Laser doppler velocimetry (LDV) has been used to assess the cutaneous pharmacodynamics of isosorbide dinitrate (ISDN) following transdermal delivery of the drug from prototypal patches. The delivery systems, which were saturated with ISDN, (a) produced various degrees of skin occlusion and (b) spanned a six-fold range of adhesiveness. The patches were applied to the ventral forearm skin of 10 healthy volunteers and the local ISDN-induced increase in local skin blood flow was determined using LDV by locating the probe in a central hole in the delivery system. Measurements were made for 1.5 hr and the pharmacodynamics were quantified by (i) the maximum LDV response and (ii) the area under the LDV response versus time curve. These parameters were not sensitive to patch occlusivity. They were significantly (P less than 0.01) dependent on patch adhesiveness, though, and decreased with increasing adhesion. Although this observation suggested that ISDN diffusion through the adhesive could determine, at least in part, the rate of drug delivery, it was subsequently demonstrated that ISDN release (in vitro, into a perfect "sink") was unaffected by the level of cross-linking in the adhesive polymer. Because the drug was present in all systems at unit thermodynamic activity, these results cannot be explained on the basis of altered ISDN partitioning at the device-stratum corneum interface. We speculate that the in vivo-in vitro discrepancy may be due to the efficiency of skin contact achieved by different adhesives: that is, the more adhesive, less flexible systems make poorer contact with the skin surface, thereby decreasing the effective surface area of drug delivery.(ABSTRACT TRUNCATED AT 250 WORDS)