Analysis of percutaneous permeation data: II. Evaluation of the lag time method

Abstract

To evaluate the lag time method, the permeation of four drugs, propranolol (PR), triamcinolone acetonide (TA), physostigmine (PHY), and tetrahydroaminoacridine (THA), was studied across two skin membranes, hairless mouse skin (HMS), and human cadaver skin (HCS). The permeation studies were conducted with vertical Franz diffusion cell in the static mode, using an infinite dose technique. The resulting permeation profile for each individual drug was analyzed by the lag time method to estimate parameters such as the steady-state flux ( J ss), lag time ( T lag), diffusion coefficient ( D) and skin/donor-phase partition coefficient ( K m). The diffusion coefficient ( D) and skin/donor phase partition coefficient ( K m) were used to regenerate the entire permeation profile (pre-steady and steady states) using an equation based on Fick's laws of diffusion. The fit of the observed data to the regenerated curve was estimated by linear regression with the observed data as the independent variable. In experiments where steady state should have been achieved at the end of the experiment (experimental duration > 3 × T lag), R 2 was greater than 0.993, the slope was unity and the y-intercept zero for the regenerated profile vs the observed data regression (e.g., PR 72 h, TA with 2% Azone (AZ) pretreatment and 2% AZ as copenetrant, PHY, and THA experiments). For the PR 27 h experiment, TA control and TA with 2% propylene glycol (PG) pretreatment experiments, the slope was significantly greater than unity for linear regression of the regenerated profile vs the observed data. In these experiments, the experimental duration was less than 3 × T lag and steady state may not have been achieved, which could have resulted in inaccurate estimates of D and K m and hence, the positive bias in the regenerated profiles. Overall, the lag time method was successful in the estimation of permeation parameters from which the entire profile could be regenerated for all the drugs studied. However, the lag time method may result in inaccurate estimates of D and K m if steady state is not achieved as demonstrated by a positive bias (slope > 1) in the regenerated profile. The approach used in this study could be used to verify whether D and K m obtained from a permeation experiment are precise, accurate and hence reliable.