TEN YEARS OF SPINAL DEFORMITY CORRECTION IN THE PRESENCE OF A CORD SYRINX

Abstract

Introduction: The natural history of scoliosis in the presence of a cord syrinx, either treated conservatively, or post surgically, is disputed. It is generally believed to be associated with a greater likelihood of rapid deformity progression pre-operatively and a much greater likelihood of intraoperative neural injury.

In this study we aimed to retrospectively assess the local experience by reviewing patients, treated over the last 10 years, in whom scoliosis has been established, by means of MR imaging, to be associated with a cord syrinx.

Methods: A retrospective cohort study was undertaken of paediatric patients attending the Leeds Spinal Unit between the years of 1997 and 2007. The entry criteria for this study were spinal deformity with MRI proven cord syrinx, in a patient without underlying tumour or other cord anomaly. Given the association with Chiari malformation this was a measured parameter rather than exclusion criteria. Other parameters assessed were mode of presentation, progression of scoliosis, details of the identified syrinx, chosen mode of treatment for the syrinx and the scoliosis (including conservative) and outcome measures (neurological function and sequelae, change in Cobb angle/deformity correction).

Results: A total of 46 patients were identified with scoliosis and an associated syrinx. The age range was from 3 to 18 years. Only 12 were male. The syrinx was associated with an Arnold Chiari malformation in 24% of patients, and located at the apex of, or local to, the maximum deformity in 73%.

The syrinx was treated surgically in 10 patients, with 80% of these achieving either deformity arrest, or no longer requiring surgical deformity correction. In the 2 patients from the same subset who did undergo deformity correction there was no neurological sequelae. Of the conservatively managed syrinxes, deformity correction with intraoperative cord monitoring was nevertheless undertaken in 31%, all without neurological sequelae. In just 4 patients (of 69%) who did not proceed to deformity correction, surgery was precluded by the inherent risks in the presence of an untreated syrinx.

Conclusion: This small series does not lend support to the literature and anecdotal evidence for significantly increased surgical risk in deformity correction without treatment for syrinx first. However, this may reflect the fact that all syrinxes likely to compromise the surgical procedure were assessed as such and treated first. In the cohort of patients whose syrinxes were treated conservatively, a significant proportion did not require subsequent deformity surgery. Identifying a syrinx by, the mandatory, MR imaging of a patient with a deformity before considering surgical correction, appears to identify a significant proportion of syrinxes which neither significantly accelerate the progression deformity, or which do not lead to poor outcome after deformity surgery.