Development and use of predictive population pharmacokinetic models for propofol dose selection in the obese

Propofol is an intravenous anaesthetic agent used for the induction and maintenance of general anaesthesia and sedation. Current strategies for dosing propofol incorporate either total body weight (TBW) or inaccurate measures of body composition in their calculations, which result in excessive doses in the obese. The global aim of this thesis was to explore the use of lean body weight (LBW) to describe propofol clearance (CL), and thus investigate its potential to be used as a propofol dosing metric across the whole adult population. The thesis begins with a review of the literature and introduction in Chapter 1, including a background on obesity, pharmacokinetic-pharmacodynamic (PKPD) analysis and covariate selection, body size metrics and a review of propofol PK and PD. The chapter concludes with a statement of the specific research aims. This is followed by a commentary in Chapter 2 on the inadequacy of current methods of propofol dosing for the obese population and encouragement for development of universal dosing strategies that consider body composition. Chapter 3 describes a simulation-estimation study in which LBW was substituted as a covariate for a nonlinear TBW relationship on CL in a prior PK model for propofol. This was performed as a preliminary study to explore the ability of LBW to describe a reported nonlinear increase in CL with TBW in the absence of original PK data. Results showed that the LBW model had similar predictive properties to the empirically derived original model, suggesting that LBW may provide a mechanistic explanation for a nonlinear increase in propofol CL with TBW. Simulations using this updated model demonstrate that LBW as a dosing metric may help to normalise plasma concentrations across a range of subject TBW. A population pharmacokinetic analysis of propofol is presented in Chapter 4. In this study, three covariate models were developed using data obtained from the Open TCI Initiative for 408 adult subjects (including 30 obese): a model including the best statistical size covariate on CL, a model including LBW to describe CL and another with TBW describing CL. Results showed that the LBW model described the data as well as the best statistical model, which included body surface area on CL, and was superior to the TBW model, which overpredicted CL in the obese subset. Given that LBW provides a mechanistic explanation for an increase in CL with size, it is proposed that LBW may be suitable for use as a linear propofol infusion dosing metric. Target controlled infusion simulations across a range of subject TBW were also performed to demonstrate the difference between infusion rates calculated using the LBW model and those calculated using two other PK models currently used in clinical practice that overdose the obese. Chapter 5 presents a pharmacodynamic analysis and simulation study that explores the use of different body size metrics including TBW, LBW and an empirical ‘corrected weight’ (CW) for dosing. CW has been suggested for propofol dose normalisation in obese patients because of concerns that a linear dose per kg TBW strategy may result in overdose. Respiratory depression upon induction of anaesthesia and time to awakening at end of dosing were compared between dose strategies, both of which cause concern for anaesthetists when dosing their obese patients. Results showed that dosing according to TBW resulted in increased respiratory depression and increased time to awakening in the obese compared to healthy TBW subjects, whereas CW and LBW dosing resulted in improved PD profiles, reducing the extent of respiratory depression and normalising time to awakening in the obese subject group. It is proposed that the success of CW for dosing obese patients is due to its correlation with LBW, but that LBW is superior as a dosing metric due to its mechanistic nature and applicability to the whole adult population. Evidence for the LBW hypothesis for drug CL arose, in part, from the findings of a literature review that specifically investigated the use of body size metrics for describing CL in the obese population. Chapter 6 builds upon this work, describing a literature survey from 2000 to 2007 that aimed to explore the use of covariates for describing CL in PK analyses performed in any patient group. In particular, the success of different body size metrics as covariates on CL is explored. The work presented in this thesis provides evidence for the use of LBW as a dosing metric for propofol and may help contribute to development of optimised dose strategies that are appropriate for use across the whole adult population. This is of great importance given the increasing prevalence of obesity worldwide