Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch Year: 2014 Scapula fractures: interobserver reliability of classification and treatment Neuhaus, Valentin ; Bot, Arjan G J ; Guitton, Thierry G ; Ring, David C Abstract: OBJECTIVES:There is substantial variation in the classification and the management of scapula fractures. The first purpose of this study was to analyze the interobserver reliability of the OTA/AO and the New International Classification of scapula fractures. The second purpose was to assess the proportion of agreement among orthopaedic surgeons on operative or nonoperative treatment. DESIGN:: Web-based reliability study SETTING:: Independent orthopaedic surgeons from several countries were invited to classify scapular fractures in an online survey. PARTICIPANTS:One-hundred and three orthopaedic surgeons evaluated 35 movies of 3DCT-reconstruction of selected scapular fractures, representing a full spectrum of fracture patterns. MAIN OUTCOME MEASUREMENTS:Fleiss’ kappa (฀) was used to assess the reliability of agreement between the surgeons. RESULTS:: The overall agreement on the OTA/AO Classification was moderate for the types (A, B, and C, ฀ = 0.54) with a 71% proportion of rater agreement (PA) as well as for the nine groups (A1 to C3, ฀ = 0.47) with a 57% PA. For the New International Classification, the agreement about the intra-articular extension of the fracture (Fossa (F), ฀ = 0.79) was substantial, the agreement about a fractured body (Body (B), ฀ = 0.57) or process was moderate (Process (P), ฀ = 0.53), however PAs were more than 81%. The agreement on the treatment recommendation was moderate (฀ = 0.57) with a 73% PA. CONCLUSIONS:The New International Classification was more reliable. Body and process fractures generated more disagreement than intra-articular fractures and need further clear definitions. DOI: https://doi.org/10.1097/BOT.0b013e31829673e2 Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-78945 Journal Article Published Version Originally published at: Neuhaus, Valentin; Bot, Arjan G J; Guitton, Thierry G; Ring, David C (2014). Scapula fractures: interobserver reliability of classification and treatment. Journal of Orthopaedic Trauma, 28(3):124-129. DOI: https://doi.org/10.1097/BOT.0b013e31829673e2 ORIGINAL ARTICLE Scapula Fractures: Interobserver Reliability of Classification and Treatment Valentin Neuhaus, MD, Arjan G. J. Bot, MD, Thierry G. Guitton, MD, PhD, and David C. Ring, MD, PhD; The Science of Variation Group Objectives: There is substantial variation in the classification and management of scapula fractures. The first purpose of this study was to analyze the interobserver reliability of the OTA/AO classification and the New International Classification for Scapula Fractures. The second purpose was to assess the proportion of agreement among orthopaedic surgeons on operative or nonoperative treatment. Design: Web-based reliability study. Setting: Independent orthopaedic surgeons from several countries were invited to classify scapular fractures in an online survey. Participants: One hundred three orthopaedic surgeons evaluated 35 movies of three-dimensional computerized tomography reconstruction of selected scapular fractures, representing a full spectrum of fracture patterns. Main Outcome Measurements: Fleiss kappa (k) was used to assess the reliability of agreement between the surgeons. Results: The overall agreement on the OTA/AO classification was moderate for the types (A, B, and C, k = 0.54) with a 71% proportion of rater agreement (PA) and for the 9 groups (A1 to C3, k = 0.47) with a 57% PA. For the New International Classification, the agreement about the intraarticular extension of the fracture (Fossa (F), k = 0.79) was substantial and the agreement about a fractured body (Body (B), k = 0.57) or process was moderate (Process (P), k = 0.53); however, PAs were more than 81%. The agreement on the treatment recommendation was moderate (k = 0.57) with a 73% PA. Conclusions: The New International Classification was more reliable. Body and process fractures generated more disagreement than intraarticular fractures and need further clear definitions. Key Words: OTA/AO fracture classification, New International Classification for Scapular Fractures, reliability, scapula (J Orthop Trauma 2014;28:124–129) Accepted for publication April 10, 2013. From the Orthopaedic Hand Service, Massachusetts General Hospital, Boston, MA. No funding was received in direct support of this study. No authors have any connections to either of the classification systems. V. Neuhaus has received a grant from the Bangerter Foundation Switzerland. A.G. J. Bot has received grants from VSB fonds, Beurs Prins Bernhard Cultuur fonds, beurs/Banning-de Jong fonds, Stichting Anna fonds, the Netherlands. T. G. Guitton had no conflicts of interest. D. C. Ring has no conflicts related to this research. Reprints: David C. Ring, MD, PhD, Orthopaedic Hand Service, Yawkey Center, Suite 2100, Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114 (e-mail: dring@partners.org). Copyright © 2013 by Lippincott Williams & Wilkins 124 | www.jorthotrauma.com INTRODUCTION Our traditional complacence with fractures of the shoulder girdle was altered by recent data showing that operative treatment of displaced clavicle fractures reduces the risk of nonunion and pain as well as improves functional results.1–3 Now, some are suggesting that more frequent operative treatment of scapula fractures should be considered.4–9 The indications for surgery are not clearly defined, and the role of classification schemes is uncertain. A New International Classification for Scapular Fractures was recently developed by a study group of 6 orthopaedic trauma surgeons hoping to develop a better fracture classification system and later to clarify the prognostic value of it for indications for operative treatment.10 This study sought to compare the OTA/AO classification with the New International Classification for Scapular Fractures. Our primary study aim was to measure the reliability of the OTA/AO classification and the New International Classification for Scapula Fractures. The second aim was to evaluate the agreement on operative treatment. PATIENTS AND METHODS Study Design Orthopaedic surgeons from 25 countries participating in the Science of Variation Group, a web-based collaborative of experienced orthopaedic surgeons, were invited to evaluate and rate 35 movies of three-dimensional computerized tomography (3DCT) reconstruction of scapular fractures in an online survey in May and June 2012.11 The movies were presented online in a random order and were assessed independently by the raters. A description of the OTA/AO12 classification and the New International Classification10 was provided for each movie. No other information (additional injuries, treatment, outcome) was made available. The raters were asked to classify the presented scapular fractures (OTA/AO and the New International Classification) and to propose operative or nonoperative treatment in young, active, and healthy patients. There was no time limit to complete the questionnaire. Raters One hundred sixty-eight (21%) of the 802 invited surgeons agreed to participate in the study (a large percentage of our collaborative do not treat scapula fractures), and 103 surgeons (61%) completed all questions (Table 1). They were not involved in the treatment of the patients presented in this study cohort and did not receive any incentives other than an acknowledgement in this article. J Orthop Trauma  Volume 28, Number 3, March 2014 J Orthop Trauma  Volume 28, Number 3, March 2014 Reliability of Scapula Fracture Classification TABLE 1. Surgeons’ Demographics Parameter All questions answered Sex Male Female Area of practice Australia Canada Europe United Kingdom United States Other Years of independent practice 0–10 More than 10 years Specialization Orthopaedic traumatology Shoulder and elbow Hand and wrist General orthopaedics or other n 103 96 7 4 4 28 3 47 17 51 52 44 23 25 11 n, number of surgeons. Fractures Under Institutional Review Board approval (protocol #: 2009-P-001019/89; Massachusetts General Hospital), a total of 457 scapular fractures were identified from a prospectively collected trauma database (from 2002 to 2011) at two level 1 trauma centers. Inclusion criteria were as follows: (1) adult patients ($18 years) and (2) adequate quality (slice thickness #2.5 mm) and completeness of computed tomographies for 3D reconstruction, leaving a cohort of 225 suitable fractures. Thirty-five fractures were selected with complete, high-quality CTs and representing a full spectrum of scapular fracture patterns. Sex, age, side, concomitant injuries, Injury Severity Score, radiological measurements (intraarticular step-off, medialization, translation, angulation, glenopolar angle, presence of a double disruption of the superior shoulder suspensory complex), and the received treatment (either operative or nonoperative) were independently of the surgeons’ ratings recorded for readers information. The radiographic measurements were performed with the Aquarius workstation (Version 4.4.6; TeraRecon, Inc., San Mateo, CA) for one institution and with the Centricity software (GE Healthcare, Buckinghamshire, United Kingdom) for the other institution by an independent experienced orthopaedic surgeon. The movies were created with Osirix13 (OsiriX Foundation/Pixmeo, Geneva, Switzerland) and were rotating (360 degrees around a vertical axis with a duration of 10 seconds) 3DCT reconstructions of the whole scapula with humerus and clavicle subtracted. The raters could replay the videos as needed. Statistical Analysis For each fracture, the most commonly proposed answers and the proportion of agreement (in percentage, PA) were presented and analyzed. The multirater agreement of the Ó 2013 Lippincott Williams & Wilkins FIGURE 1. Anterior and Y-view of the 3DCT reconstruction (fracture 11). Surgeons had problems deciding if the body and/or the process was involved. Editor’s note: A color image accompanies the online version of this article. nominal variables (OTA/AO classification; the New International Classification; recommended treatment) were calculated with the Fleiss generalized Kappa,14,15 which is a statistical chance-corrected measure for assessing multirater agreement with binary or nominal ratings. The calculated measures are presented as a value between 0 and 1 and are called Kappa value. They were interpreted in accordance to the guidelines by Landis and Koch16: 0.01–0.20 represent slight agreement, 0.21–0.40 fair agreement, 0.41–0.60 moderate agreement, 0.61–0.80 substantial agreement, and above 0.81 is considered almost perfect agreement. In another study, a value of 0.70 was considered an adequate sign of reliability.17 RESULTS OTA/AO Classification The proportion of the most proposed AO group of all answers varied between 26% and 99% for the 35 cases. The FIGURE 2. Forty-three percent of the raters classified fracture 9, which is very similar to fracture 3, as a C2 fracture (total articular; intraarticular fracture with neck) and 40% as a C3 fracture (intraarticular fracture with body). Editor’s note: A color image accompanies the online version of this article. www.jorthotrauma.com | 125 J Orthop Trauma  Volume 28, Number 3, March 2014 Neuhaus et al TABLE 2. OTA/AO Classification Fracture No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Raters (n) Most Proposed AO Type % All Answers 162 153 137 131 131 129 124 119 115 114 111 110 109 108 107 106 106 105 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 C3 A3 C2 A3 A3 A1 A3 C3 C2 A2 B1 A3 C3 A3 A1 A3 C3 A3 A3 A3 B1 C3 B1 C3 B3 A3 C3 C2 B3 C1 A3 C3 A3 C3 C3 59 95 26 99 99 99 75 70 43 94 41 78 52 82 93 88 69 35 96 84 34 40 72 56 75 88 61 60 43 64 77 69 82 69 55 Parameter Overall 3 Types (A, B, C) 9 Groups (A1 - C3) Years of practice 0–10 More than 10 years Specialization Orthopaedic traumatology Shoulder and elbow Hand and wrist Agreement New International Classification Agreement about the intraarticular extension of the fracture was substantial (k = 0.79, PA 90%), and for shoulder and hand surgeons almost perfect (k = 0.83 and k = 0.80, respectively). The agreement about a fractured body (k = 0.57, PA 82%) or process was moderate (k = 0.53, PA 80%). Another source of disagreement was fracture of the glenoid neck (Fig. 3), which was less of a problem with the OTA/AO classification. The further in-depth classification showed a fair agreement for body fractures (B1, B2, or B not applicable; k = 0.35, PA 58%) and a moderate agreement on fractures involving the fossa (F0, F1, F2, or F not applicable; k = 0.59, PA 74%) as well as process fractures (P1, P2, P3, or P not applicable; k = 0.46, PA 73%) (Table 3). Recommended Treatment Nonoperative treatment was most often recommended in 21 fractures and operative treatment in 14 fractures. Kappa PA (%) Moderate Moderate 0.54 0.47 71 57 Moderate Moderate 0.48 0.46 58 56 Moderate Moderate Moderate 0.46 0.51 0.43 56 61 53 most proposed AO group was A3 in 13 fractures, C3 in 10 fractures, B1 in 3 fractures, C2 in 3 fractures, and other groups in 6 fractures. The overall agreement on the OTA/AO classification was moderate for the 3 types of fracture (k = 0.54) with a 71% PA and for the 9 groups of fracture (k = 126 | www.jorthotrauma.com 0.47) with a 57% PA. There was a higher agreement on type A fractures (k = 0.72), and a lower one on type C (k = 0. 46) and B (k = 0.37) fractures. Although the agreements on the groups A1 (k = 0.77), A3 (k = 0.74), and A2 (k = 0.65) were highest, they were lowest for the C1 (k = 0.25), C2 (k = 0.20), and B2 (k = 0.02) fractures. Most disagreements were between B1 (anterior rim fracture), C2 (intraarticular fracture with neck), and A2 (coracoid fracture). For example, 41% of the raters classified fracture 11 (Fig. 1) as an AO type B1 (partial articular; anterior rim fracture), 28% as an A2 (extraarticular coracoid fracture), and 19% as a C2 (intraarticular fracture with neck). Another point of disagreement was between C2 (intraarticular fracture with neck) and C3 (intraarticular fracture with body) in certain circumstances (Fig. 2). The years of practice did not affect the degree of overall agreement. Shoulder surgeons were more likely to agree on the OTA/AO classification (Table 2). FIGURE 3. Glenoid neck fracture (fracture 30) caused a high disagreement in the New International Classification; it was classified as a body fracture in 63% and as an intraarticular fracture in 54%. Editor’s note: A color image accompanies the online version of this article. Ó 2013 Lippincott Williams & Wilkins J Orthop Trauma  Volume 28, Number 3, March 2014 Reliability of Scapula Fracture Classification TABLE 3. New International Classification for Scapular Fractures Fracture No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Raters (n) Body (B) % All Answers Fossa (F) % All Answers Process (P) % All Answers 162 153 137 131 131 129 124 119 115 114 111 110 109 108 107 106 106 105 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 Fractured Fractured Not fractured Fractured Fractured Not fractured Fractured Fractured Not fractured Not fractured Not fractured Fractured Fractured Fractured Not fractured Fractured Fractured Fractured Fractured Fractured Not fractured Fractured Not fractured Fractured Not fractured Fractured Fractured Not fractured Fractured Fractured Fractured Fractured Fractured Fractured Fractured 81 99 55 100 100 98 92 79 51 98 95 96 80 99 79 98 90 89 98 99 81 87 100 78 98 99 77 85 67 63 97 89 99 89 86 Intraarticular Extraarticular Intraarticular Extraarticular Extraarticular Extraarticular Extraarticular Intraarticular Intraarticular Extraarticular Intraarticular Extraarticular Intraarticular Extraarticular Extraarticular Extraarticular Intraarticular Extraarticular Extraarticular Extraarticular Intraarticular Intraarticular Intraarticular Intraarticular Intraarticular Extraarticular Intraarticular Intraarticular Intraarticular Intraarticular Extraarticular Intraarticular Extraarticular Intraarticular Intraarticular 97 98 68 99 98 100 98 96 97 99 77 93 93 96 99 97 97 61 100 97 90 94 100 97 99 99 99 98 99 54 97 97 99 97 82 Not involved Not involved Involved Not involved Not involved Involved Involved Not involved Involved Involved Involved Not involved Not involved Not involved Involved Not involved Involved Not involved Not involved Not involved Involved Involved Not involved Involved Not involved Not involved Not involved Involved Not involved Not involved Not involved Not involved Not involved Not involved Not involved 99 100 52 99 94 98 61 62 72 99 68 95 79 100 81 98 73 66 96 99 64 81 100 55 100 100 96 57 99 97 100 98 99 85 94 For Body (B) For Fossa (F) Agreement For Process (P) Parameter Agreement Kappa PA (%) Kappa PA (%) Kappa PA (%) Overall Years of practice 0–10 More than 10 years Specialization Orthopaedic traumatology Shoulder and elbow Hand and wrist Moderate 0.57 82 Substantial 0.79 90 Moderate 0.53 81 Moderate Moderate 0.59 0.57 83 80 Substantial Substantial 0.79 0.79 90 90 Moderate Moderate 0.55 0.52 82 80 Moderate Substantial Moderate 0.55 0.61 0.53 80 85 79 Substantial Almost perfect Almost perfect 0.78 0.83 0.80 89 92 90 Moderate Moderate Moderate 0.52 0.54 0.52 81 81 80 The agreement was moderate (k = 0.57) with an average PA of 73%, ranging from 52% to 98% with similar agreement for operative and nonoperative recommendations. More experienced doctors were less likely to recommend operative treatment. The specialization did not affect the treatment recommendation (Table 4). Ó 2013 Lippincott Williams & Wilkins Agreement DISCUSSION We found moderate overall agreement regarding classification of scapular fractures, better for articular than for body or process involvement. The average proportion of observers agreeing with the most popular treatment recommendation was 73%. www.jorthotrauma.com | 127 J Orthop Trauma  Volume 28, Number 3, March 2014 Neuhaus et al TABLE 4. Recommended Treatment Fracture No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Raters (n) Recommended Treatment % All Answers 162 153 137 131 131 129 124 119 115 114 111 110 109 108 107 106 106 105 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 Operative Nonoperative Nonoperative Nonoperative Nonoperative Nonoperative Nonoperative Operative Operative Operative Nonoperative Nonoperative Nonoperative Nonoperative Nonoperative Nonoperative Operative Nonoperative Nonoperative Nonoperative Nonoperative Operative Operative Operative Nonoperative Nonoperative Operative Operative Operative Operative Nonoperative Operative Nonoperative Operative Nonoperative 82 98 58 98 91 52 56 89 90 52 65 96 72 63 84 95 69 81 98 83 78 52 97 91 79 88 87 97 55 87 71 90 88 92 90 Parameter Overall Recommended treatment Years of practice 0–10 More than 10 years Specialization Orthopaedic traumatology Shoulder and elbow Hand and wrist Agreement Kappa PA (%) Moderate 0.45 73 Moderate Moderate 0.48 0.41 75 71 Moderate Moderate Moderate 0.44 0.45 0.48 72 73 74 Readers should consider several limitations. The data may not be valid outside the group of surgeons that participate in the Science of Variation Group, although we feel that the large number of surgeons of various specialties and countries improves external validity beyond that of the typical reliability study. Nearly 40% of the participating 128 | www.jorthotrauma.com surgeons did not answer all questions, which may have influenced our results. We did not measure the time surgeons spent looking at the movies, which could also correlate with agreement. In trying to present as many different fracture patterns as possible, we may have introduced a spectrum bias. For instance, intraarticular fractures were overrepresented.17 However, an overpresentation of intraarticular fractures can rather positively contribute to the study as these fractures have more impact on the treatment decision and outcome. Other downsides were that surgeons could not rotate the 3D models to their needs, and Osirix may have affected the image resolution and consequently the interpretation of the fracture patterns by rendering issues. However, all surgeons had the same kind of movies to interpret, which equalizes these problems. And last, there was no way to assess accuracy as there is no gold standard/reference classification. The OTA/AO classification distinguishes extra-articular (type A), partial articular (type B), and complete articular (type C) fractures. Although the agreement for type A scapula fractures was substantial, the agreement about types B and C was moderate or even fair for some subgroups. In comparison, the overall agreement for diaphyseal fractures was higher in one recent study and the level of experience and specialization did not affect their results.18 In our study, shoulder specialists had the best agreement. Perhaps an in-depth knowledge and greater familiarity with complex scapular anatomy and injuries may contribute to better understanding and classification of the fractures. Level of training is often associated with greater reliability when surgeons in training are observers,19 but level of experience did not affect agreement in our study of fully trained surgeons. The lesser experience of younger surgeons may be balanced by their greater familiarity with 3DCTs and greater reliance on the definitions.11 The New International Classification for Scapula Fractures distinguishes fractures extending into the body, fossa, or processes. This classification had almost perfect agreement if the fracture lines extend into the glenoid fossa and moderate agreement on body and process fractures. In comparison, the expert panel in the development study10 had comparable agreement on intraarticular (k = 0.78) but a clearly higher agreement on process (k = 0.61) or body fractures (k = 0.75), which may indicate that their intensive dispute and training about scapula fracture classification and their knowledge about the definitions improved their agreement. This new classification helped us to better understand the reasons for disagreement of the raters with the OTA/AO classification. Some surgeons advocate more frequent operative treatment of scapula fractures.4–9 In 40% of our selected fractures, more than 50% of the surgeons recommended operative treatment. The agreement on treatment recommendation was only moderate, remains controversial, and merits further study. In conclusion, the simpler New International Classification proved more reliable than the OTA/AO classification. Surgeons find it more difficult to distinguish body and process fractures than glenoid fractures. Improved definitions and training may further help to improve reliability of scapula fracture classification.20 Ó 2013 Lippincott Williams & Wilkins J Orthop Trauma  Volume 28, Number 3, March 2014 ACKNOWLEDGMENTS The Science of Variation Group authorship: Mahmoud I. Abdel-Ghany; Jeffrey Abrams; Joshua M. Abzug; Lars E. Adolfsson; George W. Balfour; H. Brent Bamberger DO; Antonio Barquet; Michael Baskies; W. Arnold Batson; Taizoon Baxamusa; Grant J. Bayne; Thierry Begue; Michael Behrman; Daphne Beingessner; Jan Biert; Julius Bishop; Mateus Borges Oliveira Alves; Martin Boyer; Drago Brilej; Peter R.G. Brink; Lance M. Brunton; Richard Buckley; Juan Carlos Cagnone; Ryan P. Calfee; Luiz Augusto B. Campinhos; Charles Cassidy; Louis Catalano III; Karel Chivers; Pradeep Choudhari; Matej Cimerman; Joseph M. Conflitti; Ralph M. Costanzo; Brett D. Crist; Brian J. Cross; Phani Dantuluri; Michael Darowish; Ramon de Bedout; Thomas DeCoster; David G. Dennison; Peter H. DeNoble; Gregory DeSilva; Thomas Dienstknecht; Scott F. Duncan; Xavier A. Duralde; Holger Durchholz; Kenneth Egol; Carl Ekholm; Nelson Elias; John M. Erickson; J. Daniel Espinosa Esparza; C. H. Fernandes; Thomas J. Fischer; Martin Fischmeister; Forigua Jaime E.; Charles L. Getz; Richard S. Gilbert; Vincenzo Giordano; David L. Glaser; Taco Gosens; Michael W. Grafe; Jose Eduardo Grandi Ribeiro Filho; Robert R.L. Gray; Lawrence V. Gulotta; Nigel William Gummerson; Eric Mark Hammerberg; Edward Harvey; R. Haverlag; Patrick D.G. Henry; Jonathan L. Hobby; Eric P. Hofmeister; Thomas Hughes; John Itamura; Peter Jebson; Richard Jenkinson; Kyle Jeray; Christopher M. Jones; Jedediah Jones; Axel Jubel; Scott G. Kaar; K. Kabir; F. Thomas D. Kaplan; Stephen A. Kennedy; Michael W. Kessler; Hervey L. Kimball; Peter Kloen; Cyrus Klostermann; Georges Kohut; G.A. Kraan; Anze Kristan; Mark I. Loebenberg; Kevin J. Malone; l. Marsh; Paul A. Martineau; John McAuliffe; Iain McGraw; Samir Mehta; Milind Merchant; Charles Metzger; S. A. Meylaerts; Anna N. Miller; Jennifer Moriatis Wolf; Joel Murachovsky; Anand Murthi; Michael Nancollas; Betsy M. Nolan; Timothy Omara; Reza Omid; Jose A. Ortiz; Joachim P. Overbeck; Richard S. Page; Alberto Pérez Castillo; Rodrigo Pesantez; Daniel Polatsch; G. Porcellini; Michael Prayson; M. Quell; Matthew M. Ragsdell; James G. Reid; J. M. Reuver; Marc J. Richard; Martin Richardson; Marco Rizzo; Sergio Rowinski; Jorge Rubio; Carlos G. Sánchez Guerrero; Wojciech Satora; Peter Schandelmaier; Johan H. Scheer; Andrew Schmidt; Todd A. Schubkegel; Leah M. Schulte; Evan D. Schumer; Benjamin W. Sears; Adam B. Shafritz; Nicholas L. Shortt; Todd Siff; Dario Mejia Silva; Raymond Malcolm Smith; Sander Spruijt; Jason A. Stein; Emilija Stojkovska Pemovska; Philipp N. Streubel; Carrie Swigart; Marc Swiontkowski; George Thomas; Eric T. Tolo; Matthias Turina; Minos Tyllianakis; Michel P. J. van den Bekerom; Huub van der Heide; M.A.J. van de Sande; P.V. van Eerten; Diederik O.F. Verbeek; David Victoria Hoffmann; A.J.H. Vochteloo; Robert Wagenmakers; Christopher J. Wall; Richard Wallensten; Daniel C. Wascher; Lawrence Ó 2013 Lippincott Williams & Wilkins Reliability of Scapula Fracture Classification Weiss; J. Michael Wiater; Brian P.D. Wills; Jeffrey Wint; Thomas Wright; Jason P. Young; Charalampos Zalavras; Robert D. Zura; Karol Zyto. REFERENCES 1. McKee RC, Whelan DB, Schemitsch EH, et al. Operative versus nonoperative care of displaced midshaft clavicular fractures: a meta-analysis of randomized clinical trials. J Bone Joint Surg Am. 2012;94:675–684. 2. Althausen PL, Shannon S, Lu M, et al. Clinical and financial comparison of operative and nonoperative treatment of displaced clavicle fractures. J Shoulder Elbow Surg. 2013;22:608–611. 3. Pandya NK, Namdari S, Hosalkar HS. Displaced clavicle fractures in adolescents: facts, controversies, and current trends. J Am Acad Orthop Surg. 2012;20:498–505. 4. Anavian J, Conflitti JM, Khanna G, et al. A reliable radiographic measurement technique for extra-articular scapular fractures. Clin Orthop Relat Res. 2011;469:3371–3378. 5. Anavian J, Khanna G, Plocher EK, et al. Progressive displacement of scapula fractures. J Trauma. 2010;69:156–161. 6. Anavian J, Wijdicks CA, Schroder LK, et al. Surgery for scapula process fractures: good outcome in 26 patients. Acta Orthop. 2009;80: 344–350. 7. Cole PA. Scapula fractures. Orthop Clin North Am. 2002;33:1–18, vii. 8. Cole PA, Gauger EM, Herrera DA, et al. Radiographic follow-up of 84 operatively treated scapula neck and body fractures. Injury. 2012;43: 327–333. 9. Cole PA, Talbot M, Schroder LK, et al. Extra-articular malunions of the scapula: a comparison of functional outcome before and after reconstruction. J Orthop Trauma. 2011;25:649–656. 10. Harvey E, Audige L, Herscovici D Jr, et al. Development and validation of the new international classification for scapula fractures. J Orthop Trauma. 2012;26:364–369. 11. Guitton TG, Ring D. Interobserver reliability of radial head fracture classification: two-dimensional compared with three-dimensional CT. J Bone Joint Surg Am. 2011;93:2015–2021. 12. Marsh JL, Slongo TF, Agel J, et al. Fracture and dislocation classification compendium—2007: Orthopaedic Trauma Association classification, database and outcomes committee. J Orthop Trauma. 2007;21:S1–S133. 13. Rosset A, Spadola L, Ratib O. OsiriX: an open-source software for navigating in multidimensional DICOM images. J Digit Imaging. 2004; 17:205–216. 14. Venkataraman G, Ananthanarayanan V, Paner GP. Accessible calculation of multirater kappa statistics for pathologists. Virchows Arch. 2006; 449:272. 15. King JE. Generalized kappa & other indices of interrater reliability. Available at: http://www.ccitonline.org/jking/homepage/interrater.html. Accessed November 1, 2012. 16. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–174. 17. Jaeger M, Lambert S, Sudkamp NP, et al. The AO Foundation and Orthopaedic Trauma Association (AO/OTA) scapula fracture classification system: focus on glenoid fossa involvement. J Shoulder Elbow Surg. 2013;22: 512–520. 18. Meling T, Harboe K, Enoksen CH, et al. How reliable and accurate is the AO/OTA comprehensive classification for adult long-bone fractures? J Trauma Acute Care Surg. 2012;73:224–231. 19. Petrisor BA, Bhandari M, Orr RD, et al. Improving reliability in the classification of fractures of the acetabulum. Arch Orthop Trauma Surg. 2003;123:228–233. 20. Buijze GA, Guitton TG, van Dijk CN, et al. Training improves interobserver reliability for the diagnosis of scaphoid fracture displacement. Clin Orthop Relat Res. 2012;470:2029–2034. www.jorthotrauma.com | 129