A Practical Limited Sampling Strategy to Predict Free Prednisolone Exposure and Glycemia in Kidney Transplant Recipients

Abstract

Despite significant interindividual variability in prednisolone pharmacokinetics and potentially serious consequences with inadequate or excessive exposure, monitoring of prednisolone levels is not employed to guide therapy. As ultrahigh-performance liquid chromatography-tandem mass spectrometry methods can now measure the active free fraction of prednisolone, this study aimed to evaluate the performance of 15 published limited sampling strategies (LSSs) for predicting free prednisolone exposure in adult kidney transplant recipients and to examine the relationship between free/total prednisolone exposure and plasma glucose. The study was performed in 11 subjects without diabetes 3-4 weeks postkidney transplantation. Area under the concentration time curve profiles of total and free prednisolone from 0 to 12 hours postdose (AUC₀₋₁₂) were determined and compared with predicted AUC₀₋₁₂ values calculated from published LSSs. Venous glucose was measured concurrently with the 13 sampling time points. The mean (±SD) age of subjects was 52 ± 12 years, 5 were men and the median (interquartile range) daily prednisolone dose was 20.0 mg (20.0-22.5). Interindividual variation in dose-adjusted free and total prednisolone exposure was 1.9- and 3.2-fold, respectively. All 15 free prednisolone LSSs exhibited good correlation (r ≥ 0.83), with bias and imprecision less than 15%. An LSS incorporating 1.25- and 3-hour samples had the highest predictive power (r = 0.97, bias 1.2%, imprecision 5.6%). Free prednisolone AUC₀₋₁₂ correlated with peak glucose levels (r = 0.65, P = 0.037), as did predicted AUC₀₋₁₂ from 14/15 LSSs. Biologically active free prednisolone exposure can be accurately predicted postkidney transplantation by LSSs incorporating 2-point concentration sampling. Peak plasma glucose concentration correlated well with prednisolone exposure.