Effects of low-frequency ultrasound on the transdermal permeation of mannitol: comparative studies with in vivo and in vitro skin.

The in vivo and the in vitro correlation of the effects of low-frequency ultrasound (low-frequency sonophoresis, LFS) on the percutaneous penetration of mannitol, a model hydrophilic permeant, was investigated using three in vitro skin models (including full-thickness and split-thickness pig skin, and heat-stripped human cadaver skin) and in vivo pig as the animal model. The central objective of this article was to identify the relevant in vitro skin models and ultrasound conditions that may be used in in vitro LFS studies to predict the effects of LFS in vivo on the transdermal delivery of hydrophilic permeants. In this article, by conducting comparative studies of the in vivo pig skin and of the three in vitro skin models under two LFS protocols (a constant ultrasound energy dose protocol, and a constant skin electrical resistance protocol), we demonstrated that: (1) under a constant ultrasound energy dose protocol (protocol A, 5 min LFS), no good correlation was observed between the in vivo skin and the in vitro skin models in terms of the measured skin permeabilities to mannitol. Moreover, the effects of LFS on the barrier functions of the in vivo pig skin, as measured by the enhancement ratio of the skin permeation rate of mannitol and by the reduction of the skin electrical resistance, are much more pronounced than those observed with the excised skin models in vitro; (2) under a constant skin electrical resistance protocol (protocol B) of LFS, a good correlation was found between the skin permeability to mannitol measured using the three in vitro skin models and that of the in vivo pig skin. This result indicates that by utilizing the skin electrical resistance as a quick indicator of the skin permeabilization state due to LFS, the three in vitro skin models can be utilized to predict the transport rate of mannitol across the in vivo skin during LFS; (3) by applying a recently developed skin porous-pathway theory, we demonstrated that within the range of LFS conditions examined, the three in vitro skin models exhibit similar transport properties to mannitol and similar skin effective pore radius values, and hence, represent equivalent skin models for the in vitro LFS studies in the case of hydrophilic permeants; (4) histological studies revealed that the LFS protocol that was shown to be efficacious in enhancing the skin penetration rate of mannitol across the in vivo pig skin, and was also utilized for the in vivo/in vitro skin comparative studies, is safe for the living skin; and (5) through measuring the skin concentration of mannitol in the presence and in the absence of the LFS treatment, we found that the LFS-induced flux enhancement outweighs the enhancement of the skin concentration of mannitol during the LFS studies both in vivo and in vitro. This result suggests that LFS represents a good method of enhancing the systemic absorption of hydrophilic permeants, while it does not significantly alter the vehicle-to-skin partition coefficient for the same class of permeants. Copyright 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association