Biomechanical and in vivo evaluation of experimental closure devices of the annulus fibrosus designed for a goat nucleus replacement model

Abstract

Promising strategies are being developed to replace or regenerate the herniated nucleus pulposus. However, clinical efficacy of these methods has still to be addressed, and the lack of appropriate annulus closure techniques is increasingly being recognised as a major limiting factor. In the current study, in vitro and in vivo evaluation of novel annulus closure devices (ACDs) was performed. These devices are intended to be used in adjunct to nucleus replacement therapies in an experimental goat study. After a standardised discectomy had been performed, different ACDs were implanted solely or in addition to a collagen nucleus replacement implant. Biomechanical effects and axial failure load were assessed in vitro and followed by in vivo evaluation in a goat model. On axial compression, the average axial failure load for ACDs with four barb rings was significantly higher compared to the implants with five barb rings. The increased range of flexion-extension and latero-flexion observed after discectomy were restored to the normal range after implantation of the implants. Positive findings with the four-ring ACD were confirmed in goats after a follow-up of 2 weeks in vivo. However, after 6 weeks most implants (n = 16) showed signs of destruction and displacement. Although there seemed to be a tendency towards better results when ACDs were placed in addition to the nucleus replacements, these differences were not statistically significant. Moreover, two endplate reactions extending into the subchondral bone were observed, most likely due to continuous friction between the ACD and the vertebrae. Although current results are encouraging first steps towards the development of an efficient ACD for animal models, further optimisation is necessary. Current results also show that one cannot rely on in vitro biomechanical studies with annulus closure techniques, and these should always be confirmed in vivo in a large animal model. © 2010 The Author(s).