SPINE Volume 39, Number 7, pp 571-578 ©2014, Lippincort Williams & Wilkins DEFORMITY Does Higher Anchor Density Result in Increased Curve Correction and Improved Clinical Outcomes in Adolescent Idiopathic Scoliosis? A. Noëlle Larson, MD,* David W. Polly, Jr., MD,+ Beverly Diamond, PhD,+ Charles Ledonio, MD,t B. Stephens Richards, III, MD,§ John B. Emans, MD,1 Daniel J. Sucato, MD, MS,§ Charles E. Johnston, MD,§ and the Minimize Implants Maximize Outcomes Study Group Study Design. Retrospective review of prospectively collected data. Objective. To determine whether anchor density is associated with curve correction and patient-reported outcomes. Summary of Background Data. There is limited information as to whether anchor density affects the results of adolescent idiopathic scoliosis surgery. Metliods. A total of 952 patients with adolescent idiopathic scoliosis met inclusion criteria (Lenke 1, 2, and 5 curves) with predominantly screw constructs (no. of screws/no, of total anchors >75%). Anchor density was defined as the number of screws, hooks, and wires per level fused, with less than 1.54 considered low density. Analysis of covariance was undertaken to determine association of anchor density with percent curve correction, Scoliosis Research Society (SRS), and Spinal Appearance Questionnaire (SAQ) scores, controlling for flexibility, fusion length, demographics, and surgeon. Results. High- compared with low-anchor density was associated with increased percent curve correction in Lenke 1 curves at 1 year (69% vs. 66% correction, P = 0.0022), controlling for percent preoperative curve flexibility, length of fusion, and sex (model, P < 0.0001 ). Similar associations held at 2-year follow-up and for Lenke 2 curves. Decreased thoracic kyphosis was found with increased anchor density for Lenke 1 and 2 curve patterns. There were no From the *Mayo Clinic, Rochester, MN; tUniversity of Minnesota, Minneapolis, MN; íPhDx Systems, Inc., Albuquerque, NM; §Texas Scottish Rite Hospital, Dallas, TX; and IChildren's Hospital, Boston, MA. A complete list of the members of the Minimize Implants Maximize Outcomes Study Group is given in the "Acknowledgment" section. Acknowledgment date: August 6, 2013. First Revision date: October 6, 2013. Second Revision date: December 10, 2013. Acceptance date: December 12, 2013. Tl?e device(s)/drug(s) is/are FDA-approved or approved by corresponding national agency for this indication. Scientific Forum/Spine Care grant from Orthopaedic Research and Fducation Foundation funds were received to support this work. Relevant financial activities outside the submitted work: grant, consultancy, grants/grants pending, payment for lectures, royalties, and stock/stock options. Address correspondence and reprint requests to David W. Polly, Jr., MD, Department of Orthopedic Surgery, University of Minnesota, 2450 Riverside Ave S, R200, Minneapolis, MN 55454; E-mail: pollydw@umn.edu DOI: 10.1097/BRS.0000000000000204 Spine associations found between anchor density and Lenke 5 curves. For Lenke 1 curve patterns at 2 years postoperatively, in the high- versus low-anchor density cohorts, there were statistically higher SRS Activity (4.3 vs. 4.2, P = 0.019), Appearance (4.3 vs. 4.1, P = 0.0005), Satisfaction (4.5 vs. 4.3, P = 0.028), and Total scores (4.3 vs. 4.2; P = 0.024). Similarly, the SAQ Appearance score at 1 year similarly was improved in the high-anchor density group (high: 14.1 vs. low: 15.0, P = 0.03) for Lenke 1 curve patterns only. Conclusion. For Lenke 1 and 2 curve patterns, improved percent correction of major coronal curve was noted in the high-screw density cohort. Although statistical significance was reached, it is unclear whether screw density resulted in clinically significant differences in patient-reported outcomes. Keywords: scoliosis, adolesent, idiopathic, implant, density, curve correction, outcomes. Level of Evidence: 3 Spine 2014;39:571-578 P edicle screws have become standard instrumentation for surgical correction of adolescent idiopathic scoliosis (AIS).'"^ Pedicle screws may have lower revision surgery rates and higher pull-out strength than hybrid or hook-rod constructs.'"" Significant variability exists, however, in the number of pedicle screws used in the surgical treatment of AIS.'''""' Suk et aF initially described pedicle screws placed in a similar array as Cotrel-Dubousset hooks but because of poor results abandoned this technique in favor of a higher anchor density construct. On the contrary, most surgeons would agree that the use of 2 pedicle screws at every level fused is not essential for good correction for most curve patterns. If proven clinically equivalent, constructs with fewer pedicle screws may have significant benefits for surgical efficiency and effective use of health care resources. With the introduction of pedicle screws, the cost of scoliosis surgery has increased because of high implant expenses,'^ which comprise on aver- age 29% of total hospital costs incurred during primary sco- liosis surgery."* Decreasing the number of implants used may lower the surgical cost. Furthermore, in pédiatrie patients, up to 9% of screws placed are malpositioned, which may rarely www.spinejournaLcom 571