Biomaterials in Drug Delivery and Tissue Engineering Lange cont drele polye-ased 5 and Enhanced Delive Drugs. Transde dali in the 0 where no transdermal pr roducts exis d to a s2 billion a business However,the successful use ery -m weig 5 To extend transdermal delivery bevond such druss.trans eded.Investigators have studied rone approaches;we dis 10D'1100 1000 mol won dered whether transfer of molecules through human skir aided ultrasound found that ult od ugh skin (termed rent arug ystudied this cavit ing res t of tra was inv e the physical chemical nature of the dru to examine the enhancement for different drugs us or the ontimal ound conditions to achieve transdermal transport human cadaver skin.For some drugs,over a 1000-fol ultr Furth many co dress th e be an s studies to und at lower fre ency.Some exar mples include insulin ance to transdermal trans ropoeitin.The levels of ins ulin and ah skin The maie is the outermost skin layer.the stratum comeum byed on he kin. which is dea skin which is lipids Further testing using diabetic rats showed that insulin ined fo e delivere d this way lowered ble ood sugar levels to norma for ultras nd-assisted transdermal trar ort- (1)tem kin an (②convecti pabar6 n thr me demonstrated that the skin resumed normal penetration ,and(④cavita ests on patients showed no negative no tem aperature change due to sonophoresis.The role of also osti s could con ection was eliminated by studying charged drugsand dhy this ling no enh ance wa between a rat's skin and an ultrasound transducer that me hanical effects were not re nsible.If cavitation interstite eophylli d gluc ie a the clark arid test sh ed that this was an In studies to da we ed an ordi tudied-enha nt dis d.At 1 numan diabe fold increased flux was observ ed for model drug estradiol while there was n effect e n The ols an widely conventiona ort for the cavitation m Finall placing und high pre atm)reduced the effec Because they ar of the skin' 6 nall and therefore have such a high surface area,they end to simple model was then developed making acroso en Enhancement rate is equal to AD'/Dl.where D is th this novel a ed This diffusivity through dis dered lipids.D is the diffusivity has led to improvements of up to about 10-20%of the al pid bila and he uld be fraction c VOL.33.NO.2,2000 ACCOUNTS OF CHEMICAL RESEARCH 97controlled release polymer-based chemotherapy was ap￾proved.36 Ultrasound Enhanced Delivery Drugs. Transdermal delivery has grown in the past 20 years from the point where no transdermal products existed to a $2 billion a year business.37 However, the successful use of transder￾mal delivery has been restricted to low-molecular-weight, highly lipophilic drugs such as nicotine or nitroglycerin.38 To extend transdermal delivery beyond such drugs, trans￾port enhancement is needed. Investigators have studied chemical, electrical, and ultrasound approaches; we dis￾cuss here our efforts in the latter area. In early studies, we found ultrasound could enhance molecular transport through polymer matrices;39 we won￾dered whether transfer of molecules through human skin might also be aided by ultrasound. We found using mannitol, inulin, and physostigmine on rats and guinea pigs that ultrasound caused enhanced drug transport through skin (termed “sonophoresis”).40 Other laboratories subsequently studied this approach for different drugs but received differing results.41-44 The extent of transport was drug dependent, but no fundamental understanding of the physical chemical nature of the drugs or the optimal ultrasound conditions to achieve transdermal transport existed. To address this issue, we began studies to understand the mechanism for the enhanced movement of molecules through skin. The major resistance to transdermal trans￾port is the outermost skin layer, the stratum corneum (which is dead skin), which is composed of lipids and keratinocytes. Human cadaver skin was used as a model system. We examined four possible release mechanisms for ultrasound-assisted transdermal transport: (1) tem￾perature, (2) convection through hair follicles, (3) me￾chanical oscillation of lipid bilayers, and (4) cavitation effects on bilayers. Measurements showed that there was no temperature change due to sonophoresis. The role of convection was eliminated by studying charged drugs and finding no enhancement. Sonophoresis was found to be inversely proportional to ultrasound frequency, suggesting that mechanical effects were not responsible. If cavitation were the mechanism, we hypothesized there should be a threshold of activity at a certain frequency. We found that above 2.5 MHzsthe cavitation threshold under the condi￾tions studiedsenhancement disappeared. At 1 MHz, a 13- fold increased flux was observed for a model drug, estradiol, while at 3 MHz there was no effect. A second set of experiments, in which gases were evacuated from skin, also eliminated the effect, providing additional support for the cavitation mechanism. Finally, placing the skin under high pressure (30 atm) reduced the effect. All these studies suggested that the mechanism consists of disordering of the skin’s lipid bilayers by cavitation. A simple model was then developed: Enhancement rate is equal to f[D1/Db], where D1 is the diffusivity through disordered lipids, Db is the diffusivity through normal lipid bilayers, and f is the fraction of bilayers disordered. These terms could be independently estimated or measured. This equation was used to predict the transdermal transport of different drugs45 (Figure 7). Since cavitation was inversely proportional to ultra￾sound frequency, we next reduced the frequency to 20 kHz to examine the enhancement for different drugs using human cadaver skin. For some drugs, over a 1000-fold enhancement was observed.46 Furthermore, many com￾pounds, even proteins, showed enhanced transdermal flux at lower frequency. Some examples include insulin, interferon, and erythropoeitin. The levels of insulin and interferon delivered by sonophoresis were in the thera￾peutic range when tested on human cadaver skin.47 Further testing using diabetic rats showed that insulin delivered this way lowered blood sugar levels to normal within 30 min.47 Histology established that the treatment did not damage skin, and water permeability experiments demonstrated that the skin resumed normal penetration within 2 h.46 Initial tests on patients showed no negative effects.48 We also examined whether diagnostic samples could be removed by this technique. By positioning a reservoir between a rat’s skin and an ultrasound transducer, interstitial fluid was extracted and theophylline, glucose, cholesterol, urea, and calcium could be measured. Studies using the Clark grid test showed that this was an accept￾able measurement route for examining sugar levels in human diabetics.48 In studies to date, we used an ordinary laboratory sonicator. Future studies will examine the possibility of designing a small portable sonicator. Porous Aerosols for Inhalation Therapy. Aerosols are widely used for lung delivery.49-51 However, conventional aerosol treatment, e.g. for asthmatics, is inefficient. Aerosol particles are small, generally on the order of 3-4 µm, and generally have a density of 1 g/cm3. Because they are so small and therefore have such a high surface area, they tend to aggregate, making aerosolization inefficient. In many cases, only 5-10% of the medication is utilized. To address this, novel aerosolizers are being designed. This has led to improvements of up to about 10-20% of the medication being utilized. In addition to low efficiency, a patient has to frequently take aerosols because, once in FIGURE 7. Model predictions and experimental observations of the enhancement rate, which is the rate of fluxes in the presence and in the absence of ultrasound, as a function of drug diffusion coefficients. Biomaterials in Drug Delivery and Tissue Engineering Langer VOL. 33, NO. 2, 2000 / ACCOUNTS OF CHEMICAL RESEARCH 97