Search

The usual recommended dose for gentamicin is 3 to 7 mg/kg/day for patients with a normal renal function while 1.7 mg/kg/day is recommended for patients undergoing chronic haemodialysis. The objectives of this study were to develop a population pharmacokinetics model (POPPK) for gentamicin, designed for patients undergoing dialysis, and to investigate the best dosing scheme for different MIC clinical breakpoints using Monte Carlo simulations.In this monocentric prospective interventional open clinical study, 23 patients (141 gentamicin samples) were included. The covariates investigated were weight, creatinine, dialysis (yes/no), dialysis flow and dialysis duration. The POPPK model was developed in Pmetrics and 1000 time-concentration profiles were simulated for 9 doses between 2 and 10 mg/kg/day, with an inter-dose period of 24, 48 or 96 h to predict the probability of having both a serum peak 8*MIC and a trough 1 mg/L for MIC values between 0.25 and 4 mg/L.A two-compartment model including the dialysis on the elimination constant and bodyweight on the volume of distribution best described the data. A 30-min gentamicin infusion of 2 mg/kg/day (for MIC = 1 mg/L) or 8 mg/kg/day (for MIC = 4 mg/L) just before dialysis eliminated by two dialysis sessions before the next administration (dose interval of at least 96 h) led to a peak 8*MIC for 90% of the simulations and a trough concentration 1 mg/L at 96 h for 92% and 34% respectively.The gentamicin dose generally used to treat infections in dialysis patients is insufficient and might be increased to 3-8 mg/kg/day just before dialysis, taking into account the type of infection.

The aim of this study was to develop routinely applicable limited sampling strategies for assessing cyclosporin (CsA) AUC(0-12 h), and possibly other exposure indices such as AUC(0-4 h) and C(max), in heart transplant patients over the first year post-transplantation.First, the individual pharmacokinetics (PKs) of 14 adult heart-transplant patients receiving Neoral were assessed at three post-transplantation periods, at the end of the first week (W1), the third month (M3) and the first year (Y1). To fit blood concentrations, a PK model specially developed for oral CsA was applied. Second, two statistical methods were compared for AUC(0-12 h) estimation using a limited sampling strategy (maximum of three blood samples): multiple regression analysis (MR) and Bayesian estimation (BE).No significant difference was observed between the individual PK parameters at M3 and Y1, so population modelling was performed taking as a whole the concentration data collected at M3 and Y1. On the contrary, a significant difference ( P0.05) was found for the C2/dose ratio between W1 and M3 and between W1 and Y1 (mean+/-SD =5.47+/-2.33; 7.78+/-1.05; 6.98+/-2.17 ml(-1 )for W1, M3 and Y1, respectively). Also, C(max)/dose and A were found significantly lower at W1 than at M3 ( P0.01 and P0.005, respectively), while lambda(1) was significantly higher at W1 than at both M3 and Y1 ( P0.01). Using three sampling times (t0 h, t1 h and t3 h), BE allowed an accurate prediction of AUC(0-12 h) (mean bias =3.06+/-12.16%; +1.50+/-1.61%; and -0.20+/-11.42% at W1, M3 and Y1, respectively), AUC(0-4 h )and C(max). MR led to satisfactory estimation of AUC(0-12 h) using only two blood samples collected 2 h and 6 h post-dose (R=0.956-0.993; bias =-5.22 to +4.41; precision =6.38 to 9.90%), but this method is unable to estimate any other exposure index and requires strict respect of sampling times, contrary to BE.Neoral monitoring based on full or abbreviated AUC is possible using BE or MR in heart transplant patients over the first year post-transplantation. BE provides a good description of the individual PK profiles and thus might be useful not only in case of potential discrepancies between C2 and clinical findings, but also for clinical trials aimed at finding optimum PK monitoring in heart recipients.