Tacrolimus in diabetic kidney transplant recipients: Pharmacokinetics and application of a limited sampling strategy

The effect of diabetes mellitus on the pharmacokinetics of tacrolimus is not well characterized. We have compared tacrolimus 12-hour steady-state concentration-time profiles in diabetic (n = 11) and demographically matched nondiabetic (n = 9) stable kidney transplant recipients and derived a limited sampling strategy for the estimation of tacrolimus area under the concentration-time curve (AUC(0-12)). Tacrolimus concentration was measured by liquid chromatography tandem mass spectrometry and acetaminophen absorption method was used to characterize gastric emptying time. Demographic and biochemical characteristics were comparable between the two groups with the exception of significantly higher glycated hemoglobin levels in patients with diabetes (P = 0.02). Time to maximum concentration (T-max) of acetaminophen was significantly longer in diabetics [D: 74.1 minute versus nondiabetics (ND): 29.3 minutes, P = 0.02]; however, tacrolimus T-max was not significantly different (D: 121 minutes versus ND: 87 minutes, P = 0.15). Median (interquartile range) of tacrolimus AUC(0-12) was 114 (101-161) mu g*hr/L in patients with diabetes and 113 (87-189) mu g*hr/L in nondiabetics (P = 0.62). The following limited sampling equation [AUC(pred) (mu g*hr/L) = 18.70 - 1.72 C-1hr - 4.09 C-2hr + 14.40 C-3hr] was derived from a training data set that included 10 patients. The correlation coefficient between model-predicted and observed AUC(0-12) values was 0.999. Mean prediction error and root mean square error of the model-predicted values derived from the patients in validation data set were 0.04 and 17.48 mu g*hr/L, respectively. In conclusion, it appears that diabetes has a modest effect on the rate but not the extent of tacrolimus absorption, and a three-point abbreviated sampling strategy common to both groups may prove useful for the estimation of tacrolimus exposure in kidney transplant recipients.