Analysis of the factors influencing anti-epileptic drug concentrations--carbamazepine.

The factors that influence carbamazepine (CBZ) serum concentrations and level:dose ratios were evaluated retrospectively on 83 consecutive routine CBZ determinations from chronically treated epileptic patients. The total level:dose ratio was lower when the drug was given in combination with phenytoin or with phenytoin plus other anti-epileptic drugs (AED) than when CBZ was given alone. The free level:dose ratio was also decreased during concomitant treatment with phenytoin. In spite of alterations to both level:dose ratios, the free fraction of CBZ was unaltered when in combination with other AEDs. There were statistically significant correlations between the CBZ dose (mg/kg/day) and the serum levels:total (r = 0.372, P less than 0.001) and free (r = 0.335, P less than 0.005). However, the wide scatter in the CBZ serum levels at each given dose were such that the relationships had no predictive values. Stepwise multivariate regression analysis (MVR) was also performed. A comparison of the normalized regression coefficients indicated that the CBZ dose, PHT dose, the time elapsed from the previous dose, the albumin concentration and the triglyceride levels were important factors which influence the CBZ total serum level. Similar analysis for the CBZ free serum level, gave a multiple correlation coefficient of 0.838, which indicates that 70.3% of the variance was explained by nine independent variables. The major factors that significantly affected the free serum level of CBZ were CBZ dose, PHT dose, the time elapsed from the previous dose and the triglyceride and albumin levels. Our observations indicate that it may be possible to predict reliably the CBZ serum concentrations for individual patients before institution of CBZ therapy.