Lumbar pedicle screws versus hooks. Results in double major curves in adolescent idiopathic scoliosis.

STUDY
DESIGN: A retrospective assessment of the effectiveness of lumbar pedicle screws versus laminar hooks in lumbar curve correction with double major curves in adolescent idiopathic scoliosis.
OBJECTIVE: To determine if pedicle screw fixation of the lumbar spine has any advantage compared with multiple laminar hook instrumentation in the treatment of double major curves in adolescent idiopathic scoliosis. SUMMARY OF BACKGROUND DATA: Although hooks have been used most commonly, pedicle screws may offer advantages in correction and maintenance of reduction of the lumbar curve in adolescent idiopathic scoliosis.
METHODS: A consecutive series of 39 patients with double major curves underwent thoracic and lumbar instrumentation by a single surgeon. Lumbar pedicle screws and hooks were used in 20 patients (Group S) and in 19 patients only lumbar hooks were used (Group H). Thoracic Cotrel-Dubousset instrumentation with hooks was the same in both groups. Preoperative age, gender, bracing, and Cobb angles were similar in both groups. Preoperative, 1-month postoperative, and latest follow-up standing posteroanterior and lateral spine radiographs were blinded to the surgeon and lumbar instrumentation covered to hide its identity. Measurements included Cobb angles, preoperative flexibility, lumbar and thoracic apical vertebral deviation, and reduction of lateral tilt and lateral displacement of the first free lumbar vertebra below the instrumentation. Percent correction, maintenance of correction at follow-up, and total levels fused were calculated.
RESULTS: The mean follow-up was 3.5 years (range, 2-8 years), which was similar for Groups H and S. Pedicle screws appear to offer some advantage in lumbar curve correction, maintenance of correction, and correction of the uninstrumented spine below the fusion when compared with the use of hooks alone. Horizontalization of the first free lumbar vertebra below the instrumentation percent correction of tilt: 62% screws vs. 11% hooks; P = 0.0003), residual tilt (8 degrees screws vs. 17 degrees hooks; P = 0.004), and loss of horizontalization at follow-up (5% screws vs. 26% hooks) were dramatically better for the group using screws. Lumbar curve correction (72% screws vs. 60% hooks; P = 0.026), loss of lumbar curve correction (5% screws vs. 13% hooks), and correction of lateral apical vertebral deviation (2.2-cm screws vs. 1.5-cm hooks or 63% vs. 31%; P = 0.013) were better when screws were used. There was no significant difference in loss of correction of the thoracic curves (35% vs. 37%) or any difference in loss of correction of lateral displacement of the thoracic apical vertebra (12% vs. 14%). There was no difference in total levels fused, operative blood loss, operative time, or ultimate patient outcome. No patients in either group had spinal imbalance at latest follow-up. There were no complications related to pedicle screw placement. Two cases of transient postoperative superior mesenteric artery syndrome (duodenal obstruction by the superior mesenteric artery) in the pedicle screw group are attributed to acute correction of the lumbar scoliosis and thoracolumbar kyphosis with resultant lordosis at the thoracolumbar junction.
CONCLUSIONS: Lumbar pedicle screws may offer greater lumbar curve correction, better maintenance of correction, and greater correction of the uninstrumented spine below double major curves. No complications were associated with the placement of pedicle screws.