Applications of Population and Other Analytical Approaches to Pharmacokinetic and Pharmacokinetic/Pharmacodynamic Modelling

Rosario, Maria C. P (1999) Applications of Population and Other Analytical Approaches to Pharmacokinetic and Pharmacokinetic/Pharmacodynamic Modelling. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 13832105.pdf] PDF
Download (6MB)


In spite of being in use for more than 50 years, the pharmacokinetics of gentamicin still remain controversial. Advances in analytical techniques during the 70s made it possible to identify a long elimination phase of gentamicin. At that time several studies were published documenting accumulation of gentamicin for a long period of time and the inadequacy of a mono-exponential decline model to describe the elimination phase. Since then, several studies investigating the pharmacokinetics of gentamicin have been carried out. These studies involved either a general patient population or specific subgroups of the patient population. It is only in the last decade that the first report documenting an increased volume of distribution in patients with cancer was published. To date only eight studies have been performed to investigate the volume of distribution of gentamicin. They have all used a mono-exponential decline model and the parameters were estimated either by a Bayesian technique, by the Sawchuck-Zaske method or by the non-parametric expectation maximization (NPEM) algorithm. The nonlinear mixed effects approach has never been employed to characterize the population pharmacokinetics of gentamicin. Most of these studies enrolled a small number of patients and the results obtained were compared with values assumed for the rest of population. A matched control group was used in only a few studies. Therefore the conclusion drawn by these investigators was contradictory. In this thesis the population pharmacokinetics of gentamicin were investigated in patients with cancer using a nonlinear mixed effects approach (NONMEM) using data collected from routine therapeutic drug monitoring. The data were best fit with a bi-exponential disposition model with a combined residual error structure. Several covariates were examined for a possible influence on the pharmacokinetics of gentamicin. The best covariate model related clearance to estimates of creatinine clearance (minimum creatinine value 60 mumol L-1) and volume of the central compartment to body surface area and albumin concentration. Since gentamicin is mainly eliminated by glomerular filtration, creatinine clearance estimates are often used to determine initial doses of gentamicin. However, there are limitations to the value of creatinine concentration as a measure of renal function. Disease states that affect the normal production of creatinine will yield estimates of creatinine clearance that do not represent the renal function of the patient, putting patients at risk of overdosing and toxicity. Several strategies have been proposed in the literature to overcome this problem. In this study the influence of low creatinine concentration by deriving three new covariates corresponding to three different minimum creatinine values of 60, 70 and 88.4 mumol L-1 was investigated. In this study it was found that setting concentrations less than or equal to 60 mumol L-1 to 60 mumol L-1 or concentrations less than or equal to 70 mumol L-1 to 70 mumol L-1 was superior to using either actual creatinine concentration or minimum value of 88.4 mumol L-1. The usefulness of linear regression as a complementary technique in the covariate model building process was evaluated. It performed well when the relationship between individual parameter estimates and covariate was close to a straight line. The best model obtained in the population analysis was then used to simulate concentration-time profiles in two simulated "patients" with different clinical characteristics according to two published nomograms: one that used a 'once daily' regimen and one based on traditional target ranges. In the context of published nomograms this analysis indicated that both nomograms achieve satisfactory concentrations in cancer patients. These results confirm the wide interpatient variability in aminoglycoside pharmacokinetics and the need for dosage optimisation of this drug. (Abstract shortened by ProQuest.).

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Additional Information: Adviser: Peter A Meredith
Keywords: Pharmaceutical sciences
Date of Award: 1999
Depositing User: Enlighten Team
Unique ID: glathesis:1999-75506
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 19 Nov 2019 19:37
Last Modified: 19 Nov 2019 19:37

Actions (login required)

View Item View Item


Downloads per month over past year