Renal Function in Relation to Gentamicin and Dabigatran Clearance and Dosing
Dose individualisation is a key principle in prescribing. An important reason for this is the variability of drug clearance between individuals. Clinicians require tools that reflect drug clearance to provide guidance in this regard. In the setting of drugs that are renally cleared, these tools are often equations that estimate renal function, based on demographic and renal biomarker data from the individual. There is a range of such equations, including those that exclusively employ either creatinine or cystatin C as the renal biomarker, and those that use both. The identification of the best equation, or the best way of using a widely used equation, in relation to pharmacokinetics, may ultimately lead to improved patient outcomes. We performed a survey and found that most prescribers at a large local general medicine department used creatinine in the context of either the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, or the Cockcroft-Gault equation, to guide the dosing of renally cleared drugs. In this thesis, these and other renal function equations that have been developed using creatinine and/or cystatin C assays traceable to modern reference standards were examined in relation to gentamicin and dabigatran. In relation to the prediction of gentamicin clearance, a group of 240 hospitalised patients were studied. The Cockcroft-Gault equation using total body weight, and the creatinine-only CKD-EPI equation with adjustment for individual body surface area (BSA), were the best of the equations that exclusively used creatinine. In a separate group of 260 hospitalised individuals given gentamicin, the CKD-EPI equation employing both creatinine and cystatin C was found to be superior to both the creatinine-only and cystatin C-only CKD-EPI equations in the prediction of gentamicin clearance. Adjusting for individual BSA was found to improve the performances of all the cystatin C-based equations, especially at the extremes of size. However, we were unable to identify a subgroup where the cystatin C-only CKD-EPI equation outperformed the creatinine-only equation. In relation to dabigatran, we audited the dosing of its prodrug, dabigatran etexilate, at a local tertiary hospital, and found 54% of 204 patients were dosed either excessively or inadequately in relation to renal function. The Cockcroft-Gault equation was only 79% concordant in terms of dabigatran etexilate dosing guidance with either of the MDRD Study or CKD-EPI equations. This highlights an area of uncertainty, in that it is unclear which renal function equation is best for guiding the dosing of dabigatran etexilate. Further, 75% of patients were co-prescribed drugs that potentially interact, pharmacokinetically or pharmacodynamically with dabigatran etexilate. To support further dabigatran clinical studies, a method based on liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) was developed to measure dabigatran concentrations in human plasma. This dabigatran assay was used in a study of 52 individuals treated with dabigatran etexilate for atrial fibrillation (AF). We were unable to demonstrate a statistically significant difference in the explained variances of the renal function equations for dabigatran concentrations, and calculated that data from around 680 patients would be required to show the superiority of the CKD-EPI creatinine-cystatin C equation over the Cockcroft-Gault equation. We corroborated the previously reported explained variances of four coagulation assays assessed for plasma dabigatran concentrations in our real-world setting. Even if the best renal function equation for estimating dabigatran clearance was identified, the use of this alone would not account for drug-drug interactions involving dabigatran etexilate that are unrelated to renal clearance. Using the published data, including our own, we constructed a target steady-state trough plasma dabigatran concentration range for use in the setting of AF. This was converted to a target range of trough thrombin time at steady-state (TTtrough,ss). Around 50% of the 52 individuals in our dabigatran study had TTtrough,ss outside this range, demonstrating the potential of this target range for changing management. We have thus examined the renal function equations as tools for assessing the pharmacokinetics of gentamicin and dabigatran, as exemplars of renally cleared drugs. Further, we propose the use of another tool, the TTtrough,ss, for guiding dabigatran dosing.
Advisor: Begg, Evan James; Barclay, Murray Lindsay
Degree Name: Doctor of Philosophy
Degree Discipline: Department of Medicine, University of Otago, Christchurch
Publisher: University of Otago
Keywords: renal; dabigatran; gentamicin; clearance; monitoring; atrial fibrillation; dosing; blood coagulation; anticoagulants; stroke; thromboembolism; creatinine; cystatin C; pharmacokinetics
Research Type: Thesis