Cadaveric fibula, locking plate, and allogeneic bone matrix for anterior cervical fusions after cervical discectomy for radiculopathy or myelopathy.

OBJECT: The authors have previously reported that the results of using cadaveric fibula and locking plate (CF/LP) fusion following anterior cervical discectomy (ACD) for cervical spondylotic radiculopathy and myelopathy are superior to those obtained using autologous iliac crest (AIC) grafts in the short term. The long-term results of using this construct are important in substantiating this improvement. The authors report on 246 consecutive patients (54% smokers) who underwent ACD with CF/LP fusion (175 with allogeneic bone matrix [ABM]) and compare them with 111 consecutive patients in whom AIC fusions (49% smokers) were performed by the same surgeons.
METHODS: The study is a retrospective nonrandomized analysis, and chi-square statistics were used. Bone densitometric studies of AIC grafts and CF grafts were performed. A paired t-test was used for statistical analysis of the results. Disease in the group of patients undergoing CF/LP fusion included soft-disc herniation with radiculopathy in 14, soft-disc hemiation with myelopathy in seven, cervical spondylotic radiculopathy in 144, cervical spondylotic myelopathy in 75, AIC graft collapse pseudarthrosis in five, and ACD with no fusion collapse/kyphosis in one. Operations consisted of single-level CF/LP fusion in 142 patients and multilevel CF/LP fusion in 104. Perioperative complications in the CF/LP group included three cases of transient hoarseness. There were no transfusions, infections, neurological injuries, or deaths. The mean hospital length of stay was 1.2 days (28% outpatient and 66% 23-hour stay). The mean follow-up period was 60 months (range 12-94 months). Ten patients were lost to follow up after 1 year. Long-term complications included one traumatic plate fracture and one symptomatic pseudarthrosis with plate fracture. At 1 year and beyond, in 245 (99.6%) of 246 patients radiographically documented fusion with no motion at the fused level on flexion-extension films was demonstrated. There was no kyphosis, no symptomatic screw plate backout, and no CF/LP graft collapse (100% in the ABM group). In the 111 consecutive patients with AIC fusions, however, there was a 17% graft-related complication rate. There were significantly fewer graft-related complications in the CF/LP group (p < 0.001). There was no difference in neurological outcome for any of the groups. In the groups undergoing single-level ACD there was a significantly greater chance of complete relief of neck pain CF/LP fusion compared with those undergoing AIC fusion (p < 0.02). There was a significantly greater chance of AIC collapse with the passage of time compared with CF graft (p < 0.02). Time until return to work was shorter for the CF/LP group by 5 to 6 weeks (p < 0.02). There was a higher rate of radiographically documented pseudarthrosis in the AIC group (p < 0.006). The mean bone densitometry for the CF/LP group was 0.7 g/cm2, significantly greater than that of the AIC group, which was 0.2 g/cm2 (paired t-test p < 0.001).
CONCLUSIONS: When fusion is necessary following ACD, the results of CF/LP fusion are significantly superior in the first 5 years after surgery compared with those for AIC fusions. It remains to be determined if demineralized ABM has a significant effect in enhancing fusion.