Introduction and Aims: Surgical correction of pelvic obliquity is an important component of spinal instrumentation for neuromuscular scoliosis, though instrumentation to the pelvis has high reported complication rates. This study evaluates the results of pelvic fixation during surgical correction of neuromuscular scoliosis in a consecutive series of 62 children and adolescents.

Method: A retrospective chart and radiographic review of 62 consecutive patients treated with spinal fusions to the pelvis as treatment for neuromuscular scoliosis was performed. Follow-up ranged from two to seven years. Diagnoses included cerebral palsy (36 patients), muscular dystrophy (16 patients), myelomeningocele (three patients), spinal muscular atrophy (three patients) and other disorders (four patients). Mean age at surgery was 13.5 years. Pelvic fixation techniques used included Luque-Galveston or iliosacral screw fixation. Correction of deformity in each patient was assessed with Cobb angle measurements of scoliosis, thoracic kyphosis, and lumbar lordosis. Pelvic obliquity and coronal decompensation was also assessed.

Results: The Luque-Galveston spinal instrumentation technique was used in 54 patients and iliosacral screw fixation was used in eight patients. Seventeen patients had an additional anterior release and fusion without instrumentation. The mean Cobb angle measured 73 degrees pre-operatively and 31 degrees (mean correction 59%) post-operatively. The mean Cobb angle on latest follow-up was 33 degrees (loss of correction 12%). Thoracic kyphosis remained essentially unchanged, as did lumbar lordosis (56 pre-op and 61 on follow-up). Pelvic obliquity corrected from a mean of 16 degrees pre-operatively to eight degrees on most recent follow-up. Mean pre-operative coronal decompensation measured 135mm, and follow-up decompensation measured 46mm. Eleven patients with Galveston fixation exhibited the ‘windshield-wiper’ sign, with a radiolucency of 2mm or more, though most were asymptomatic. Wound infection was observed in 6% (3/54) of the patients who underwent Galveston instrumentation and 50% (4/8) who had iliosacral screws. In patients treated with Galveston fixation, three had symptomatic prominant hardware and one had hardware breakage for an overall mechanical failure rate of 7% (4/54). In contrast, two patients with iliosacral screws had construct breakage and pseudoarthrosis for a mechanical failure rate of 25% (2/8), though the numbers in the iliosacral screw group are small.

Conclusions: In this series, Galveston pelvic fixation during spinal instrumentation treatment of neuromuscular scoliosis was associated with satisfactory results and with less complications than generally reported in the literature. This technique is recommended as the preferred method for pelvic fixation in severe neuromuscular scoliosis associated with pelvic obliquity.

Introduction and Aims: Thoracolumbar adolescent idiopathic scoliosis may be treated surgically with anterior or posterior spinal instrumentation, with little evidence in the literature to suggest superiority of either technique. The purpose of this study is to compare anterior vs. posterior instrumentation in a well-defined population of patients with thoracolumbar adolescent idiopathic scoliosis.

Method: Medical records and radiographs of all patients undergoing spinal instrumentation for the treatment of adolescent idiopathic scoliosis with primary thoraco-lumbar curves, defined as curve apices between T10 and L2, between 1993 and 2001 were reviewed. Fusions extending above T7 were excluded from the study. The study group consists of 12 patients treated with anterior spinal instrumentation and 16 with posterior instrumentation. Various radiographic and outcome measures were compared between groups.

Results: The anterior group had 75% correction of the primary Cobb angle compared to 56% in the posterior group (P = 0.019). An average of 3.8 vertebral levels in the anterior and 6.7 in the posterior procedures were fused (P < 0.001). Less blood loss was observed in the anterior group (P = 0.007), with fewer transfusions as well (P < 0.001). The anterior group produced more lumbar lordosis (p=0.03) than the posterior group. In the anterior group there was a 0% rate of revision surgery (0/12), whereas the posterior group had a 31% revision rate (5/16), which was a significant difference (p=0.047).

This study comparing anterior versus posterior instrumentation is unique in that it is limited to thoracolumbar curves. While earlier series of anterior instrumentation revealed high rates of hardware failure and pseudoarthrosis, this series found no instance of either in the anterior group. In addition, concern over anterior compression causing kyphosis at the thoracolumbar junction proved unwarranted, and in fact the anterior instrumented group had improved lumbar lordosis compared to the posterior.

Conclusions: In this series limited to thoracolumbar idiopathic scoliosis surgery, anterior instrumentation had a significantly improved Cobb angle, less levels fused, more lumbar lordosis, and less transfusions when compared to posterior instrumentation. Patients undergoing anterior instrumentation had a lower rate of revision surgery compared to those with posterior instrumentation.