Pedicle Screw Instrumentation for Adult Idiopathic Scoliosis An Improvement Over Hook/Hybrid Fixation

Study Design. A matched cohort comparison of adult idiopathic scoliosis (AdIS) patients treated with all pedicle screw constructs compared to hook/hybrid constructs. Objective. To compare clinical and radiographic results of AdIS treatment using all pedicle screw constructs versus hook/hybrid constructs. Summary of Background Data. Pedicle screw instrumentation has demonstrated excellent clinical efficacy in the treatment of pediatric spinal deformity. No prior reports have compared the outcomes of pedicle screw only constructs to hook/hybrid constructs in the treatment of AdIS. Methods. We analyzed 34 consecutive patients undergoing posterior-only correction for AdIS, using pedicle screw instrumentation at minimum 2-year follow-up. Thirty-four matching patients (11 with anterior releases) were selected from a cohort of 58 patients treated with hook/hybrid constructs based on similar age, curve type, magnitude, and fusion levels. Results. Significantly greater curve correction was seen in the pedicle screw compared to the hook/hybrid group (56 vs. 40%, P < 0.01). Coronal and sagittal imbalance were equivalent between the groups (P = 0.91 and 0.23, respectively). Thoracic kyphosis (T5-T12) was maintained in the pedicle screw patients but significantly increased in the hybrid/hook patients over time (P < 0.05). Scoliosis Research Society outcome scores significantly improved in both groups. Blood loss was equivalent but operative time was longer in the hook/hybrid patients. No pedicle screw patients were revised for instrumentation complications with 1 lumbosacral nonunion revised at 5 years postoperative (3% revision rate). Eight of 58 patients among the hook/hybrid cohort underwent 9 revisions for instrumentation failure (n = 3) or nonunion (n = 6) (14% revision rate; P = 0.04). Conclusion. Pedicle screw correction of AdIS is safe and effective. Compared to hook/hybrid constructs, these patients displayed significantly improved correction of the major curve (even in the absence of anterior releases), maintenance of thoracic kyphosis, and a lower revision rate. Similar SRS scores and blood loss were noted. The hook/hybrid patients had significantly longer operative times. The use of BMP-2 in 7 patients and TLIF/ALIF also in 7 of the pedicle screw patients may have decreased the need for revision surgery up to the latest follow-up.