[Multiple-level instrumented anterior cervical fusion: a risk factor for pseudoarthrosis? A prospective study with a...

This prospective study with minimal 3-year follow-up was performed to compare fusion rates, course of fusion, collapse incidence and occurrence of subsidence in one- and two-level instrumented anterior cervical fusions (ACDF) and thus to proof the hypothesis that use of internal fixation decreases the risk of non-union in bi-segmental ACDFs to the same level that can be expected in mono-segmental procedure.

In 79 consecutive patients operated upon by the Smith-Robinson technique for degenerative process of cervical spine in one or two levels was applied single instrumentation system in order to ensure ideal condition for solid bone fusion of 113 grafts (45 in one and 68 in two levels). All the patients were invariably followed for a minimum of 3 years.

Radiological criteria were used for evaluation of intervertebral fusion, graft collapse and its subsidence and results were statistically analyzed using M-L Chi-square test for the comparison of fusion and collapse incidence and further Chi-square test for the analysis of fusion course. All these figures were calculated at the level of significance 0.05 (alpha=0.05).

Overall, no significant difference was observed in achieving solid bone fusion 3 years after the surgery in one- and two-level procedures (95.6% vers. 92.6%, p=0.522), neither the bone graft collapse rate was of significant difference (2.2% vers. 7.6%, p=0.208). In single-level group the time to bone fusion was significantly shorter (p<0.001). When pooling the data into autologous and allogenic graft subgroup, there was observed no statistically significant difference in achieving union in autologous subgroup (100% vers. 90.9%, p=0.142); in allogenic subgroup this situation was similar: no significant difference in fusion rate (93.3% vers. 93.5%, p=0.980) was observed. In both auto- and allogenic subgroups monosegmentally implanted grafts fused more readily (p<0.001). There was no case of graft subsidence in any investigated group.

Our prospective study did not find any statistically significant difference in graft collapse and fusion rate when comparing one- and two-level instrumented ACDFs 3 years after the surgery. Plating system used in our patients brings more stability to operated segments and thus presumably prevents micromotions in postoperative period. Micromovements seems to be the major risk factor for non-union in non-instrumented multilevel cervical fusion. Other risk factor that should be considered in non-instrumented procedure is increase in compressive forces that are also partially eliminated by the semirigid internal fixation. Significantly delayed time to union observed in two level fusions shows most probably on increased number of surfaces that must be consolidated during the bone-healing process.

This study demonstrates that internal fixation used in multilevel ACDF decreases risk of pseudoarthrosis to the same level that can be expected in monosegmental procedures.