Value of treating primary causes of syrinx in scoliosis associated with syringomyelia

Abstract

Study design: Retrospective chart and radiologic analysis.

Objectives: To analyze the value and risks of neurosurgical procedures for syringomyelia and of fusion procedures for scoliosis.

Summary of background data: There are few reports on the benefit of neurosurgical procedures on scoliosis, as well as the risk of fusion procedures on scoliosis related to syringomyelia.

Methods: The 105 patients who had both scoliosis and syringomyelia were subdivided into three groups: (I) 59 patients without congenital scoliosis or myelomeningocele, (II) 20 patients with congenital scoliosis and syringomyelia, and (III) 26 patients with myelomeningocele and syringomyelia. The study evaluated: (1) The effect of either suboccipital craniectomy or direct shunting on the syrinx, the curvature, and the neurologic status; (2) the neurologic risk of scoliosis fusion surgery; and (3) the curve status after fusion surgery.

Results: In Group I (patients without congenital scoliosis or myelomeningocele), suboccipital craniectomy resulted in curve improvement in seven, worsening in three, and no change in two, whereas direct syrinx shunting gave curve improvement in none, curve worsening in six, and no change in two. In Groups II and III, no patient had curve improvement after neurosurgery. Group III had the highest rate of subsequent neurosurgery (50%). The subsequent neurosurgery was nearly always suboccipital craniectomy or detethering of the cord. Three of 38 patients (8%) had neurologic worsening when scoliosis fusion surgery was done without previous syrinx decompression.

Conclusions: In patients without myelomeningocele or congenital scoliosis, but with Arnold-Chiari malformation and syringomyelia, suboccipital craniectomy gave the best chance for syrinx reduction and scoliosis improvement, particularly in children younger than 10 years. Syrinx shunting improved none of the scolioses. For syrinxes in patients with congenital scoliosis or myelomeningocele, neither neurosurgical procedure resulted in curve improvement, as other causes of scoliosis (vertebra anomalies, paralysis) remained untreated. Patients with myelomeningocele require a multipronged surgical approach to address all causes of syrinx, thus minimizing the potential need for repeat neurosurgery. Scoliosis correction without prior syrinx decompression carries a high neurologic risk.