Failure of resorbable plates and screws in an ovine model of anterior cervical discectomy and fusion

Abstract

Background context: Containment plates are often placed anteriorly in anterior cervical discectomy and fusion (ACDF) to provide stability and prevent migration of the interbody device or autograft. The main advantage of a bioresorbable plate over the typical metallic plate is that it will resorb after fusion has occurred, thus mitigating any long-term instrumentation-related complications. Furthermore, the plates are radiolucent, allowing complete visualization of the fusion site and eliminating imaging artifact.

Purpose: To evaluate radiographic fusion, mechanical success rates, and histologic characteristics of a bioresorbable containment plate and screws in a 3-month ovine model of ACDF.

Study design: An in vivo prospective analysis of resorbable anterior cervical plates and screws for use in ACDF in an ovine model.

Methods: Six sheep underwent C2-C3 and C4-C5 discectomies. Fusions were performed using a polyetheretherketone cage filled with autograft bone. A polymeric plate (70/30 poly-dl-lactic acid), and four screws were placed over an intervertebral disc spacer at each of these two levels. After 3 months, the animals were euthanized and radiographically imaged. Radiographs were analyzed for fusion and instrumentation failures. Functional spinal units were harvested for histologic processing and evaluation.

Results: Radiographic fusion was noted in three of the 12 levels with no evidence of device failure at any of the levels. However, at necropsy, it was observed that six of the 12 specimens had either a broken screw or a cracked plate. These gross observations were confirmed within the histologic sections. Fusion was verified histologically at C2-C3 in three of the six sheep; none of the fusions were successful at C4-C5. Histologic analysis also found that the tissue surrounding the plate and disc spacer consisted of vascularized fibrous tissue with islands of active woven bone. Inflammatory cells were rarely observed.

Conclusions: Although the bioresorbable plates and screws did not elicit an iatrogenic tissue response, a high percentage of them failed mechanically. This phenomenon was difficult to observe radiographically, as the radiolucent markers were not able to convey these instrumentation failures. Additionally, there was only a 25% fusion rate. These findings suggest that resorbable implant materials with the current biomechanical and chemical properties are inadequate for cervical fusion. The results of this study strongly suggest that radiographic outcomes alone may not be adequate and that gross or histologic methods should accompany radiographs in studies of bioresorbable materials in animal models.