Abstracts

Little information of LTG pharmacokinetics is available in elderly patients. The aims of this study were to 1) estimate the oral clearance (CL/F) of LTG and 2) identify and quantify covariate effects that significantly affect CL/F., Data from VA Cooperative Study 428, a randomized, double-blind, parallel group, monotherapy comparison study of carbamazepine (CBZ), gabapentin (GBP), and lamotrigine (LTG), were used for the analysis (N=593). Patients 60 years and older with newly diagnosed seizures were randomly assigned to one of three treatments. Data from the LTG arm were used in this study. Thirty six patients on PHT at enrollment were gradually withdrawn during the 6-week titration of LTG. Clinical evaluations including blood samples were obtained at enrollment, biweekly to week 8, monthly to week 28, and bimonthly to the end of the study(week 52). Analyses were performed using NONMEM. A one-compartment model with first-order absorption and elimination was used. The absorption rate constant (Ka) could not be estimated and were fixed to the literature values of 3.5 hr^-1. Inter-individual variability was modeled using an exponential error model. The residual unexplained variability was modeled using a combined additive and proportional error model. Covariate selection was done by forward selection (p[lt]0.01; [chi]²) and backward elimination (p[lt]0.001; [chi]²) using the likelihood ratio test as the model selection criterion., The data consisted of 875 observations from 148 patients. It was found that weight, BUN/creatinine ratio, and phenytoin (PHT) use were significant covariates for clearance in the final model. Race was inititially significant but found to be correlated to BUN/creatinine ratio. The final regression model was determined to be:
CL/F = [(0.0332 x BUN/creatinine ratio) + (0.0268 x weight)] x 1.59(if using phenytoin)
Therefore, CL/F was estimated to be 2.64 L/hr, for patients weighing 80 kg, and a BUN/creatinine ratio 15:1. CL/F increased 59% in patients using PHT. Interindividual variability for CL/F of LTG was 34.2%. Percent CV and standard deviation for residual unexplained variability were 19.8% and 0.31 mcg/ml respectively., In our study, weight, BUN/creatinine ratio, and PHT use were significantly predictive of clearance. The effect of BUN/creatinine ratio on CL/F of LTG may be due to a surrogate effect of race. PHT, a known enzyme inducer, increases CL/F of LTG in elderly patients more than 50%. The results from this study may be useful for individualizing dose regimens in this population based on patient-specific covariates., (Supported by NIH NINDS P50-NS16308 and the VA Cooperative Study 428.)