2.Department of Pharmaceutical Bioscience, Uppsala University, Uppsala, Sweden; 3.Infectious Diseases and Microbiology Unit, UCL Institute of Child Health, London, UK; 4.Strathclyde Institute of Pharmacy and Biomedical Sciences, University of Strathclyde, Glasgow, UK; 5.Pharmacy Department, Western Infirmary, Glasgow, UK
Objectives: While several studies have examined the pharmacokinetics (PK) of tobramycin in patients with cystic fibrosis (CF) (1-3), there is no common consensus on the following questions: do the PK of tobramycin differ in patients with and without CF; if so, can PK differences be explained by patient characteristics?
Methods: To answer these questions, a PK meta-analysis was undertaken of data from adults and children with and without CF, who received tobramycin by short intravenous infusion or bolus. A population PK approach was employed, utilising NONMEM 7, to establish PK parameters and identify potential covariates to explain between subject variability. Data were available from 4 published studies (3-6) and collected at 3 additional sites (Royal Children's Hospital, Brisbane Australia; Cincinnati Children's Hospital, Cincinnati, USA; and Gartnavel General Hospital, Glasgow, UK). Data from 732 patients were included providing 5605 tobramycin concentration-time points from 0.17 to 15 hours post-infusion.
Results: Tobramycin disposition was best described by a two-compartment model with first-order elimination. Patient gender significantly influenced tobramycin clearance (CL) and volume of the central compartment (Vc). CL was 7.4 L/h/70kg and 9.5 L/h/70kg and Vc was 21.4 L/70kg and 27.3 L/70kg in females and males, respectively. Inter-compartmental clearance (Q) and peripheral volume of distribution (Vp) were 0.486 L/h and 4.9 L respectively. Lean body weight (LBW) was superior to total body weight as a descriptor of CL (allometrically scaled with estimated exponent of 0.85) and of Vc (fixed exponent of 1). Patient age (modelled using a piece-wise linear model with a breakpoint at 18 years) and creatinine clearance (modelled using a power model) were also included as significant covariates in the final model. CF as an independent disease specific factor was not significant at any stage during covariate model building on CL, Vc or Vp. Model residual error was 20.5%. The final model showed excellent predictive properties in a VPC and no flaws in GOF plots.
Conclusions: The PK of tobramycin does not differ significantly in CF patients compared to patients without CF when subject age, LBW, gender and renal function are taken into consideration. Any differences in tobramycin dosing between CF and non-CF patients should be based on differences in expected pathogen sensitivity and can be optimised using an efficacy/toxicity utility function(6).
References:
[1] Touw DJ, et al. J Cyst Fibros. 2007;6:327-33.
[2] Campbell D, et al. Ther Drug Monit. 1999 ;21:281-8.
[3] Hennig S, et al. . Br J Clin Pharmacol. 2008 ;65:502-10.
[4] Massie J, et al. Journal of paediatrics and child health. 2006 ;42:601-5.
[5] Aarons L, et al. Br J Clin Pharmacol. 1989 ;28:305-14.
[6] Standing JF, et al. http://wwwpage-meetingorg/?abstract=1401; Marseille, France, 2008.
Reference: PAGE 21 () Abstr 2415 [www.page-meeting.org/?abstract=2415]
Poster: Infection