Limitations of predicting in vivo biostability of multiphase polyurethane elastomers using temperature-accelerated degradation testing

Polyurethane biostability has been the subject of intense research since the failure of polyether polyurethane pacemaker leads in the 1980s. Accelerated in vitro testing has been used to isolate degradation mechanisms and predict clinical performance of biomaterials. However, validation that in vitro methods reproduce in vivo degradation is critical to the selection of appropriate tests. High temperature has been proposed as a method to accelerate degradation. However, correlation of such data to in vivo performance is poor for polyurethanes due to the impact of temperature on microstructure. In this study, we characterize the lack of correlation between hydrolytic degradation predicted using a high temperature aging model of a polydimethylsiloxane-based polyurethane and its in vivo performance. Most notably, the predicted molecular weight and tensile property changes from the accelerated aging study did not correlate with clinical explants subjected to human biological stresses in real time through 5 years. Further, DMTA, ATR-FTIR, and SAXS experiments on samples aged for 2 weeks in PBS indicated greater phase separation in samples aged at 85°C compared to those aged at 37°C and unaged controls. These results confirm that microstructural changes occur at high temperatures that do not occur at in vivo temperatures. In addition, water absorption studies demonstrated that water saturation levels increased significantly with temperature. This study highlights that the multiphase morphology of polyurethane precludes the use of temperature accelerated biodegradation for the prediction of clinical performance and provides critical information in designing appropriate in vitro tests for this class of materials.