Copyright © 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Supine Lateral Bending Radiographs Predict the Initial In-brace Correction of the Providence Brace in Patients With Adolescent Idiopathic Scoliosis Søren Ohrt-Nissen, MD, Dennis W. Hallager, MD, Martin Gehrchen, MD, PhD, and Benny Dahl, MD, PhD, DSc Study Design. Retrospective, cross-sectional. Objective. To determine the initial curve correction of the providence brace (PB) and to determine to what extend the in- brace Cobb angle corresponds to the curve seen on supine lateral bending radiographs (SLBR). Summary of Background Data. SLBR are used to assess curve flexibility in patients undergoing surgical treatment for adolescent idiopathic scoliosis (AIS). A low rate of in-brace correction (IBC) has been associated with a higher risk of curve progression, but to what extent SLBR can be used to predict IBC before initiating bracing treatment is unknown. Methods. All patients with AIS treated with the PB from January 1, 2006 to December 31, 2013 with a major curve of 25 to 45 degrees8 were included. Cobb angle on SLBR before treatment and on initial standing, in-brace radiograph (IBR) were measured twice for each patient by one observer 30 days apart. Using a repeated measure mixed effect model, mean difference and 95% limits of agreement (LOA) between Cobb angles on each type of radiograph were estimated. Correction index (CI) was calculated as: curve flexibility (%)/curve correction (%). Results. A total of 127 patients were included. Mean long- standing Cobb angle was 358 (SD: 68), which was reduced to mean 138 (SD: 8) on IBR (P < 0.05). No difference in curve correction between curve types was found when adjusting for flexibility using correction index (P ¼ 0.77). Overall mean difference between SLBR and IBR was 0.28 (LOA  108). Mean difference for thoracic curves was 0.28 (LOA  88), for thoraco- lumbar/lumbar curves 0.98 (LOA  108) and for double major curves 0.48 (LOA  16). Conclusion. SLBR provide a close estimation to the expected in-brace correction with a mean difference of less than one degree. SLRB could potentially serve as prognostic parameter for curve progression before initiating brace treatment. Key words: adolescent idiopathic scoliosis ;bland-Altman plot, correction index, curve flexibility, curve type, initial in-brace correction: repeated measurements, limits of agreement, providence brace, supine lateral bending radiograph. Level of Evidence: 3 Spine 2016;41:798–802 A dolescent idiopathic scoliosis (AIS) is a three-dimen- sional deformity of the spine affecting about 2 to 3% of adolescents. The majority of curves are minor and do not require intervention, however, approxi- mately 0.5 to 1.5% of adolescents present with curves more than 20 degrees for which treatment may be required. 1–3 The only viable nonoperative treatment in such cases involves bracing of the spine until skeletal maturity and is typically indicated in the immature patients with a curve size between 25 and 45 degrees. 4,5 Various bracing principles exist and the most frequently described is the full-time thoracolumbosacral orthosis (TLSO). This brace is designed to be used > 23 hours a day and has shown to be superior to observation in a randomized clinical trial. 4 However, var- ious studies have reported a high rate of noncompliance and an associated poor outcome. 6–8 To increase compliance, braces designed for nighttime use only have been devel- oped. 9–11 Among the most frequently reported are the Charleston and the providence brace (PB). The PB is a hyper corrective brace that uses an acrylic frame to apply direct, lateral, and rotational forces to the trunk. 9 Both the radio- graphic and clinical outcome has been described in the literature and initial long-term results with the PB are comparable with that of the full-time TLSO. 9,11–13 From the Spine Unit, Department of Orthopedic Surgery, Rigshospitalet, University of Copenhagen, Denmark. Acknowledgment date: September 1, 2015. First revision date: October 15, 2015. Acceptance date: October 25, 2015. The article submitted does not contain information about medical device(s)/ drug(s). Globus Medical, Medtronic, and K2 M grant funds were received in support of this work. No relevant financial activities outside the submitted work. Address correspondence and reprint requests to Søren Ohrt-Nissen, MD, Spine Unit, Department of Orthopedic Surgery, Rigshospitalet, University of Copenhagen, Copenhagen, East Denmark; E-mail: ohrtnissen@gmail.com DOI: 10.1097/BRS.0000000000001519 798 www.spinejournal.com May 2016 SPINE Volume 41, Number 9, pp 798–802 ß 2016 Wolters Kluwer Health, Inc. All rights reserved DIAGNOSTICS