Anterior only fusion for scoliosis in patients with myelomeningocele

A series of patients with single major scoliosis curvatures attributable to spina bifida treated by anterior only spinal fusion was studied for 2 year s to determine whether the infection rate could be decreased, adequate corr ection and pelvic balance could be provided, and posterior surgery could be avoided in these patients. Anterior surgery alone was performed for thorac olumbar scoliosis greater than 45 degrees if the compensatory thoracic curv e was less than 40 degrees and there was no significant junctional kyphosis . Fourteen patients were treated at a mean age of 11.9 years (range, 7-16 y ears), with a mean curve of 64 degrees (range, 51 degrees-85 degrees), and motor levels distributed from T10-L4. Thirteen patients had prior neurosurg ery for tether, syrinx, or Arnold-Chiari malformation. The spine was fused over a mean of seven vertebrae. A 3/16 inch Texas Scottish Rite Hospital ro d was used most commonly (10 patients). Blood loss averaged 1100 cc. The me an curve correction was 57% at 40 months after surgery. Loss of correction occurred primarily by adding on outside the instrumented area. Mean pelvic obliquity was improved from 16 degrees to 9 degrees. There was one superfic ial infection. Results were good in five patients, fair in four, and poor i n five. Failures were attributable to proximal decompensation in two patien ts who required revision surgery (two), neurologic deterioration in two, an d screw pullout in one. Both patients with decompensation had syringomyelia . Both patients with neurologic deterioration had large curves (>75 degrees ). Both patients recovered after rod removal. Retrospectively, by eliminati ng patients with syrinx or with a curve greater than 75 degrees, all poor r esults would be eliminated. Anterior only fusion and instrumentation may ha ve significant advantages, but only for selected patients with thoracolumba r curves less than 75 degrees, compensatory curves less than 40 degrees, no increased kyphosis, and no syrinx. Quadriceps function should be monitored . On the basis of this preliminary experience, continued use of this approa ch using stricter selection seems warranted.