Revision strategies for salvaging or improving failed cylindrical cages.

Abstract

This is a review of 20 patients who experienced failure of threaded interbody fusion cages and underwent surgical correction. To review the causes and possible treatment strategies for failed cylindrical cages. Intraoperative complications have been described in the past; however, management of the postoperative patient with failure of interbody fusion devices has not been described. In 20 patients with failed threaded titanium fusion cages (18 Bagby and Kuslich Devices [BAK; Sulzer-Spine Tech, Minneapolis, MN], 2 Ray Threaded Fusion Cages [Ray TFC; Surgical Dynamics, Norwalk, CT) who underwent revision surgery, all had failure before successful arthrodesis was achieved. Eight of the original titanium cages had been inserted anteriorly (7 laparoscopically), and 12 had been inserted for posterior interbody lumbar fusion. Before the revision surgery, five of the implants were thought to be solid by the referring surgeon, but pseudarthrosis was clearly present in all. In addition, 14 other explanted BAK devices were subjected to undecalcified histologic preparation, quantitative histomorphometry, and histopathologic analysis. The average length of time before revision surgery (implant duration) was 31.8 weeks (range, 1-156 weeks). The most common revision procedure was posterior exploration of the symptomatic nerve root with foraminotomy for unrecognized lateral recess stenosis (11 cases) or excision of iatrogenically herniated intervertebral disc fragments (4 cases). However, four cages inserted through posterior exposure during an interbody lumbar fusion procedure had to be removed because of migration into the spinal canal. In nine cases posterior pedicle screw instrumentation was necessary in addition to posterolateral fusion using iliac crest bone grafting. All 20 cages failed because of surgical technique rather than an intrinsic defect in fusion cage technology. The factors associated with failure of the original insertion procedure were failure to achieve adequate distraction of the anulus fibrosis; undersized cages, especially when placed through the posterior interbody lumbar fusion approach; cerebrospinal fluid leakage or pseudomeningocele; Type 2 diabetes mellitus; the use of local bone graft rather than iliac crest inside the cage; anterior insertion in an excessively lateral position resulting in symptoms of a far lateral disc herniation; and failure to identify the spinal midline during an anterior approach.