Surface erosion, drug release devices

The devices prepared in our study were almost cylindrical in shape because the two smaller dimensions were very similar (0.5 and 0.7 cm). The devices were equivalent to a cylinder both in terms of suriace area and volume, with a radius of 0.33 cm and a length of 1.3 cm, and they could thus be classified as erodible cylindrical devices with dispersed drug. Therefore, Eqs. (1) and (2) are applicable and should describe the release profile. To determine if drug release from the devices was controlled by surface erosion, the release profiles were fitted to Eq. (1). Since all the parameters in Eq. (1) except B (the erosion rate) are known, nonlinear regression was used on the release profiles to obtain the optimized values of B. 

Convincing evidence for a surface erosion process is shown in Fig. 8, which shows the concomitant release of the incorporated marker, methylene blue, release of the anhydride excipient hydrolysis product, succinic acid, and total weight loss of the device. According to these data, the release of an incorporated drug from an anhydride-catalyzed erosion of poly (ortho esters) can be unambiguously described by a polymer surface erosion mechanism. 

Surface erosion not only leads to zero-order drug release from devices that maintain a constant surface area, but has other important consequences. Among these are the following (1) the rate of drug release is directly proportional to drug loading, (2) the lifetime... 

Because swelling and consequent bulk erosion induced by the water-soluble salt is not desirable, use of the low-water-solubility, sUghtly acidic salt calcium lactate was investigated (30). By using this excipient it was hoped that a lowering of the pH within the surface layers of the device would take place and release of the drug would be controlled by polymer erosion confined to the surface layers of the device. In these experiments norethindrone was replaced by the currently favored steroid levonorgestrel. 

Although it was possible to achieve constant in vitro release of levonorgestrel for up to 410 days at which point the experiment was discontinued, release of the drug was not controlled by surface erosion of the polymer but instead the device underwent bulk erosion and release of levonorgestrel was completely controlled by its rate of dissolution. Bulk erosion of the rod-shaped device was evident by scanning electron microscopy as shown in Fig. 17. 

In their study of branched PSA, Maniar et al. (1990) found that the molecular architecture of branched polymers affects the release kinetics in a variety of ways. They found that the branched polymers degraded faster than linear PSA of comparable molecular weight (Maniar et al., 1990). They also noted that drug (morphine) release profiles were more characteristic of bulk erosion than surface erosion An initial lag time during which very little drug was released was associated with the time required for water to swell the polymer. This was followed by a period of relatively fast release, which tapered off as the device disintegrated. The polymer matrix lost its mechanical integrity before the release experiment was complete (Maniar et al., 1990). Despite the increase... 

In non-porous polymeric systems the rate of drug release is dictated by the device surface area which is linked directly to its shape. Drug release from polymeric systems with a variety of geometries has been described. " Zero-order release kinetics may be more easily achieved with slab or rod geometries than spheres. The rate of release from spheres may result from polymer diffusion or erosion. " " " Diffusion-mediated release has been studied extensively and described mathematically. " ... 

Drug release rate for matrices undergoing bulk erosion is nonlinear and difficult to predict because it is determined by a combination of diffusion and erosion. However, drug release from devices undergoing surface erosion is predictable and can lead to zero order-kinetics provided diffusional release of the drug is minimal and the overall surface area of the device remains essentially constant. 

Because surface erosion results in constant and predictable rate of drug release, this type of erosion is clearly preferrable to bulk erosion. However, to achieve surface erodibility, a system must be devised in which the rate of polymer degradation at the surface of a device is very much faster than the rate of degradation in the interior. 

A much more desirable erosion mechanism is surface erosion, where hydrolysis is confined to a narrow zone at the periphery of the device. Then, if the drug is weU-immobihzed in the matrix so that drug release due to diffusion is minimal, the release rate is completely controlled by polymer erosion, and an ability to control erosion rate would translate into an ability to control dmg delivery rate. For a polymer matrix that is very hydrophobic so that water penetration is limited to the surface (thus Hmiting bulk erosion), and at the same time, allowing polymer hydrolysis to proceed rapidly, it should be possible to achieve a drug release rate that is controlled by the rate of surface erosion. Two classes of biodegradable polymers successfully developed based on this rationale are the polyanhydrides [31] and poly (ortho esters) [32], the latter of which is the subject of this chapter. 

When a water-insoluble drug such as hydrocortisone is physically dispersed in the polymer and the polymer fabricated into thin disks which are then placed in a buffer at pH 7.4, release kinetics shown in Fig. 1 are obtained. Clearly, this system exhibits excellent zero order drug release kinetics with concomitant linear rate of polymer weight loss. The simultaneous polymer erosion and drug release is a clear indication that erosion occurs at the surface of the device and that drug release is controlled by erosion of the polymer. 

Surface erosion has been further demonstrated by noting a linear relationship between surface area and drug release rate shown in Fig. 12 and a linear relationship between device thickness and lifetime, shown in Fig. 13. 

After investigating a number of linear poly (ortho esters) a material prepared from 3,9-(bis ethylidene 2,4,8,10-tetraoxaspiro [5,5] undecane) and 1,6-hexa-nediol was selected as the best material [42]. Figure 23 shows results of a study where both 5FU release and weight loss were determined. The data show that with this particular system, concomitant drug release and polymer erosion has been achieved. Further, because the molecular weight of the residual polymer remains unchanged, the hydrolysis process is confined to the outer surface of the device and surface erosion has been achieved. 

---