Rigid Fixation Improves Outcomes of Spinal Fusion for C1-C2 Instability in Children with Skeletal Dysplasias

Background: Upper cervical instability is common in many skeletal dysplasias, and surgical treatment can be difficult because of small, fragile osseous elements. In this study of children with skeletal dysplasia and upper cervical instability, we compared fusion rates and complications between (1) patients treated with no instrumentation or with wiring techniques and (2) those who underwent rigid cervical spine instrumentation. We also sought to characterize the presentation and common parameters of upper cervical instability in this population. Methods: A multicenter study identified twenty-eight children with skeletal dysplasia who underwent surgery from 2000 through 2011 for C1-C2 instability and were followed for a minimum of two years. Fourteen children were treated with no instrumentation or with instrumentation with wires or cables (nonrigid-fixation group) and fourteen were treated with screws (or hooks) and rods (rigid-fixation group). All patients received autograft, and twenty (twelve in the nonrigid group and eight in the rigid group) were treated with a halo-body jacket. Results: Fourteen children had C1-C2 fusion, and fourteen had occipitocervical fusion. Eleven (39%) underwent spinal cord decompression. The nonunion rate was significantly higher in the nonrigid-fixation group (six of fourteen) than in the rigid-fixation group (zero of fourteen; p = 0.0057). Complications, including nonunion, occurred in nine patients in the nonrigid group and one patient in the rigid group. However, two of the complications in the nonrigid-fixation group were vertebral artery bleeding episodes that actually occurred during an attempt at rigid fixation (the fixation was subsequently done with wiring). No new neurologic deficits were observed. Five of the seven children with a preoperative neurologic deficit showed at least partial recovery, with significant improvement in the Japanese Orthopaedic Association upper-extremity score (p = 0.047). Conclusions: The nonunion rate is relatively high after patients undergo spinal fusion for C1-C2 instability with nonrigid instrumentation, even if a halo-body jacket is applied. Rigid fixation with screws and rods improves fusion rates.