Autologous versus allogenic bone grafts in instrumented anterior cervical discectomy and fusion: a prospective study with respect to bone union pattern

Background The purpose of this prospective semi-randomised comparative study was to compare fusion rates, course of fusion, and occurrence of collapse and subsidence of autologous and allogenic bone grafts in instrumented anterior cervical fusion. The number of fused levels and the smoking status were investigated as potential factors influencing the bone-healing process. No similar prospective study on instrumented anterior cervical discectomy and fusion was found in the literature.Methods Seventy-nine consecutive patients were operated on using the Smith–Robinson technique with a single instrumentation system at one or two levels. Seventy-six cadaverous fibular bone grafts and 37 autologous iliac-crest bone grafts were inserted. All patients were followed up for at least 2 years.Results The radiographs obtained during the follow-up were analysed, and showed no statistical difference in fusion and collapse rate between autografts and allografts. Allografts showed significantly longer time to union. No case of graft migration was observed. No difference was found between fusion and collapse rate with respect to the number of fused levels in general, but greater time to union was seen in two-level fusions. When one- and two-level subgroups were compared, there was no evidence of any significant difference in fusion or collapse rates between autografts and allografts, and the healing process took longer in allogenic grafts. Smoking status did not alter any of the fusion or collapse rates, or the course of bone fusion.Conclusions This study demonstrates that allografts are suitable substitutes for autografts in instrumented ACDF. Prolonged time to union observed in allogenic bone grafts does not seem to be an important factor in instrumented procedures. Two-level grafting does not imply a significantly lower fusion rate, but longer time to union can be expected than with single-level instrumented procedures in both allograft and autograft subgroups. Our relatively small number of patients may not have been sufficient to decipher significant differences between smokers and non-smokers in the rate or course of fusion as previously reported.