Background: Although interbody cages are widely used, there is little histological documentation of the tissue within cages in the human spine. The purpose of this study was to describe the contents of retrieved, clinically failed, interbody cages from human patients, with special reference to the influence of graft type on the viability of bone in the cages.

Methods: Seventy-eight cages that had been retrieved from forty-eight patients were analyzed. There were eight carbon-fiber cages and seventy threaded metal cages. Of the sixty-seven cages for which information about grafting was available, fifty-six had been packed with autograft only, six had local autograft mixed with demineralized bone matrix, four had allograft, and one had demineralized bone matrix only. The indications for cage retrieval included a failed fusion, malposition or migration of the cage, trauma (a compression fracture at the fusion site), low-back pain, progressive spondylosis, nerve-root impingement, and/or infection. The cages had been in situ for an average of twenty-two months. Undecalcified sections through the center of each plastic embedded cage were reviewed, and the approximate areas occupied by viable bone, necrotic bone, fibrocartilage, hyaline cartilage, fibrous tissue, and graft substitute were visually estimated. Debris particles were estimated by a semiquantitative scoring system.

Results: Seventy-one of the seventy-eight cages showed evidence of vascular ingrowth and areas of histologically viable bone, representing incorporating bone graft. The average area occupied by viable bone was 44% (range, 0% to 80%). In some cages, relatively large fragments of cortical bone graft were associated with only minimal new-bone formation. Fibrocartilage occupied up to 50% of the available area in these failed cages. Some cages also contained small fibrocartilage seams connecting segments of bone in a pattern that suggested motion in vivo. In thirty-one of the seventy-eight cages, =5% of the available area was occupied by hyaline cartilage, probably from vertebral end plates or facet joints.

Conclusions: While this study was not designed to test the efficacy of cages or of bone graft, the prevalence of hyaline and fibrocartilage in these failed cages illustrates the importance of graft and graft-site preparation to maximize bone-graft incorporation.

Level of Evidence: Therapeutic study, Level IV (case series [no, or historical, control group]). See Instructions to Authors for a complete description of levels of evidence.