Efficacy of Spinal Instrumentation and Fusion in the Prevention of Post-Laminectomy Spinal Deformity in Children with Intramedullary Spinal Cord Tumors

Purpose: To examine the effectiveness of single-stage laminectomy, spinal cord decompression, and spinal instrumentation and fusion for the prevention of post-laminectomy spinal deformity in children with intramedullary spinal cord tumors.

Methods: 129 children with spinal cord tumors (excluding neurofibromas and schwannomas) have been treated at our institution since 1970. Of this group, 47 had biopsy-proven intramedullary tumors. The remainder had extramedullary tumors. 16 children with intramedullary tumors were excluded from this study due to: death within two years of diagnosis (8), diagnosis within the last two years (3), lack of sufficient clinical follow-up (5). Clinical charts and radiographs were reviewed for treatment description, pre-operative spinal alignment, post-operative spinal alignment, and subsequent medical or surgical treatment. Significant spinal deformity was defined as the presence of cervical kyphosis, thoracic kyphosis > 60º, or an unbalanced scoliosis.

Results: The study group consisted of 31 children, 19 males and 12 females, with mean age at presentation of 9 years (range 2.5-14.9 years) and mean follow-up of 6.5 years (range 2-22.7 years). The diagnoses represented are: astrocytoma (18), ependymoma (7), and ganglioglioma (6). An average of 6.7 laminae per patient were removed for tumor removal (range 2-15). The initial surgical treatment consisted of laminectomy only in 16 children, laminoplasty in four, and laminectomy plus spinal fusion in 11. Of the eleven children with spinal fusion at the time of tumor removal, six had spinal instrumentation placed while five had in situ fusion with bone graft only. Significant spinal deformity developed in eight of 16 children with laminectomy only, three of four children with laminoplasty, and three of 11 children with spinal fusion (1/6 with spinal instrumentation, 2/5 with in situ fusion). While not statistically significant, spinal fusion of any type (p=0.14) and spinal instrumentation (p=0.12) showed a trend towards reduced incidence of post-laminectomy spinal deformity compared with laminectomy and laminoplasty. Other features of children with increased risk of post-laminectomy spinal deformity were removal of greater than three laminae and use of radiation therapy (RT), however, these also did not reach the threshold for statistical significance. 12 of 24 children with greater than three laminae removed developed significant spinal deformity compared with only two of seven with three or fewer laminae removed (p=0.32). Ten of 19 children receiving RT following surgery developed significant spinal deformity compared with four of 12 not receiving RT (p=0.29). We could not demonstrate a relationship between age, gender, specific diagnosis, location of laminectomy, or presence of syrinx with the development of significant spinal deformity.

Conclusions: Post-laminectomy spinal deformity is a frequent problem in the pediatric population. This study demonstrates the effectiveness of spinal instrumentation and fusion for the prevention of post-laminectomy spinal deformity in children with intramedullary spinal cord tumors. Although risks include increased operative time, blood loss, and complications associated with spinal instrumentation and fusion, we believe that in most children, the benefits such as ease of exposure, prevention of second general anesthesia, and prevention of spinal deformity outweigh these risks.