Comparison between allograft plus demineralized bone matrix versus autograft in anterior cervical fusion. A prospective multicenter study

Abstract

Study design: This study analyzed the fusion results of an allograft-demineralized bone matrix composite versus autograft in a prospective series of patients undergoing surgery for cervical disc disease.

Objectives: To determine the fusion rates of allograft-demineralized bone matrix composite in anterior cervical fusion as compared with the gold standard autograft.

Summary of background data: For the anterior cervical fusion, the use of freeze-dried allograft is well documented in the literature, citing its effectiveness and inferior fusion rates. The use of demineralized bone matrix in conjunction with freeze-dried allograft in anterior cervical fusion has not been reported.

Methods: This study was done in a prospective fashion in two medical centers. One group received autograft from the anterior iliac crest, whereas others received freeze-dried allograft augmented with demineralized bone matrix (Grafton, Osteotech, Inc., Shrewsbury, New Jersey). For the autograft group, the standard Smith-Robinson grafting technique was used. For the allograft composite group, demineralized bone matrix was pasted onto the freeze-dried allograft and into the disc space before graft insertion. The autograft group consisted of 38 patients with age ranging 26-71 years (mean, 46.1 years) and follow-up periods of 12-33 months (mean, 18.4 months). There were 19 one-level, 17 two-level, and two three-level fusions. Similarly, the allograft group consisted of 39 patients with age ranging 28-80 years (mean, 48.0 years) with follow-up period of 12-31 months (mean, 17.5 months). There were 19 one-level, 16 two-level, and four three-level fusions. Clinical and radiographic follow-up evaluations were completed at 3-month intervals. Radiographs taken 12 months after surgery were analyzed blindly.

Results: Pseudarthrosis developed in 46.2% of patients (33.3% of levels) in the allograft-demineralized bone matrix group compared with 26.3% (22% of levels) in the autograft group (P = 0.11 for patients, P = 0.23 for levels). For patients undergoing two-level fusions, 37.5% of allograft-demineralized bone matrix failed compared with 23.5% of autografts. For single-level fusions, 47.4% of allograft patients developed a pseudarthrosis compared with 26.3% in the autograft group. Graft collapse of > or = 3 mm was noted in 11% of the autograft group versus 19% in the allograft-demineralized bone matrix group (P = 0.32). Graft collapse of > or = 2 mm occurred in 24.4% of autograft patients compared with 39.7% of the allograft-demineralized bone matrix group (P = 0.09). Smokers had an increased rate of pseudarthrosis (47.1%) compared with nonsmokers (27.9%, P = 0.13).

Conclusions: The study revealed that the allograft-demineralized bone matrix construct gives a higher rate of graft collapse and pseudarthrosis when compared with autograft in a prospective series, although the differences were not statistically significant. The pseudarthrosis rate in the series may be high because of the large percentage of smokers and radiographic evaluation techniques. For the purpose of solid radiographic fusion, the use of autograft is recommended in anterior cervical surgery until other acceptable osteoinductive materials are developed.