The AO Foundation and Orthopaedic Trauma Association (AO/OTA) scapula fracture classification system: focus on glenoid fossa involvement Martin Jaeger, MD a , Simon Lambert, FRCSEd(Orth) b , Norbert P. S € udkamp, MD a , James F. Kellam, MD c , Jan Erik Madsen, MD d , Reto Babst, MD e , Jonas Andermahr, MD f , Wilson Li, MD g , Laurent Audig  e, PhD h, * a Department of Orthop€ adie und Traumatologie, Universit€ atsklinikum Freiburg, Albert-Ludwigs-Universit€ at, Freiburg, Germany b Shoulder and Elbow Service, The Royal National Orthopaedic Hospital, Stanmore, UK c Department of Orthopaedic Surgery, Carolinas Medical Center, Charlotte, NC, USA d Orthopaedic Department, Oslo University Hospital, Ullevaal, Norway e Klinik f€ ur Unfallchirurgie, Luzerner Kantonsspital, Luzern, Switzerland f Department for Orthopedic and Trauma Surgery, Hospital of the University of Bonn, Mechernich, Germany g Department of Orthopaedics & Traumatology, Queen Elizabeth Hospital, Kowloon, Hong Kong h AO Clinical Investigation and Documentation, D€ ubendorf, Switzerland Background: Fractures of the glenoid frequently require surgical treatment. A comprehensive and reliable scapula classification system involving the glenoid fracture patterns is needed to describe the underlying pathology. The AO Scapula Classification Group introduces an appropriate novel system that is presented along with its inter-rater reliability and accuracy. Materials and Methods: An iterative consensus process (involving a series of face-to-face meetings and agreement studies) with an international group of 7 experienced shoulder surgeons was used to specify and evaluate a scapular fracture classification system with a focus on fracture patterns of the glenoid fossa. The last evaluation was conducted on a consecutive collection of 120 scapular fractures documented by both plain radiographs and computed tomography scans including 3-dimensional surface rendering. Inter-rater reliability was analyzed with k statistics, and accuracy was estimated by latent class modeling. Results: Of 120 scapular fractures, 46 involved the glenoid (38%), with 38 classified as F1 articular rim fractures. The overall median sensitivity and specificity in identifying these fractures were 95% and 93%, respectively. Surgeons’ accuracy in classifying F1 fractures ranged from 86% to 100% (median, 94%). Subsequently, classification of simple F1 fractures resulted in a proportion of 36% of anterior rim fractures, 19% of posterior rim fractures, and 45% of short oblique fractures, with accuracies ranging from 85% to 98%. Conclusion: This new system for scapular glenoid fractures has proved to be sufficiently reliable and accu- rate when applied by experienced shoulder surgeons. Further validation of the most detailed system, as well Institutional review board/ethics committee approval: not required. *Reprint requests: Laurent Audig e, PhD, AO Clinical Investigation and Documentation, Stettbachstrasse 6, CH-8600 D€ ubendorf, Switzerland. E-mail address: laurent.audige@aofoundation.org (L. Audig e). J Shoulder Elbow Surg (2012) -, 1-9 www.elsevier.com/locate/ymse 1058-2746/$ - see front matter Ó 2012 Journal of Shoulder and Elbow Surgery Board of Trustees. http://dx.doi.org/10.1016/j.jse.2012.08.003