Prévia do material em texto
Orthopedic Traumatology Manish K. Sethi ● A. Alex Jahangir William T. Obremskey Editors Mohit Bhandari ● Mitchel B. Harris Michael D. McKee ● Steven A. Olson Paul Tornetta, III ● Roy W. Sanders Andrew H. Schmidt Section Editors Orthopedic Traumatology An Evidence-Based Approach Editors Manish K. Sethi Department of Orthopedic Surgery and Rehabilitation Vanderbilt University Medical Center Nashville, TN, USA William T. Obremskey Department of Orthopedic Surgery and Rehabilitation Vanderbilt University Medical Center Nashville, TN, USA A. Alex Jahangir Department of Orthopedic Surgery and Rehabilitation Vanderbilt University Medical Center Nashville, TN, USA ISBN 978-1-4614-3510-5 ISBN 978-1-4614-3511-2 (eBook) DOI 10.1007/978-1-4614-3511-2 Springer New York Heidelberg Dordrecht London Library of Congress Control Number: 2012940415 © Springer Science+Business Media New York 2013 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, speci fi cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on micro fi lms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied speci fi cally for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a speci fi c statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com) v Orthopedic surgeons who manage the injured patient have always sought the best information on which to make treatment decisions. In the past, this has generally consisted of large textbooks with a comprehensive listing of all articles on a particu- lar injury, with an author’s expert opinion and treatment recommendations synthe- sizing this information. With the evolution of higher quality clinical research methods, orthopedic trauma surgeons have been in the lead and the paradigm has shifted. Because levels of evidence are now routinely inserted into the medical literature and has scienti fi c presentations, the desire to have the strongest case for treatment decisions has increased. As the number of Level I and Level II studies has proliferated in the orthopedic trauma literature, the need for aggregation of this information into easily accessible formats has escalated. This need is speci fi cally for surgeons treating patients at the point of care. Drs. Sethi, Jahangir, and Obremskey have taken the next step in the synthesis of higher levels of evidence for the practicing orthopedic surgeon. The fi eld of ortho- pedic trauma surgery is wonderfully diverse in the types and locations of skeletal and soft tissue injury. This is one of the attractions of a career in orthopedic trauma- tology: the fact that while some injuries are common and managed frequently, oth- ers are extremely rare. The authors have taken the practical approach of taking the most commonly seen orthopedic injures, in which the most comprehensive evi- dence base exists, and aggregating the information in a useful format. They have organized the chapters around common clinical scenarios taken from real-life expe- rience. The case scenarios are then followed by a synthesis of the best clinical litera- ture focusing on Level I and Level II studies. The scienti fi c technique of meta-analysis and the structured literature review are strictly employed to provide the reader with the best recommendations for management of an individual injury. Those principles of literature synthesis using scienti fi c methodology are explained in an introductory chapter to enhance the utility of the individual chapters. The progression of clinical research and orthopedic traumatology has de fi nitely moved towards collaborative focused research using multicenter teams. Each year the number and quality of clinical research articles improves. This book is a wonder- ful fi rst step in synthesizing this information into a format useful to the individual Foreword vi Foreword surgeon working with the patient and their family in making treatment decisions. As the clinical research and datasets expand, the work will need to expand. I expect that this work will be frequently revised on a regular basis to help all of us in the orthopedic community to deliver the best care for our patients. I enthusiastically recommend this book to orthopedic surgeons everywhere. Marc Swiontkowski vii Part I Evidence-Based Medicine in Orthopedic Trauma Surgery Mohit Bhandari 1 Introduction to Evidence-Based Medicine ............................................ 3 Clary Foote and Mohit Bhandari Part II Spine Trauma Mitchel B. Harris 2 Cervical Spine Clearance ....................................................................... 23 Andrew K. Simpson and Mitchel B. Harris 3 Cervical Spine Fracture Dislocation ..................................................... 41 Kevin R. O’Neill, Jesse E. Bible, and Clinton James Devin 4 Lumbar Burst Fractures ........................................................................ 55 Robert Greenleaf and Mitchel B. Harris Part III Upper Extremity Trauma Michael D. McKee 5 Scapula Fractures ................................................................................... 71 Peter A. Cole and Brian W. Hill 6 Clavicle Fractures ................................................................................... 87 Christopher R. Geddes and Michael D. McKee 7 Proximal Humerus Fracture .................................................................. 103 Daniel J. Stinner, Philipp N. Streubel, and William T. Obremskey 8 Humeral Shaft Fractures........................................................................ 129 Bill Ristevski and Jeremy Hall Contents viii Contents 9 Distal Humerus Fractures ...................................................................... 141 Andrew Jawa and David Ring 10 Distal Radius Fractures .......................................................................... 151 Cameron T. Atkinson, Philipp N. Streubel, and Jeffry Watson Part IV Acetabular, Hip and Pelvic Trauma Steven A. Olson 11 Acetabular Fractures in the Elderly ...................................................... 169 John C. Weinlein, Edward A. Perez, Matthew I. Rudloff, and James L. Guyton 12 Pelvic Ring Injury I................................................................................. 185 Damien G. Billow and Steven A. Olson 13Pelvic Ring Injury II ............................................................................... 195 Matthew D. Karam and David C. Templemen 14 Femoral Neck Fractures in the Elderly ................................................. 207 Dave Polga and Robert T. Trousdale 15 Intertrochanteric Femur Fractures ....................................................... 219 Hassan R. Mir and George J. Haidukewych Part V Lower Extremity Trauma Paul Tornetta, III 16 Diaphyseal Femur Fractures .................................................................. 235 Manish K. Sethi, Kyle Judd, A. Alex Jahangir, and William T. Obremskey 17 Distal Femur Fractures........................................................................... 247 A. Alex Jahangir and William M. Ricci 18 Knee Dislocations .................................................................................... 261 Samuel N. Crosby Jr., Manish K. Sethi, and William T. Obremskey 19 Tibial Plateau Fractures ......................................................................... 277 Jodi Siegel and Paul Tornetta III 20 Closed Diaphyseal Tibia Fractures........................................................ 291 Marlis T. Sabo and David W. Sanders 21 Open Diaphyseal Tibia Fractures .......................................................... 303 Scott P. Ryan, Christina L. Boulton, and Robert V. O’Toole ixContents Part VI Foot and Ankle Trauma Roy W. Sanders 22 Pilon Fractures ........................................................................................ 323 David P. Barei 23 Ankle Fractures ....................................................................................... 345 Conor P. Kleweno and Edward K. Rodriguez 24 Calcaneus Fractures ............................................................................... 359 Theo Tosounidis and Richard Buckley 25 Talus Fractures ........................................................................................ 373 Hassan R. Mir and Roy W. Sanders Part VII Polytrauma, Infection, and Perioperative Management of the Orthopedic Trauma Patient Andrew H. Schmidt 26 Damage Control ...................................................................................... 389 Laurence B. Kempton and Michael J. Bosse 27 DVT Prophylaxis in Orthopedic Trauma ............................................. 405 Keith D. Baldwin, Surena Namdari, and Samir Mehta 28 The Infected Tibial Nail .......................................................................... 417 Megan A. Brady and Brendan M. Patterson 29 Perioperative Optimization in Orthopedic Trauma ............................ 431 Clifford Bowens Jr. and Jesse M. Ehrenfeld Index ................................................................................................................. 445 xi Cameron T. Atkinson , M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Keith D. Baldwin , M.D., M.S.P.T., M.P.H. Orthopedic Surgery , Hospital of the University of Pennsylvania , Philadelphia , PA , USA David P. Barei , M.D., F.R.C.S.(C) Department of Orthopedics , Harborview Medical Center, University of Washington , Seattle , WA , USA Mohit Bhandari , M.D., Ph.D., F.R.C.S.C. Division of Orthopedic Surgery , McMaster University , Hamilton , ON , Canada Jesse E. Bible , M.D., M.H.S. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Damien G. Billow , M.D. Department of Orthopedic Surgery and Rehabilitation, Vanderbilt University Medical Center , Nashville , TN , USA Michael J. Bosse , M.D. Orthopedic Surgery , Carolinas Medical Center, Charlotte , NC , USA Christina L. Boulton , M.D. Department of Orthopedic Traumatology , R Adams Cowley Shock Trauma Center , Baltimore , MD , USA Clifford Bowens Jr., M.D. Orthopedic Anesthesia, Department of Anesthesiology , Vanderbilt University School of Medicine , Nashville , TN , USA Megan A. Brady , M.D. Department of Orthopedic Surgery , MetroHealth , Cleveland , OH , USA Richard Buckley , M.D., F.R.C.S. Department of Surgery , Foothills Hospital, University of Calgary , Calgary , AB , Canada Peter A. Cole , M.D. Department of Orthopedic Surgery , Regions Hospital, University of Minnesota , St. Paul , MN , USA Contributors xii Contributors Samuel N. Crosby Jr. M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Clinton James Devin , M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Jesse M. Ehrenfeld , M.D., M.P.H. Biomedical Informatics, Department of Anesthesiology, Vanderbilt University School of Medicine , Nashville , TN , USA Clary Foote , M.D. McMaster University , Hamilton , ON , Canada Christopher R. Geddes , M.D., M.Sc. Division of Orthopedic Surgery , Department of Surgery, University of Toronto, St. Michael’s Hospital , Toronto , ON , Canada Robert Greenleaf , M.D. Reconstructive Orthopedics , Lumberton , NJ , USA James L. Guyton , M.D. Department of Orthopedics Surgery , Regional Medical Center-Memphis, Methodist Germantown Hospital, University of Tennessee/ Campbell Clinic , Germantown , TN , USA George J. Haidukewych , M.D. Orlando Health Level One Orthopedics , Orlando , FL , USA Jeremy Hall , M.D. Department of Surgery/Orthopedic Surgery , St. Michael’s Hospital , Toronto , ON , Canada Mitchel B. Harris , M.D., F.A.C.S. Orthopedic Trauma, Department of Orthopedic Surgery , Harvard Medical School, Brigham and Women’s Hospital , Boston , MA , USA Brian W. Hill , M.D. Department of Orthopedic Surgery , Regions Hospital, University of Minnesota , St. Paul , MN , USA A. Alex Jahangir , M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Andrew Jawa , M.D. Department of Orthopedic Surgery , Boston University Medical Center , Boston , MA , USA Kyle Judd , M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Matthew D. Karam , M.D. Department of Orthopedics and Rehabilitation , University of Iowa Hospitals , Iowa City , IA , USA Laurence B. Kempton , M.D. Orthopedic Surgery , Carolinas Medical Center , Charlotte , NC , USA Conor P. Kleweno , M.D. Department of Orthopedic Surgery , Beth Israel Deaconess Medical Center, Brigham and Women’s Hospital, Massachusetts General Hospital , Boston , MA , USA xiiiContributors Michael D. McKee , M.D., F.R.C.S(c) Division of Orthopedic Surgery, Department of Surgery , University of Toronto, St. Michael’s Hospital , Toronto , ON , Canada Samir Mehta , M.D. Orthopedic Trauma and Fracture Service, Department of Orthopedic Surgery , Hospital of the University of Pennsylvania , Philadelphia , PA , USA Hassan R. Mir , M.D. Department of OrthopedicSurgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Surena Namdari , M.D., M.Sc Department of Orthopedic Surgery , University of Pennsylvania , Philadelphia , PA , USA Kevin R. O’Neill , M.D., M.S. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center East , Nashville , TN , USA Robert V. O’Toole , M.D. Department of Orthopedic Traumatology , R Adams Cowley Shock Trauma Center , Baltimore , MD , USA William T. Obremskey , M.D., M.P.H. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Steven A. Olson , M.D. Department of Orthopedic Surgery , Orthopedic Trauma Duke University Medical Center, Duke University Hospital , Durham , NC , USA Brendan M. Patterson , M.D. Department of Orthopedic Surgery , MetroHealth , Cleveland , OH , USA Edward A. Perez , M.D. Department of Orthopedic Surgery , University of Tennessee/Campbell Clinic , Memphis , TN , USA Dave Polga , M.D. Department of Orthopedic Surgery , Marsh fi eld Clinic , Marsh fi eld , WI , USA William M. Ricci , M.D. Department of Orthopedic Surgery , Barnes-Jewish Hospital , St. Louis , MO , USA David Ring , M.D., Ph.D. Department of Orthopedic Surgery , Harvard Medical School, Massachusetts General Hospital , Boston , MA , USA Bill Ristevski , M.D., M.Sc. Division of Orthopedic Surgery, Department of Surgery, Hamilton General Hospital , Hamilton , ON , Canada Edward K. Rodriguez , M.D., Ph.D. Department of Orthopedic Surgery , Beth Israel-Deaconess Medical Center, Harvard Medical School , Boston , MA , USA Matthew I. Rudloff , M.D. Department of Orthopedic Surgery , University of Tennessee/Campbell Clinic , Memphis , TN , USA Scott P. Ryan , M.D. Department of Orthopedic Surgery , Tufts University Medical Center , Boston , MA , USA xiv Contributors Marlis T. Sabo , M.D., M.S.C., F.R.C.S.C. Department of Surgery , Victoria Hospital , London , ON , Canada David W. Sanders , M.D., M.S.C., F.R.C.S.C. Department of Surgery , Victoria Hospital , London , ON , Canada Roy W. Sanders , M.D. Department of Orthopedics , Florida Orthopedic Institute, Tampa General Hospital , Tampa , FL , USA Andrew H. Schmidt , M.D. Department of Orthopedic Surgery , Hennepin County Medical Center, University of Minnesota , Minneapolis , MN , USA Manish K. Sethi , M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA Jodi Siegel , M.D. Department of Orthopedic Surgery , UMass Memorial Medical Center, University of Massachusetts Medical School , Worcester , MA , USA Andrew K. Simpson , M.D., M.H.S. Harvard Combined Orthopedic Surgery , Massachusetts General Hospital , Boston , MA , USA Daniel J. Stinner , M.D. Department of Orthopedics and Rehabilitation , San Antonio Military Medical Center , Fort Sam Houston , TX , USA Philipp N. Streubel , M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA David C. Templemen , M.D. Department of Orthopedic Surgery , Hennepin County Medical Center , Minneapolis , MN , USA Paul Tornetta III, M.D. Department of Orthopedic Surgery , Orthopedic Trauma, Boston Medical Center , Boston , MA , USA Theo Tosounidis , M.D. Department of Surgery , Foothills Hospital, University of Calgary , Calgary , AB , Canada Robert T. Trousdale , M.D. Department of Orthopedic Surgery , Mayo Medical Center, Mayo Medical School , Rochester , MN , USA Jeffry Watson , M.D. Department of Orthopedic Surgery and Rehabilitation , Vanderbilt University Medical Center , Nashville , TN , USA John C. Weinlein , M.D. Department of Orthopedic Surgery , University of Tennessee/Campbell Clinic , Memphis , TN , USA Part I Evidence-Based Medicine in Orthopedic Trauma Surgery Section editor—Mohit Bhandari 3M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_1, © Springer Science+Business Media New York 2013 Keywords Evidence-based medicine Evidence-based orthopedics Introduction Hierarchy of evidence Hierarchy of research studies Parallel trial design Factorial design Introduction The science of addressing orthopedic problems that confront orthopedic surgeons everyday requires a rigorous methodology to guide investigation and provide valid answers. The term “evidence-based medicine”, fi rst coined by Dr. Gordon Guyatt at McMaster University has become the standard for clinical investigation and critical appraisal. It has been de fi ned as the conscientious and judicious use of current best available evidence as the basis for surgical decisions [ 1– 3 ] . Application of the evidence does not occur in isolation but rather with integration of surgical expertise and clinical circumstances, as well as with societal and patient values [ 4, 5 ] (Fig. 1.1 ). In addition, identifying and applying best available evidence requires a comprehen- sive search of the literature, a critical appraisal of the validity and quality of avail- able studies, astute consideration of the clinical situation and factors that may in fl uence applicability, and a balanced application of valid results to the clinical problem [ 6 ] . C. Foote , M.D. (*) McMaster University , 2-18 Hill Street , Hamilton , ON , Canada L8P 1W7 e-mail: clary.foote@medportal.ca M. Bhandari , M.D., Ph.D., F.R.C.S.C. Division of Orthopedic Surgery , McMaster University , Hamilton , ON , Canada Chapter 1 Introduction to Evidence-Based Medicine Clary Foote and Mohit Bhandari 4 C. Foote and M. Bhandari I n 2000, Marc Swiontkowski introduced the evidence-based orthopedics (EBO) section of the Journal of Bone Surgery (JBJS) with a focus on higher levels of evi- dence such as randomized control trials (RCTs) which recognized the de fi ciency of controlled studies in the orthopedic literature [ 7 ] . In 2003, the JBJS adopted EBM and the hierarchy of evidence for grading all clinical papers. Also during that year, Dr. Bhandari initiated the evidence-based orthopedic trauma section in the Journal of Orthopedic Trauma (JOT) [ 8 ] . Since then, the EBO initiative has grown into a global initiative and has become the common language at international orthopedic meetings. The American Orthopedic Society has recognized and incorporated EBO for utilization into clinical guidelines [ 9 ] . Paramount to the understanding of “best available evidence” are the concepts of hierarchy of evidence, meta-analyses, study design, and precision of results. A familiarity with these concepts will aid the orthopedic surgeon in identifying, understanding, and incorporating best evidence into their practice. We begin here with an overview of the hierarchy of surgical evidence with attention paid to study designand methodological quality. Some of the common instruments to measure study quality are described, and we direct our readership to adjunctive educational resources. Finally, we conclude by clarifying misconceptions of EBO to reinforce its underpinning principles that help the reader interpret the surgical evidence presented in this text. Hierarchy of Research Studies To understand the concept of best evidence a surgeon must fi rst be knowledgeable about the hierarchy of surgical evidence. The hierarchy can be thought of as a classi fi cation system to provide a common language for communication and a basis Fig. 1.1 The triumvirate of evidence-based orthopedics (EBO) to improve best practice in orthopedics. Reproduced with permission from Tilburt JC, et al. J Eval Clin Pract. 2008;14:721–5 51 Introduction to Evidence-Based Medicine for review of available evidence. Research studies range from very high quality to low quality which are largely based on the study design and methodological quality [ 10 ] . In general, high-quality studies minimize bias and thus increase our con fi dence in the validity of results. Bias can be de fi ned as systematic error in a research study that impacts outcome such that it differs from the truth [ 11 ] . There are several avail- able systems to formulate the level of evidence of a given study. The Oxford Centre for Evidence-Based Medicine has published hierarchies for therapeutic, prognostic, harm, prevalence, and economic analyses [ 12 ] . For each of aforementioned subcat- egories, there is a hierarchy of evidence with unique clinical signi fi cance [ 13, 14 ] . JBJS has incorporated the Oxford System in order to develop a hierarchy for orthopedic studies (Table 1.1 ). For the purposes of this text, when we refer to the “hierarchy” or “level of evidence,” we will be referring to this table. In orthopedic traumatology, therapeutic studies are of central importance. For instance, they may tell us the healing and complication rates of reamed versus unreamed technique for intramedullary nailing of tibial shaft fractures [ 15 ] . When evaluating a study of a surgical or therapeutic intervention one must identify the study design as an initial step to identify best evidence [ 16 ] . The highest level of evidence lies in RCTs and systematic reviews or meta-analyses of high-quality RCTs [ 17, 18 ] . These are referred to as level I trials [ 2 ] . The process of randomiza- tion is the best research tool to minimize bias by distributing known and unknown prognostic variables uniformly between treatment groups [ 19, 20 ] . Available evi- dence suggests that non-randomized studies tend to overestimate [ 21 ] or underesti- mate [ 22 ] treatment effects. Reviews of RCTs use rigorous methodology to improve sample size and precision of study results and are therefore considered the highest level of evidence when reviewed studies are of suf fi cient methodological quality (Table 1.1 ) [ 23 ] . Reviews may statistically combine results (meta-analyses) when trial reporting allows or provide a qualitative overview of the results of included studies (systematic reviews) [ 24 ] . Unrandomized prospective studies such as cohort studies (also known as prospective comparative studies) provide weaker empirical evidence, as they are prone to several biases [ 22 ] . For instance, treatment allocation is uncontrolled and therefore treatment cohorts may differ in prognosis from the outset due to selection bias (Table 1.2 ) [ 25 ] . Retrospective case–control studies assess past characteristics and exposures in cases as compared with controls. These studies are subject to several types of bias including selection and recall bias (Table 1.2 ). Matching treatment and control groups for known prognostic variables (e.g., age, gender, functional level) may partially control for confounding variables but rarely suf fi ciently negates them. One can also “overmatch” groups such that the groups are so closely matched that the exposure rates between cohorts are analo- gous [ 26 ] . In addition, the retrospective structure can lead to imprecise data collec- tion and differential patient follow-up [ 27 ] . At the bottom of the evidence hierarchy are case reports, series, and expert opinion. Case series are uncontrolled, unsystem- atic studies with a role mainly in hypothesis generation for future investigation and provide very little utility in guiding care. These reports are usually single-surgeon and single-center experiences which further impairs generalizability. Ta bl e 1. 1 Jo ur na l o f B on e an d Jo in t S ur ge ry Am hi er ar ch y of o rth op ed ic ev id en ce [ 6 6 ] Ty pe s o f s tu di es Th er ap eu tic st ud ie s— in v es tig at in g th e re su lts o f t re at m en t Pr og no sti c stu di es — in v es tig at in g th e ef fe ct o f a p at ie nt c ha ra ct er ist ic o n th e ou tc om e of d ise as e D ia gn os tic st ud ie s— in v es tig at in g a di ag no sti c te st Ec on om ic a nd d ec isi on a na ly se s— de v el op in g an e co no m ic o r d ec isi on m o de l Le v el I H ig h- qu al ity a ra n do m iz ed c on tro lle d tr ia l w ith st at ist ic al ly si gn i fi ca n t d iff er en ce o r n o s ta tis tic al ly si gn i fi ca n t d iff er en ce b u t n ar ro w c o n fi d en ce in te rv al s Sy ste m at ic re v ie w b o f L ev el I ra nd om iz ed co n tr ol le d tri al s ( an d s tud y r esu lts w ere ho m og en eo us c ) H ig h- qu al ity p ro sp ec tiv e st ud y d (al l p ati en ts we re en rol led at th e sa m e po in t i n th ei r d ise as e w ith ³ 80 % fo llo w -u p of e nr ol le d pa tie nt s) Sy ste m at ic re v ie w b o f L ev el I stu di es Te st in g of p re v io us ly d ev el op ed di ag no sti c cr ite ria in se rie s o f co n se cu tiv e pa tie nt s ( wi th u n iv er sa lly a pp lie d re fe re nc e “g ol d” st an da rd ) Sy ste m at ic re v ie w b o f L ev el I stu di es Se ns ib le c os ts an d al te rn at iv es ; v al ue s o bt ai ne d fro m m an y stu di es ; m u lti w ay se ns iti v ity a na ly se s Sy ste m at ic re v ie w b o f L ev el I stu di es Le v el II Le ss er - qu al ity ra nd om iz ed c on tro lle d tri al (e. g., <8 0% fo llo w -u p, n o bl in di ng , o r im pr op er ra nd om iz at io n) Pr os pe ct iv e d co m pa ra tiv e st ud y e Sy ste m at ic re v ie w b o f L ev elII st ud ie s o r Le v el I stu di es w ith in co ns ist en t r es ul ts R et ro sp ec tiv e f st ud y U nt re at ed c on tro ls fro m a ra nd om iz ed co n tr ol le d tri al Le ss er - qu al ity p ro sp ec tiv e st ud y (e. g., pa tie nt s e nr ol le d at d iff er en t p oi nt s in th ei r d ise as e or < 80 % fo llo w -u p) Sy ste m at ic re v ie w b o f L ev el II st ud ie s D ev el op m en t o f d ia gn os tic c rit er ia o n th e ba sis o f c on se cu tiv e pa tie nt s (w ith un ive rs al ly a pp lie d re fe re nc e “ go ld ” sta nd ar d) Sy ste m at ic re v ie w b o f L ev el II st ud ie s Se ns ib le c os ts an d al te rn at iv es ; v al ue s o bt ai ne d fro m li m ite d stu di es ; m u lti w ay se ns iti v ity a na ly se s Sy ste m at ic re v ie w b o f L ev el II st ud ie s Le v el II I Ca se –c on tro l s tu dy g R et ro sp ec tiv e f co m pa ra tiv e st ud y e Sy ste m at ic re v ie w b o f L ev el II I s tu di es Ca se –c on tro l s tu dy g St ud y of n on co ns ec ut iv e pa tie nt s (w ith ou t c on sis ten tly ap pli ed re fe re nc e “g ol d” st an da rd ) Sy ste m at ic re v ie w b o f L ev el II I s tu di es A na ly se s b as ed o n lim ite d al te rn at iv es an d co sts ; p oo r e sti m at es Sy ste m at ic re v ie w b o f L ev el II I st ud ie s Le v el IV Ca se se rie s h Ca se se rie s Ca se –c on tro l s tu dy Po or re fe re nc e sta nd ar d N o se ns iti v ity a na ly se s Le v el V Ex pe rt o pi ni on Ex pe rt o pi ni on Ex pe rt o pi ni on Ex pe rt o pi ni on Th is ch ar t w as a da pt ed fr om m at er ia l p ub lis he d by th e Ce nt re fo r E vi de nc e- Ba se d M ed ic in e, O xf or d, U K . F o r m o re in fo rm at io n, p le as e se e w w w . ce bm .n et a A c o m pl et e as se ss m en t o f t he q ua lit y of in di v id ua l s tu di es re qu ire s c rit ic al a pp ra isa l o f a ll as pe ct s o f t he st ud y de sig n b A c om bi na tio n of re su lts fr om tw o o r m o re p rio r s tu di es c S tu di es p ro v id ed c on sis te nt re su lts d S tu dy w as s ta rt ed b ef or e th e fi r st p at ie nt e nr ol le d e P at ie nt s t re at ed o ne w ay (e .g. , w ith ce me nte d h ip art hro pla sty ) c om pa red w ith pa tie nts tr ea ted an oth er wa y (e. g., w ith ce me ntl es s hi p ar th ro pl as ty ) a t th e s am e i ns titu tio n f S tu dy w as s ta rt ed a fte r t he fi rs t p at ie nt e nr ol le d g P at ie nt s i de nt i fi ed fo r t he st ud y on th e b as is of th ei r o ut co m e ( e.g ., f ai le d to ta l h ip ar th ro pl as ty ), c all ed “c ase s,” a re c o m pa re d w ith th os e w ho d id n ot h av e th e o ut co m e ( e.g ., h ad a su cc ess ful to ta l h ip a rth ro pl as ty ), c all ed “c on tro ls” h P at ie nt s t re at ed o ne w ay w ith n o co m pa ris on g ro up o f p at ie nt s t re at ed a no th er w ay 71 Introduction to Evidence-Based Medicine Study Quality and the Hierarchy of Evidence When placing a study into the surgical hierarchy one must also consider study qual- ity. In general, studies drop one level if they contain methodological problems (Table 1.1 ) [ 12, 28 ] . RCTs are only considered level I evidence when they have proper institution of safe-guards against bias (Table 1.3 ), high precision (narrow con fi dence intervals), and high levels of patient follow-up; lesser quality RCTs are assigned to level II evidence. Several instruments have been validated to assess the quality of RCTs which include the Jadad (range 0–5), Delphi list (range 0–9), and numeric rating scale (NRS; range 1–10). For example, the Jadad Scale is the sim- plest and most widely utilized instrument to assess methodological quality of clini- cal trials (Table 1.4 ) [ 29 ] . In the orthopedic literature it has been used to evaluate the quality of research in a particular fi eld [ 30– 32 ] , set a minimum standard for included papers in a systematic review or meta-analysis [ 33 ] , or for critical appraisal of an individual paper. The Jadad Scale contains three main areas of assessment: random- ization, blinding, and loss to follow-up (Table 1.4 ). In addition, quality scoring sys- tems exist for observational studies (i.e., cohort and case–control) such as the Newcastle–Ottawa Scale for Cohort Studies [ 34 ] . For cohort studies, this tool assesses the rigor of cohort selection and comparability, ascertainment of exposure, outcome assessment (e.g., blinded assessment), and follow-up. From this, we have summarized crucial methodological elements of quality studies in Table 1.3 . Although the actual validated instruments need not be used rigorously in everyday orthopedics, these quality criteria should be of central concern to the orthopedic surgeon in assessing the validity of results of published studies. Table 1.2 De fi nitions of bias types in therapeutic studies Types of biases De fi nition Selection bias Treatment groups differ in measured and unmeasured characteristics and therefore have differential prognosis due to systematic error in creating intervention groups [ 35 ]. Recall bias Patients who experience an adverse outcome are more likely to recall exposure than patients who do not sustain an adverse outcome [ 27, 70 ]. Detection bias Biased assessment of outcome. May be in fl uenced by such things as prior knowledge of treatment allocation or lack of independent af fi liation within a trial [ 25 ]. Performance bias Systematic differences in the care provided to cohorts are independent of the intervention being evaluated [ 25, 71 ]. Attrition bias Occurs when those that drop out of a study are systematically different from those that remain.Thus, fi nal cohorts may not be representative of original group assignments [ 2, 67 ]. Expertise bias Occurs when a surgeon involved in a trial has differential expertise (and/or convictions) with regard to procedures in a trial where trial outcomes may be impacted by surgeon competency and/or beliefs rather than interventional ef fi cacy [ 72 ]. 8 C. Foote and M. Bhandari Table 1.3 Some essential methodological components of high-quality studies Item Study design Description A priori de fi ned study protocol RCT and observational A protocol is critical to establish a priori primary and secondary outcomes which will require speci fi c considerations, resources, and sample size. A priori outcomes maximize the bene fi ts of cohort assignment (e.g., randomization) and limit overanalyzing trial data that leads to a higher rate of identifying signi fi cant differences by chance alone. Prospective RCT and observational Studies started before the fi rst patient enrolled to improve cohort assignments, blinding, precision of data collection, completeness of follow-up, and study directness. Power analysis RCT or observational Determination of the appropriate sample size to detect a pre-speci fi ed difference of clinical signi fi cance between cohorts. Based on standard deviation measurements from previous reputable studies. Ensures that a study has suf fi cient power to detect a clinically signi fi cant difference. Exclusion and inclusion criteria RCT and observational De fi ning the study population of interest and limiting patient factors which may confound outcomes greatly improves the generalizability of study results. Clinically relevant and validated outcome measures RCT and observational The ef fi cacy of an intervention should be based on outcomes that are important to patients using instruments validated in capturing this clinical information. Blinding RCT and observational Surgeon blinding may not be possible, but blinding patients, outcome assessors, data analysts, authors of the results section, and outcomes’ adjudicators are imperative to protect against detection and performance biases. Randomization RCT Safe-guard against selection bias by ensuring equal distribution of prognostic characteristics between cohorts. Concealment RCT Investigators must be blinded to treatment allocation of patients to protect against undue in fl uence on treatment allocation (i.e., selection bias). Complete follow-up RCT and observational Complete follow-up of all patients should always be sought [ 67 ] . Appreciable risk of attrition bias exists when follow-up is less than 80% [ 68 ]. Expert-based design RCT A surgeon with expertise in one of the procedures being evaluated in a trial is paired with a surgeon with expertise in the other procedure. Subjects are then randomized to a surgeon, who performs only one of the interventions (i.e., the procedure that he/she has expertise and/or a belief that it is the superior procedure) [ 69 ] A safe-guard for expertise bias. 91 Introduction to Evidence-Based Medicine Recently, the Consolidated Standards of Reporting Trials (CONSORT) Group published updated guidelines on how to report RCTs [ 35 ] . Previous systematic review of the surgical literature has reported poor compliance of surgical RCTs with its recommendations and endorsed educational initiatives to improve RCT report- ing [ 36 ] . Although a thorough review of this document is beyond the scope of this chapter, it suf fi ces to say that it serves as an excellent overview to aid in planning, executing, and reporting RCTs. Randomized Surgical Trials: An Overview of Speci fi c Methodologies RCTs are considered the optimal study design to assess the ef fi cacy of surgical inter- ventions [ 28 ] . RCTs in the orthopedic literature have been described as explanatory (also called mechanistic) or pragmatic [ 37 ] . The explanatory trial is a rigorous study design that involves patients who are most likely to bene fi t from the intervention and asks the question of whether the intervention works in this patient population who receive treatment. Pragmatic trials include a more heterogeneous population, usually involve a less rigorous protocol and question whether the intervention works to whom it was offered [ 38 ] . The explanatory trial measures the ef fi cacy of the intervention under ideal conditions, whereas the pragmatic trial measures the effectiveness of the intervention in circumstances resembling daily surgical practice. For that reason prag- matic trials have been said to be more generalizable but this comes at the cost of reduced study power due to patient heterogeneity which results in a larger range of treatment effects (increased noise). Explanatory and pragmatic approaches should be thought of as a continuum, and any particular trial may have aspects of each. The opti- mal trial design depends on the research question, the complexity of the intervention, and the anticipated bene fi t of the new intervention to the patient. Randomized trials are best suited to assess interventions with small-to-medium treatment effects. The smaller the anticipated effect, the more an investigator should consider optimizing the partici- pant pool and intervention to provide clean results (explanatory trial) [ 38, 39 ] . Orthopedic surgery trials pose many methodological challenges to researchers. These include dif fi culties with recruitment of an adequate number of patients, blinding, Table 1.4 Jadad Scale for assessment of methodological quality of a clinical trial [ 29 ] Primary questions: 1. Was the study described as randomized? 2. Was the study described as double blind? 3. Was there a description of withdrawals and dropouts? Two addition points can be given if the following criteria are met : 4. The method of randomization was described in the paper, and that method was appropriate 5. The method of blinding was described, and it was appropriate One point is deducted for each of the following criteria: The method of randomization was described, but was inappropriate The method of blinding was described, but was inappropriate Jadad Score 0 (poor quality) to 5 (high quality) 10 C. Foote and M. Bhandari differential cointervention, and outcome assessment. These dif fi culties are re fl ected in the quality of the current orthopedic literature. A previous review of orthopedic RCTs showed that a high percentage failed to report concealment of allocation, blinding, and reasons for excluding patients [ 40– 42 ] . The results of these RCTs may be misleading to readers and there is a growing consensus that larger trials are required [ 43 ] . A recent RCT has shown that many of these problems can be circumvented with multicenter surgical RCTs that include strict guidelines for cointervention and contain a blinded adjudication committee to determine outcomes [ 44 ] . The orthopedic community generally agrees that RCTs are the future of orthope- dic research, but there have been many arguments against them. These include ethi- cal assertions about patient harm which include: (1) surgeons performing different operations at random where they may be forced to perform a procedure at which they are less skilled and comfortable performing; (2) conducting RCTs which involve withholding care such as in a placebo-controlled trial; and (3) inability to blind sur- geons and the dif fi culty in blinding patients unless a sham RCT is conducted [ 25 ] . Although sham RCTs that facilitate patient blinding have been published, many eth- ics committees continueto deny its use on the basis of potential harm to patients who receive sham treatment [ 45, 46 ] . To help answer the question of harm in sham RCTs, several authors are currently conducting a systematic review looking at outcomes in sham trials (unpublished study). On another note, new innovative designs have emerged to address some of the ethical problems with surgical RCTs. The Expertise-Based Design In surgical trials the ethical dilemma can present if the surgeon believes one inter- vention is superior or has more expertise with one procedure, but is forced to per- form the other procedure due to random patient allocation. In such a circumstance, it is unethical for the surgeon to be involved in the trial. To address this problem, Dr. Devereaux has published extensively on the expertise-based design where the patient is randomized to one of the two groups of surgeons and not to the procedure itself. This is in contrast to the parallel RCT where surgeons perform both proce- dures in random order. This avoids the aforementioned ethical dilemma and also minimizes performance bias where the results of the trial may be heavily impacted by surgeon experience or comfort. The downside of expertise-based design is that in some research areas, such as trauma surgery, both surgeon groups need to be avail- able at all times to perform their designated intervention. This may limit feasibility in small centers with scarce resources. Parallel Trial Design The most commonly utilized and simplest design is the parallel randomized trial. Participants are assigned to one of two or more treatment groups in a random order. The most basic of these involves two treatments groups – a treatment and control arm. 111 Introduction to Evidence-Based Medicine Trials can have more than two arms to facilitate multiple comparisons, but this requires larger sample sizes and increases the complexity of analysis. Factorial Design The factorial trial enables two or more interventions to be evaluated both individually and in combination with one another. This trial design is thought to be economical in some settings because more than one hypothesis (and treatment) can be tested within a single study. For example, Petrisor et al. [ 47, 48 ] conducted a multicenter, blinded randomized 2 × 3 factorial trial looking at the effect of irrigation solution (castile soap or normal saline) and pressure (high versus low versus very low pressure lavage) on outcomes in open fracture wounds. The corresponding 2 × 3 table is shown in Table 1.5 . From this table the investigator wound compare the 1,140 patients receiving soap with the 1,140 who received saline solution. Concurrently, comparison can be made between each of the pressure categories with 760 participants. With factorial designs there may be interaction between the interventions. That is, when treatments share a similar mechanism of action, the effect of one treatment may be in fl uenced by the presence of the other. If the treatments are commonly coadministered in surgical practice (such as the aforementioned lavage study), then this trial design is ideal, as it allows for assessment of the interaction to identify the optimal treatment combination. Treatment interactions may be negative (antagonistic) or positive (synergistic), which reduce or increase the study power, respectively. This consequently affects sample size, and therefore potential interactions should be considered in the design phase of the study. Other Randomized Designs In surgical trials the unit of randomization is often the patient or the limb of interest [ 15, 47 ] . In other words, when we randomize to one treatment versus another, we are usually talking about randomizing patients. In some circumstances, however, randomizing patients may not be feasible or warranted. When the intervention is at an institutional or department level, such as with implementation of a new Table 1.5 A 2 × 3 factorial trial table from the fl uid lavage in open fracture wounds (FLOW) randomized trial Gravity fl ow pressure Low pressure High pressure Total Soap solution 380 380 380 1,140 Saline 380 380 380 1,140 Total 760 760 760 2,280 This study had a target sample size of 2,280 participants and was designed to assess the impact of irrigation solution (soap or saline = 2 categories) and lavage pressure (gravity fl ow, low, and high pressure = 3 categories) in open fracture wounds [ 73 ] 12 C. Foote and M. Bhandari process, guideline, or screening program, patient randomization is dif fi cult and often impossible. This is for several reasons: (1) surgeons or health care practitio- ners are unlikely to use a new guideline for one patient and not the other; (2) patients randomized to different interventions will often educate each other (a process called contamination); and (3) department wide programs are often expensive and challenging to implement, so running multiple programs is not practical or eco- nomical. In these circumstances, it is best to randomize institutions, departments, or geographical areas. This process is called cluster randomization. For instance, if one were to implement a chewing tobacco cessation program among major league base- ball players, it would make more sense to randomize teams to the cessation program rather than individual players. Two important aspects of cluster trials are: (1) par- ticipants within clusters are more similar with regard to prognostic factors than between clusters and (2) a suf fi cient number of clusters must be available to provide prognostic balance and suf fi cient power. In general, because patients within clusters are similar, there is a reduction power and an increased required sample size of cluster trials. In the analysis, one can compare the outcomes of entire clusters or individuals. Individual patient analysis requires an estimate of patient similarity (called an intraclass correlation coef fi cient). The more similar the participants are within clusters, the higher the intraclass correlation coef fi cient, and the required sample size is consequently greater to reach signi fi cance. In crossover trials patients are randomized to a treatment and then receive the other treatment after a designated period of time. Each participant serves as their own control when a within patient analysis is conducted. These studies have signi fi cant power but are rarely conducted in orthopedic surgery because they require chronic diseases with treatments that are quickly reversible once stopped. For example, Pagani et al. [ 49 ] conducted a crossover trial assessing the gait correc- tion of 4-valgus and neutral knee bracing in patients with knee OA. All patients performed gait and stair climbing assessments without an orthosis and then were randomized to one of the two bracing arms for 2 weeks followed by crossover to the other bracing arm for 2 weeks. Because of the power of this analysis, they demon- strated a statistically signi fi cant improvement in gait mechanics with 4-valgus bracing with only 11 patients. Special Considerations Within the Hierarchy In addition to reviews of level II studies [ 50 ] , reviews of high-quality RCTs with inconsistent results [ 51 ] are also regarded as level II evidence (Table 1.1 ). For instance, Hopley et al. performed a meta-analysis comparing total hip arthroplasty (THA) to hemiarthroplasty (unipolar and bipolar) which included seven RCTs, three quasi-randomized, and eight retrospective cohort studies. This review reported reduced reoperation rates and better functional improvements after THA than hemi- arthroplasty. However, from review of this study’s forest plot of randomized studies, 131 Introduction to Evidence-BasedMedicine one can see that there is a wide range in point estimates leading to imprecision within their pooled effect size (Fig. 1.2 ). This analysis encountered methodological issues such as lack of concealment, heterogeneity of study inclusion criteria, and type of hemiarthroplasty; all of these factors would negatively affect this meta- analysis’s rating within the hierarchy. In addition, the included review of retrospec- tive cohort studies would be regarded as level III evidence (Fig. 1.2 ; Table 1.1 ). Fig. 1.2 Sample forest plot that shows the point estimates and 95% con fi dence intervals of individual primary studies and pooled effect sizes represented as a relative risk ( diamond ). This meta-analysis provided separate pooled effect sizes for each type of study design and an overall pooled estimate shown at the bottom . Estimates to the left favor total hip arthroplasty and to the right hemiarthro- plasty. Reproduced with permission from Hopley C, Stengel D, Ekkernkamp A, et al. Primary total hip arthroplasty versus hemiarthroplasty for displaced intracapsular hip fractures in older patients: systematic review. BMJ. 2010;340:c2332 14 C. Foote and M. Bhandari Grades of Recommendation: From the Bench to the Operating Room The quality of best available evidence and the magnitude of treatment effect reported play a central role in the strength of clinical practice recommendations. A recom- mendation for or against an intervention is based on a comprehensive systematic review of available evidence, evaluation of the methodological quality of available studies, and focus group discussion of subspecialty experts to achieve consensus. In 2004, the Grading of Recommendation Assessment, Development, and Evaluation (GRADE) Working Group developed a system for scoring the quality of evidence (Table 1.6 ) [ 52 ] . This scoring system places more weight on studies with better design, higher methodological quality, and larger treatment effects, but also consid- ers factors such as directness [ 53 ] . The GRADE Criteria are applied to all critical outcomes. Once the evidence is “graded” and several factors such as calculation of baseline risk in the target population, feasibility of the proposed intervention, and a bene fi t versus harm assessment are completed, a recommendation level is assigned which includes one of the following (1) do it; (2) probably do it; (3) toss up; (4) probably do not do it; and (5) do not do it [ 38, 39 ] . These recommendations guide surgeons by suggesting that most (items 1 and 4) or many (items 2 and 4) well- informed surgeons would make a particular decision, based on systematic review of the literature. The GRADE approach provides a basic foundation for translating evidence into practice and serves as a useful communication tool for clinicians and Table 1.6 Modi fi ed GRADE quality assessment criteria [ 53 ] Quality of evidence Study design Lower if a Higher if a High Randomized trial Study quality : −1 Serious limitations −2 Very serious limitations −1 Important inconsistency Directness : −1 Some uncertainty −2 Major uncertainty −1 Sparse data −1 High probability of Reporting bias Strong association : +1 Strong, no plausible confounders, consistent and direct evidence b +2 Very strong, no major threats to validity and direct evidence c +1 Evidence of a dose– response gradient +1 All plausible confound- ers would have reduced the effect Moderate Quasirandomized trial Low Observational study Very low Any other evidence a 1 = move up or down one grade (e.g., from high to moderate). 2 = move up or down two grades (e.g., from high to low). The highest possible score is high (4) and the lowest possible score is very low (1). Thus, for example, randomized trials with a strong association would not move up a grade b A relative risk of >2 (<0.5), based on consistent evidence from two or more observational studies, with no plausible confounders c Available studies provide direct comparisons between alternative treatments in similar participant populations 151 Introduction to Evidence-Based Medicine review panels. However, even valued input and consensus from expert panels does not replace a sound understanding of the available evidence (e.g. from a critical appraisal of a meta-analyses) and good clinical judgment. Hence, we return to the essence of EBO which considers best available evidence, clinical judgment, patient values, and clinical circumstances when making treatment decisions (Fig. 1.1 ). Evidence-Based Orthopedics: Advances and Misconceptions EBM has been recognized as one of the top 15 medical discoveries of the last 160 years. In the past decade it revolutionized clinical research and care by provid- ing the basis for the development of clinical trials, systematic review, and validated outcomes. International standards have been developed such as the Oxford Centre for Evidence-Based Medicine, the Cochrane Collaboration, and Britain’s Center for Review, which are providing updated systematic reviews of the effects of medical and surgical care [ 54 ] . In orthopedics, JBJS has fully incorporated the hierarchy of evidence into all published manuscripts, and this has been utilized in Annual Meetings of the American Academy of Orthopedic Surgeons (AAOS) [ 55 ] . As a consequence, the overall quality of clinical trials and systematic reviews in orthope- dics appears to be improving [ 23, 56 ] . Improving the validity of orthopedic studies is only one facet of EBO in its pur- suit to improving standards in orthopedic practice. EBO also requires a willingness of an orthopedic society, for example, the AAOS in this case, to incorporate best evidence into practice [ 57 ] . Traditionally, there has been a resistance to perform well-designed studies in orthopedics and misconceptions about the practice of EBO [ 58, 59 ] . In contrast, a recent international cross-sectional survey among International Hip Fracture Research Collaborative (IHFRC) Surgeons revealed that most sur- geons are willing to change their practice based on large-scale clinical trial results [ 60 ] . Thus, it appears that orthopedists are recognizing the need for higher standards to ensure best care for patients with musculoskeletal conditions. Despite the global movement of EBO, misconceptions about it exist. There have been criticisms that EBO only gives information about the average patient and that simple application of trial results is analogous to “cook-book” medicine [ 16, 61 ] . The approach of EBO is actually exactly the opposite. EBO utilizes a bottom-up approach which begins with a surgical problem and incorporates best available evi- dence, surgical expertise and experience, the clinical context, and patient prefer- ences. Surgical expertise and a working understanding of EBO are essential to appreciate if the available evidence applies well to the individual patient and clinical circumstances, and if so, how it should be applied. For example, if one were to encounter the 65-year-old marathon runner with a displaced femoral neck fracture after a fall, one must consider the available evidence of improved outcomes of THA as compared to hemiarthroplasty and internal fi xation, the current limitations of this literature, the patient’s functional status and physiologic age, and patient prefer- ences and expectations with regard to the complication pro fi le and functional out- comes of these procedures [ 51, 62, 63 ] . 16 C. Foote and M. Bhandari Some have equated EBO with RCTs and meta-analysis. On the contrary, EBO proposes to use the most appropriate studydesign and methodology to answer the surgical question with maximal validity. RCTs are more effective when the condi- tion is common rather than when it is rare. For instance, many conditions in ortho- pedic oncology are too scarce to permit an RCT but EBO advocates that studies in this fi eld institute as many safe-guards as possible to limit bias, to focus on out- comes that are important to patients, and to perform systematic review when pos- sible [ 64 ] . In addition, evaluation of diagnostic ef fi cacy is best answered by cross-sectional studies rather than RCTs. Questions regarding biomechanics and prosthetic wear-properties are often best addressed by studies in basic science. Despite this, randomized trials have claimed much of the focus of EBO because of their important role in providing valid outcomes for surgical interventions (Table 1.1 ). Thus, it is important to keep in mind that many factors determine the ideal study design that best answers the clinical problem. Such considerations include the type of question being asked (e.g., therapeutic ef fi cacy, diagnosis), fre- quency of the condition, ethics of intervention, the quality and uncertainties of available evidence, and surgical equipoise. Closing Comments Ultimately, becoming an evidence-based orthopedic surgeon is not a simple task. One must understand the hierarchy of evidence, from meta-analysis of RCTs to clinical experience. In making surgical decisions a surgeon should know the strength of best available evidence and the corresponding degree of uncertainty. The process of exploration of evidence to answer speci fi c questions is equally critical. The abil- ity to search the available literature, evaluate the methodological quality of studies to identify best evidence, determine the applicability of this information to the patient, and appropriately store this information for further reference requires edu- cation and practice. For educational modules on these topics we direct you toward several additional resources to this text including Clinical Research for Surgeons [ 25 ] , the Users’ Guide of the Medical Literature [ 2 ] , the JBJS Users’ Guides to the Surgical Literature [ 65 ] , and the Journal of Orthopedic Trauma Evidence-based Orthopedic Trauma Summaries [ 8 ] . References 1. Hoppe DJ, Bhandari M. Evidence-based orthopaedics: a brief history. Indian J Orthop. 2008;42:104–10. 2. G Guyatt DR (editor). Users’ guide to the medial literature: a manual for evidence-based clinical practice. Chicago, IL: American Medical Association Press; 2001. 3. Guyatt GH, Sackett DL, Cook DJ. Users’ guides to the medical literature. II. How to use an article about therapy or prevention. B. What were the results and will they help me in caring for my patients? Evidence-Based Medicine Working Group. JAMA. 1994;271:59–63. 171 Introduction to Evidence-Based Medicine 4. Sackett DL. Evidence-based medicine. Semin Perinatol. 1997;21:3–5. 5. Tilburt JC. Evidence-based medicine beyond the bedside: keeping an eye on context. J Eval Clin Pract. 2008;14:721–5. 6. Guyatt GH, Sackett DL, Sinclair JC, et al. Users’ guides to the medical literature. IX. A method for grading health care recommendations. Evidence-Based Medicine Working Group. JAMA. 1995;274:1800–4. 7. Swiontkowski MF Wright JG. Introducing a new journal section: evidence-based orthopae- dics. J Bone Joint Surg. 200;82:759. 8. Bhandari M, Sanders RW. Where’s the evidence? Evidence-based orthopaedic trauma: a new section in the Journal. J Orthop Trauma. 2003;17:87. 9. Weber KL. The AAOS clinical practice guidelines. J Am Acad Orthop Surg. 2009;17:335–6. 10. Phillips BBC, Sackett DL, Badenoch D, Straus S, Haynes B, et al. Levels of evidence and grades of recommendation. Oxford: Oxford-Centre For Evidence Based Medicine: GENERIC; 1998 11. Schunemann HJ, Bone L. Evidence-based orthopaedics: a primer. Clin Orthop Relat Res. 2003;213:117–32. 12. Group OLoEW. The Oxford 2011 levels of evidence. Oxford: Oxford Centre for Evidence- Based Medicine; 2011. 13. Sackett DL, Rosenberg WM, Haynes RB. Evidence based medicine: how to practice and teach EBM. New York: Churchill Livingstone; 1997. 14. Sackett DL. Evidence-based medicine and treatment choices. Lancet. 1997;349:570; author reply 572–3. 15. Bhandari M, Guyatt G, Tornetta III P, et al. Study to prospectively evaluate reamed intramedu- ally nails in patients with tibial fractures (S.P.R.I.N.T.): study rationale and design. BMC Musculoskelet Disord. 2008;9:91. 16. Sackett DL, Rosenberg WM, Gray JA, et al. Evidence based medicine: what it is and what it isn’t. BMJ. 1996;312:71–2. 17. Schulz KF, Grimes DA. Unequal group sizes in randomised trials: guarding against guessing. Lancet. 2002;359:966–70. 18. Biedermann R, Martin A, Handle G, et al. Extracorporeal shock waves in the treatment of nonunions. J Trauma. 2003;54:936–42. 19. Thoma A, Farrokhyar F, Bhandari M, et al. Users’ guide to the surgical literature. How to assess a randomized controlled trial in surgery. Can J Surg. 2004;47:200–8. 20. Urschel JD, Goldsmith CH, Tandan VR, et al. Users’ guide to evidence-based surgery: how to use an article evaluating surgical interventions. Evidence-Based Surgery Working Group. Can J Surg. 2001;44:95–100. 21. Miller JN, Colditz GA, Mosteller F. How study design affects outcomes in comparisons of therapy. II surgical. Stat Med. 1989;8:455–66. 22. Bhandari M, Tornetta III P, Ellis T, et al. Hierarchy of evidence: differences in results between non-randomized studies and randomized trials in patients with femoral neck fractures. Arch Orthop Trauma Surg. 2004;124:10–6. 23. Bhandari M, Morrow F, Kulkarni AV, et al. Meta-analyses in orthopaedic surgery. A system- atic review of their methodologies. J Bone Joint Surg Am. 2001;83-A:15–24. 24. Kuzyk PR, Saccone M, Sprague S, et al. Cross-linked versus conventional polyethylene for total hip replacement: A meta-analysis of randomised controlled trials. J Bone Joint Surg Br. 2011;93:593–600. 25. Bhandari M, Joensson A. Clinical research for surgeons. New York: Thieme; 2009. 26. Bowling A. Research methods in health: investigating health and health services research. Philadelphia: Open University Press; 2000. 27. Hartz A, Marsh JL. Methodologic issues in observational studies. Clin Orthop Relat Res. 2003;213:33–42. 28. Brighton B, Bhandari M, Tornetta P III, et al. Hierarchy of evidence: from case reports to randomized controlled trials. Clin Orthop Relat Res. 2003;213:19–24. 29. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17:1–12. 18 C. Foote and M. Bhandari 30. Singh JA, Murphy S, Bhandari M. Assessment of the methodologic quality of medical and surgical clinical trials in patients with arthroplasty. J Rheumatol. 2009;36:2642–54. 31. Boutron I, Tubach F, Giraudeau B, et al. Methodological differences in clinical trials evaluat- ing nonpharmacological and pharmacological treatments of hip and knee osteoarthritis. JAMA. 2003;290:1062–70. 32. Gummesson C, Atroshi I, Ekdahl C. The quality of reporting and outcome measures in ran- domized clinical trials related to upper-extremity disorders. J Hand Surg Am. 2004;29:727–34; discussion 727–35. 33. Jacobs WC, Clement DJ, Wymenga AB. Retention versus sacri fi ce of the posterior cruciate ligament in total knee replacement for treatment of osteoarthritis and rheumatoid arthritis. Cochrane Database Syst Rev. 2005:CD004803.34. Wells GA Shea B, O’Connell D, et al. The Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa: Ottawa Health Research Institute; 2011. 35. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for report- ing parallel group randomized trials. Ann Intern Med. 2010;152:726–32. 36. Agha R, Cooper D, Muir G. The reporting quality of randomised controlled trials in surgery: a systematic review. Int J Surg. 2007;5:413–22. 37. Bhandari M. Evidence-based orthopedics. London: BMJ Books; 2011. 38. Jadad AR, McQuay HJ. A high-yield strategy to identify randomized controlled trials for sys- tematic reviews. Online J Curr Clin Trials. 1993;Doc No 33:[3973 words; 3939 paragraphs]. 39. Moher D, Pham B, Jones A, et al. Does quality of reports of randomised trials affect estimates of intervention ef fi cacy reported in meta-analyses? Lancet. 1998;352:609–13. 40. Bhandari M, Richards RR, Sprague S, et al. Quality in the reporting of randomized trials in surgery: is the Jadad Scale reliable? Control Clin Trials. 2001;22:687–8. 41. Chan S, Bhandari M. The quality of reporting of orthopaedic randomized trials with use of a checklist for nonpharmacological therapies. J Bone Joint Surg Am. 2007;89:1970–8. 42. Bhandari M, Richards RR, Sprague S, et al. The quality of reporting of randomized trials in the Journal of Bone and Joint Surgery from 1988 through 2000. J Bone Joint Surg Am. 2002;84- A:388–96. 43. Foote CJ, Sprague S, Schemitsch EH, et al. Future perspectives: the need for large clinical tri- als. J Orthop Trauma. 2011;25 Suppl 1:S95–8. 44. Bhandari M, Guyatt G, Tornetta III P, et al. Randomized trial of reamed and unreamed intramedullary nailing of tibial shaft fractures. J Bone Joint Surg Am. 2008;90:2567–78. 45. Kallmes D, Buchbinder R, Jarvik J, et al. Response to “randomized vertebroplasty trials: bad news or sham news?”. AJNR Am J Neuroradiol. 2009;30:1809–10. 46. Noonan P. Randomized vertebroplasty trials: bad news or sham news? AJNR Am J Neuroradiol. 2009;30:1808–9. 47. Petrisor B, Sun X, Bhandari M, et al. Fluid lavage of open wounds (FLOW): a multicenter, blinded, factorial pilot trial comparing alternative irrigating solutions and pressures in patients with open fractures. J Trauma. 2011;71:596–606. 48. Petrisor B, Jeray K, Schemitsch E, et al. Fluid lavage in patients with open fracture wounds (FLOW): an international survey of 984 surgeons. BMC Musculoskelet Disord. 2008;9:7. 49. Pagani CH, Bohle C, Potthast W, et al. Short-term effects of a dedicated knee orthosis on knee adduction moment, pain, and function in patients with osteoarthritis. Arch Phys Med Rehabil. 2010;91:1936–41. 50. Petrisor B, Lisson S, Sprague S. Extracorporeal shockwave therapy: a systematic review of its use in fracture management. Indian J Orthop. 2009;43:161–7. 51. Hopley C, Stengel D, Ekkernkamp A, et al. Primary total hip arthroplasty versus hemiarthro- plasty for displaced intracapsular hip fractures in older patients: systematic review. BMJ. 2010;340:c2332. 52. Atkins D, Eccles M, Flottorp S, et al. Systems for grading the quality of evidence and the strength of recommendations I: critical appraisal of existing approaches The GRADE Working Group. BMC Health Serv Res. 2004;4:38. 191 Introduction to Evidence-Based Medicine 53. Atkins D, Briss PA, Eccles M, et al. Systems for grading the quality of evidence and the strength of recommendations II: pilot study of a new system. BMC Health Serv Res. 2005;5:25. 54. Handoll HH, Sherrington C, Mak JC. Interventions for improving mobility after hip fracture surgery in adults. Cochrane Database Syst Rev. 2011;3:CD001704. 55. Viveiros H, Mignott T, Bhandari M. Evidence-based orthopaedics: is it possible? J Long Term Eff Med Implants. 2007;17:87–93. 56. Dijkman BG, Abouali JA, Kooistra BW, et al. Twenty years of meta-analyses in orthopaedic surgery: has quality kept up with quantity? J Bone Joint Surg Am. 2010;92:48–57. 57. Hurwitz S. Evidence-based medicine in orthopaedic surgery – a way to the future. Iowa Orthop J. 2003;23:61–5. 58. Colton C. Statistical correctness. J Orthop Trauma. 2000;14:527–8. 59. Rudicel S, Esdaile J. The randomized clinical trial in orthopaedics: obligation or option? J Bone Joint Surg Am. 1985;67-A:1284–93. 60. Dijkman BG, Kooistra BW, Pemberton J, et al. Can orthopedic trials change practice? Acta Orthop. 2010;81:122–5. 61. Davidoff F, Haynes B, Sackett D, et al. Evidence based medicine. BMJ. 1995;310:1085–6. 62. Bhandari M, Sprague S, Schemitsch EH. Resolving controversies in hip fracture care: the need for large collaborative trials in hip fractures. J Orthop Trauma. 2009;23:479–84. 63. Bhandari M, Devereaux PJ, Swiontkowski MF, et al. Internal fi xation compared with arthro- plasty for displaced fractures of the femoral neck. A meta-analysis. J Bone Joint Surg Am. 2003;85-A:1673–81. 64. Hickey M, Farrokhyar F, Deheshi B, et al. A systematic review and meta-analysis of intrale- sional versus wide resection for intramedullary grade I chondrosarcoma of the extremities. Ann Surg Oncol. 2011;18:1705–9. 65. Bhandari M, Devereaux PJ, Montori V, et al. Users’ guide to the surgical literature: how to use a systematic literature review and meta-analysis. Can J Surg. 2004;47:60–7. 66. JBJS. “Levels of Evidence for Primary Research Question.” Retrieved April 27, 2012, from http://jbjs.org/public/instructionsauthors.aspx - LevelsEvidence; 2012. 67. Sprague S, Leece P, Bhandari M, et al. Limiting loss to follow-up in a multicenter randomized trial in orthopedic surgery. Control Clin Trials. 2003;24:719–25. 68. Toerien M, Brookes ST, Metcalfe C, et al. A review of reporting of participant recruitment and retention in RCTs in six major journals. Trials. 2009;10:52. 69. Devereaux PJ, Bhandari M, Clarke M, et al. Need for expertise based randomised controlled trials. BMJ. 2005;330:88. 70. Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. Am J Ophthalmol. 2000;130:688. 71. Colditz GA, Miller JN, Mosteller F. How study design affects outcomes in comparisons of therapy I: medical. Stat Med. 1989;8:441–54. 72. Bednarska E, Bryant D, Devereaux PJ. Orthopaedic surgeons prefer to participate in expertise- based randomized trials. Clin Orthop Relat Res. 2008;466:1734–44. 73. Fluid lavage of open wounds (FLOW): design and rationale for a large, multicenter collabora- tive 2 × 3 factorial trial of irrigating pressures and solutions in patients with open fractures. BMC Musculoskelet Disord. 2010;11:85. Part II Spine Trauma Section editor—Mitchel B. Harris 23M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_2, © Springer Science+Business Media New York 2013 Keywords Cervical spine clearance • Cervical spine trauma initial management • Clinical assessment of cervical spine injury • The Advanced Trauma Life Support (ATLS) protocol • Asymptomatic • Temporarily non-assessable • Symptomatic • Obtunded GB: 25-Year-Old Male with Neck Pain Case Presentation GB is a 25-year-old male who presents after an all terrain vehicle (ATV) accident. At the scene the patient demonstrates a GCS score of 12 complaining of chest pain and is placed in a cervical collar. The patient presents to the local emergency room via EMS. On primary survey the patient demonstrates afl ail chest and is hemody- namically unstable. In the trauma bay the patient is intubated and stabilized hemo- dynamically. His CXR demonstrates multiple rib fractures and a hemothorax. A left-sided chest tube is placed. Secondary survey is negative. Past medical and surgical history are unremarkable. The patient has no allergies and negative family history. A. K. Simpson , M.D., M.H.S. (*) Harvard Combined Orthopedic Surgery , Massachusetts General Hospital , 87 W Cedar St. Apt. 1B , Boston , MA 02114 , USA e-mail: aksimpson@partners.org M. B. Harris , M.D., F.A.C.S. Orthopedic Trauma, Department of Orthopedic Surgery , Harvard Medical School, Brigham and Women’s Hospital , 75 Francis St , Boston , MA 02115 , USA e-mail: mbharris@partners.org Chapter 2 Cervical Spine Clearance Andrew K. Simpson and Mitchel B. Harris 24 A.K. Simpson and M.B. Harris On physical examination the patient is moving all four extremities spontaneously. Consultation is requested regarding clearance of the Cervical Spine. X-rays and CT scan images are demonstrated (Figs. 2.1 , 2.2 , and 2.3 ). Fig. 2.1 ( a ) AP radiograph C-spine, ( b ) Lateral radiograph C-spine Fig. 2.2 ( a ) Coronal CT spine, ( b ) Sagittal CT spine 252 Cervical Spine Clearance Interpretation of Clinical Presentation This case presents the challenges of cervical spine clearance in the multiply injured, cognitively compromised trauma patient. Cervical spine injuries occur in approxi- mately 1% of the more than 13 million blunt trauma patients who present annually to emergency rooms in the United States and Canada [ 1, 2 ] . The timely (early) detection of these injuries is important as undetected injuries have a higher inci- dence of subacute neurologic compromise [ 3 ] . Additionally, prolonged spinal immobilization can cause several complications including respiratory deterioration, dysphagia, and venous thromboembolism [ 4 ] . Therefore, the primary goal of cervi- cal spine evaluation and clearance in trauma patients is to accurately and rapidly con fi rm the absence of a clinically signi fi cant spinal injury, while at the same time utilizing evaluation methods sensitive enough to accurately identify those patients that speci fi cally require immobilization or operative stabilization. Initial Management and ATLS The objective in the initial management of trauma patients is to identify limb and life-threatening conditions/injuries and initiate interventions while maintaining strict spinal precautions. The Advanced Trauma Life Support (ATLS) protocol, developed by the American College of Surgeons, presents a reproducible algorithm for rapidly identifying and initiating interventions for limb and life-threatening injuries [ 5 ] . ATLS protocol dictates that all patients should be presumed to have a cervical spine injury until proven otherwise. “Spine precautions” should be initiated Fig. 2.3 ( a ) Extension radiograph C-spine, ( b ) Flexion radiograph C-spine 26 A.K. Simpson and M.B. Harris at the scene of the injury by emergency medical services personnel and include application of a hard cervical collar, maintenance of the patient in a supine position, and log roll precautions. Spine precautions should then be maintained until a sec- ondary evaluation is performed by a quali fi ed examiner, which can be an emergency room physician, trauma team surgeon, or spine specialist [ 5 ] . Clinical Assessment and Examination Clinical assessment and examination of the trauma patient is perhaps the most criti- cal aspect of cervical spine clearance. Spinal evaluation often begins concurrently with other resuscitative measures in the emergency department. The assessment generally begins with receiving the following: “scene of the accident” reports from EMS personnel regarding injury mechanism, initial condition and complaints of the patient, GCS at the scene, as well as vitals and associated injuries. The medical evaluation begins with the observation of spontaneous extremity movement; however, complete spine evaluation and neurological examination should quickly follow required resuscitation. The fi rst priority is assessment of a patient’s mental status and his/her ability to provide a history and examination, as further management is predicated upon not only the examination itself, but also the reliability of the examination. The ATLS algorithm provides guidelines for cervical spine clearance based on the patient’s symptomatology, ability to provide an objective and reliable examination, and hemodynamic stability. Based on these criteria, patients are categorized in one of four groups: (1) asymptomatic, (2) temporarily non-assessable, (3) symptomatic, and (4) obtunded. Patients who are awake, alert, sober, and without distracting inju- ries can provide evaluating clinicians with the most useful information. The ideal assessment begins with a thorough patient history addressing pre-injury function, preexisting spinal conditions, speci fi c recollection of the trauma, and presence of pain or tenderness throughout the head, neck, or thoracolumbar spine. Speci fi c ques- tioning regarding motor and sensory function, even transient weakness, numbness, or paresthesias is important as it may be indicative of an unstable spine injury. Examination of the spine begins with inspection and palpation of the neck and back. Palpation should proceed in the midline from the occiput to the sacrum. Findings which may indicate injury include focal tenderness, crepitus, and step-off or gapping of the spinous processes. Palpable manifestations of spinal injury are often subtle, and while the absence of any tenderness is a helpful fi nding, the pres- ence of tenderness, however, is non-speci fi c. A comprehensive neurologic assessment is required for all patients with sus- pected spinal injury. This includes sensory and motor testing of all four limbs [ 6 ] . An accurate and reliable neurological examination is often challenging in the trauma patient with extremity injuries. It is vital that the sensory examination include rectal examination and examination of perineal sensation. Sacral sensory or 272 Cervical Spine Clearance anal sphincter function, in addition to distal lower extremity motor function, is important in predicting recovery of neurologic status in the “incomplete spinal cord injury” [ 7, 8 ] . In the asymptomatic, awake, and alert patient, with a normal neurological exami- nation, range of motion (ROM) can be assessed with the Canadian cervical spine functional test, which has been shown to have very high sensitivity for unstable cervical spine injuries [ 2, 9– 11 ] . In this test, rotation and subsequently fl exion and extension are tested. First, the patient is asked to rotate his or her head 45° to the left and right. If no symptoms are elicited, the patient is then asked to fl ex and extend their head. Patients who are asymptomatic, awake, and alert, and have a normal neurologic and functional ROM testing can be clinically cleared without radio- graphic evaluation [ 12– 16 ] . However, in patients with midline cervical tenderness or painful motion, as well as in those patients who have distracting injuries or an altered mental status, a screening cervical spine commuted tomography (CT) scan is indicated [ 17 ] . In the initial management of patient GB, he was placed in a cervical collar by Emergency Medical Services. In the trauma bay, his fl ail chest and hemodynamic instability necessitated intubation. Based on his injuries and altered mental status, patient GB would be classi fi edin the obtunded or otherwise unable to clinically eval- uate category. The prevalence of clinically occult cervical spine injuries in obtunded patients is 8%, which is three times more likely than in alert trauma patients [ 18 ] . The imaging presented in this case is a cervical CT and dynamic radiographs. The cervical spine CT demonstrates overall normal alignment without listhesis and no signi fi cant boney injury. A closer look at the CT between C6 and C7 demon- strates a subtle crack in the most anterior superior aspect of C7 and a disruption in the ossi fi ed annulus at the posteroinferior aspect of C6, which may represent a liga- mentous injury. Perhaps the greatest debate in the evaluation of the cervical spine in obtunded patients is whether a normal CT scan is suf fi cient for cervical spine clearance, or whether magnetic resonance imaging (MRI) or another imaging study is also warranted. Many studies have documented that somewhere between 10 and 30% of patients with a “normal” cervical CT scan will have abnormalities demonstrated on MRI [ 19– 21 ] . The clinical signi fi cance of these abnormalities has not been entirely borne out. Speci fi cally, MRI of the acutely post-traumatic spine often demonstrates some abnormalities, though the majority of these abnormalities do not require inter- vention or immobilization. As a result, some protocols allow for “C-spine clear- ance” in obtunded patients with negative CT scans while others require an additional study such as an MRI, a normal and reliable physical examination, or adequate dynamic studies prior to cervical spine clearance. The latter often occurs in the subacute stages, in outpatient clinic follow-up. Patient GB underwent dynamic radiographs. Dynamic lateral cervical spine radiographs may demonstrate cervical instability, which would manifest as listhesis or relative anteroposterior translation of adjacent vertebral bodies. The fl exion- extension fi lms presented here do not demonstrate any listhesis or other signs of instability. 28 A.K. Simpson and M.B. Harris Declaration of Speci fi c Diagnosis GB is a 25-year-old multiply injured patient who was placed in a fi eld collar and subsequently intubated secondary to a fl ail chest and hemodynamic instability. GB requires cervical spine clearance and is currently pharmacologically obtunded. Brainstorming: What Are the Treatment Goals and the Surgical Options? Treatment goals consist of the following objectives: 1. Immobilization of the spine with a cervical collar and log roll precautions initiated at the scene. 2. Presumption of a cervical spine injury until proven otherwise. 3. Accurately and rapidly con fi rm the absence of a clinically signi fi cant cervical spine injury where one does not truly exist. 4. Early detection and treatment of the acute cervical spine injury, especially those associated with an incomplete neurologic de fi cit or those felt to be clinically unstable. Diagnostic options include: 1. Clearance without imaging in selected group of asymptomatic patients 2. Plain radiographs 3. Dynamic radiographs 4. Upright radiographs 5. Computed tomography (CT) 6. Magnetic resonance imaging (MRI) Evaluation of the Literature In order to identify relevant publications on cervical spine clearance, a PubMed search was performed with the following keywords: “cervical spine,” “trauma,” and “clearance.” This search identi fi ed 159 articles which were reviewed. The search was limited to articles published from 1975 to the present. Detailed Review of the Pertinent Articles Cervical spine clearance is required in all trauma patients with a reasonable mecha- nism of injury, and as such is a common issue facing emergency medicine and trauma surgeons. The following discussion addresses the epidemiology of cervical 292 Cervical Spine Clearance injuries, ATLS guidelines, classi fi cation of patients, and corresponding diagnostic and management protocols. Epidemiology of Cervical Spine Injuries The reported incidence of cervical spine injuries in blunt trauma patients who pres- ent to emergency rooms ranges from 1 to 6% in various studies, with even higher numbers reported in the multiply injured patient [ 2, 22, 23 ] . The cervical spine is the most commonly injured region of the spinal column, accounting for more than half of all spine injuries. There are several risk factors for cervical spine injury in trauma patients. Cervical spine injuries occur more commonly in patients greater than 50 years of age, after higher energy mechanisms, and in those patients with associ- ated head injuries [ 24 ] . Cervical spinal cord injuries are relatively uncommon, occurring in less than 50 cases per million persons per year [ 23 ] . Spinal cord injury predominately occurs in younger male patients with an average age of 25 years and a 4:1 male to female gender ratio [ 25, 26 ] . Classi fi cation of Patient Types and Corresponding Management Anderson et al. recently reviewed cervical spine clearance in blunt trauma patients and classi fi ed patients based on their symptoms and ability to provide a reliable evaluation [ 27 ] . Patients are acutely categorized into one of four groups: (1) asymptomatic, (2) temporarily nonassessable, (3) symptomatic, and (4) obtunded. Each of these patient groups has its own algorithm for cervical spine clearance. Asymptomatic Large prospective clinical studies and meta-analyses have demonstrated that clear- ance of the cervical spine in asymptomatic patients can be accomplished based on clinical evaluation alone and does not require imaging studies [ 13, 27 ] . Stiell et al. prospectively studied a cohort of almost 9,000 trauma patients, asking physicians to document 20 clinical fi ndings prior to radiographic evaluation of the C-spine to determine clinical and exam fi ndings predictive of cervical spine injury or lack of injury. They derived the resultant model, known as the Canadian C-spine rule, which was found to have 100% sensitivity for clinically signi fi cant cervical spine injuries. Anderson et al. performed a meta-analysis of prospective studies evaluat- ing the clinical clearance of the cervical spine in asymptomatic patients and found an overall sensitivity of 98.1% for the detection of cervical injury based on clinical exam, including a functional ROM test. 30 A.K. Simpson and M.B. Harris There are two main protocols for cervical clearance in the asymptomatic patient that have been and continue to be widely utilized: (1) National Emergency X-Radiography Utilization Study (NEXUS) Low-Risk Criteria (NLC) and (2) Canadian Cervical-Spine Rule (CCR). The NEXUS method requires that all of the following criteria are met for a patient to be cleared without imaging: awake patient with normal level of alertness; no history, signs, or laboratory evidence of distracting injury; no evidence of intoxication; no focal neurologic de fi cit; no cervi- cal spine pain or midline tenderness. The NEXUS criteria were established in 1992, and demonstrated 99.6% sensitivity and 12.9% speci fi city for cervical spine injury in a large US validation study [ 12, 28– 30 ] . The central drawback of the NEXUS system is its low speci fi city, meaning that although it is a good screening test in terms of detecting injury, there are a large number of patients receiving diagnostic imaging that do not have any injury. Given that such a large number of blunt trauma patients present to emergency departments around the world, this equates to a large number of unnecessary radiographic studies and exaggerated health care costs. The Canadian C-Spine Rule (CCR) was developed froma multicenter study fol- lowing almost 9,000 blunt trauma patients [ 31– 33 ] . As mentioned above, the CCR was established by having physicians prospectively evaluate trauma patients for 20 different criteria, looking at the correlation between these criteria and fi ndings on cervical imaging studies to determine which clinical and exam fi ndings were predic- tive of radiographically proven cervical injury. The CCR is based on three high risk criteria, fi ve low-risk factors, and an active ROM test. The fi rst criteria evaluate if the patient is high enough risk that imaging should be obtained. These criteria include: age greater than 65, dangerous injury mechanism, and extremity paresthe- sias. Patients who satisfy these criteria require imaging. The second criteria are low- risk factors that, if present, suggest that the patient is low enough risk to proceed with active ROM testing. Low-risk factors include low speed, rear-end motor vehicle collision, patient comfortably sitting up in the emergency room, history of ambulat- ing after the injury, delayed onset of neck pain, and absence of midline cervical- spine tenderness. In patients with no high-risk factors and the presence of at least one of the low-risk criteria, active ROM testing is performed by asking the patient to turn their head 45° to the left and right. Patients who can perform adequate active ROM without pain can have their cervical spines cleared clinically without imaging. Stiell et al. reported the sensitivity and speci fi city of CCR testing as 100 and 42%, respectively, demonstrating a signi fi cantly higher speci fi city than the NEXUS pro- tocol [ 13 ] . Steill et al. published another prospective cohort study in the New England Journal of Medicine in 2003 comparing the NEXUS protocol to the Canadian C-Spine Rule in over 8,000 patients. 394 physicians performed CCR and NEXUS criteria and the sensitivity and speci fi city of these tests were compared for detection of cervical spine injury. Results con fi rmed that the CCR was both more sensitive (99.4 vs. 90.7%) and more speci fi c (45.1 vs. 36.8%) than the NEXUS protocol [ 11 ] . Como et al. and Hadley et al. have both performed systemic reviews of cervical spine clearance, with review of over 50 articles from MEDLINE published between 312 Cervical Spine Clearance 1998 and 2007, and found that there was Level I evidence supporting removal of cervical collar without imaging studies in awake, alert, asymptomatic patients who have full painless active ROM [ 16, 34 ] . Temporarily Nonassessable Temporarily nonassessable patients have a transient inability to provide a reliable examination. These patients are expected to resume their baseline cognitive func- tion and be evaluable within 24–48 hours. Harris et al. reported on this group of patients and found that most common factors in the category of temporarily non- assessable are drug/alcohol intoxication, concussion, or pain from distracting injuries [ 35 ] . Given that patients’ cognitive recovery is unpredictable, this group of patients is inherently poorly-de fi ned and treatment recommendations therefore remain vague. Once the patient has regained adequate cognitive function and is deemed clinically asymptomatic after a comprehensive examination, he or she can be treated as such and cleared clinically without imaging. If a patient is presumed to regain his or her normal baseline cognitive function within the 24–48 h time period but cervical clearance becomes more urgent, such as when surgical intervention is required for other injuries, the individual can be evaluated as an obtunded patient with advanced imaging. The most common guideline for this patient popu- lation is to have them undergo a multi-detector CT scan and/or maintain their cervical collar. Symptomatic Symptomatic patients are those with neck pain, tenderness, or neurologic symp- toms. This subgroup requires spinal imaging. The various imaging options include static and dynamic plain radiographs, CT, and MRI. Each of these various imaging methodologies has speci fi c advantages and disadvantages, not only in regard to sensitivity and speci fi city but also in terms of timing and cost. Plain radiography is one of the earliest imaging modalities and is currently read- ily available, fast, and low cost in comparison to other types of studies. In patients with suspected spine trauma, plain radiographic evaluation has been shown to have a wide range of reported sensitivity for cervical spine injuries ranging from 31.6 to 52% depending on the study [ 36– 40 ] . Gale et al. adopted a protocol where 848 patients underwent AP and lateral plain radiographs, as well as head CT visualizing down to C2. The sensitivity for plain radiographic detection of upper cervical spine injuries demonstrated on CT was 32%. In a meta-analysis of MEDLINE articles from 1995 through 2004 [ 37 ] , Holmes et al. found pooled sensitivity data from the seven included studies and determined that cervical spine radiographs detected 52% 32 A.K. Simpson and M.B. Harris of injuries compared to 98% for CT [ 36 ] . The most dif fi cult areas to assess on plain radiography are from the occiput to the axis and at the cervicothoracic junction where a signi fi cant amount of traumatic injuries occur. In the vast majority of trauma centers, plain radiographic evaluation of the trau- matic spine has been supplanted by multi-detector CT scans and their reconstruc- tions. The sensitivity for detection of cervical spine fractures utilizing modern MDCT scanners has been reported as high as 99.3 and 99.7% [ 41, 42 ] . Brown et al. retrospectively looked at 3,500 trauma patients, 236 of which sustained a spine fracture and found that spiral CT detected 99.3% of all injuries [ 41 ] ; Henessy et al. prospectively looked at 402 patients who underwent spiral CT and dynamic radiog- raphy and found that 99.7% of clinically signi fi cant cervical spine injuries were demonstrated on CT [ 42 ] . With these improved sensitivities, CT scans have been demonstrated to be cost effective in high-risk blunt trauma patients and are now considered the primary imaging study in these patients. This is even further emphasized in the polytrauma patient requiring CT evaluation of the chest, abdo- men, or pelvis [ 43, 44 ] . The one noteworthy exception to this idea would be the patient in extremis that is going directly to the operating room. In order to rule out a major injury (fracture dislocation or occipito-cervical dissociation), a single lat- eral radiograph or lateral fl uoroscopic image in the OR will provide a signi fi cant amount of information and facilitate the identi fi cation of a catastrophic cervical spine injury that may bene fi t from an early reduction maneuver. There is very little utility for dynamic studies in the acute setting as recent litera- ture has con fi rmed that fl exion-extension radiographs provide no new information in the presence of a negative CT scan. Khan et al. demonstrated in 311 blunt trauma patients with negative CT scans that fl exion-extension fi lms performed in the acute setting did not show a single case of instability and management was therefore not affected [ 45 ] . Some authors have recently investigated the utility of fl exion- extension CT scans performed at the end of a typical cervical CT series. However, to date, these dynamic CT scans have again not provided any clear bene fi t [ 46 ] . MRI provides an effective means of evaluating the disc-ligament complex, the posterior ligaments, and the neural elements of the cervical spine. MRI is an extremely sensitive study (near 100%) for detection of cervical spine injuries, but its speci fi city and positive predictive value for clinically signi fi cant and/or unstable injuries isless than perfect [ 20 ] . Vaccaro et al. studied the utility of MRI and found it very useful in patients with neurologic de fi cits, altering treatment in a quarter of these patients [ 47 ] . Patients admitted with isolated upper cervical spine fractures were prospectively enrolled and evaluated with MRI performed within 48 hours of their trauma. The authors found that MRI affected management for one in four of the patients with a neurologic de fi cit. In that same study, however, MRI did not provide any additional information to alter treatment in patients who were neurologically intact. All symptomatic trauma patients require diagnostic imaging, more speci fi cally with MDCT or MRI. CT scans have essentially supplanted plain radiographs in this setting and are the preferred primary imaging modality. MRI is particularly useful in symptomatic patients with neurologic de fi cit or suspected ligamentous injury based on other imaging. 332 Cervical Spine Clearance Obtunded Cervical spine clearance in the obtunded patient is the most controversial aspect and focuses mainly on whether an MRI is necessary in addition to a negative MDCT. Though an exceedingly rare incident, isolated neurologic injury does occur and can lead to signi fi cant morbidity [ 48 ] . The percentage of acute trauma patients who are obtunded at the time of evaluation ranges from about 20 to 30% [ 49, 50 ] . Additionally, Milby et al. found that cervical spine injury is nearly three times more common in the obtunded trauma patient than in the awake, alert patient who can provide a reli- able evaluation (7.7 vs. 2.8%) [ 18 ] . Although extended cervical immobilization is possible, it is not without complications and discomfort, including occipital and submental ulcers and exacerbation of intracranial pressure. Furthermore, cervical orthoses may not provide suf fi cient immobilization to protect the spinal cord in cases of severe soft tissue injuries of the cervical spine [ 51– 58 ] . Patients who cannot provide a reliable cervical spine evaluation require cervical spine imaging. There have been numerous large series which have shown CT to be effective as a single and isolated modality for clearance of the cervical spine in obtunded patients. Harris et al. and Hogan et al. have both published large series, which found that CT detected all clinically signi fi cant injures [ 59, 60 ] . In the Harris study, only one minor injury was missed and in the Hogan study 4 of the 366 patients with negative CT demonstrated isolated ligamentous injury on MRI, all of which were deemed stable. Tomycz et al. reviewed their series of 690 patients over 4 years and found that no patients with a negative CT had an acute unstable injury missed or developed delayed instability [ 61 ] . The ability to detect potential unsta- ble ligamentous injury by CT alone, however, does vary depending upon who is interpretting the study. Simon et al. found that when spine specialists reviewed CT scans in obtunded patients read as normal by emergency room radiologists, they deemed nearly 20% of the CT scans to be concerning for ligamentous injury. A subsequent MRI performed on those cases identi fi ed a ligamentous injury in 17/22 patients [ 62 ] . There is also a body of literature supporting the utility of MRI as an adjunct for its increased detection of ligamentous injuries. Menaker et al. performed a study looking at 734 patients, 203 of which were obtunded, where MRI and CT were in obtunded blunt trauma patients and found that 8.9% of patients with a normal CT scan had an abnormal MRI. Based on the MRI fi ndings, 7.9% had a change in their management and 1% of the patients (with a normal CT) were found to have an unstable cervical spine requiring surgical fi xation [ 63 ] . Stassen et al. utilized a pro- tocol requiring both CT and MRI in obtunded patients and described excellent results [ 21 ] . Over a 1 year period, 52 patients underwent a CT scan and MRI, and of the approximately 80% of patients who had a negative CT scan, 30% demonstrated ligamentous injury on MRI. These injuries required prolonged collar immobiliza- tion and repeat subacute evaluation. They reported no incidence of missed C-spine injuries or cervical collar related complications. Though requiring both CT and MRI for clearance is certainly a comprehensive method for spinal injury detection, 34 A.K. Simpson and M.B. Harris it remains unclear whether the bene fi ts of this protocol outweigh the medical risks of prolonged immobilization awaiting MRI or the fi nancial risks of obtaining MRI in the millions of obtunded patients seen in hospitals every year. In the case of GB, cervical spine CT was obtained and demonstrated normal alignment and no gross injury. In many institutions he would have been cleared based on that CT scan, but in this case dynamic radiographs were obtained to detect potential ligamentous instability. The utility of dynamic radiographs performed acutely in obtunded patients has been shown to be of minimal bene fi t and also carries the potential risk of injury in obtunded patients, who may not have protective re fl exes or the ability to verbalize pain or other potentially harmful symptoms during the examination [ 64 ] . Though dynamic radiography can visualize the mid-subaxial cervical spine rather well, the cervicothoracic junction, as with all plain radiography, is poorly visualized on the lateral projection, making lower cervical instability dif fi cult to detect [ 65 ] . Anglen et al. reviewed 837 patients, the majority of which were obtunded, who underwent CT and subsequent dynamic radiography as part of their clearance protocol. In the three cases where dynamic fi lms demonstrated ligamentous injury not seen on CT, no surgical intervention was required, thus concluding that dynamic fi lms were nei- ther clinically helpful nor cost effective [ 66 ] . Some authors advocate the use of dynamic fl uoroscopy performed by the physician; potentially supplemented with SSEP neuromonitoring [ 67 ] . Although dynamic imaging can demonstrate subtle instability not seen on CT or static radiographs, it has not been shown to be superior to MRI. Upright cervical radiographs are commonly utilized to assess alignment in patients with traumatic instability undergoing non-operative management. They have not, however, demonstrated any utility for detection of subtle ligamentous instability in obtunded patients with negative CT scans, as exempli fi ed by Harris et al. in a retrospective review of 367 patients who underwent both studies. This retrospective review found no bene fi t to upright radiographs in the presence of a negative CT scan [ 58 ] . Obtunded patients like patient GB present a diagnostic challenge for cervical spine clearance. Despite a signi fi cant volume of research, no single standard algo- rithm has proven to be superior. MDCT has proven to be the most appropriate initial screening study, as plain radiographs are unable to detect a signi fi cant number of unstable injuries. Further, many of these patients also require CT evaluation of their chest, abdomen, pelvis, or head, and therefore concomitant cervical CT is ef fi cient and relatively cost effective. Therefore, obtunded patients, with a negative CT scan can be cleared in the majority of cases. Alternatively, a negative CT scan can be followed by cervical MRI to rule out ligamentous injury and associated potential instability. The latter is more costly and time consuming but does decrease the like- lihood of an occult soft tissue injury missed with CT alone. Both options are sup- ported in the literature, and choosing a particular algorithm may be based on availability and ef fi ciency of MRI at a particular institution or preferences of trauma and spinespecialists at that particular institution. Ultimately, there will be a very minute population of obtunded patients with negative CT scans who will have an unstable cervical injury. In determining which algorithm is appropriate for patient GB, the cost effectiveness of obtaining MRI on all obtunded patients in order to 352 Cervical Spine Clearance detect these extremely rare occurrences should be considered, but has not yet been borne out in the current literature. Early involvement of spine specialists is encour- aged as their ability to interpret CT scans and identify subtle fi ndings indicative of instability may decrease the number of missed unstable injuries in institutions where CT only is the standard for cervical spine clearance [ 61 ] . Literature Inconsistencies The greatest challenge in the literature is in regards to the clearance of the cervical spine in obtunded patients and whether a negative CT scan is suf fi cient for clearance or whether subsequent MRI or other imaging is required. Both methods of cervical clearance are supported by prospective large series. The challenges in determining an unequivocal standard algorithm seem secondary to the dif fi culty de fi ning an accurate percentage of unstable injuries that go undetected on CT scan, the morbid- ity associated with these missed injuries, and de fi ning the cost-effectiveness of requiring MRI in all obtunded patients in order to detect this population of patients. Evidentiary Table and Cervical Spine Clearance Algorithm De fi nitive Treatment Plan In the case presented, GB was initially moving all extremities but is now obtunded secondary to intubation for respiratory dysfunction. The imaging presented in this case is a cervical CT and dynamic radiographs. The cervical spine CT demonstrates overall normal alignment without listhesis and no signi fi cant boney injury. A more discerning radiologist or consultant, however, may recognize that between C6 and C7 there appears to be a subtle crack in the most anterior superior aspect of C7 and a disruption in the ossi fi ed annulus at the posteroinferior aspect of C6. This potential discrepancy demonstrates why the sensitivity of cervical CT is variable based upon whom is interpreting the study. As these CT fi ndings could indicate associated liga- mentous injury, further imaging to test ligamentous stability should be obtained prior to clearance. One option would be to obtain an MRI, whereas another would be to continue a cervical collar until dynamic radiographs could be performed in the subacute setting after return of cognitive function. In the case presented, dynamic radiographs were obtained. Based on the literature discussed, dynamic cervical radiographs should not be obtained acutely in an obtunded patient and are most useful as a physiologic test for instability when obtained in the subacute set- ting. The dynamic radiographs in this case demonstrate no evidence of instability and the patient’s cervical spine can thus be clinically cleared (Table 2.1 ). 36 A.K. Simpson and M.B. Harris References 1. McCaig LF, Ly N. National Hospital Ambulatory Medical Care Survey: 2000 emergency department summary. Advance data from vital and health statistics. No. 326. Hyattsville, Md.: National Center for Health Statistics, 2002. (DHHS publication no. (PHS) 2002-1250 02-0259.) 2. Roberge RJ, Wears RC. Evaluation of neck discomfort, neck tenderness, and neurologic de fi cits as indicators for radiography in blunt trauma victims. J Emerg Med. 1992;10:539–44. 3. Marshall LF, Knowlton S, Gar fi n SR, et al. Deterioration following spinal cord injury. A mul- ticenter study. J Neurosurg. 1987;66:400–4. 4. Morris CG, McCoy E. Clearing the cervical spine in unconscious polytrauma victims, balanc- ing risks and effective screening. Anaesthesia. 2004;59:464–82. 5. American College of Surgeons: American College of Surgeons Committee on Trauma. Advanced trauma life support: ATLS. 7th ed. Chicago, IL: American College of Surgeons; 2004. Table 2.1 Evidentiary table Author (year) Description Summary of results Level of evidence Harris et al. (2008) Retrospective cohort study 367 Obtunded blunt trauma patients underwent CT and were followed clinically and with upright fi lms to detect instability. Upright radiographs did not identify any injuries missed by CT. One patient with normal CT had a central cord contusion later identi fi ed by MRI and treated non-operatively. III Hogan et al. (2005) Retrospective cohort study 366 Obtunded blunt trauma patients underwent both CT and MRI. CT had negative predictive value of 98.9% for ligamentous injury and 100.0% for unstable cervical spine injury. III Tomycz et al. (2008) Retrospective cohort study 690 Patients underwent concomitant cervical CT and MRI. 38 of 180 patients (26.2%) found to have a negative CT demonstrated acute fi ndings, though none of these 38 patients with acute fi ndings required surgical intervention. III Menaker et al. (2008) Retrospective cohort study 203 Patients with unreliable examination underwent concomitant cervical CT and MRI. 18 patients (8.9%) with negative CT had an abnormal MRI, 14 of which required prolonged cervical immobiliza- tion and 2 of which required surgery. III Stassen et al. (2006) Retrospective cohort study 52 Obtunded blunt trauma patients underwent both CT and MRI. 25% of patients with a negative CT had a positive MRI for ligamentous injury and were treated in hard collar for 6 weeks. None of these patients ultimately required surgical intervention. III 372 Cervical Spine Clearance 6. ASIA. Standards for neurological classi fi cation of spinal injury. Chicago: American Spinal Injury Association; 1996. 7. Stauffer ES. Neurologic recovery following injuries to the cervical spinal cord and nerve roots. Spine. 1984;9:532–4. 8. Kiwerski J, Weiss M. Neurological improvement in traumatic injuries of cervical spinal cord. Paraplegia. 1981;19:31–7. 9. Duane TM, Dechert T, Wolfe LG, et al. Clinical examination and its reliability in identifying cervical spine fractures. J Trauma. 2007;62:1405–8. 10. Gonzalez RP, Fried PO, Bukhalo M, et al. Role of clinical examination in screening for blunt cervical spine injury. J Am Coll Surg. 1999;189:152–7. 11. Stiell IG, Clement CM, McKnight RD, et al. The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med. 2003;349:2510–8. 12. Hoffman JR, Mower WR, Wolfson AB, et al. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma: National Emergency X-Radiography Utilization Study Group. N Engl J Med. 2000;343:94–9. 13. Stiell IG, Wells GA, Vandemheen KL, et al. The Canadian C-spine rule for radiography in alert and stable trauma patients. JAMA. 2001;286:1841–8. 14. Tontz W, Anderson PA, Resnick DK. Clearance of the asymptomatic cervical spine: a meta- analysis. Spine J. 2006;6:60S. 15. Daffner RH, Hackney DB. ACR appropriateness criteria on suspected spine trauma. J Am Coll Radiol. 2007;4:762–75. 16. Hadley MN. Radiographic assessment of the cervical spine in asymptomatic trauma patients. Neurosurgery. 2002;50(3 suppl):S30–5. 17. France JC. Update on the appropriate radiographic studies for cervical spine Evaluation and clearance in the polytraumatized patient. Curr Orthop Pract. 2008;19:411–5. 18. Milby AH, Halpern CH, Guo W, et al. Prevalence of cervical spinal injury in trauma. Neurosurg Focus. 2008;25:E10. 19. Menaker J, Philp A, Boswell S,et al. Computed tomography alone for cervical spine clearance in the unreliable patient: are we there yet? J Trauma. 2008;64:898–903. 20. Muchow RD, Resnick DK, Abdel MP, et al. Magnetic resonance imaging (MRI) in the clear- ance of the cervical spine in blunt trauma: a meta-analysis. J Trauma. 2008;64:179–89. 21. Stassen NA, Williams VA, Gestring ML, et al. Magnetic resonance imaging in combination with helical computed tomography provides a safe and ef fi cient method of cervical spine clearance in the obtunded trauma patient. J Trauma. 2006;60:171–7. 22. Grossman MD, Reilly PM, Gillett T, et al. National survey of the incidence of cervical spine injury and approach to cervical spine clearance in U.S. trauma centers. J Trauma. 1999; 47:684–90. 23. Burrney RE, Maio RF, Maynard F, et al. Incidence, characteristics, and outcome of spinal cord injury at trauma centers in North America. Arch Surg. 1993;128:596–9. 24. Blackmore CC, Emerson SS, Mann FA, et al. Cervical spine imaging in patients with trauma: determination of fracture risk to optimize use. Radiology. 1999;211:759–65. 25. Parsons KC. The impact of spinal cord injury on long term survival. J Insur Med. 1991; 23:227. 26. Parsons KC, Lammertse DP. Rehabilitation in spinal cord disorders. 1. Epidemiology, preven- tion, and system of care of spinal disorders. Arch Phys Med Rehabil. 1991;72:S293–294. 27. Anderson PA, Muchow RD, Munoz A, et al. Clearance of the asymptomatic cervical spine: a meta-analysis. J Orthop Trauma. 2010;24:100–6. 28. Hoffman JR, Schriger DL, Mower W, et al. Low-risk criteria for cervical-spine radiography in blunt trauma: a prospective study. Ann Emerg Med. 1992;21:1454–60. 29. Mahadevan S, Mower WR, Hoffman JR, et al. Interrater reliability of cervical spine injury criteria in patients with blunt trauma. Ann Emerg Med. 1998;31:197–201. 30. Hoffman JR, Wolfson AB, Todd K, et al. Selective cervical spine radiography in blunt trauma: methodology of the National Emergency X-Radiography Utilization Study (NEXUS). Ann Emerg Med. 1998;32:461–9. 38 A.K. Simpson and M.B. Harris 31. Bandiera G, Stiell IG, Wells GA, et al. The Canadian C-Spine Rule performs better than unstructures physician judgement. Ann Emerg Med. 2003;42:395–402. 32. Stiell IG, Wells GA, McKnight RD, et al. Canadian C-Spine Rule study for alert and stable trauma patients. I. Background and rationale. Can J Emerg Med. 2002;4:84–90. 33. Idem. Canadian C-Spine Rule study for alert and stable trauma patients. II. Study objectives and methodology. Can J Emerg Med. 2002;4:185–93. 34. Como JJ, Diaz JJ, Dunham CM, et al. Practice management guidelines for identi fi cation of cervical spine injuries following trauma: update from the eastern association for the surgery of trauma practice management guidelines committee. J Trauma. 2009;67:651–9. 35. Harris MB, Sethi RK. The initial assessment and management of the multiple-trauma patient with an associated spine injury. Spine (Phila Pa 1976). 2006;31(11 suppl):S9–S15. 36. Holmes JF, Akkinepalli R. Computed tomography versus plain radiography to screen for cer- vical spine injury: a metaanalysis. J Trauma. 2005;58:902–5. 37. Gale SC, Gracias VH, Reilly PM, et al. The inef fi ciency of plain radiography to evaluate the cervical spine after blunt trauma. J Trauma. 2005;59:1121–5. 38. Bachulis BL, Long WB, Hynes GD, et al. Clinical indications for cervical spine radiographs in the traumatized patient. Am J Surg. 1987;153:473–8. 39. Radiographic assessment of the cervical spine in symptomatic trauma patients. Neurosurgery. 2002;50:S36–S43. 40. MacDonald RL, Schwartz ML, Mirich D, et al. Diagnosis of cervical spine injury in motor vehicle crash victims: how many X-rays are enough? J Trauma. 1990;30:392–7. 41. Brown CV, Antevil JL, Sise MJ, et al. Spiral computed tomography for the diagnosis of cervi- cal, thoracic, and lumbar spine fractures: its time has come. J Trauma. 2005;58:890–5. 42. Hennessy D, Widder S, Zygun D, et al. Cervical spine clearance in obtunded blunt trauma patients: a prospective study. J Trauma. 2010;68:576–82. 43. McCulloch PT, France J, Jones DL, et al. Helical computed tomography alone compared with plain radiographs with adjunct computed tomography to evaluate the cervical spine after high energy trauma. J Bone Joint Surg Am. 2005;87:2388–94. 44. Grogan EL, Morris Jr JA, Dittus RS, et al. Cervical spine evaluation in urban trauma centers: lowering institutional costs and complications through helical CT scan. J Am Coll Surg. 2005;200:160–5. 45. Khan SN, Erickson G, Sena MJ, Gupta MC. Use of fl exion and extension radiographs of the cervical spine to rule out acute instability in patients with negative computed tomography scans. J Orthop Trauma. 2011;1:51–6. 46. Wadhwa R, Shamieh S, Haydel J, et al. The role of fl exion and extension computed tomogra- phy with reconstruction in clearing the cervical spine in trauma patients: a pilot study. J Neurosurg Spine. 2011;14:341–7. 47. Vaccaro AR, Kreidl KO, Pan W, et al. Usefulness of MRI in isolated upper cervical spine fractures in adults. J Spinal Disord. 1998;11:289–93. 48. Chiu WC, Haan JM, Cushing BM, et al. Ligamentous injuries of the cervical spine in unreli- able blunt trauma patients: incidence, evaluation, and outcome. J Trauma. 2001;50:457–63. 49. Chiu WC, Haan JM, Cushing BM, et al. Ligamentous injuries of the cervical spine in unreli- able blunt trauma patients: incidence, evaluation, and outcome. J Trauma. 2001;50:457–63. discussion 464. 50. Iida H, Tachibana S, Kitahara T, et al. Association of head trauma with cervical spine injury, spinal cord injury, or both. J Trauma. 1999;46:450–2. 51. Webber-Jones JE, Thomas CA, Bordeaux Jr RE. The management and prevention of rigid cervical collar complications. Orthop Nurs. 2002;21:19–25. 52. Laylock B. Solving the problem of pressure ulcers resulting from cervical collars. Ostomy Wound Manage. 1996;42:26–8. 30, 32–3. 53. Liew SC, Hill DA. Complication of hard cervical collars in multi-trauma patients. Aust N Z J Surg. 1994;64:139–40. 54. Powers JA. Multidisciplinary approach to occipital pressure ulcers related to cervical collars. J Nurs Care Qual. 1997;12:46–52. 392 Cervical Spine Clearance 55. Watts D, Abrahams E, MacMillan C, et al. Insult after injury: pressure ulcers in trauma patients. Orthop Nurs. 1998;17:84–91. 56. Papadopoulos MC, Chakraborty A, Waldron G, et al. Lesson of the week: exacerbating cervi- cal spine injury by applying a hard collar. BMJ. 1999;319:171–2. 57. Ho AM, Fung KY, Joynt GM, et al. Rigid cervical collar and intracranial pressure of patients with severe head injury. J Trauma. 2002;53:1185–8. 58. Horodyski M, Dipaola CP, Conrad BP, et al. Cervical collars are insuf fi cient for immobilizing an unstable cervical spine injury. J Emerg Med. 2011;41:513–9. 59. Harris TJ, Blackmore CC, Mirza SK, et al. Clearing the cervical spine in obtunded patients. Spine. 2008;33:1547–53. 60. Hogan GJ, Mirvis SE, Shanmuganathan K, Scalea TM. Exclusion of unstable cervical spine injury in obtunded patients with blunt trauma: is MR imaging needed when multi-detector row CT fi ndings are normal? Radiology. 2005;237:106–13. 61. Tomycz ND, Chew BG, Chang YF, et al. MRI is unnecessary to clear the cervical spine in obtunded/comatose trauma patients: the four-year experience of a level I trauma center. J Trauma. 2008;64:1258–63. 62. Simon JB, Schoenfeld AJ, Ketz JN, et al. Are “normal”multidetector computed tomographic scans suf fi cient to allow collar removal in the trauma patient? J Trauma. 2010;68:103–8. 63. Menaker J, Philp A, Boswell S, Scalea TM. Computed tomography alone for cervical spine clearance in the unreliable patient: are we there yet? J Trauma. 2008;64:898–903. 64. Bolinger B, Shartz M, Marion D. Bedside fl uoroscopic fl exion and extension cervical spine radiographs for clearance of the cervical spine in comatose trauma patients. J Trauma. 2004;56:132–6. 65. Spiteri V, Kotnis R, Singh P, et al. Cervical dynamic screening in spinal clearance: now redun- dant. J Trauma. 2006;61:1171–7. 66. Anglen J, Metzler M, Bunn P, et al. Flexion and extension views are not cost-effective in a cervical spine clearance protocol for obtunded trauma patients. J Trauma. 2002;52:54–9. 67. Cox MW, McCarthy M, Lemmon G, et al. Cervical spine instability: clearance using dynamic fl uoroscopy. Curr Surg. 2001;58:96–100. 41M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_3, © Springer Science+Business Media New York 2013 Keywords Cervical spine trauma Operative management of cervical spine fracture Closed reduction and external orthosis Long-term outcomes MJ: 43-Year-Old Female with Neck Pain and Paralysis Case Presentation MJ is a 43-year-old female who presents to the emergency department after a fall from a standing height. On presentation, the patient complains of severe neck pain and bilateral upper and lower extremity weakness. Her primary survey demonstrates a clear airway and hemodynamic stability. She denies any LOC and maintains a GCS of 15. Her secondary survey is negative. On physical examination, the patient is awake and alert. She demonstrates 5/5 strength in her deltoids, 5/5 strength of the biceps, 1/5 strength of the triceps, and 0/5 strength distally in the upper extremities. She has 2/5 gastroc strength and 1/5 quad strength. Sensation is intact at the C6 Level but is diminished below. She dem- onstrates an intact biceps, patella, and Achilles re fl exes, but has an absent triceps re fl ex. She has diminished rectal tone and demonstrates an intact bulbocavernosus re fl ex and anal sensation. Cervical spine CT and MRI C-spine images are included below in Figs. 3.1 and 3.2 . K. R. O’Neill , M.D., M.S. J. E. Bible , M.D., M.H.S. C.J. Devin, M.D. (*) Department of Orthopedic Surgery and Rehabilitation , Vanderbilt Medical Center , Suite 4200, South tower Medical Center East – 1215 21st Avenue South , Nashville , TN 37232 , USA e-mail: kevin.oneill@vanderbilt.edu; clinton.j.devin@vanderbilt.edu Chapter 3 Cervical Spine Fracture Dislocation Kevin R. O’Neill , Jesse E. Bible , and Clinton James Devin 42 K.R. O’Neill et al. Interpretation of Clinical Presentation Cervical spine trauma accounts for roughly half of the 11,000 spinal cord injuries that occur annually in North America. Injuries to the subaxial cervical spine, involv- ing levels C3-7, constitute two-thirds of cervical spine fractures and more than three-fourths of spine dislocations [ 1 ] . Despite this prevalence, the optimal medical and surgical treatments are not clear. Patients with injury mechanisms that place them at risk of a spine injury should be immobilized at the scene with a cervical collar and rigid backboard. The entire spine should be evaluated, as noncontiguous spinal trauma occurs in 19% of patients with cervical spine injuries (8% cervical, 8% thoracic, and 6% lumbar) [ 2 ] . Further, patients with cervical spine trauma should undergo a complete trauma team evaluation, as 57% of these patients present with extraspinal injuries (34% Fig. 3.1 ( a ) Axial CT spine, ( b ) Axial CT spine, ( c ) Coronal CT spine, ( d ) Coronal CT spine, ( e ) Sagittal CT spine 433 Cervical Spine Fracture Dislocation head and neck, 17% intrathoracic, 10% intraabdominal/pelvic, and 30% nonspinal orthopedic conditions) [ 2 ] . The standard Advanced Trauma Life Support protocol supports the use of plain radiographs as a screening tool with computed tomography (CT) employed as an adjunct. However, recent evidence suggests that obtaining a helical CT alone (sen- sitivity 98%) is superior to this protocol (sensitivity 45%) [ 3 ] . The additional use of magnetic resonance imaging (MRI) provides valuable information in evaluating for intervertebral disk and posterior ligamentous complex (PLC) disruption, spinal cord compression, and intraparenchymal injury. A fat-suppressed T2-weighted sagittal sequence has been shown to be a highly sensitive, speci fi c, and accurate method of evaluating PLC injury [ 4 ] . Fig. 3.2 ( a ) T1 Sagittal MRI spine, ( b ) T2 Sagittal MRI spine, ( c ) T1 Axial MRI spine 44 K.R. O’Neill et al. Neurologic assessments of motor and sensory function are made according to American Spinal Injury Association (ASIA) guidelines. Spinal cord injuries are classi fi ed according to ASIA guidelines as follows: [ 5 ] A. Complete: No sensory or motor function is preserved in sacral segments S4–S5. B. Incomplete: Sensory, but not motor, function is preserved below the neurologic level and extends through sacral segments S4–S5. C. Incomplete: Motor function is preserved below the neurologic level, and more than half of the key muscles below the neurologic level have muscle grade less than 3. D. Incomplete: Motor function is preserved below the neurologic level, and more than half of the key muscles below the neurologic level have muscle grade greater than or equal to 3. E. Normal: Sensory and motor functions are normal. Cervical spine injuries have historically been classi fi ed based on the presumed mechanism of injury as determined by utilizing plain radiographs. However, this classi fi cation scheme does not provide information relating to stability of the injured spine and therefore is less useful in determining treatment. A more recent classi fi cation, the subaxial injury classi fi cation (SLIC), has been developed that assesses subaxial cervical spine trauma based on injury morphology, integrity of the discoligamentous complex, and neurologic status of the patient [ 6 ] . Points are awarded for each component, and the total score can be used to help guide surgical versus nonsurgical treatment. This patient presents with a C6-7 bilateral facet dislocation with incomplete ASIA grade B C5 spinal cord injury. The sagittal CT image (Fig. 3.1 ) demonstrates 50% anterolisthesis of C6 on C7. Axial and sagittal MRI images (Fig. 3.2 ) demon- strate a disrupted C6-7 discoligamentous complex. There is retropulsion of disk material causing cord compression. Increased signal within the spinal cord at this level can be seen on the T2 image sequences. There is complete disruption of the PLC. This injury would be classi fi ed as a fl exion-distraction injury with a SLIC total score of 8, indicating likely need for surgical stabilization. Declaration of Speci fi c Diagnosis This patient presents with a C6-7 bilateral facet dislocation with an incomplete C7 spinal cord injury secondary to a fall. Brainstorming: What Are the Treatment Goals and the Surgical Options? Treatment Goals: 1. Reduce the dislocation 2. Provide cervical stabilization 453 Cervical Spine Fracture Dislocation 3. Minimize secondary spinal cord injury 4. Maximize potential for neurologic recovery Treatment Discussion and Options:1. Optimal medical management 2. Closed reduction and external orthosis 3. Open posterior reduction and fusion 4. Open anterior discectomy, reduction, and fusion 5. Combined anterior and posterior reduction and fusion 6. Intraoperative considerations (a) Positioning (b) Neurophysiological monitoring Evaluation of the Literature In order to identify relevant articles in the treatment of this patient with a cervical facet fracture-dislocation and spinal cord injury, PubMed searches were conducted on articles published between 1975 and the present. Search headings included “Cervical Vertebrae,” “Spinal Injuries,” and “Spinal Cord Injuries”. This search yielded 341 abstracts in the English literature. These abstracts were reviewed, and relevant articles were selected. Select references from these articles were also reviewed and relevant articles were also included. Detailed Review of Pertinent Articles Medical Management Medical management of spinal cord injuries is directed at mitigating secondary injury and starts with maintenance of spinal cord perfusion. Maintenance of mean arterial pressures greater than 85 mmHg has been shown to improve neurological outcome [ 7 ] . This generally requires invasive hemodynamic monitoring in an inten- sive care setting and may require intubation. A variety of pharmacologic agents have been investigated for use in spinal cord injuries in an attempt to mitigate secondary injury and improve neurological out- come. However, no such agent has been found to be clinically effective. The use of steroids remains controversial as the three National Acute Spinal Cord Injury Study (NASCIS) trials [ 8– 10 ] have been widely criticized for errors of randomization, clinical endpoints, reliability of data collection, and de fi nitions of functional motor levels [ 11 ] . In a meta-analysis combining the NASCIS studies with others that investigated steroid use in spinal cord injury, the use of steroids was found to remain unproven and experimental [ 12 ] . 46 K.R. O’Neill et al. Hypothermia has also been investigated, as it slows basic enzymatic activity and reduces energy requirements and therefore may have a neuroprotective effect. Animal studies of acute traumatic spinal cord injury have yielded inconclusive results [ 13 ] . Although systemic hypothermia may hold some increased risks of coagulopathy, sepsis, and cardiac dysrhythmia, the only clinical trial thus far found no difference in complications between 48 h of 33°C intravascular hypothermia and control groups [ 14 ] . However, little can be concluded from the 14 complete cervical spine injury patients included in this phase 1 trial, of which only three were ASIA C or higher at fi nal 1 year follow-up. Conservative Management Closed Reduction and Magnetic Resonance Imaging It is important to relieve spinal cord compression as soon as possible. Immediate closed reduction using skeletal traction with sequentially increased weight is usu- ally effective in restoring cervical alignment [ 15 ] . However, closed reductions can result in increased disk herniation [ 16 ] . The risk of signi fi cant neurologic deteriora- tion resulting from this is rare [ 17 ] . A prereduction MRI has been advocated in patients with an unreliable exam due to altered mental status. The main concern is identifying disk material in the spinal canal that may cause increased cord compres- sion and potential further neurological deterioration, following a closed reduction [ 18 ] . Others believe that this risk is so low that the delay of obtaining an MRI prior to reduction is not warranted [ 15 ] . The mental and neurological status of the patient should be considered in this decision. In the awake and alert patient who is intact or has an incomplete injury, a closed reduction can safely be performed provided that a contraindication to trac- tion does not exist. The main contraindications to the use of a closed reduction attempt in this population are a skull fracture or concomitant upper cervical disloca- tion [ 16 ] . If neurologic deterioration occurs during the reduction attempt, the trac- tion should be removed and an open reduction performed. If an awake exam is not possible or the patient is in spinal shock, neurologic deterioration resulting from traction would not be recognized. Prereduction MRI may be useful in these cases to evaluate for extruded disk material that might cause such deterioration following a closed reduction. If the patient has a complete injury, there is no risk of further dete- rioration and an immediate closed reduction should be performed. External Orthosis Bilateral facet fracture-dislocations are unstable injuries and are best treated with operative fi xation. As such, there is no data available regarding conservative treat- ment with orthoses of these injuries. However, it is useful to consider available evidence in the treatment of unilateral facet fracture-dislocations. In a retrospective 473 Cervical Spine Fracture Dislocation review of 34 patients with unilateral facet dislocations or fracture-dislocations, the surgical group ( n = 10) in comparison to a nonoperative group ( n = 24) was found to be more likely to attain anatomic reductions (60% vs. 25%) and bony fusion (100% vs. 46%), and less likely to have signi fi cant chronic pain (10% vs. 42%) and transla- tion on fl exion–extension views (20% vs. 38%) [ 19 ] . However, none of the differ- ences between groups were statistically signi fi cant ( p > 0.05), and the methodology was not clearly described. In another retrospective study of unilateral facet fracture- dislocations, 18 patients treated with a cervical collar or halo had failures related to either inability to hold the reduction or persistent neurologic de fi cit [ 20 ] . Patients treated with posterior reduction and fusion had signi fi cantly improved SF-36 PCS scores compared to the nonoperative group [ 21 ] . It is reasonable to assume that the bene fi ts of surgery would be even greater in the more unstable case of bilateral facet dislocations. To help determine whether operative intervention or conservative treatment is indicated, an evidence-based algorithm has been developed based on the SLIC classi fi cation [ 22 ] . Combining available studies with expert opinion, this study sug- gested guidelines to help with clinical decision making. With regard to facet dislocations or fracture-dislocations in particular, four points are awarded for the injury pattern and two points for disruption of the discoligamentous complex. An additional four points would be allotted based on the neurologic status of the patient, with surgical treatment being recommended for scores greater than 5. Surgical Management Intubation Intubation and patient positioning must be carefully considered in cases of unstable cervical spine injuries. Cervical motion should be minimized in order to prevent further injury to the spinal cord. Although inline intubation is the quickest technique for airway control, it produces signi fi cant cervical motion with an average of 22.5° during intubation with a Macintosh laryngoscope. Instead, use of a fi ber-optic guided system or a Bullard laryngoscope should strongly be considered as these produce signi fi cantly less cervical motion (5.5 and 3.4°, respectively) [ 23 ] . Patient Positioning It is also important to limit cervical spine motion when turning a patient from supine to prone on the operating table. Although patients can be manually rolled into a prone position, utilization of a rotating Jackson spine table has been shown to result in two to three times less cervicalangular motion [ 24 ] . In this cadaveric study, cer- vical motion was signi fi cantly reduced in all three motion planes compared to the manual method. Additionally, use of a cervical collar during these maneuvers was shown to signi fi cantly reduce motion. 48 K.R. O’Neill et al. Neuromonitoring Intraoperative neuromonitoring (IOM) is commonly used in order to avert compli- cations and optimize outcomes. In one retrospective study, IOM was shown to be helpful in preventing a postoperative de fi cit in 5.2% of patients [ 25 ] . Somatosensory- evoked potentials (SSEPs), in which the distal extremities are stimulated and record- ings are made at the scalp, provide continuous intraoperative assessments of dorsal column sensory pathways. In ideal situations, SSEPs can require 5 min to detect neurologic changes. Transcranial motor-evoked potentials (tcMEPs), where cranial stimulation is provided and compound motor action potentials (CMAP) are recorded in distal muscle groups, assess motor pathways. Motor-evoked potentials detect a change much quicker due to the higher metabolism rate of the anterior horn cells. The combination of tcMEPs and SSEPs permits assessment of both ascending sen- sory and descending motor pathways. Preposition and postposition neuromonitoring are commonly performed to ensure that there is not a neurological change during positioning. Additionally, unless a complete neurological injury exists with no SSEP or MEPs detected, con- tinuous intraoperative monitoring is commonly performed. If a signi fi cant change is detected (SSEP amplitude decreases >50% or MEP amplitude decreases >75%), then the fi rst step should be to elevate the mean arterial pressure and ensure that the patient is adequately oxygenated, followed by removal of hardware and assessment of inadequate decompression if there is no improvement. Surgical Timing There has been considerable interest in determining the optimal timing of surgical decompression following spinal cord injury. A multicenter prospective cohort study, named the Surgical Trial in Acute SCI Study (STASCIS), is currently underway to investigate this question. Current literature seems to provide some support for early (<24–72 h) surgical decompression [ 26 ] . In an evidenced-based review, 19 papers were identi fi ed involving animal models of SCI. In 11 of these studies, improved outcomes were found with early surgical decompression. Additionally, 22 clinical studies were evaluated. Drawing conclusions from these studies regarding the effect of early decompression is complicated by the various de fi nitions of “early” in the literature, which ranged from 8 h to 4 days. However, none of the studies demon- strated increased complications or a worse clinical outcome with early surgery in the medically stable patient. Several studies demonstrated improved outcomes in groups treated with early decompression, with shorter hospitalization, shorter length of ICU care, decreased mortality, decreased complications, and improved neuro- logic outcome. Based on this review, it was concluded that patients with acute SCI can safely undergo early decompression once medically stable, with the potential for improved neurological outcome. 493 Cervical Spine Fracture Dislocation Surgical Approach In the absence of a herniated disk at the level of dislocation, a facet dislocation or fracture-dislocation can be reduced and stabilized using either an anterior or poste- rior approach. Available literature does not indicate a clearly superior approach. In a prospective study of 52 patients with spinal cord injury and unstable subaxial cervical spine injuries, no difference was found when comparing anterior and pos- terior approaches in terms of fusion rates, alignment, neurologic recovery, or long- term complaints of pain [ 27 ] . All patients in this study received MPSS and had a closed reduction prior to surgery. Radiographic failure and loss of reduction may occur more often using an ante- rior approach alone. In a retrospective radiographic review of 87 patients with either unilateral or bilateral facet dislocations or fracture-dislocations treated with anterior cervical discectomy and fusion (ACDF) using a static anterior plate, 13% demon- strated radiographic failure de fi ned as translation of at least 4 mm or increased kyphosis of at least 11° [ 28 ] . In contrast, another retrospective radiographic review of 65 patients whose facet dislocation or fracture-dislocation were managed with single segment posterior cervical instrumentation (plate or wire fi xation) and fusion, 3.5% ( n = 2) demonstrated radiographic failure [ 29 ] . The integrity of the PLC is an important consideration. In a cadaveric biomechanical study using a corpectomy model, anterior fi xation with a static plate and PEEK cage was found to adequately stabilize the cervical motion segment provided that that PLC was intact. If the PLC was disrupted, anterior combined with segmental posterior fi xation was required to achieve adequate stability [ 30 ] . If the dislocation is associated with a herniated disk, an anterior approach is preferred, as the disk can then be directly removed before the reduction is per- formed. This approach is supported by an algorithm based on the SLIC classi fi cation [ 31 ] . In one study, various treatment options were compared in patients with facet fracture-dislocations. A retrospective arm consisted of 12 patients treated with hard cervical collar, 6 patients with halo, and 11 patients with posterior fusion using either wire or lateral mass screw and rod fi xation. A prospective group of 18 patients treated with ACDF using a static plate and iliac crest bone graft was compared to this cohort [ 20 ] . Patients treated with a cervical collar or halo had failures related to either inability to hold the reduction or persistent neurologic de fi cit. The poste- rior fusion group had 45% failure due to inadequate reduction ( n = 5), persistent radiculopathy ( n = 2), or progressive kyphosis ( n = 3). The study did not indicate whether these failures occurred with wire versus lateral mass fi xation. The anterior fusion group demonstrated 100% success, as de fi ned by the ability to achieve and maintain reduction, maintain neurological status, and prevent the need for second- ary surgery. Another retrospective study of 22 patients with bilateral facet fracture- dislocations con fi rmed these excellent results using ACDF with bone graft and static plate fi xation, with all patients demonstrating radiographic union at 32 months follow-up [ 32 ] . 50 K.R. O’Neill et al. Literature Inconsistencies Most of the studies used to drive current treatment strategies related to this case are based off expert opinion, retrospective studies, and under-powered prospective series. More prospective randomized data is needed to better guide decision making. Evidentiary Table and Selection of Treatment Method The key studies in treating MJ are listed in Table 3.1 . Based on the available litera- ture, MJ would undergo an attempted closed reduction immediately. Within 24 h of presentation, she would undergo anterior open reduction and ACDF, followed by posterior lateral mass screw and rod fi xation. De fi nitive Treatment Plan The patient would initially be placed in a rigid cervical orthosis and kept in an ICU environment with maintenance of the mean arterial pressure above 85 mmHg. The authors do not believe that clear bene fi ts have been demonstrated by current phar- macologic interventions aimed at improving neurologic recovery. As a result, no such interventions, including steroid administration, would be initiated. Table 3.1 Evidentiary table Author (year) DescriptionSummary of results Quality of evidence Hurlbert (2001) Meta-analysis Meta-analysis combining results of nine studies on effects of steroids in acute spinal cord injury, showing steroids to have known risk and unproven bene fi t. Level I Furlan et al. (2011) Meta-analysis 19 animal studies and 22 clinical studies included on the timing of surgical decompression in spinal cord injury, suggesting early decompression within 24 h if medically feasible. Level III Lifeso et al. (2000) Prospective, nonrandomized 18 patients with unstable cervical spine fractures successfully treated with anterior stabilization, compared retrospectively to 11 patients treated posteriorly with 45% failure, and 18 patients treated nonoperatively with 100% failure. Level III 513 Cervical Spine Fracture Dislocation Because the patient is awake and alert, a closed reduction attempt would be initiated as soon as possible with Gardner-Wells tongs in order to minimize the time of cord compression. Ten pounds of traction would initially be applied, followed by the incremental addition of 10 pounds until reduction was achieved. After each addition of weight, a neurologic and radiographic evaluation with a lateral radio- graph would be performed. Traction would be discontinued if the patient developed worsening neurological complaints, if radiographs demonstrated overdistraction of the injured or adjacent level, or if there was evidence of occipitocervical distraction. Following reduction, the patient would be kept in 20 pounds of traction and the cervical collar would be kept in place. An MRI would then be obtained to localize any continued areas of cord compression and assist in surgical planning. Early surgical stabilization would be performed within 24–48 h after injury. An anterior surgical approach would be used in this case because of the herniated disk at the injured level. In the operating room, pre- and postpositioning and intraopera- tive SSEPS and MEPS would be obtained. A lateral fl uoroscopic view would be taken prior to prepping and draping to insure adequate visualization. The mean arte- rial pressure would be kept above 85 and careful attention would be given to main- taining oxygenation. A standard anterior cervical discectomy and fusion would be performed. Caspar pins would be asymmetrically placed in a divergent manner to facilitate the reduc- tion and disk removal. An autograft or allograft would be placed in the disk space taking care not to overdistract, and a static anterior plate would be applied. Given that there is injury to the PLC, the patient would be turned to a prone position with posterior segmental fi xation and fusion with autograft or allograft (Fig. 3.3 ). If a reduction was unsuccessful from an anterior approach, the disk space would be left Fig. 3.3 ( a ) Postoperative lateral radiograph cervical spine, ( b ) Postoperative AP radiograph cervical spine 52 K.R. O’Neill et al. empty, the patient would be rolled into the prone position and a posterior reduction and stabilization would be performed. The patient would then be rolled back into a supine position for placement of an anterior graft and a static plate. A hard cervical collar would be placed until stability was veri fi ed with fl exion and extension radio- graphs. The patient would remain in the ICU to maintain spinal cord perfusion. References 1. Goldberg W, Mueller C, Panacek E, et al. Distribution and patterns of blunt traumatic cervical spine injury. Ann Emerg Med. 2001;38:17–21. 2. Miller CP, Brubacher JW, Biswas D, et al. The incidence of noncontiguous spinal fractures and other traumatic injuries associated with cervical spine fractures: a ten year experience at an academic medical center. Spine. 2011;36:1532–40. 3. McCulloch PT, France J, Jones DL, et al. Helical computed tomography alone compared with plain radiographs with adjunct computed tomography to evaluate the cervical spine after high- energy trauma. J Bone Joint Surg Am. 2005;87:2388–94. 4. Lee HM, Kim HS, Kim DJ, et al. Reliability of magnetic resonance imaging in detecting posterior ligament complex injury in thoracolumbar spinal fractures. Spine. 2000;25:2079–84. 5. Ditunno JF, Young W, Donovan WH, et al. The international standards booklet for neurologi- cal and functional classi fi cation of spinal cord injury. American Spinal Injury Association. Paraplegia. 1994;32:70–80. 6. Patel AA, Hurlbert RJ, Bono CM, et al. Classi fi cation and surgical decision making in acute subaxial cervical spine trauma. Spine. 2010;35:S228–34. 7. Vale FL, Burns J, Jackson AB, et al. Combined medical and surgical treatment after acute spinal cord injury: results of a prospective pilot study to assess the merits of aggressive medical resuscitation and blood pressure management. J Neurosurg. 1997;87:239–46. 8. Bracken MB, Collins WF, Freeman DF, et al. Ef fi cacy of methylprednisolone in acute spinal cord injury. JAMA. 1984;251:45–52. 9. Bracken MB, Shepard MJ, Collins WF, et al. A randomized, controlled trial of methylpredni- solone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engl J Med. 1990;322:1405–11. 10. Bracken MB, Shepard MJ, Holford TR, et al. Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA. 1997;277:1597–604. 11. Nesathurai S. Steroids and spinal cord injury: revisiting the NASCIS 2 and NASCIS 3 trials. J Trauma. 1998;45:1088–93. 12. Hurlbert RJ. The role of steroids in acute spinal cord injury: an evidence-based analysis. Spine. 2001;26:S39–46. 13. Kwon BK, Mann C, Sohn HM, et al. Hypothermia for spinal cord injury. Spine J. 2008; 8:859–74. 14. Levi AD, Casella G, Green BA, et al. Clinical outcomes using modest intravascular hypo- thermia after acute cervical spinal cord injury. Neurosurgery. 2010;66:670–7. 15. Grant GA, Mirza SK, Chapman JR, et al. Risk of early closed reduction in cervical spine sub- luxation injuries. J Neurosurg. 1999;90:13–8. 16. Vaccaro AR, Falatyn SP, Flanders AE, et al. Magnetic resonance evaluation of the interverte- bral disc, spinal ligaments, and spinal cord before and after closed traction reduction of cervi- cal spine dislocations. Spine. 1999;24:1210–7. 17. Wimberley DW, Vaccaro AR, Goyal N, et al. Acute quadriplegia following closed traction reduction of a cervical facet dislocation in the setting of ossi fi cation of the posterior longitudi- nal ligament: case report. Spine. 2005;30:E433–8. 533 Cervical Spine Fracture Dislocation 18. Hart RA. Cervical facet dislocation: when is magnetic resonance imaging indicated? Spine. 2002;27:116–7. 19. Beyer CA, Cabanela ME, Berquist TH. Unilateral facet dislocations and fracture-dislocations of the cervical spine. J Bone Joint Surg Br. 1991;73:977–81. 20. Lifeso RM, Colucci MA. Anterior fusion for rotationally unstable cervical spine fractures. Spine. 2000;25:2028–34. 21. Dvorak MF, Fisher CG, Aarabi B, et al. Clinical outcomes of 90 isolated unilateral facet frac- tures, subluxations, and dislocations treated surgically and nonoperatively. Spine. 2007;32:3007–13. 22. Vaccaro AR, Hulbert RJ, Patel AA, et al. The subaxial cervical spine injury classi fi cation sys- tem: a novel approach to recognize the importance of morphology, neurology, and integrity of the disco-ligamentous complex. Spine. 2007;32:2365–74.23. Wahlen BM, Gercek E. Three-dimensional cervical spine movement during intubation using the Macintosh and Bullard laryngoscopes, the bon fi ls fi brescope and the intubating laryngeal mask airway. Eur J Anaesthesiol. 2004;21:907–13. 24. DiPaola MJ, DiPaola CP, Conrad BP, et al. Cervical spine motion in manual versus Jackson table turning methods in a cadaveric global instability model. J Spinal Disord Tech. 2008;21:273–80. 25. Wiedemayer H, Fauser B, Sandalcioglu IE, et al. The impact of neurophysiological intraopera- tive monitoring on surgical decisions: a critical analysis of 423 cases. J Neurosurg. 2002;96:255–62. 26. Furlan JC, Noonan V, Cadotte DW, et al. Timing of decompressive surgery of spinal cord after traumatic spinal cord injury: an evidence-based examination of pre-clinical and clinical stud- ies. J Neurotrauma. 2011;28:1371–99. 27. Brodke DS, Anderson PA, Newell DW, et al. Comparison of anterior and posterior approaches in cervical spinal cord injuries. J Spinal Disord Tech. 2003;16:229–35. 28. Johnson M, Fisher C, Boyd M, et al. The radiographic failure of single segment anterior cervi- cal plate fi xation in traumatic cervical fl exion distraction injuries. Spine. 2004;29(24): 2815–20. 29. Elgafy H, Fisher C, Zhao Y, et al. The radiographic failure of single segment posterior cervical instrumentation in traumatic cervical fl exion distraction injuries. Topics in spinal cord injury rehabilitation. 2006;12(2):20–9. 30. Karam YR, Dahdaleh NS, Magnetta MJ, et al. Biomechanical comparison of anterior, poste- rior, and circumferential fi xation after one-level anterior cervical corpectomy in the human cadaveric spine. Spine. 2011;36:E455–60. 31. Dvorak MF, Fisher CG, Fehlings MG, et al. The surgical approach to subaxial cervical spine injuries: an evidence-based algorithm based on the SLIC classi fi cation system. Spine. 2007;32:2620–9. 32. Razack N, Green BA, Levi AD. The management of traumatic cervical bilateral facet fracture- dislocations with unicortical anterior plates. J Spinal Disord. 2000;13:374–81. 55M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_4, © Springer Science+Business Media New York 2013 Keywords Brace management of lumbar spine fracture • Lumbar spine trauma • Operative management of lumbar spine fracture: anterior and/or posterior approach • Thoracic/lumbar spine burst fracture HS: 21-Year-Old Male with Lower Back Pain Case Presentation HS is a 21-year-old male who presents status post fall from 20 ft, transferred via EMS. On presentation, the patient demonstrates a GCS of 15, denies any LOC, and complains of severe lower back pain. On primary survey, HS demonstrates a patent airway and is hemodynamically stable. His secondary survey is otherwise negative. He complains of right lower extremity radicular pain and feels that he is not able to urinate. On physical exam, the patient is awake and alert. He demonstrates 5/5 strength in the upper and lower extremities with normal re fl exes. He has normal rectal tone with no clonus or Babinski sign. Radiographs, CT, and MRI of the Lumbar Spine are demonstrated below in Figs. 4.1 , 4.2 , and 4.3 . R. Greenleaf , M.D. (*) Reconstructive Orthopedics , 737 Main Street, Suite 6 , Lumberton , NJ 08048 , USA e-mail: Bob.greenleaf1@gmail.com M. B. Harris , M.D., F.A.C.S. Orthopedic Trauma, Department of Orthopedic Surgery , Harvard Medical School, Brigham and Women’s Hospital , 75 Francis Street , Boston , MA 02115 , USA Chapter 4 Lumbar Burst Fractures Robert Greenleaf and Mitchel B. Harris Fig. 4.1 ( a ) AP radiograph lumbar spine, ( b ) Lateral radiograph lumbar spine Fig. 4.2 ( a ) Axial CT spine, ( b ) Axial CT spine, ( c ) Sagittal CT spine, ( d ) Sagittal CT spine, ( e ) Coronal CT spine 574 Lumbar Burst Fractures Interpretation of Clinical Presentation This is a young patient with a primary complaint of low back pain, with associated right leg pain after falling from a height. With this type of mechanism of injury, care must be taken during the initial evaluation to rule out any occult life threatening inju- ries. Airway, breathing, and circulatory status must be fully assessed. Further, a com- prehensive secondary survey evaluating all extremities must be carefully performed as the spine injury may be a distracting injury. Hsu et al. [ 1 ] in a large retrospective review of thoracolumbar spine fractures in blunt trauma patients found a 68% inci- dence of associated injuries with a 25% incidence of major injuries (Abbreviated injury scale ³ 3). Axial loading injuries to the lower extremities, such as pelvic, tibial plateau, and calcaneus fractures, commonly occur. As with any injury to the spine, strict immobilization must be observed until the stability of the injury and a full assessment of neurological function is performed. In the thoracolumbar spine, this consists of observing logroll precautions with the patient remaining in a fl at supine position. A logroll should be performed in the secondary survey to look for signs of trauma on the back. While doing this logroll, the spinous processes should be pal- pated throughout the entire length of the spine. Abnormalities such as swelling or bogginess of the soft tissue overlying the spine, ecchymosis, palpable gaps between spinous processes, or areas of focal tenderness should be noted. Positive fi ndings help localize the area of injury and give clinical cues regarding the extent of posterior liga- mentous complex injury, which will be discussed in a later section. A digital rectal exam and perianal sensation evaluation are also conveniently performed at this time. After performing the initial trauma evaluation, focus may turn to the spinal injury. A careful history is taken from the patient, observers, or EMS personnel. Fig. 4.3 ( a ) T1 Sagittal MRI spine, ( b ) T2 Sagittal MRI spine, ( c ) Sagittal MRI spine 58 R. Greenleaf and M.B. Harris Severely injured patients may arrive to your facility intubated or obtunded, making the history provided the best assessment of neurological status. Furthermore, the mechanism of injury is often associated with certain spinal injury patterns. Distraction injuries may follow rapid deceleration which occurs during motor vehi- cle accidents. HS fell from a height, which is associated with compression across the spinal elements and particularly the anterior weight-bearing column. Next, a careful neurologic assessment should be performed. The primary goal of acute spine care is preservation of neurological function. A progressive neurological de fi cit is the most widely accepted indication for urgent surgical intervention. Detecting such progression is only possible with a thorough initial assessment followed by clear and accurate documentation of subsequent neurological deterioration. De fi ning a progressive neurological injury is often best accomplished by having the same individual repeat serial neurological exams over a series of hours. The spinal cord terminates variably between T11 and L2 in adults as the conus medullaris (CM), which then forms the cauda equina (CE) and the spinal nerve roots. Trauma to these structures near the thoracolumbar junction may result in a variety of neurological injuries with a large range of severity and a combination of upper and lower motor neuron fi ndings. Radiculopathy due to injury to select nerve roots results in a dermatomal pattern of paresthesias with or without myotomal weakness and/or hyporefl exia. Damage to the bladder and bowel control centers in the CM affect bowel and/or bladder function. A more diffuse injury pattern in the lower extremities of paresthesias, weakness, and abnormal re fl exes may indi- cate injury to the CM or CE. The bulbocavernosus re fl ex should be checked to evaluate for spinal shock. Lastly, the presence of rectal tone and the ability to con- tract the external anal sphincter, which is under voluntary control, as well as perineal pinprick sensation should be assessed. These tests assess for sacral sparing. Sacral sparing demonstrates that some nervous function is traversing the zone of injury and improves neurologic prognosis. On presentation, HS complains of right lower extremity radicular pain. Focal extremity pain should be clinically evaluated with appropriate imaging. However, in the setting of radicular pain and a concomitant spinal injury, insult to local nerve roots at the site of injury should be suspected. Careful documentation of the nature and location of the pain as well as any associated motor, sensory, or re fl ex abnor- malities is imperative. The location and pattern of the pain should be correlated with radiographic images to ascertain which nerve roots are involved. The bladder should be evaluated for urinary retention by a bladder scan or uro- dynamic or cystometric testing. If the bladder volume is greater than 400–500 ml, a bladder re fl ex injury should be suspected and an indwelling Foley catheter should be placed. Dif fi culty urinating immediately after an acute spinal fracture may or may not portend a neural tissue injury. Pain from the injury, medications, and supine posi- tioning all make routine voiding more dif fi cult. Moreover, given the patient’s intact rectal tone with no other signs of cauda equina or conus medullaris syndrome makes an incomplete spinal cord or CE injury less likely. Otherwise, the patient has dem- onstrated normal motor strength and re fl exes in his lower extremities. 594 Lumbar Burst Fractures The patient’s radiographic imaging is consistent with a thoracolumbar burst-type fracture. It is important to note that the patient has the appearance of six lumbar vertebrae, or a lumbarized sacral segment. For our purposes, we will count down from cephalad to caudad starting at the fi rst level (L1) without a rib. The fracture is therefore at L2. Also, there is mild anterolisthesis of L5 on S1 which is likely chronic but should not be overlooked. The CT scan demonstrates a burst-type fracture. There may also be associated fractures into the posterior elements. On the AP fi lm, there is subtle widening of the interpedicular distance which is typical of burst fractures and often seen more clearly on coronal CT images. On the lateral view, there is approximately 50% loss of the L2 vertebral body height at its most compressed point, measured in relation to the average height of the adjacent vertebral bodies. There is clear retropulsion of fracture fragments compromising the patency of the spinal canal. An accurate assessment of the amount of canal compro- mise is not possible on plain radiographs. Lastly, the overall sagittal alignment of the thoracolumbar spine is easily assessed on plain lateral radiographs. The normal lumbar lordosis transitions into thoracic kyphosis from T10 through L2, resulting in an overall neutral alignment. The local kyphosis angle is useful to measure trau- matic kyphosis and consists of the Cobb angle using the superior endplate of the adjacent cephalad vertebral body and the inferior endplate of the caudal vertebral body. This patient has roughly neutral alignment. Angulation or abnormal kyphosis may also be manifested by splaying or widening of the spinous processes on the lateral view. HS does not have widening of the spinous processes on any of his images; however, abnormalities may not be as evident on postinjury supine fi lms as compared to images taken with the patient upright. On CT imaging, we can con fi rm multiple facts including: approximately 50% loss of body height, neutral sagittal alignment, and comminution of the entire verte- bral body on axial imaging. There is 30–40% canal compromise seen on the CT axial images secondary to the retropulsed fragments. Facet joint orientation appears grossly normal on both sagittal and axial images which suggest no translation or distraction at this level. The sagittal MRI images show extensive signal changes within the L2 vertebral body consistent with the fracture as well as boney retropulsion into the canal. The spinal cord terminates as the CM at the middle of the L1 body. There is no signal change in the CM. The integrity of the posterior ligaments, including the posterior longitudinal ligament (PLL), ligamentum fl avum (LF), interspinous ligaments, and supraspinous ligament (SL) all appear grossly intact. Declaration of Speci fi c Diagnosis HS is a 21-year-old male with an isolated L2 burst fracture with associated radicular pain from nerve root insult and possibly a neurogenic bladder. 60 R. Greenleaf and M.B. Harris Brainstorming: What Are the Treatment Goals and the Options? Treatment goals consist of the following objectives: 1. Prevent further neurological injury and optimize ability for neurological recovery. 2. Ensure long-term fracture stability, minimizing number of segments included in fusion if necessary. 3. Mobilize patient to avoid morbidity associated with immobilization. 4. Minimize both acute and chronic pain. Treatment options include the following: 1. Conservative/nonoperative treatment (a) Brace (b) No brace 2. Operative treatment (a) Posterior approach Long segment • Short segment • (b) Anterior approach Evaluation of the Literature A PubMed search was performed to fi nd relevant articles. Keywords included “thora- columbar spine” and subheadings “fracture” and “injury.” Articles were limited from 1975 to present. 905 article abstracts were reviewed and 110 articles were read. Classi fi cation of Thoracolumbar Spine Trauma There may be no pathology in Orthopedic Surgery that has been the focus of more classi fi cation systems than thoracolumbar spine fractures. Early systems fell out of favor for many reasons, including: lack of reproducibility or validity, arduous com- plexity, failing to account for neurological injury, and for being simply descriptions of the injury and not offering prognosis or treatment guidance [ 2– 5 ] . The Spine Trauma Study Group devised and published the Thoracolumbar Injury Classi fi cation and Severity Score (TLICS) in 2005 [ 4 ] . This system takes into account fracture morphology, neurologic status, and posterior ligament integrity. A severity score is produced, with a score of 3 points or less suggesting nonoperative treatment, 5 or more points suggesting that surgical intervention is indicated, and 4 points suggest- ing that operative or nonoperative treatment may be indicated. The system further 614 Lumbar Burst Fractures guides the surgical approach, anterior versus posterior, based on neurologic status and ligament integrity [ 4 ] . The TLICS has improved interobserver reliability, repro- ducibility, and validity compared to previous systems [ 3, 6, 7 ] . Joaquim et al. [ 6 ] retrospectively reviewed 49 cases of thoracolumbar spine injuries treated surgically. When compared to surgical decision making using the AO classi fi cation and neuro- logic status, the authors found that the TLICS also suggested surgery in 47 of the 49 patients. Two patients with burst fractures and no neurologic de fi cit or ligament injury were surgically treated in this study, but scored only two pointson the TLICS. They conclude that the TLICS can be used to classify thoracolumbar injuries and accurately predict surgical management. In patient HS, the TLICS score is 2 points for compression and burst fracture morphology, 0 points for intact ligaments, and 2 points for nerve root injury. This totals 4 points, which suggests that both operative and nonoperative treatment should each be considered. Conservative/Nonoperative Care Reports of successful nonoperative treatment of thoracolumbar burst fractures are well published in the spinal literature [ 8– 18 ] . Thomas et al. [ 15 ] performed a sys- tematic review of studies evaluating operative and nonoperative treatment for tho- racolumbar burst fractures. The authors conclude that there is a lack of evidence demonstrating the superiority of one approach over the other as measured using generic and disease-speci fi c health-related quality of life scales. They state that there is no scienti fi c evidence linking posttraumatic kyphosis to clinical outcomes and comment that there is a strong need for improved clinical research methodology to be applied to this patient population. However, this study includes only those studies involving spine fracture patients without neurologic de fi cits. The authors feel that radiculopathy such as that experienced in our patient does not represent a neurologic de fi cit; gross motor weakness or re fl ex abnormalities qualify as de fi cits. Nonoperative treatment of burst fractures with true neurologic de fi cits remains controversial. Dai et al. [ 19 ] published one of the few studies commenting speci fi cally on nonoperative outcomes in patients with burst fractures and neurologic de fi cits. The authors found that no patients experienced further neurologic deterioration, while 93% experienced complete neurologic recovery. The authors did fi nd a cor- relation between loss of kyphosis correction/ fi nal kyphosis angle and pain. Incorporation of neurologic injury into a treatment algorithm was fi rst attempted in the TLICS as described above. In a study of 41 patients with thoracolumbar burst fractures treated nonopera- tively and followed with serial CT scans, Mumford et al. [ 17 ] published several interesting fi ndings. These include: (1) on average, two-thirds of the retropulsed bone fragments causing canal-narrowing resorb; (2) delayed neurologic de fi cits rarely develop with nonoperative treatment (1/41 cases); (3) severity of initial or fi nal kyphotic deformity does not correlate with clinical outcomes, and; (4) nonop- erative treatment results in acceptable clinical outcomes. 62 R. Greenleaf and M.B. Harris If nonoperative treatment is chosen for thoracolumbar injuries, immobilization with either extension casting or a thoracolumbar-sacral orthosis (TLSO) is fre- quently used. The main theoretical bene fi ts of casting or bracing include pain con- trol and prevention of fracture collapse with progressive kyphosis and spinal canal narrowing. However, a growing body of literature questions the need for immobili- zation after thoracolumbar spine fractures [ 12, 20, 21 ] . Giele et al. [ 12 ] performed a systematic review of the best available literature comparing bracing with nonbrac- ing for any thoracolumbar fractures. There are no randomized trials and only seven retrospective studies, which satis fi ed inclusion criteria. The authors conclude that in the present literature there is no evidence for the effectiveness of bracing in patients with traumatic thoracolumbar fractures, but the lack of high-quality studies prevents relevant conclusions from being drawn. To address this lack of a well-designed study comparing bracing with nonbrac- ing, Bailey et al. [ 20 ] are currently conducting a randomized, multicenter trial in acute trauma patients with AO type A3 burst fractures between T11 and L3, such as in our patient. Exclusion criteria included neurologic de fi cit and initial kyphosis angle >35°. In a 1-year interim analysis, the authors have found no signi fi cant dif- ference in any outcome measure. They contend that a thoracolumbar burst fracture, in exclusion of an associated posterior ligamentous complex injury, is inherently a very stable injury and may not require a brace. Operative Treatment Consistent with treatment for all spine injuries, the most compelling indication for surgical treatment remains neurologic de fi cits. This is particularly true in the pres- ence of either persistent nervous tissue compression or a grossly unstable pattern of injury. The concept of instability in thoracolumbar trauma has been debated for decades. Historically, popular parameters suggesting instability include vertebral body height loss, canal compromise of retropulsed bone ³ 50%, and 30° or more of kyphosis [ 22, 23 ] . These parameters do not take into account neurologic status or ligamentous instability. Newer classi fi cation systems like the TLICS include both these factors given more recent widespread availability of MRI. The main goal of operative intervention is fi xation of unstable injuries which may prevent nervous tissue insult as well as early and late deformity. Suggested bene fi ts also include more rapid mobilization with avoidance of the morbidity asso- ciated with immobilization, pain relief, shorter hospital stay, earlier return to work, and less reliance on orthoses [ 11, 14, 24– 26 ] . Posterior Approach Posterior instrumentation is the most common fi xation technique for thoracolumbar spine fractures. The posterior approach and use of instrumentation to the thora- columbar spine are most familiar to spine surgeons. The surgeon relies on reduction 634 Lumbar Burst Fractures of the burst fracture via prone positioning in relative thoracolumbar extension and ligamentotaxis. Three general techniques of fi xation include long, short, and minimally invasive. For this discussion, short segment fi xation refers to pedicle instrumentation extending no more than one level above and below the fractured level while long segment fi xation is anything longer. In general, maximal mechanical stabilization of unstable thoracolumbar burst fractures, particularly with compromised bone qual- ity, requires instrumentation at least two levels above and below the level of injury. The biomechanical advantage of long fusions is countered by the negative effects of fusing multiple motion segments. The Load Sharing Classi fi cation (LSC) devised by Gaines et al. [ 27 ] in 1994 speci fi cally evaluates short segment fi xation in burst fractures. A point system based on the severity of fracture comminution and degree of correctable kyphosis predicts failure of short segment fi xation due to lack of anterior column support. The authors suggest that anterior column reconstruction is necessary if an injury totals greater than 6 points. An in vitro biomechanical study [ 28 ] validated the classi fi cation system. We cannot accurately determine the Load Sharing Classi fi cation score for H.S. without comparative lateral radiographs to determine sagittal alignment. Direct comparison between long and short fusions has been published. In 2005, Tezeren and Kuru [ 25 ] randomized 18 burst-fracture patients into long or short instrumentation and fusion with autologous bone graft. The patients with the long fusion had signi fi cantly improved radiographic parameters at fi nal follow-up (kyphosis and vertebral body height loss), but there was no clinical difference in the outcomes between the two groups. Short versus long fi xation was also compared with reference to the LSC in another retrospective study [ 29 ] . Altay et al. found that outcomes after fi xation of fractures scoring 7 or more points according to the LSC were signifi cantly better with long fi xation while fractures scoring 6 or less points did equally well with short or long fusion. The authors concluded that short segment fi xation is adequate alone if the LSC is 6 points or less. Placement of pedicle screws into the fractured segment during short fi xation has been described and reviewed in multiple articles. Wang et al. [ 30 ] concisely describes the technique in which a short pedicle screw is placed into the pedicle of the frac- tured vertebra, leaving the screw slightly prouder than the cephalad and caudal screws. During rod insertion, this segment is forced anteriorly to push the body and fracture fragments anteriorly, which in turn helps create local lordosis. Mahar [ 31 ] employed this technique in a cadaveric biomechanical study as well as a retrospec- tive postoperative chart review and found signi fi cantly improved biomechanical stiffness in the constructs with the pedicle screw at the fractured level. Alternate Methods of Fixation The authors do not have suf fi cient experience using minimally invasive instrumenta- tion techniques nor kyphoplasty for acute fracture care to advocate their use. However, due to their growing popularity, these techniques warrant mention [ 32– 34 ] . 64 R. Greenleaf and M.B. Harris Anterior Approach The anterior approach to the thoracolumbar spine typically requires an “approach” surgeon depending on the spine surgeon’s training. Anterior procedures generally include a corpectomy and thorough decompression of the canal along with place- ment of a graft or cage, and stabilization with spanning instrumentation. Many ante- rior techniques and constructs have been described with most giving satisfactory results [ 35– 41 ] . However, the indications for an anterior approach remain controversial. Anterior surgery is ideal in cases of incomplete neurologic de fi cit with persistent neural tissue compression or canal compromise, severe vertebral body comminution in conjunction with a kyphotic deformity, or a delay in fi xation greater than 4 days after injury. Parker et al. [ 42 ] retrospectively reviewed 46 patients with thoracolum- bar spine fractures treated surgically, with short segment posterior instrumentation and fusion, and with 3 months of bracing for an average of 5.5 years. The authors based their approach on the Load Sharing Classi fi cation. They conclude that inju- ries with a score of 7 or greater lacked enough anterior support for short posterior fi xation and required anterior reconstruction. Special consideration for anterior decompression and reconstruction should be given in patients with canal compromise and incomplete neurologic de fi cits. Bradford et al. [ 43 ] studied 59 patients with acute incomplete neurologic de fi cits in 59 patients after thoracolumbar trauma. In this study, 20 patients treated with ante- rior decompression and reconstruction were compared to 39 patients receiving pos- terior fi xation. Normal bowel and bladder function returned in 69% of anteriorly treated patients versus 33% of posteriorly treated patients, leading the authors to conclude that the boney stenosis in this group of patients should be addressed via an anterior approach. Literature Inconsistencies Clearly, the main controversy in this case is whether or not early operative stabiliza- tion of thoracolumbar burst fractures without evidence of posterior ligamentous injury leads to improved long-term outcomes. Early outcome measures may lend support to either argument: nonoperative treatment avoids the morbidity and complications asso- ciated with surgery, whereas surgery offers the bene fi ts of improved radiographic parameters and immediate post-op biomechanical stability. Well-designed prospective randomized trials do not exist, and existing studies are fl awed due to heterogeneity of treatment protocols, short follow-up, insuf fi cient power, or inappropriate outcomes measures. Further, even when operative fi xation is deemed necessary, the approach to fi xation remains controversial. Some surgeons feel that complete anterior decompres- sion with reconstruction offers the best risk–bene fi t pro fi le [ 44 ] , while others believe that given modern instrumentation and techniques, posterior stabilization alone is suf fi cient. The great heterogeneity of injury patterns and severity, patient characteris- tics, and surgeon preferences make direct and accurate comparison dif fi cult. 654 Lumbar Burst Fractures Another signi fi cant inconsistency is the association between clinical outcomes and pain with the appearance of radiographic kyphosis. While most studies suggest that the fi nal sagittal alignment does not correlate with pain, some reports do fi nd an association. Further, our increased awareness of how critical overall sagittal balance is in deformity surgery suggests that the long-term effects of posttraumatic kyphosis should not be ignored. Finally, use of an orthosis, typically a TLSO, for nonoperative treatment is still a common practice in many centers. Early reports by Bailey et al. suggest that bracing is not necessary, but appropriate long-term follow-up is needed. Evidentiary Table and Selection of Treatment Method The key studies guiding the treatment of HS are listed in Table 4.1 . Based on the available literature, the authors would choose to initially attempt nonoperative treat- ment with close observation. No brace would be initially applied; however, if the patient was too uncomfortable to mobilize secondary to pain, a brace would be pro- vided in an attempt to improve comfort and facilitate mobilization. We would obtain baseline standing thoracolumbar radiographs with particular attention to the degree of local kyphosis and vertebral body collapse. Indications for delayed operative intervention include progressively worsening kyphosis to greater than 35° with unre- lenting pain or development of any new neurologic de fi cits. If surgery was needed, two valid options would be considered. The chosen surgical approach in this case would be based on three parameters: (1) progressive kyphosis and pain, (2) align- ment correction with prone positioning, (3) intact bowel and bladder function. If HS demonstrates correctable alignment and no bowel or bladder dysfunction, he would be treated with a posterior approach. If posterior stabilization was performed, a short construct would likely do well because of the mild vertebral body comminution and fracture displacement along with this patient’s young age and good bone stock. However, the major determinant of approach options in HS would be the residual bladder dysfunction or perineal numbness, in which case the anterior approach would be advocated. This would directly decompress the neural elements, restore alignment more adequately than indirect posterior fi xation, and preserve more motion segments. Predicting Outcomes The authors anticipate that within 1–2 days of injury, H.S. will be mobilizing with assistive devices and will have normal bladder function. His right leg radicular pain may persist for 2–3 weeks but will gradually subside as the local irritation and in fl ammation from the trauma subsides. Narcotic pain medications will be necessary for mobilization for at least 4–6 weeks. We would predict that the kyphosis angle and vertebral body height loss will increase 5–10° and 10–15%, respectively, over the fi rst 4–6 weeks but then stabilize and heal uneventfully (Table 4.1 ). 66 R. Greenleaf and M.B. Harris References 1. Hsu JM, Joseph T, Ellis AM. Thoracolumbar fracture in blunt trauma patients: guidelines for diagnosis and imaging. Injury. 2003;34:426–33. 2. Patel AA, Dailey A, Brodke DS, et al. Thoracolumbar spine trauma classi fi cation:the Thoracolumbar Injury Classi fi cation and Severity Score system and case examples. J Neurosurg Spine. 2009;10:201–6. Table 4.1 Evidentiary table: a summary of the quality of evidence for our treatment of thoracolumbar burst fractures Author (year) Description Summary of results Quality of evidence Mumford (1993) Retrospective case series 2-year follow-up after nonoperative treatment of neurologically intact burst fracture: (1) 49% excellent, 17% good, 22% fair, 12% poor clinical outcomes; (2) average 8% progression of fracture collapse; (3) 2/3 of fragments in canal resorbed by 1 year on CT; (4) 1/41 patients suffered neuro decline during nonoperative treatment. Level IV Wood (2003) Prospective randomized trial 3½-year follow-up, 24 burst fractures treated operatively vs. 23 treated nonoperatively: no signi fi cant difference in fi nal kyphosis, canal narrowing, return to work, pain scores, SF-36, and Oswestry scores. Less disability reported in the nonoperative group. Level II Thomas (2006) Meta-analysis Literature search of all relevant studies concerning thoracolumbar burst fracture without neurological de fi cit: (1) lack of evidence demonstrating superiority of operative vs. nonoperative treatment using generic and disease speci fi c quality of life scales; (2) no evidence of correlation between posttraumatic kyphosis and clinical outcomes. Level III Dai (2008) Retrospective series 7.2-year follow-up, 127 burst fractures patients with and without neurological de fi cits. (1) No patients had neurological decline; (2) all pts with initial neurological de fi cit improved; (3) A correlation was found between fi nal kyphosis and Load Sharing Classi fi cation grade and pain. Level IV Bailey (2009) Prospective, randomized, multicenter equivalence trial Interim analysis of 69 patients randomized to TLSO or no TLSO. No signi fi cant difference in Roland-Morris Disability, pain, functional outcomes, sagittal alignment, hospital stay, and complications. Bracing for AO type A3 burst fractures offers no early bene fi t but fi nal long-term 2-year outcomes are yet to be determined. Level II 674 Lumbar Burst Fractures 3. Patel AA, Whang PG, Brodke DS, et al. Evaluation of two novel thoracolumbar trauma classi fi cation systems. Indian J Orthop. 2007;41:322–6. 4. Vaccaro AR, Lehmann RA, Hurlbert RJ, et al. A new classi fi cation of thoracolumbar injuries: the importance of injury morphology, the integrity of the posterior ligamentous complex, and neurologic status. Spine. 2005;30:2325–33. 5. Wood KB, Khanna G, Vaccaro AR, et al. Assessment of two thoracolumbar fracture classi fi cation systems as used by multiple surgeons. J Bone Joint Surg Am. 2005;87:1423–9. 6. Joaquim AF, Fernandes YB, Cavalcante RA, et al. Evaluation of the thoracolumbar injury classi fi cation system in thoracic and lumbar spinal trauma. Spine. 2011;36:33–6. 7. Whang PG, Vaccaro AR, Poelstra KA, et al. The in fl uence of fracture mechanism and morphology on the reliability and validity of two novel thoracolumbar injury classi fi cation systems. Spine. 2007;32:791–5. 8. Agus H, Kayali C, Arslantas M. Nonoperative treatment of burst-type thoracolumbar vertebra fractures: clinical and radiological results of 29 patients. Eur Spine J. 2005;14:536–40. 9. Cantor JB, Lebwohl NH, Garvey T, et al. Nonoperative management of stable thoracolumbar burst fractures with early ambulation and bracing. Spine. 1993;18:971–6. 10. Chow GH, Nelson BJ, Gebhard JS, et al. Functional outcome of thoracolumbar burst fractures managed with hyperextension casting or bracing and early mobilization. Spine. 1996;21(18): 2170–5. 11. Denis F, Armstrong GW, Searls K, et al. Acute thoracolumbar burst fractures in the absence of neurologic de fi cit. A comparison between operative and nonoperative treatment. Clin Orthop Relat Res. 1984;189:142–9. 12. Giele BM, Wiertsema SH, Beelen A, et al. No evidence for the effectiveness of bracing in patients with thoracolumbar fractures. Acta Orthop. 2009;80:226–32. 13. Mirza SK, Chapman JR, Anderson PA. Functional outcome of thoracolumbar burst fractures managed with hyperextension casting or bracing and early mobilization. Spine. 1997;22: 1421–2. 14. Shen WJ, Liu TJ, Shen YS. Nonoperative treatment versus posterior fi xation for thoracolum- bar junction burst fractures without neurologic de fi cit. Spine. 2001;26(9):1038–45. 15. Thomas KC, Bailey CS, Dvorak MF, et al. Comparison of operative and nonoperative treat- ment for thoracolumbar burst fractures in patients without neurological de fi cit: a systematic review. J Neurosurg Spine. 2006;4(5):351–8. 16. Wood K, Buttermann G, Mehbod A, et al. Operative compared with nonoperative treatment of a thoracolumbar burst fracture without neurological de fi cit. A prospective, randomized study. J Bone Joint Surg Am. 2003;85-A(5):773–81. 17. Mumford J, Weinstein JN, Spratt KF, Goel VK. Thoracolumbar burst fractures. The clinical ef fi cacy and outcome of nonoperative management. Spine. 1993;18(8):955–70. 18. Shen WJ, Shen YS. Nonsurgical treatment of three-column thoracolumbar junction burst frac- tures without neurologic de fi cit. Spine. 1999;24(4):412–5. 19. Dai LY, Jiang LS, Jiang SD. Conservative treatment of thoracolumbar burst fractures: a long- term follow-up results with special reference to the load sharing classi fi cation. Spine. 2008;33(23):2536–44. 20. Bailey CS, Dvorak MF, Thomas KC, et al. Comparison of thoracolumbosacral orthosis and no orthosis for the treatment of thoracolumbar burst fractures: interim analysis of a multicenter randomized clinical equivalence trial. J Neurosurg Spine. 2009;11(3):295–303. 21. Folman Y, Gepstein R. Late outcome of nonoperative management of thoracolumbar vertebral wedge fractures. J Orthop Trauma. 2003;17(3):190–2. 22. Vaccaro AR, Kim DH, Brodke DS, et al. Diagnosis and management of thoracolumbar spine fractures. Instr Course Lect. 2004;53:359–73. 23. McAfee PC, Yuan HA, Lasda NA. The unstable burst fracture. Spine. 1982;7(4):365–73. 24. Marco RA, Meyer BC, Kushwaha VP. Thoracolumbar burst fractures treated with posterior decompression and pedicle screw instrumentation supplemented with balloon-assisted verte- broplasty and calcium phosphate reconstruction Surgical technique. J Bone Joint Surg Am. 2010;92(1):67–76. 68 R. Greenleaf and M.B. Harris 25. Tezeren G, Kuru I. Posterior fi xation of thoracolumbar burst fracture: short-segment pedicle fi xation versus long-segment instrumentation. J Spinal Disord Tech. 2005;18(6):485–8. 26. Hartmann F, Gercek E, Leiner L, Rommens PM, et al. Kyphoplasty as an alternative treatment of traumatic thoracolumbar burst fractures Magerl type A3. Injury. 2010: 10. 27. McCormack T, Karaikovic E, Gaines RW. The load sharing classi fi cation of spine fractures. Spine. 1994;19(15):1741–4. 28. Wang XY, Dai LY, Xu HZ, Chi YL. The load-sharing classi fi cation of thoracolumbar fractures: an in vitro biomechanical validation. Spine. 2007;32(11):1214–9. 29. Altay M, Ozkurt B, Aktekin CN, et al. Treatment of unstable thoracolumbar junction burst fractures with short- or long-segment posterior fi xation in magerl type a fractures. Eur Spine J. 2007;16(8):1145–55. 30. Wang ST, Ma HL, Liu CL, et al. Is fusion necessary for surgically treated burst fractures of the thoracolumbar and lumbar spine?: a prospective, randomized study. Spine. 2006;31(23): 2646–53. 31. Mahar A, Kim C, Wedemeyer M, et al. Short-segment fi xation oflumbar burst fractures using pedicle fi xation at the level of the fracture. Spine. 2007;32(14):1503–7. 32. Wild MH, Glees M, Plieschnegger C, Wenda K. Five-year follow-up examination after purely minimally invasive posterior stabilization of thoracolumbar fractures: a comparison of mini- mally invasive percutaneously and conventionally open treated patients. Arch Orthop Trauma Surg. 2007;127(5):335–43. 33. Merom L, Raz N, Hamud C, et al. Minimally invasive burst fracture fi xation in the thora- columbar region. Orthopedics. 2009;32(4):273. 34. Fuentes S, Metellus P, Fondop J, et al. Percutaneous pedicle screw fi xation and kyphoplasty for management of thoracolumbar burst fractures. Neurochirurgie. 2007;53(4):272–6. 35. Wood KB, Bohn D, Mehbod A. Anterior versus posterior treatment of stable thoracolumbar burst fractures without neurologic de fi cit: a prospective, randomized study. J Spinal Disord Tech. 2005;18(Suppl):S15–23. 36. Bohlman HH, Kirkpatrick JS, Delamarter RB, Leventhal M. Anterior decompression for late pain and paralysis after fractures of the thoracolumbar spine. Clin Orthop Relat Res. 1994;300:24–9. 37. Crutcher Jr JP, Anderson PA, King HA, Montesano PX. Indirect spinal canal decompression in patients with thoracolumbar burst fractures treated by posterior distraction rods. J Spinal Disord. 1991;4(1):39–48. 38. Kaneda K, Taneichi H, Abumi K, et al. Anterior decompression and stabilization with the Kaneda device for thoracolumbar burst fractures associated with neurological de fi cits. J Bone Joint Surg Am. 1997;79(1):69–83. 39. McCullen G, Vaccaro AR, Gar fi n SR. Thoracic and lumbar trauma: rationale for selecting the appropriate fusion technique. Orthop Clin North Am. 1998;29(4):813–28. 40. Ren Z, Jing Z, Pei F. Anterior decompression and reconstruction with internal fi xation for severe thoracolumbar burst fracture. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi. 2006;20(4):397–9. 41. Sasso RC, Renkens K, Hanson D, et al. Unstable thoracolumbar burst fractures: anterior-only versus short-segment posterior fi xation. J Spinal Disord Tech. 2006;19(4):242–8. 42. Parker JW, Lane JR, Karaikovic EE, Gaines RW. Successful short-segment instrumentation and fusion for thoracolumbar spine fractures: a consecutive 41/2-year series. Spine. 2000;25(9):1157–70. 43. Bradford DS, McBride GG. Surgical management of thoracolumbar spine fractures with incomplete neurologic de fi cits. Clin Orthop Relat Res. 1987;(218):201–16. 44. Whang PG, Vaccaro AR. Thoracolumbar fractures: anterior decompression and interbody fusion. J Am Acad Orthop Surg. 2008;16(7):424–31. Part III Upper Extremity Trauma Section editor—Michael D. McKee 71M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_5, © Springer Science+Business Media New York 2013 Keywords Nonoperative management of scapula fracture Surgical treatment of scapula fracture Surgical approach for scapula Judet incision WT Is a 19-Year-Old Male with Shoulder Pain Case Presentation WT is a 19-year-old male who presents to the emergency department via EMS complaining of severe shoulder pain after a high-speed motorcycle accident in which he was the helmeted driver. He denies any loss of consciousness. On primary survey, he demonstrates a GCS (Glascow Coma Score) of 15, a patent airway, and is hemodynamically stable. On secondary survey, he demonstrates severe pain with passive range of motion of the right shoulder, but his exam is otherwise negative. On physical examination, the patient demonstrates a strong radial pulse in the right upper extremity and 5/5 strength throughout the right arm. He has full painless range of motion about the right wrist and elbow. On examination of his shoulder, he demonstrates ecchymosis over the posterior aspect of the scapula and severe tender- ness to palpation. He is not able to tolerate passive range of motion of the shoulder. Radiographs of W.T.’s shoulder reveal a right scapula neck and comminuted body fracture with multiple ipsilateral rib fractures (Fig. 5.1 ). Transcapular Y radio- graph (Fig. 5.2 ) and computed tomography imaging with 3D reconstruction con fi rms P. A. Cole , M.D. (*) B. W. Hill , M.D. Department of Orthopedic Surgery , Regions Hospital, University of Minnesota , Mail Stop 11503L, 640 Jackson Street , St. Paul , MN 55101 , USA e-mail: peter.a.cole@heathpartners.com Chapter 5 Scapula Fractures Peter A. Cole and Brian W. Hill 72 P.A. Cole and B.W. Hill Fig. 5.1 ( a ) AP and ( b ) scapula Y radiograph demonstrate a scapular neck fracture with marked angulation of the scapular body Fig. 5.2 The angulation and “z” type deformity can be appreciated on the scapula Y radiograph 735 Scapula Fractures this diagnosis and depicts, a “z” type deformity of the body of the scapula visualized by the lateral border landmarks of the scapula (Figs. 5.3 and 5.4 ). Interpretation of Clinical Presentation The patient’s radiographic fi ndings in addition to symptoms and physical examina- tion are consistent with a displaced right scapula neck and comminuted body fracture of the scapula which is distinguished by a segmental fracture of the lateral border, and multiple ipsilateral rib fractures. The patient’s presentation is typical of many who present with a severe scapula fracture. Fig. 5.4 ( a ) 3D CT reconstruction of the scapula. ( b ) The 3D CT reconstruction is useful to aid in assessment of the glenopolar angle (GPA) and fracture pattern [ 11 ]. (b) The glenopolar angle (GPA) measured 20 degrees Fig. 5.3 ( a ) Axial and ( b ) coronal CT scan of the scapula fracture clarify the pattern of injury 74 P.A. Cole and B.W. Hill The majority of scapula fractures are the result of high energy trauma such as motor vehicle accidents, which account for 52–70% of such injuries [ 1, 2 ] . Though scapula fractures represent only about 1% of all fractures [ 3 ] , they occur with about the same frequency as distal femur and calcaneus fractures according to recent epidemiologic studies. Isolated scapula fractures are uncommon because it takes great energy to fracture the scapula. Concomitant injuries are estimated to occur in 61–90% of scapula fractures [ 4, 5 ] . In addition, up to 53% of scapula fractures are associated with hemo- or pneumothorax [ 6 ] , and 50% are associated with ipsilateral extremity fractures [ 4 ] . Our patient W.T. is fortunate because another 15% of scapula fracture patients sustain signi fi cant head injury [ 4 ] . It is imperative that these poten- tially life threatening injuries be ruled out when a scapula fracture is detected in the emergency room during initial radiographic studies. A secondary survey, repeated the next day will be helpful in detecting initially missed injuries. Multiple level rib fractures in W.T.’s case underscore the severity of this trauma (Fig. 5.5 ). The patient history should identify a few key items. The fi rst is a review of symptoms to detect other injuries. Concomitant ipsilateral neurovascular injuries are common and demand an appropriate physical assessment of the brachial plexus and distal extremity perfusion. It is also important to understand the patient’s baseline function based on recreation and occupational activities. Since highly displaced extra-articular scapula fractures can manifest with long-term symptoms and functional de fi cits, the patient should be aware of possible limitations.Overhead loss of motion and muscular fatigue may not be an issue for a 50-year- old sedentary factory inspector, but it may be for the homemaker whose main diversion is tennis at the country club several times a week. The drooped shortened shoulder may bother a body-image-conscious person. The patient presenting with a scapula fracture should always be disrobed and undergo a thorough secondary survey. A detailed neurological exam, assessment of Fig. 5.5 ( a ) AP radiograph of the left shoulder illustrates the severity of the injury and ( b ) concomitant rib fractures ( green arrows ) 755 Scapula Fractures pulses in the extremities, and a skin check should be performed. Skin abrasions, which occur over the prominent shoulder elements, should also be noted. If surgery is chosen, there should be a delay until re-epithelialization occurs to decrease the risk of infection. Our protocol is a simple soap and washcloth cleansing 2–3 times/ day until resolved. Imaging of a scapula fracture should include three X-ray views: an anteroposte- rior scapula view (Grashey View), an axillary view, and a scapula Y view of the shoulder. Due to the fact that scapula fractures are most often the result of high energy trauma, 10% are missed or delayed in the primary diagnostic survey [ 7, 8 ] . If on plain radiographs, there is a displaced fracture which could warrant surgery, then a CT scan should be obtained. Of course the screening spiral trauma CT scan should be reviewed to assess for all fractures and will detect displaced injuries occasionally missed on screening chest X-ray or shoulder fi lms. This CT modality has been shown to be a much better test, with increased sensitivity for detection [ 9 ] . A formal 3D CT reconstruction of the shoulder (or scapula) should be obtained to make accurate measurements of displacement [ 10, 11 ] . In addition to the imaging required, in cases where there is a delay of greater than 2 weeks, an EMG-Nerve Conduction Study is recommended to detail any brachial plexus, axillary and suprascapular nerve lesions which occur with many displaced scapula fractures [ 12 ] . Declaration of Speci fi c Diagnosis W.T. is a 19-year-old male who presents with a high energy injury that consists of a right displaced scapula neck (Ada & Miller IIC [ 13 ] ) and comminuted body frac- ture, with concomitant multilevel ipsilateral rib fractures. Brainstorming: What Are the Treatment Options? Treatment Goals are the same as for any other fracture: 1. Restore length, alignment, and rotation 2. Render stability to allow for rehab 3. Promote maximal function 4. Minimize risk and complications Treatment Options Nonoperative 1. Sling and physical therapy when pain subsides 2. Benign neglect 76 P.A. Cole and B.W. Hill Surgical 1. Several posterior surgical approach options A. Judet incision I. With elevation of the entire muscular fl ap (deltoid, infraspinatus, teres minor) II. With elevation of a subcutaneous fl ap III. With or without elevation of deltoid B. Intermuscular Interval Exposure I. Between teres minor and infraspinatus 2. Fracture Reduction 3. Plate and screw fi xation Evaluation of the Literature A PubMed search was conducted. Keywords included: “scapula fracture” and sub- headings: “surgical treatment” and “conservative treatment,” with limits from 1975 until present. 704 abstract entries were reviewed, and from this list, 72 articles were selected. Review of Pertinent Articles There are limitations of the existing literature on scapula fractures. Classi fi cation systems used are varied, and there is no consensus as to the optimal system. The AO/ OTA has recently modi fi ed its classi fi cation of scapula fractures [ 14 ] , but few publi- cations with outcomes exist using this classi fi cation scheme. Further, the OTA sys- tem was developed without clinical or radiographic observations and is fl awed by not including combination injuries, or all fracture patterns which occur in trauma. Few studies can be compared because of the differences in the descriptive terms dictating the management of the scapula fracture. Authors commonly use descriptors such as displaced, medialized, angulated, and/or shortened; yet these terms have not been de fi ned or validated until very recently [ 10 ] . Most literature relies on the Ideberg or modi fi ed Ideberg classi fi cation [ 15, 16 ] for intra-articular fractures as well as the Ada–Miller [ 13 ] classi fi cation system for extra-articular fractures. Recently, the importance of computed tomography and the role of 3D reconstruction have aided orthopedists in the understanding of scapula fracture patterns [ 11 ] . To date, no ran- domized prospective trials have been published on clinical outcomes of scapula frac- tures. Furthermore, no studies clearly stratifying clinical outcome as a function of the amount of angulation and displacement have been performed. Rather, there are 775 Scapula Fractures multiple retrospective studies and prospective clinical series of operative and nonoperative treatment that describe good outcomes, though clearly subsets of patients in many nonoperative series report some poor and fair clinical results. This data dearth leaves a lack of clear evidence for operative indications. Due to the lack of published evidence, the decision to operate is often based on the surgeon’s train- ing and knowledge of surgical approaches. Such surgeons generally apply the ratio- nale which they apply to all other fractures: restoration of length, alignment, rotation and stability. Below is a discussion of the most pertinent literature. Conservative/Nonoperative Treatment The French and others have chronicled the sequence of the diagnosis and surgical management of scapula fractures. Beginning with the fi rst recorded operation in 1913 by Albin Lambotte, contemporary treatment approaches to extra-articular scapula fractures have largely been characterized by benign neglect [ 17 ] . In 1984, Armstrong and Van der Spuy documented 62 patients with scapula fractures that were treated nonoperatively [ 18 ] . Fractures were strati fi ed simply by fracture loca- tion with range of motion being the main outcome assessed. They found good results could be obtained with fractures of the body and spine, while neck and glenoid frac- tures should be considered for surgery [ 18 ] . Bozkurt et al. countered that patients with scapula neck fractures could indeed be treated conservatively and that the gle- nopolar angle (GPA) less than 20 degrees was more indicative of prognosis rather than fracture type [ 19 ] . In 1993, Goss theorized that if 2 or more “breaks” in the superior shoulder suspensory complex (SSSC) occur, it would create an unstable shoulder girdle as a result of the discontinuity between the axial and appendicular skeleton [ 20 ] . The theory has been challenged by some authors [ 21, 22 ] in that not all double disruptions are unstable and thus do not warrant operative intervention. Outcomes of minimal to moderately displaced extra-articular fractures managed nonoperatively are good. Gosens et al. [ 23 ] described the outcomes of 22 patients that sustained scapula body fractures and were managed nonoperatively. Disabilities of the Arm, Shoulder and Hand (DASH) score, the Simple Shoulder Test (SST) score, and the range-of-motion were the outcomes assessed. There was no effort in this study to measure the amount of displacement; nor was there any exclusion criterion for nondisplaced fractures. The mean DASH score of the 22 patients was 17.5, compared with a published normative value forthe general population of 10 points. This marginal increase is less than what is considered clinically signi fi cant. They further strati fi ed the results to those with isolated scapula fractures and those with multiple injuries ( n = 8). The mean DASH in patients with multiple injuries was 34.9, and they noted a signi fi cant decrease in range of motion compared to the uninjured arm. In another large recent retrospective study, Schofer et al. assessed functional outcomes of 51 scapula fractures treated nonoperatively [ 24 ] . Seventy- three percent of the fractures were isolated to the body. Again, the degrees of dis- placement were not assessed, nor were nondisplaced fractures excluded. In addition, the follow-up rate of this study was 37%. The authors reported that 84% of the 78 P.A. Cole and B.W. Hill patients were rated as having good outcomes based on the Constant Score. On follow-up examination, they found restricted range of movement in all directions, lower peak torque values, and lower mean power output in all planes of movement during isokinetic testing. Their conclusion was that scapula fractures heal with a good functional result despite measurable restrictions [ 24 ] . Dimitroulias et al. [ 25 ] published a study early in 2011 using a prospective database of 32 patients treated conservatively for scapula fractures. Only fractures with substantial displacement, de fi ned as a fracture with at least 100% and/or 1 cm displacement, were included in the study. The main outcome was the DASH questionnaire and no objective measurements such as strength and range of motion were assessed. They reported that the mean increase (i.e., more disability) in DASH from preinjury to fi nal follow-up was 10.2. Noted in this study were the follow-up (65%) and modest numbers of associated injuries (65%) as compared with other scapula fracture literature. The authors did note that a high ISS and presence of rib fractures are associated with a less favorable outcome. While this study does not refute the merits of surgical fi xation for severely displaced scapula fractures, it does serve as a testament to the scapula’s ability to compensate for displacement and help delineate optimal management of scapula fractures. Surgical Treatment As previously stated, minimally displaced extra-articular fractures are best managed nonoperatively. Displaced fractures, however, can disrupt the shoulder’s normal mechanisms of stability and motion. Scapula neck malunions have been shown in biomechanical models to shorten the length of rotator cuff muscles, which results in a loss of predicted force as well as altered normal muscle activation during abduction [ 26 ] . Furthermore, the compressive force against the glenoid changes to shear forces as a function of the aberrancy of vector [ 13 ] . Possibly the greatest support for surgical treatment of displaced scapula frac- tures has come from suboptimal results of nonoperative treatment. Ada and Miller noted that when they followed 24 patients, managed nonoperatively with displaced, intra-articular, or comminuted scapula spine fractures, signi fi cant disability was found in patients with displaced scapula spine and neck fractures: (1) pain at rest in 50–l00%, (2) weakness with exertion in 40–60%, and (3) pain with exertion in 20–66% [ 13 ] . Nordqvist et al. reported that 32% (7/22) of their scapula neck frac- tures had poor or fair outcomes with 48% (23/48) percent of all their fractures stud- ied having radiographic deformity [ 27 ] . Multiple authors have suggested operative criteria for scapula fractures based upon personal experience and case series outcome reviews [ 13, 27– 29 ] . Operatively treated extra-articular fracture surgery indications are limited to four reports which de fi ne radiologic operative criteria for displacement and angulation [ 13, 30– 32 ] . Hardegger et al. described their operative results for displaced scapula fractures after an average follow-up of 6.5 years [ 28 ] . The authors achieved 79% good or 795 Scapula Fractures excellent results in a series of 37 patients treated operatively, although only fi ve cases were included that were “severely displaced or unstable” scapula neck frac- tures [ 28 ] . Bauer et al. followed 20 patients who were treated with open reduction internal fi xation (ORIF) of their scapula fracture [ 2 ] . They used similar “descrip- tive” indications recommending ORIF for patients that had grossly displaced acro- mion or coracoid process fractures, displaced fractures of the anatomical neck, unstable fractures of the surgical neck, and displaced fractures of the glenoid [ 2 ] . Several recent key studies collectively suggest a range of indications for surgery [ 29– 34 ] : (1) ³4 mm step-off of an articular glenoid fracture, (2) ³20–25 mm dis- placement of the glenohumeral joint, (3) ³25–45° of angular deformity in the semi- coronal plane as seen in the scapula Y view, (4) shortening of ³25 mm, (5) ³10 mm double lesions of the superior shoulder suspensory complex (SSSC), or (6) GPA £ 20°–22°. Thus far, the published outcomes after operative treatment seem promising. Most of the data available are derived from smaller retrospective studies, but two system- atic reviews offer further support as to the safety of scapula surgery [ 5, 35 ] . Zlowodski et al.’s review included 22 studies with 520 scapula fractures [ 35 ]. Of those operatively treated ( n = 140), there was a reported infection rate of 3.5% and secondary surgical procedures in 23.5% of the cases reported (8 manipulations under anesthesia, 7 hardware removals, 3 irrigation and debridements, 2 hematoma evacuations, 2 revision fi xations, and 1 arthrodesis of the glenohumeral joint). Lantry et al. performed a systematic review of 17 studies which included 243 opera- tively treated scapula fractures [ 5 ]. They reported infection as the most common complication (4.2%); however, only one case required repeat surgery for resolution. Lantry et al. noted that good to excellent results (using a variety of outcomes) were obtained in approximately 85% of the cases. A 2.4% rate of nerve injury was also noted, however the authors acknowledged that it was unclear whether these were related to the initial trauma or surgical intervention. With an operative inclusion criteria of 100% translation, 30° angular deformity, or scapula fragment penetration through the thoracic wall, Bartincek et al. reported a Constant Score of greater than 90 of 100 points in 19/20 scapula fractures treated surgically [ 36 ] . Because of the high incidence of other injuries, scapula fracture treatment is often delayed. In a retrospective study of 22 patients with an average delay of 30 days before surgery, Herrera et al. reported favorable outcomes following surgical intervention [ 31 ] . Extra-articular fractures were included in the presence of 15 mm displacement, 25° angular deformity, or double disruptions of the SSSC with 10 mm displacement. Functional outcomes were attained on 14 patients with a mean DASH score of 14; in addition, the injured shoulder had 93% of the range of motion and 76% of strength of the uninjured shoulder. From this study, it was noted that the surgery becomes technically more dif fi cult with greater time from injury, but good clinical results could still be achieved. Surgical management of symptomatic scapula malunions has also been pub- lished [ 37 ] , perhaps providing the most compelling evidence to date regarding deformity and function. Cole et al. reported on a cohort of patients that were ini- tially treated nonoperatively and presented with complaints of debilitating pain and 80 P.A. Cole and B.W. Hill weakness. Pre- and post-reconstructive range of motion, strength,and function were all documented and demonstrated signi fi cant improvements. Mean follow-up was 39 months (range, 18–101 months), and all fi ve patients were pain free. The study cohort also experienced a mean improvement in DASH scores of 29 points [ 37 ] . However, while it is clear that a minority of patients may experience symptoms fol- lowing scapula malunion, the incidence of this problem is unknown. As operative management becomes more common, surgical approaches have been rede fi ned as well. Classically, the most common approach is the posterior (Judet) approach, which utilizes a dissection of the infraspinatus from the infraspi- natus fossa [ 38 ] . Obremskey et al. further modi fi ed this approach to limit muscle dissection to an interval between the infraspinatus and teres minor: to access the lateral border, but still allow for fi xation of scapula body or neck fractures [ 38 ] . At least one study has reported promising early results using this approach [ 30 ] . Jones et al. reported on 37 scapula fractures surgically treated with limited windows. The authors reported no infections, hematomas, or dehiscences (incision or muscle). In addition, no complaints of postoperative fl ap numbness or hypesthesia were noted. Based on preoperative planning, the surgeon’s approach depends on the desire for limited or complete exposure to the posterior scapula. Limited inter-muscular windows are favored to spare soft tissue elevation. These are access windows to address fracture displacement at the lateral border, acromial spine, and vertebral border. The extensile approach is used to expose the entire infraspinatus fossa by elevating the infraspinatus and teres minor with an elevator. This muscular fl ap can be elevated laterally as far as the suprascapular artery and nerve allow without excessive retraction. The posterior glenoid rim, lateral border, neck, spine and ver- tebral border are all exposed in this approach. Note that the entire subscapularis- muscular sleeve on the anterior surface of the scapula is preserved, maintaining blood supply to scapula body. In this context, the approach is biologically appro- priate as long as the neurovascular pedicle is respected―a claim substantiated by the presence of only one case of nonunion after ORIF reported in the literature. It is recommended to utilize an extensile approach for fractures over 2 weeks old or for complex body and neck fracture patterns. The exposure allows the surgeon total control of the fracture at multiple simultaneous sites to effect the reduction and take down of intervening callus. It will not allow for exposure of the articular glenoid because the fl ap cannot be retracted suf fi ciently to expose the glenoid adequately. Literature Inconsistencies The literature on scapula fractures and its management is limited to small retrospec- tive cohort studies. Multiple publications exist with reported surgical indications; however, none of these studies have compared operative versus nonoperative treat- ment. In addition, no level I or II studies are available to support surgical treatment indications for scapula fractures. Current information is based on retrospective 815 Scapula Fractures cohorts and systematic reviews. There is a need for prospective randomized control trials, or more realistically prospective prognostic studies to further aid in treatment decisions. Evidentary Table and Selection of Treatment Method The key studies to support particular treatment options for our patient W.T. are sum- marized in Table 5.1 . From the current literature, it is the authors’ opinion that W.T. would bene fi t from operative treatment. The patient meets several of the surgical indi- cations, and early fi xation would allow for early and more aggressive rehabilitation. De fi nitive Treatment Plan When addressing a patient with a scapula fracture, the surgeon must look at the radiographic imaging, but also devote attention to the patient’s age, associated injuries, and occupation, and take the physical activity level into account. A strong argument can be made that operative intervention would result in improved outcome for W.T. The general indications for surgical treatment of scapula fractures includes at least one of the following: (a) Intra-articular gap/step off ³3–10 mm, (b) displacement of the glenoid to the lateral border ³ 10–25 mm, (c) glenopolar angle £20–22°, (d) angulation ³ 30–45° on scapula Y radiograph [ 10, 13, 19, 22, 30, 31 ] . From the case history, W.T. meets several of these indications, using even the most conservative measurements. There are two exit points on the lateral border, which can make the true degree of displacement dif fi cult to appreciate. By “reducing” the segmental lateral border fragment, as shown in Fig. 5.6 , displacement is measured at 25 mm. The glenopolar angle of WT equaled 20°. As described in the literature, patients with a GPA of <20° suffered from severe glenoid rotational malalignment and had worse long-term outcomes when treated conservatively [ 19, 22, 29 ] . Compounding this with the knowledge that W.T. is young and will certainly demand overhead function post-injury, he is a good candidate for open reduction internal fi xation of the scapula. Postoperatively, because the scapula is now surgically stabilized, the physician must direct the effort at regaining motion. This should be the primary goal for the fi rst 4 weeks. Immediate active and passive range of motion is instituted. It is the authors’ preference to use an indwelling interscalene catheter for the fi rst 48–72 h to promote early gains in motion. Also, the use of pulleys and push–pull sticks used in the opposite extremity are a great resource and aid in rehabilitating upper extremity range of motion. Strength training is normally begun at 4 weeks, beginning with light weights (3–5 lbs). Follow-up for the surgically treated patient should be at 2, 6, and 12 weeks with routine radiographic imaging (AP, Scapula Y, Axillary). At 6 months and beyond, standard AP radiographs should be suf fi cient, and likely are not even necessary unless untoward events mandate radiographs. 82 P.A. Cole and B.W. Hill Ta bl e 5. 1 Ev id en tia ry ta bl e: a su m m ar y o f t he q ua lit y of ev id en ce fo r n on op er at iv e v er su s o pe ra tiv e fi x at io n af te r s ca pu la fr ac tu re A ut ho r ( ye ar) D es cr ip tio n Le v el o f ev id en ce N o. o f pa tie nt s N o. o f pa tie nt s fo llo w ed % P at ie nt s fo llo w ed Le ng th o f F U m ea n (ra ng e) (m os ) Su m m ar y No no pe ra tiv e A rm str on g & Va n d er S pu y (19 84 ) R et ro sp ec tiv e IV 62 52 83 .9 % Le ss fa v o ra bl e re su lts w ith fr ac tu re s o f n ec k an d gl en oi d. In y ou ng a nd fi t p at ie nt s O RI F m ay b e in di ca te d in th es e fra ct ur es . A da & M ill er (1 99 1) R et ro sp ec tiv e IV 11 3 24 21 .2 % > 15 In di ca tio ns fo r s u rg ic al m an ag em en t s ho ul d in cl ud e so m e sc ap ul a ne ck a nd sp in e f ra ct ur es . Edw ar ds e t a l. (20 00 ) R et ro sp ec tiv e IV 36 20 55 .6 % 28 (9 –7 9) Fl oa tin g sh ou ld er in jur ies ar e n ot as un sta ble as pr ev io us ly th ou gh t a nd m ay h av e go od fu nc tio na l o ut co m es w ith n on op er at iv e tr ea tm en t o f m in im al ly d isp la ce d in jur ies . B oz ku rt et a l. (20 05 ) R et ro sp ec tiv e IV 18 0. 0% 25 D ec re as ed G PA £ 20 ° m ay b e m or e re lia bl e th an fra ct ur e ty pe fo r d ys fu nc tio n fo llo w in g sc ap ul a fra ct ur es . v an N oo rt et a l. (20 05 ) R et ro sp ec tiv e IV 24 13 54 .2 % 66 (1 9.2 –1 44 ) N on op er at iv e tr ea tm en t o f s ca pu la n ec k fra ct ur es in th e ab se nc e of n eu ro lo gi c di sa bi lit y an d ip sil at er al sh ou ld er in jur y c an yi eld go od / ex ce lle nt fu nc tio na l o ut co m es . G os en s e t a l. (20 09 ) R et ro sp ec tiv e IV 26 22 84 .6 % 63 (4 1– 85 ) In sc ap ul a fra ct ur es w ith a ss oc ia te d in jur ies fu nc tio na l o ut co m e sc or es a re p oo re r w ith co n se rv at iv e tr ea tm en t v er su s iso la te d sc ap ul a fra ct ur es tr ea te d no no pe ra tiv el y. Sc ho fe r e t a l. (20 09 ) R et ro sp ec tiv e IV 13 7 51 37 .2 % 65 (1 3– 12 0) Pa tie nt s m an ag ed c o n se rv at iv el y af te r s ca pu la fra ct ur es su ffe r f ro m si gn i fi ca tio n lim ita tio ns in RO M . D im itr ou lia s e t a l. (20 11 ) Pr os pe ct iv e IV 49 32 65 .3 % 15 (6 –3 3) N on op er at iv e tre at m en t m ay b e s ati sfa ct or y, alt ho ug h in cr ea se d IS S, an d th e p re se nc e o f r ib fr ac tu re s ad ve rs ely af fe cts th e c lin ica l o ut co m e. 835 Scapula Fractures O pe ra tiv e H ar de gg er e t a l. (19 84 ) R et ro sp ec tiv e IV 37 33 89 .2 % 78 (1 8– 18 0) G le no id fr ac tu re -d isl oc at io ns , u n st ab le fr ac tu re s o f th e sc ap ul a ne ck , a nd d isp la ce d ap op hy se al fra ct ur es m ay n ee d an at om ic al re po sit io n if la te di sa bi lit y is to b e av o id ed . B au er e t a l. (19 95 ) R et ro sp ec tiv e IV 25 20 80 .0 % 73 .2 (1 2– 13 2) Ea rly o pe ra tiv e tr ea tm en t a nd u nd er sta nd in g of th e pa th op hy sio lo gy o f t he p ol yt ra um a p ati en t i s cr iti ca l f or d isp la ce d sc ap ul a fra ct ur e m an ag em en t. H er re ra e t a l. (20 09 ) Pr os pe ct iv e re gi str y IV 22 16 72 .7 % 26 .4 (1 2– 72 ) M alu ni on o f t he sc ap ul a c an b e p re v en te d by su rg ica l tre at m en t i n pa tie nt s w ith d ela ye d pr es en tat io n. Jo ne s e t a l. (20 09 ) R et ro sp ec tiv e IV 37 37 10 0. 0% > 12 M od i fi ed Ju de t a pp ro ac h al lo w s fo r e x ce lle nt sc ap ul a or g le no id fr ac tu re v isu al iz at io n w hi le pr es er vi ng ro ta to r c uf f f un ct io n. B ar to ni ce k & F ric (20 11 ) R et ro sp ec tiv e IV 22 22 10 0. 0% 26 (1 2– 48 ) St ab le fi xa tio n o f t he la te ra l b oa rd er is k ey to re st or e an at om ic al re la tio ns hi p of th e sc ap ul a an d ob ta in g oo d re su lts . Co le e t a l. (20 11 ) Pr os pe ct iv e IV 5 5 10 0. 0% 39 (1 8– 10 1) Co rre ct iv e re co n st ru ct io n fo llo w in g sc ap ul a m al un io n ca n de cr ea se sy m pt om s a nd im pr ov e sh ou ld er fu nc tio n. 84 P.A. Cole and B.W. Hill Predicting Long-Term Outcomes Prior literature has suggested the bene fi ts of operative treatment. Ada and Miller recommended ORIF for displaced scapula fractures after following eight patients treated operatively, with all reporting good clinical results and no residual pain at an average of 15 months post surgery [ 13 ] . Fifty percent of this same cohort with comminuted scapula spine fractures treated nonoperatively had residual pain. More recent publications have continued to show promising results. In a systematic review of 163 cases, 83% of the patients achieved excellent or good results after operative treatment of scapula fractures [ 5 ] . Outcomes of the analysis showed 70% of patients can expect good pain relief and 143° of abduction at an average of 53 months follow-up. The most common complication in this analysis was infec- tion (4.2%) followed by nerve injuries (2.4%). The authors noted though that it was dif fi cult to distinguish iatrogenic nerve injuries versus injuries accumulated at the time of trauma. Hardware removal, including clavicle plates, occurred in 7.1% of the patients due to local discomfort or mechanical failure [ 5 ] . Limitations on the time from injury to surgery continue to increase with satisfactory results; however, the surgery becomes more technically demanding [ 37 ] . Prospective studies with longer-term follow-up after operative management may be lacking, but the studies reviewed lead us to believe that W.T. will bene fi t from operative treatment and have a good result. Fig. 5.6 3D CT reconstruction of the scapula with an illustration demonstrates alignment of the lateral border and the true lateral border offset 855 Scapula Fractures References 1. McGahan JP, Rab GT, Dublin A. Fractures of the scapula. J Trauma. 1980;20(10):880–3. 2. Bauer G, Fleischmann W, Dussler E. Displaced scapular fractures: indication and long-term results of open reduction and internal fi xation. ArchOrthop Trauma Surg. 1995;114(4):215–9. 3. Court-Brown CM, Aitken SA, Forward DR, O’Toole RV. The epidemiology of fractures. In: Bucholz RW, editor. Fractures in adults. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2009. 4. Baldwin KD, Ohman-Strickland P, Mehta S, Hume E. Scapula fractures: a marker for concomitant injury? A retrospective review of data in the national trauma database. J Trauma. 2008;65(2):430–5. 5. Lantry JM, Roberts CS, Giannoudis PV. Operative treatment of scapular fractures: a system- atic review. Injury. 2008;39(3):271–83. 6. McLennan JG, Ungersma J. Pneumothorax complicating fracture of the scapula. J Bone Joint Surg Am. 1982;64(4):598–9. 7. Tadros AM, Lunsjo K, Czechowski J, Abu-Zidan FM. Causes of delayed diagnosis of scapular fractures. Injury. 2008;39(3):314–8. 8. Harris RD, Harris Jr JH. The prevalence and signi fi cance of missed scapular fractures in blunt chest trauma. AJR Am J Roentgenol. 1988;151(4):747–50. 9. Kwon YW, Powell KA, Yum JK, Brems JJ, Iannotti JP. Use of three-dimensional computed tomography for the analysis of the glenoid anatomy. J Shoulder Elbow Surg. 2005;14(1): 85–90. 10. Anavian J, Con fl itti JM, Khanna G, Guthrie ST, Cole PA. A reliable radiographic measurement technique for extra-articular scapular fractures. Clin Orthop Relat Res. 2011;469(12): 3371–8. 11. Armitage BM, Wijdicks CA, Tarkin IS, et al. Mapping of scapular fractures with three-dimensional computed tomography. J Bone Joint Surg Am. 2009;91(9):2222–8. 12. Robinson LR. Role of neurophysiologic evaluation in diagnosis. J Am Acad Orthop Surg. 2000;8(3):190–9. 13. Ada JR, Miller ME. Scapular fractures. Analysis of 113 cases. Clin Orthop Relat Res. 1991;(269):174–80. 14. Marsh JL, Slongo TF, Agel J, et al. Fracture and dislocation classi fi cation compendium – 2007: Orthopaedic trauma association classi fi cation, database and outcomes committee. J Orthop Trauma. 2007;21(10 Suppl):S1–133. 15. Ideberg R, Grevsten S, Larsson S. Epidemiology of scapular fractures. Incidence and classi fi cation of 338 fractures. Acta Orthop Scand. 1995;66(5):395–7. 16. Mayo KA, Benirschke SK, Mast JW. Displaced fractures of the glenoidfossa. Results of open reduction and internal fi xation. Clin Orthop Relat Res. 1998;(347):122–30. 17. Lambotte A. Chirurgieopératoire des fractures. Paris: Masson; 1913. 18. Armstrong CP, Van der Spuy J. The fractured scapula: importance and management based on a series of 62 patients. Injury. 1984;15(5):324–9. 19. Bozkurt M, Can F, Kirdemir V, Erden Z, Demirkale I, Basbozkurt M. Conservative treatment of scapular neck fracture: the effect of stability and glenopolar angle on clinical outcome. Injury. 2005;36(10):1176–81. 20. Goss TP. Double disruptions of the superior shoulder suspensory complex. J Orthop Trauma. 1993;7(2):99–106. 21. Edwards SG, Whittle AP, Wood 2nd GW. Nonoperative treatment of ipsilateral fractures of the scapula and clavicle. J Bone Joint Surg Am. 2000;82(6):774–80. 22. Ramos L, Mencia R, Alonso A, Ferrandez L. Conservative treatment of ipsilateral fractures of the scapula and clavicle. J Trauma. 1997;42(2):239–42. 23. Gosens T, Speigner B, Minekus J. Fracture of the scapular body: functional outcome after conservative treatment. J Shoulder Elbow Surg. 2009;18(3):443–8. 86 P.A. Cole and B.W. Hill 24. Schofer MD, Sehrt AC, Timmesfeld N, Stormer S, Kortmann HR. Fractures of the scapula: long- term results after conservative treatment. Arch Orthop Trauma Surg. 2009;129(11):1511–9. 25. Dimitroulias A, Molinero KG, Krenk DE, Muf fl y MT, Altman DT, Altman GT. Outcomes of nonoperatively treated displaced scapular body fractures. Clin Orthop Relat Res. 2011;469(5):1459–65. 26. Chadwick EK, van Noort A, van der Helm FC. Biomechanical analysis of scapular neck malunion – a simulation study. Clin Biomech. 2004;19(9):906–12. 27. Nordqvist A, Petersson C. Fracture of the body, neck, or spine of the scapula. A long-term follow-up study. Clin Orthop Relat Res. 1992;(283):139–44. 28. Hardegger FH, Simpson LA, Weber BG. The operative treatment of scapular fractures. J Bone Joint Surg Br. 1984;66(5):725–31. 29. Romero J, Schai P, Imhoff AB. Scapular neck fracture – the in fl uence of permanent malalign- ment of the glenoid neck on clinical outcome. Arch Orthop Trauma Surg. 2001;121(6):313–6. 30. Jones CB, Cornelius JP, Sietsema DL, Ringler JR, Endres TJ. Modi fi ed judet approach and minifragment fi xation of scapular body and glenoid neck fractures. J Orthop Trauma. 2009;23(8):558–64. 31. Herrera DA, Anavian J, Tarkin IS, Armitage BA, Schroder LK, Cole PA. Delayed operative management of fractures of the scapula. J Bone Joint Surg Br. 2009;91(5):619–26. 32. Khallaf F, Mikami A, Al-Akkad M. The use of surgery in displaced scapular neck fractures. Med Princ Pract. 2006;15(6):443–8. 33. Anavian J, Wijdicks CA, Schroder LK, Vang S, Cole PA. Surgery for scapula process fractures. Acta Orthop. 2009;80(3):344–50. 34. Pace AM, Stuart R, Brownlow H. Outcome of glenoid neck fractures. J Shoulder Elbow Surg. 2005;14(6):585–90. 35. Zlowodzki M, Bhandari M, Zelle BA, Kregor PJ, Cole PA. Treatment of scapula fractures: systematic review of 520 fractures in 22 case series. J Orthop Trauma. 2006;20(3):230–3. 36. Bartonicek J, Fric V. Scapular body fractures: results of operative treatment. Int Orthop. 2011;35(5):747–53. doi: 10.1007/s00264-010-1072-y . 37. Cole PA, Talbot M, Schroder LK, Anavian J. Extra-articular malunions of the scapula: a com- parison of functional outcome before and after reconstruction. J Orthop Trauma. 2011; 25(11):649–56. 38. Obremskey WT, Lyman JR. A modi fi ed judet approach to the scapula. J Orthop Trauma. 2004; 18(10):696–9. 39. van Noort A, van Kampen A. Fractures of the scapula surgical neck: outcome after conserva- tive treatment in 13 cases. Arch Orthop Trauma Surg 2005;125(10):696–700. 87M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_6, © Springer Science+Business Media New York 2013 Keywords Operative fi xation of clavicle fracture • Plate fi xation • Intramedullary fi xation • Nonoperative management of clavicle fractures • Midshaft clavicle fracture BP: 31-Year-Old Female with Shoulder Pain Case Presentation BP is a 31-year-old female who presents to the emergency department via EMS complaining of severe shoulder pain after falling at soccer practice. She denies any loss of consciousness. On primary survey, she demonstrates a GCS of 15, a patent airway, and is hemodynamic stability. She has severe pain on palpation of her right shoulder, but a secondary survey is otherwise negative. Her past medical history is unremarkable. She takes no medications and has no allergies. On physical examination, she has a strong radial pulse in the right upper extrem- ity and 5/5 strength throughout the right arm. She has normal capillary re fi ll and sensation in her hand and a full painless range of motion about the right wrist and elbow. On examination of her right shoulder, she has ecchymosis, extreme tender- ness with palpation of the clavicle, and pain on passive range of motion of the shoulder. Sensation over the anterior aspect of the deltoid is intact. Radiographs of the right shoulder and AP chest are demonstrated in Fig. 6.1 . C. R. Geddes , M.D., M.Sc. • M.D. McKee, M.D., F.R.C.S(c) (*) Division of Orthopedic Surgery , Department of Surgery,University of Toronto, St. Michael’s Hospital , 55 Queen Street East Suite 800 , Toronto , ON , Canada M5C 1R6 e-mail: chris.geddes@utoronto.ca; mcvieemesmh.ca; mckeem@smh.ca Chapter 6 Clavicle Fractures Christopher R. Geddes and Michael D. McKee 88 C.R. Geddes and M.D. McKee Interpretation of Clinical Presentation A healthy active 31-year-old female (BP) presents with severe right shoulder pain after a fall at soccer practice. The physical exam suggests a closed, isolated injury to the right shoulder girdle. Differential diagnosis at this stage includes: fractures of the clavicle, scapula, proximal humerus, and/or rib, ligamentous injury of the glenohumeral, sternoclavicular (SC), and/or acromioclavicular (AC) joints; and dislocation of the glenohumeral, SC, and/or AC joints. The radiographs con fi rm the diagnosis of a midshaft, or middle-third, clavicle fracture with complete displace- ment and approximately 2 cm of axial shortening and comminution. Clavicle fractures are common with an estimated incidence of 2–5% of all fractures. In a review of 690 clavicle fractures, 82% involved the middle-third Fig. 6.1 ( a – c ) AP chest and AP right shoulder radiographs revealing a right displaced, comminuted, diaphyseal clavicle fracture 896 Clavicle Fractures segment of bone [ 1 ] . Patients presenting with a fractured clavicle are typically male (2:1 male to female), between the age of 10 and 40, and injured in a road-traf fi c accident, fall from height, or sporting activity, especially cycling [ 2 ] . BP presents with no obvious life- or limb-threatening injuries. However the initial evaluation of any trauma patient necessitates a thorough assessment and reas- sessment of their condition. Due to the proximity of the clavicle to the subclavian vessels and brachial plexus, a detailed neurological (sensory, motor, and re fl exes) and vascular examination (pulses, temperature, and capillary re fi ll) is important in the physical examination. The injured area is carefully inspected focusing on the presence of swelling, deformity, tenting of the skin, or any open wounds that may indicate an open or impending-open fracture. A note should be made of any previous scars or old inci- sions. Clavicle fractures are occasionally missed in the multiply injured or obtunded trauma population, and care should be taken to evaluate each patient thoroughly. In a recent study of 692 patients admitted to a level one trauma center over 13-months, 17 (2.5%) had missed injuries of which two were clavicle fractures [ 3 ] . Open mid- shaft clavicle fractures, while rare and often from high-energy mechanisms of injury, have a strong correlation with serious injuries involving the head, chest, spine, and upper extremity [ 4, 5 ] . In a study by Taitsman et al., 1,740 (2.6%) of the 67,679 trauma admissions over a 13-year period had clavicle fractures, with only 24/1,740 (1.4%) being open fractures [ 5 ] . A complete examination of the shoulder girdle and neck should be carried out and documented. Care should be taken to protect the cervical spine. Spine- precautions can be discontinued only after clearance by the standard trauma pro- tocol. Following inspection, palpation of the clavicle, scapula, and proximal humerus should be performed. Firm pressure on the clavicle, applied adjacent to the fracture site, may elicit crepitus or palpable motion of the fragments and pain. Each joint around the shoulder (glenohumeral, AC, and SC joints) should be pal- pated and manipulated to evoke signs of tenderness, instability, or apprehension, and compared with the opposite side. Examination of the shoulder may be limited by pain in the clavicle. However, it is important to document whether the gle- nohumeral joint is reduced and if the axillary nerve is functioning. The length of the injured clavicle is measured from the SC joint to the AC joint and compared with the opposite side. Any amount of shortening should be evaluated both clini- cally and radiographically. Figure 6.1 illustrates a typical appearance of a midshaft clavicle fracture with an oblique orientation, and approximately 2 cm of displacement with predominantly anteroinferior comminution. Dedicated clavicle radiographs are important in the characterization and classi fi cation of clavicle fractures [ 6 ] . Standard radiographs include anteroposterior (AP) and 15–30° AP cephalic tilt Zanca views. An estimated length of displacement is quanti fi ed on calibrated, standardized radiographs, recog- nizing the limitations of this technique [ 7 ] . The direction of fracture displacement normally occurs in the coronal plane, due in part to the pull of the sternocleidomastoid muscle tilting the medial clavicular 90 C.R. Geddes and M.D. McKee fragment superiorly and posteriorly, while gravity and the pull of the pectoralis major muscle on the humerus displaces and rotates the lateral fragment inferiorly and anteriorly. In lateral third fractures, the attachments of the conoid and trapezoid coracoclavicular ligaments may avulse a segment of bone from the inferolateral aspect sometimes referred to as a conoid process or tubercle avulsion fracture. In Fig. 6.1 , the conoid and trapezoid ligaments are most likely intact and attached to the distal-third fragment, as the coracoclavicular space appears normal. The sterno- clavicular and acromioclavicular joints are aligned. There are no obvious fractures of the proximal humerus, scapula, or ribs, and no evidence of a pneumothorax. Initial treatment of patient BP should include a broad arm sling for comfort to her injured right upper extremity. The subsequent management of a closed, displaced clavicle fracture in a young, healthy, active individual such as BP remains contro- versial. Our changing understanding of the natural history of clavicle fractures drives much of this controversy, as studies suggest a higher than previously reported risk of nonunion and symptomatic malunion with nonoperative treatment of these injuries [ 8, 9 ] . The nonoperative management of acute midshaft clavicle fractures has recently been reviewed by Lenza and colleagues in a systematic review of randomized con- trolled trials [ 10 ] . Two-hundred and thirty-four patients in two trials compared a fi gure-of-eight bandage to an arm sling with no signi fi cant difference in functional outcomes [ 11, 12 ] . A third trial of 120 patients looked at the use of therapeutic ultrasound as compared with placebo and found no difference in outcomes [ 13 ] . All three trials were underpowered, and the authors conclude that current evidence is insuf fi cient to determine which method of conservative management is the most appropriate. To date, no method of splinting or bracing has been shown to be effec- tive in preventing shortening or malunion [ 14 ] . Additionally, there is no evidence to support performing a closed reduction in the emergency room. Recent literature supports initial operative management with plate fi xation for active, healthy, young adult patients such as BP, to improve the rate of union, and to decrease the likelihood of symptomatic malunions, albeit with the added risk of surgical complications [ 14 ] . In 2005, a systematic review of 2,144 fractures con fi rmed improved outcomes in the operative management of midshaft clavicle fractures [ 15 ] . In 2007, the Canadian Orthopedic Trauma Society (COTS) published a multicenter randomized prospective clinical trial of 132 patients with displaced midshaft clavicle fractures, comparing operative treatment with plate fi xation to nonoperative treatment with a sling [ 14 ] . Outcomes included standard- ized clinicalevaluations, radiographs, and the completion of the Constant shoul- der (CS) score and the Disability of the Arm, Shoulder, and Hand (DASH) score. Patients in the operative group did better on all outcome measures, with signi fi cantly lower rates of nonunion and symptomatic malunion. However, nine patients in the operative group had hardware-related complications, including three wound infections. 916 Clavicle Fractures Declaration of Speci fi c Diagnosis BP is a 31-year-old healthy female presenting with a closed, displaced, comminuted, midshaft fracture of the right clavicle. Brainstorming: What Are the Treatment Goals and the Options? Treatment Goals Consist of the Following Objectives: 1. Rapid return to function and/or work 2. Union of fracture 3. Recovery of strength and range of motion of shoulder 4. Minimize the risk of complications (nonunion and symptomatic malunion) Treatment Options Include Conservative/nonoperative treatment 1. Sling for comfort, early motion 2. Figure-of-eight bandage (or splint), early motion Surgical treatment 1. Plate fi xation 2. Intramedullary (IM) fi xation Evaluation of the Literature To identify publications relating to the treatment of midshaft clavicle fractures, PubMed ( http://pubmed.gov/ , U.S. National Library of Medicine, (NLM ® )) was searched. Using the Medical Subject Heading (MeSH) Database search function, the major headings “Fractures, Bone” and “Clavicle” were added to the search builder. The keywords “midshaft” or “middle” were added to the search: (“Clavicle”[Mesh]) AND (“Fractures, Bone”[Mesh]) AND (“midshaft” OR “mid- shaft” OR “middle”). Next, the search was limited to include articles published from 1975 to July 2011, in the English language, and involving adult subjects aged 19–44 (Adult). This search yielded 335 results. All 335 titles and/or abstracts were reviewed 92 C.R. Geddes and M.D. McKee and from this list 48 abstracts were reviewed in detail. An additional limit of “Clinical Trial” yielded 28 articles of which 7 were selected as high quality studies relating to the treatment of displaced, midshaft clavicle fractures in adults [ 14, 16– 21 ] . Detailed Review of Pertinent Articles Diaphyseal clavicle fractures have traditionally been treated nonoperatively and were considered to be benign injuries with little long-term functional impairment [ 22, 23 ] . However, a number of recent clinical trials, described below, have evalu- ated both conservative and operative treatment strategies for patients with displaced clavicle fractures, and we are left with many questions as to what is the best treat- ment option in the adult population [ 24– 27 ] . The following section focuses on the best evidence to guide treatment decisions for our patient, BP. Conservative/Nonoperative Management For hundreds of years variations of slings and bandages were applied to the upper extremity in an attempt to obtain and maintain a closed reduction of midshaft clav- icle fractures. Currently, the two most common nonoperative treatment protocols recommend a simple, Velpeau-style, broad-arm sling and a fi gure-of-eight bandage for comfort with an early range of motion protocol [ 22, 24, 27 ] . Andersen et al. conducted a randomized, clinical trial with 61 patients comparing a sling and fi gure-of-eight bandage in two treatment groups [ 22 ] . The authors found the simple sling to be better tolerated, with potentially fewer complications than the fi gure-of-eight bandage. There were no differences in the functional, radiographic, and cosmetic results between the two groups. A second retrospective cohort study of 136 patients had similar fi ndings of no difference in radiographic clavicle shortening or clinical outcomes using a Constant–Murley Score between a sling and fi gure-of- eight bandage [ 24 ] . There is no evidence that a closed reduction maneuver can main- tain alignment of displaced clavicle fractures for any signi fi cant period of time. Nonoperative management of clavicle fractures avoids potential surgical complica- tions of infection; hardware prominence or failure; refracture after hardware removal; hypertrophic or dysesthetic scars; and the need for reoperation. There are also case reports of the uncommon but serious risk of intraoperative neurovascular injury [ 28 ] . Historically most clavicle fractures have been treated nonoperatively. This is in part due to two in fl uential articles from the 1960s. Neer [ 23 ] and Rowe [ 1 ] reported a nonunion rate of 0.1% in 2,235 patients and 0.8% in 690 patients, respectively, with midshaft clavicle fractures treated with closed reduction. Critics of these stud- ies suggest that a large number of adolescents were included–a population in which fractures generally heal without sequelae [ 29 ] . Additionally, these studies had 936 Clavicle Fractures signi fi cant numbers of patients lost to follow-up, and assessments at the fi nal visits were not rigorous [ 1, 23 ] . Recent literature has highlighted the risks of nonoperative treatment of displaced midshaft fractures of the clavicle, including the possibility of developing a symptomatic malunion and an increased risk of nonunion [ 9, 14 ] . Complications associated with malunion of the clavicle can include musculosk- eletal pain related to muscle weakness and shoulder impingement, cosmetic con- cerns such as a drooping shoulder and a “bump” deformity and, less commonly, neurologic symptoms from brachial plexus irritation. In 1997, Hill and colleagues found that patients with displaced fractures with greater than 2 cm of initial shorten- ing had a higher risk of nonunion and decreased patient satisfaction when treated nonoperatively [ 9 ] . In this consecutive case series of 242 clavicle fractures, 52 patients with middle-third fractures were reviewed. Sixteen patients (31%) were dissatis fi ed with their clinical outcome following nonoperative management. A prospective long-term follow-up study by Nowak et al. found that 96 of 208 patients (46%) treated nonoperatively did not feel fully recovered at their 9- to 10-year follow-up [ 30 ] . Results correlated comminution and displacement with pain and poor satisfaction on patient survey. Additionally, 27% of patients were unhappy with the appearance of their shoulder. Lazarides et al. conducted a retrospective study from 1998 to 2001 reviewing 272 patients with clavicle fractures treated nonoperatively [ 31 ] . They found that 34 patients (25.8%) were dissatis fi ed with the results of their nonoperative manage- ment. Initial shortening of the clavicle greater than 1.8 cm in males and 1.4 cm in females was signi fi cantly associated with poor satisfaction on patient survey. Our patient, BP, is an active, healthy adult presenting with a shortened (approx. 2 cm) and displaced midshaft clavicle fracture. The goals of treatment are to achieve rapid return to function, early union, and recovery of strength and range of motion while minimizing the risk of complications. Nonoperative management may delay BP’s return to activity and is associated with a risk of a higher rate of patient dis- satisfaction and worse functional outcome scores. If she were to select nonoperative management, she could expect some degree of radiographic malalignment, with or without symptoms; asymmetric appearance of the right shoulder, due to shortening and depression of shoulder girdle; and an increased likelihood of delayed union or nonunion in the range of 15–20% [ 8, 9, 14 ] . Operative Management: Plate Fixation The use of plate osteosynthesis to treat midshaft clavicle fractureshas increased in popularity for completely displaced fractures with greater than 2 cm of shortening in young, active adults. However, the only absolute indications for open reduction and internal fi xation are in cases of open fractures and fractures with associated upper extremity neurovascular compromise. The remaining relative indications include fractures with tenting of the skin, and those with associated scapulothoracic dissociation or displaced glenoid fractures. Other relative indications include the 94 C.R. Geddes and M.D. McKee need for quicker return to activity (faster healing) and prevention of secondary symptomatic malunions [ 14, 17, 31 ] . Recent studies with long-term follow-up and patient-oriented outcome measures have estimated the incidence of nonunion and symptomatic malunion in displaced midshaft clavicle fractures to be approximately 15–20% [ 8, 9, 14, 15 ] . Hill and colleagues found a nonunion rate of 15% in fractures managed nonoperatively, 2% in those managed with plate fi xation, and 2.2% for those managed with IM pinning [ 9 ] . Zlowodzki et al. completed a systematic review of the English literature identi- fying 22 studies and 2,144 patients [ 15 ] . The nonunion rate for nonoperative treat- ment was 5.9% for all midshaft clavicle fractures and 15.1% for displaced midshaft fractures. They identi fi ed the following risk factors associated with nonunion after nonoperative treatment of clavicle fractures: fracture displacement (relative risk = 2.3), fracture comminution (relative risk = 1.4), female gender (relative risk = 1.4), and advancing age. The 2007 COTS trial compared nonoperative treatment to open reduction inter- nal fi xation with a plate [ 14 ] . 132 patients were randomized to standard sling (65) or small fragment plate (67). They found improved CS and DASH scores in the operative fi xation group at all time points up until 52 weeks follow-up ( P = 0.001 and P < 0.01, respectively). Time to fracture union was 16.4 weeks in the operative group and 28.4 weeks in the nonoperative group ( P = 0.001). There were seven non- unions in the nonoperative group and two nonunions in the operative group ( P = 0.042). Symptomatic malunions requiring further treatment were present in 9 of 49 patients (18.3%) treated nonoperatively at 1 year of follow-up, but none were present in the operative group ( P = 0.001). The overall complication rate for the operative group was 17.7% versus 32.6% complication rate for the nonoperative group (nonunions and symptomatic malunions). Complications unique to the operative group included infection (three cases) and the need for hardware removal due to prominent implants ( fi ve cases). There were no major neurovascular complications reported. This trial pro- vides Level I evidence regarding the overall improved outcomes that can be achieved in select patients (active, healthy individuals between 16 and 60 years of age) with completely displaced (mean displacement of 2 cm) midshaft clavicle fractures [ 14 ] . Another study by Mirzatolooei compared nonoperative treatment (24) to open reduction internal fi xation with a plate (26) in a randomized clinical trial of 60 patients [ 21 ] . One patient in each group had a nonunion, and rates of malunion were signi fi cantly lower in the operative group (4 or 15.4%) than the nonoperative group (19 or 79.2%). The nonunion in the operative group occurred secondary to infection. This study reported signi fi cantly better DASH and CS scores in the operative fi xation group at 12 months, with lower rates of pain, weakness, and limitation of motion. The individual characteristics of the patient or fracture that will bene fi t from operative treatment have not been clearly de fi ned. The COTS trial showed that 33 of 49 patients (67.4%) with 100% displaced clavicle fractures treated nonopera- tively healed with results essentially the same as the operative group [ 11 ] . Possible predictors of poor outcome based on current literature include marked displacement 956 Clavicle Fractures and/or shortening at the fracture site, fracture comminution, female gender, and advanced age. In the case of BP, she has three of four risk factors for a worse out- come with nonoperative management. Several biomechanical and clinical papers have focused on the type of plate (reconstruction versus 3.5 mm small fragment plate) and anatomic location of fi xation (anterior or superior) for midshaft clavicle fractures with and without com- minution. In a study by Drosdewech et al., the biomechanics of anterior and supe- rior locations of plate fi xation were compared to IM fi xation in 20 cadaveric midshaft clavicle fractures [ 32 ] . More rigid plates (dynamic compression plates (DCP) and locked compression plates (LCP)) placed on the superior clavicle for simulated unstable midshaft clavicular fractures resisted higher bending and torque loads than did less rigid reconstruction plates or IM devices. A cadaveric study by Harnroongroj et al. concluded that stability against a bending moment was improved, with supe- rior plating in fractures without an inferior cortical defect and anterior plating in fractures with an inferior cortical defect [ 33 ] . In 2006, Collinge and colleagues looked at a consecutive clinical series of 80 patients with midshaft clavicle fractures treated with anterior-inferior plate fi xation [ 34 ] . Patients were evaluated with both clinical and radiographic examinations, the American Shoulder and Elbow Surgeons Shoulder Assessment, and the Short Form- 36 outcomes questionnaire. At 2 years follow-up, they had similar complication rates related to failure of fi xation, infection, and nonunion, but concluded that anterior-inferior plating has the potential advantage of avoiding infraclavicular neurovascular structures. Other studies have evaluated the biomechanics of precontoured plates compared to standard plates and have shown no difference [ 35 ] . Vanbeek and colleagues ret- rospectively reviewed 52 displaced midshaft clavicle fractures treated with plate fi xation and found lower rates of prominent hardware in those treated with precon- toured plates (9/28 patients) than those treated with noncontoured plates (9/14) [ 36 ] . The percentage of patients who underwent hardware removal was less in the pre- contoured group (3/28) than the noncontoured group (3/14). Timing of surgery is also a consideration in the management of acute displaced midshaft clavicle fractures. Potter and colleagues compared objective outcomes in 15 patients who underwent delayed operative intervention for nonunion and malunion with 15 patients who had immediate open reduction and internal fi xation [ 37 ] . The delayed group had a mean time from fracture to operative intervention of 63 months (range, 6–67 months). They measured differences in strength, endur- ance, and patient outcome scores. The mean duration from operation to testing was 33 months in the delayed group and 25 months in the acute group, with a minimum of 12 months. They found no difference in satisfaction with the proce- dure, shoulder strength, and DASH scores. The acute fi xation group had signi fi cantly better Constant Shoulder scores and endurance in forward fl exion. This study suggests that the outcome after delayed reconstruction will on average be slightly inferior to what might have been obtained with acute fi xation with the added potential complications associated with delayed fracture surgery, such as the need for bone grafting. 96 C.R. Geddes and M.D. McKee For patient BP to decide on operative versus nonoperative management of her acute clavicle fracture, the potential bene fi ts of operative fi xation must be discussed in relation to risks of potential complications.Criticisms of initial operative man- agement of clavicle fractures include the potential for a hypertrophic scar or regional dysesthesia from injury to the supraclavicular nerves. Also, there is the risk of infec- tion, hardware prominence, and the required local soft tissue trauma required for exposure of the bone fragments [ 16, 20 ] . Potentially, some of these complications are avoidable with IM techniques, yet the superiority of IM fi xation remains to be clinically proven [ 31 ] . Operative Management: Intramedullary Fixation IM fi xation is proposed to be a less invasive, alternative to plate fi xation for mid- shaft clavicle fractures [ 16 ] The goal of IM fi xation for displaced clavicle fractures is to maintain fracture reduction (length, angulation, and rotation) while reducing the amount of periosteal and soft tissue dissection from the bone. Additionally, the IM devices can be inserted under fl uoroscopic guidance and may have the advan- tage of a smaller incision and scar. Many different variations on IM devices have been available for approximately 40 years. More recently, IM stabilization with tita- nium elastic nails has increased in popularity [ 38 ] . Three randomized trials were identi fi ed in the literature pertaining to IM fi xation of displaced midshaft clavicle fractures [ 16– 18 ] . Smekal et al. reported on 60 patients randomized to sling (30 patients) or elastic titanium IM pin fi xation (30 patients) and followed the groups until 2 years post-injury [ 17 ] . The operative group had a faster time to union, lower DASH, and higher CS scores. Delayed union was identi fi ed in 6 of 30 patients (20%) in the nonoperative group. In 2009, a prospective randomized study by Judd and colleagues of 57 patients (military personnel) randomized to IM fi xation (29) or sling (28) found no signi fi cant difference between operative and nonoperative groups at 1 year [ 18 ] . The operative group did, however, demonstrate signi fi cantly higher Single Assessment Numeric Evaluation and L’Insalata scores at 3 weeks. The complication rate was greater in the operative group including: nonunion, refracture, infection, and prominent hardware. Nearly half of the patients in the operative group lost some of the original reduction. Similarly, a 2007 report by Strauss et al. reported a 50% postoperative complication rate in 16 patients treated with IM fi xation over a 10-year period [ 39 ] . If surgery is the treatment of choice for patient BP, the preferred surgical tech- nique (IM or plate fi xation) remains controversial. Ferran et al. randomized 32 patients, 15 to plate fi xation (uncontoured low contact dynamic compression plates (LCDCP)) and 17 to locked IM fi xation (Rockwood pin). Twelve-month follow-up demonstrated 100% union in both groups [ 16 ] . No signi fi cant difference was detected in CS scores or Oxford scores at any time. Two IM pins (12%) were symp- tomatic postoperatively and 8 plates (53%) were removed (3 super fi cial infec- tions, 2 prominent plates, 2 for discomfort). 976 Clavicle Fractures No single technique has become standard in the IM nailing of displaced midshaft clavicle fractures, and no de fi nitive recommendations can be made. There appears to be a high rate of postoperative complications, with some studies reporting up to 50% complications including implant breakage, temporary brachial plexus palsy, skin breakdown over the pin, and implant protrusion from the lateral clavicle [ 39 ] . Like any other unlocked device, IM nails do not hold length or rotation well in com- minuted fractures [ 18 ] . High-quality randomized clinical trials are required to eval- uate the use of IM devices for the treatment of clavicle fractures. The role for IM fi xation has yet to be well de fi ned. Literature Inconsistencies A number of well-designed randomized clinical trials provide Level I evidence for the treatment of displaced, midshaft clavicle fractures. However, further studies need to evaluate the risks and bene fi ts of different operative and nonoperative treat- ment options particularly relating to the use of IM fi xation. New studies should meet current standards in the planning, execution, and reporting of randomized clinical trials, and should require an adequate sample size to allow for clear inter- pretation of the results. Improving the quality of studies will empower patients and clinicians to make truly informed choices about treatment strategies. Evidentiary Table and Selection of Treatment Method Table 6.1 summarizes important clinical trials relating to the treatment of displaced midshaft clavicle fractures. Based on a review of the evidence, it is the authors’ opinion that the best treatment for our 31-year-old, active, female patient, BP, is primary fi xation with plate osteosynthesis using a strong, precontoured plate in the superior anatomic position with lag screw fi xation. The treatment goals for BP of rapid return to function (or work), early union of fracture, recovery of strength and range of motion, with minimal complications, are best achieved with plate fi xation. While IM fi xation with an elastic stable IM nail may prove to have similar outcome results to plating with a more cosmetic result, the clinical evidence and biomechanical literature currently supports plate fi xation as a more predictable and superior outcome, especially in a North American population. De fi nitive Treatment Plan The surgical goal for managing a displaced midshaft clavicle fracture is to restore alignment of the shoulder girdle. This is achieved with anatomic reduction of the fracture with reconstitution of clavicular length and rotation. The time from injury 98 C.R. Geddes and M.D. McKee Table 6.1 A summary of the evidence for operative versus nonoperative treatment of midshaft clavicle fractures in adults Author (year) Description Summary of results Levels of evidence Canadian Orthopedic Trauma Association (COTS) (2007) Randomized clinical trial 132 patients randomized sling (65) or plate fi xation (67). The CS and DASH scores were signi fi cantly better at all time points for the operative group ( p < 0.01). Nonunion and symptomatic malunion were higher in nonoperative group. 3 infections, 5 hardware removal surgeries in operative group. I Smekal et al. (2009) Randomized clinical trial 68 patients randomized to sling or elastic titanium intramedullary (IM) pin fi xation. 60 patients (30 in each group) were followed for 2 years post-injury. The operative group had faster time to union, lower DASH, and higher CS scores. Delayed union occurred in 6 of 30 patients in the nonoperative group. I Judd et al. (2009) Randomized clinical trial 57 patients (military personnel) randomized to Haigie pin fi xation (29) or sling (28) and followed for 1 year. Both the SANE and L’Insalata scores were signi fi cantly higher at 3 weeks; there was no difference in outcome scores at any other point. Higher complication rate in operative group. I Ferran et al. (2010) Randomized clinical trial 32 patients, 15 randomized to plate fi xation, and 17 to locked IM fi xation followed for 12 months. 100% union in both groups. No difference in CS or Oxford scores. 2 IM pins symptomatic, and 8 plates removed (3 due to super fi cial infection). II Kulshrestha et al. (2011) Prospective cohort study 73 patients (military and civilian) allocated to plate fi xation or nonoperative treatment groups. Followed for 18 months. 100% union rate. 2 symptomatic malunions in the operative group, 8 nonunions, and 10 symptomatic malunions in the nonoperative group. CS score signi fi cantlybetter in the operative group. 6 patients had hardware removed due to irritation (4) or failure (2). II (continued) 996 Clavicle Fractures Table 6.1 (continued) Author (year) Description Summary of results Levels of evidence Mirzatolooei (2011) Randomized clinical trial 60 patients, randomized into open plate fi xation (26) and nonoperative treatment (24) and followed for 12 months. 3 patients declined surgery and 7 patients were lost to follow-up. 1 nonunion in each group. 4 malunions in the operative group and 19 malunions in the nonoperative group. 1 infection in the operative group. DASH and CS scores signi fi cantly better in the operative group. I CS Constant Shoulder Score, DASH Disability of the Arm, Shoulder and Hand, DCP limited contact dynamic compression plate, LCP locking compression plate, RECON reconstruction plate, SANE Single Assessment Numeric Evaluation to surgery, the amount of displacement and comminution, and the body habitus of the patient all contribute to the dif fi culty of the procedure. Severely comminuted midshaft clavicle fractures may require a bridge-plating technique rather than lag screws and compression plating. The surgical plan is preoperatively designed after reviewing the history and physical examination and radiographs, and then discuss- ing options with the patient. Radiographic comparison between injured and intact sides helps determine the amount of initial shortening, and the reduction required intraoperatively. Position the patient for the procedure in the beach chair position with a small bump under the posteromedial aspect of the injured shoulder. Prepare and drape the clavicle area in a sterile manner. Rotate and gently tilt the patient’s head away from the operative fi eld. Secure the patient and pad any pressure points. If desired, the arm can be free draped to help achieve the reduction. Preoperatively mark bony landmarks to allow proper placement of the incision. Expose the clavicle through a 5–10 cm incision centered over the fracture site. With experience, a smaller incision can be used and is preferred. Identify and protect where possible visible branches of the supraclavicular nerves. Separate skin from the deep fascial layer to produce a “mobile window” and facilitate the operation. Dissect fascia and periosteum from the bone ends with care to preserve soft tissue to any comminuted fragments of bone. A self-retainer is often useful to maintain visualization. Expose and debride fractured bone ends to clear interposed hematoma and soft tissues. Use Kirschner wires and reduction clamps when possible to obtain provi- sional fi xation while inserting the lag screw. Careful assessment of length and rota- tion is important in restoring normal clavicle anatomy. Place a strong, precontoured, low-pro fi le clavicle-speci fi c plate along the superior surface of the clavicle and hold it in place with reduction clamps. A minimum of three screws and a lag screw or 100 C.R. Geddes and M.D. McKee four screws on either side of the fracture is preferred while drilling and using the depth-gauge, care should be taken to not plunge into the infraclavicular neurovascu- lar structures or pleural space. Assess stability of the construct and irrigate the wound with sterile saline solu- tion. With the wound fi lled with saline wash, ask the anesthesia team to perform a Valsalva maneuver, thus increasing intrathoracic pressure and assessing pleural integrity prior to closure (bubbles from the wound bed may signify a pneumothorax). Two-layer closure over the superiorly positioned clavicle plate is critical. Close the deep fascia over the plate with interrupted or running absorbable sutures. Close the skin with a subcuticular closure, horizontal-mattress sutures, or staples. Apply a nonadhesive dressing to the wound and reinforce with a dry gauze dressing. Apply an arm sling for patient comfort. The postoperative protocol includes an upright clavicle radiograph to assess fi xation. Perform a postoperative (maximal inspiration) chest radiograph and serial clinical examinations if a pneumothorax is suspected. No evidence-based literature exists to guide acute postoperative management of clavicle fractures, and the following suggestions are based on the current standard of care. Start early gentle unrestricted range-of-motion exercises of the shoulder, usually at 1–2 weeks. Begin strength and resistance exercises at 6 weeks if the patient has a favorable clinical examination and the radiographs show evidence of healing. Return to sports is determined on a case-by-case basis, but usually can begin by 8–12 weeks if the patient is asymptomatic and the fracture is clinically and radiographically healed. Patients who have dif fi culties complying with postoperative protocols may require prolonged immobilization and can have a delayed return to sports. Predicting Long-Term Outcomes Long-term studies of clinical outcomes after primary fi xation of displaced midshaft clavicle fractures are limited in the literature. Schemitsch et al. [ 40 ] published a follow- up of the COTS study group comparing the clinical outcomes (DASH and CS scores) at 1 and 2 years post-injury and found a plateau effect with no signi fi cant change in scores after 1 year in either group. Additionally, the DASH and CS scores continued to be signi fi cantly better in the operative than the nonoperative cohorts at 2 years. The authors conclude that this plateau-effect can be used to counsel patients. Following acute operative fi xation of her displaced midshaft clavicle fracture, BP can expect to improve up to 1 year and then reach a steady state in terms of functional outcome. References 1. Rowe CR. An atlas of anatomy and treatment of midclavicular fractures. Clin Orthop Relat Res. 1968;58:29–42. 2. Postacchini F, Gumina S, De Santis P, et al. Epidemiology of clavicle fractures. J Shoulder Elbow Surg. 2002;11:452–6. 1016 Clavicle Fractures 3. Malhotra AK, Martin N, Jacoby M, et al. What are we missing: results of a 13-month active follow-up program at a level I trauma center. J Trauma. 2009;66:1696–703. 4. Gottschalk HP, Browne RH, Starr AJ. Shoulder girdle: patterns of trauma and associated inju- ries. J Orthop Trauma. 2011;25:266–71. 5. Taitsman LA, Nork SE, Coles CP, et al. Open clavicle fractures and associated injuries. J Orthop Trauma. 2006;20:396–9. 6. Robinson CM. Fractures of the clavicle in the adult. Epidemiology and classi fi cation. J Bone Joint Surg Br. 1998;80:476–84. 7. Smekal V, Deml C, Irenberger A, et al. Length determination in midshaft clavicle fractures: validation of measurement. J Orthop Trauma. 2008;22:458–62. 8. Robinson CM, Court-Brown CM, et al. Estimating the risk of nonunion following nonopera- tive treatment of a clavicular fracture. J Bone Joint Surg Am. 2004;86:1359–65. 9. Hill JM, McGuire MH, Crosby LA. Closed treatment of displaced middle-third fractures of the clavicle gives poor results. J Bone Joint Surg Br. 1997;79:536–41. 10. Lenza M, Belloti JC, Andriolo RB, Gomes dos Santos JB, Faloppa F. Conservative interven- tions for treating middle third clavicle fractures in adolescents and adults. Cochrane Database Syst Rev. 2009. Issue 2. Art. No.: CD007121. doi: 10.1002/14651858.CD007121.pub2 . 11. Andersen K, Jensen PO, Lauritzen J. Treatment of clavicular fractures. Figure-of-eight ban- dage versus a simple sling. Acta Orthop Scand. 1987;58:71–4. 12. Hoofwijk AGM, van der Werken C. Closed treatment of clavicular fractures: a prospective randomized trial. Acta Orthop Scand. 1986;57:482–3. 13. Lubbert PH, Van der Rijt RH, Hoorntje LE,Van der Werken C. Low-intensity pulsed ultra- sound (LIPUS) in fresh clavicle fractures: a multi-centre double blind randomised controlled trial. Injury. 2008;39:1444–52. 14. Canadian Orthopaedic Trauma Society. Nonoperative treatment compared with plate fi xation of displaced midshaft clavicular fractures. A multicenter, randomized clinical trial. J Bone Joint Surg Am. 2007;89:1–10. 15. Zlowodzki M, Zelle BA, Cole PA, Evidence-Based Orthopaedic Trauma Working Group, et al. Treatment of acute midshaft clavicle fractures: systematic review of 2144 fractures: on behalf of the Evidence-Based Orthopaedic Trauma Working Group. J Orthop Trauma. 2005;19:504–7. 16. Ferran NA, Hodgson P, Vannet N, et al. Locked intramedullary fi xation vs. plating for displaced and shortened midshaft clavicle fractures: a randomized clinical trial. J Shoulder Elbow Surg. 2010;19:783–9. 17. Smekal V, Irenberger A, Struve P, et al. Elastic stable intramedullary nailing versus nonopera- tive treatment of displaced midshaft clavicular fractures – a randomized, controlled, clinical trial. J Orthop Trauma. 2009;23:106–12. 18. Judd DB, Pallis MP, Smith E, et al. Acute operative stabilization versus nonoperative manage- ment of clavicle fractures. Am J Orthop. 2009;38:341–5. 19. Lubbert PH, van der Rijt RH, Hoorntje LE, et al. Low-intensity pulsed ultrasound (LIPUS) in fresh clavicle fractures: a multi-centre double blind randomised controlled trial. Injury. 2008;39:1444–52. 20. Kulshrestha V, Roy T, Audige L. Operative versus nonoperative management of displaced midshaft clavicle fractures: a prospective cohort study. J Orthop Trauma. 2011;25:31–8. 21. Mirzatolooei F. Comparison between operative and nonoperative treatment methods in the management of comminuted fractures of the clavicle. Acta Orthop Traumatol Turc. 2011;45:34–40. 22. Andersen K, Jensen PO, Lauritzen J. Treatment of clavicular fractures. Figure-of-eight bandage versus a simple sling. Acta Orthop Scand. 1987;58:71–4. 23. Neer 2nd CS. Nonunion of the clavicle. JAMA. 1960;172:1006–11. 24. Rasmussen JV, Jensen SL, Petersen JB, et al. A retrospective study of the association between shortening of the clavicle after fracture and the clinical outcome in 136 patients. Injury. 2011;42:414–7. 25. McKee MD. Clavicle fractures in 2010: sling/swathe or open reduction and internal fi xation? Orthop Clin North Am. 2010;41:225–31. 102 C.R. Geddes and M.D. McKee 26. Lenza M, Belloti JC, Gomes dos Santos JB, Matsumoto MH, Faloppa F. Surgical interventions for treating acute fractures or non-union of the middle third of the clavicle. Cochrane Database Syst Rev. 2009. Issue 4. Art. No.: CD007428. doi: 10.1002/14651858.CD007428.pub2 . 27. Postacchini R, Gumina S, Farsetti P, et al. Long-term results of conservative management of midshaft clavicle fracture. Int Orthop. 2010;34:731–6. 28. Ring D, Holovacs T. Brachial plexus palsy after intramedullary fi xation of a clavicular fracture. A report of three cases. J Bone Joint Surg Am. 2005;87:1834–7. 29. Smekal V, Oberladstaetter J, Struve P, et al. Shaft fractures of the clavicle: current concepts. Arch Orthop Trauma Surg. 2009;129:807–15. 30. Nowak J, Holgersson M, Larsson S. Can we predict long-term sequelae after fractures of the clavicle based on initial fi ndings? A prospective study with nine to ten years of follow-up. J Shoulder Elbow Surg. 2004;13:479–86. 31. Lazarides S, Za fi ropoulos G. Conservative treatment of fractures at the middle third of the clavicle: the relevance of shortening and clinical outcome. J Shoulder Elbow Surg. 2006; 15:191–4. 32. Drosdowech DS, Manwell SE, Ferreira LM, et al. Biomechanical analysis of fi xation of middle third fractures of the clavicle. J Orthop Trauma. 2011;25:39–43. 33. Harnroongroj T, Vanadurongwan V. Biomechanical aspects of plating osteosynthesis of transverse clavicular fracture with and without inferior cortical defect. Clin Biomech. 1996;11:290–4. 34. Collinge C, Devinney S, Herscovici D, et al. Anterior-inferior plate fi xation of middle-third fractures and nonunions of the clavicle. J Orthop Trauma. 2006;20:680–6. 35. Goswami T, Markert RJ, Anderson CG, et al. Biomechanical evaluation of a pre-contoured clavicle plate. J Shoulder Elbow Surg. 2008;17:815–8. 36. Vanbeek C, Boselli KJ, Cadet ER, et al. Precontoured plating of clavicle fractures: decreased hardware-related complications? Clin Orthop Relat Res. 2011;468:3337–43. 37. Potter JM, Jones C, Wild LM, et al. Does delay matter? The restoration of objectively mea- sured shoulder strength and patient-oriented outcome after immediate fi xation versus delayed reconstruction of displaced midshaft fractures of the clavicle. J Shoulder Elbow Surg. 2007;16:514–8. 38. Smekal V, et al. Elastic stable intramedullary nailing is best for mid-shaft clavicular fractures without comminution: results in 60 patients. Injury. 2010;42:324–9. 39. Strauss EJ, Egol KA, France MA, et al. Complications of intramedullary Hagie pin fi xation for acute midshaft clavicle fractures. J Shoulder Elbow Surg. 2007;16:280–4. 40. Schemitsch LA, Schemitsch EH, Veillette C, Zdero R, McKee MD. Function plateaus by one year in patients with surgically treated displaced midshaft clavicle fractures. Clin Orthop Relat Res. 2011;469:3351–5. 103M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_7, © Springer Science+Business Media New York 2013 Keywords Nonoperative management of proximal humerus fracture • Operative management of proximal humerus fracture: plate • Operative management of proximal humerus fracture: intramedullary nail • Closed reduction and percutaneous pinning • Proximal humerus hemiarthroplasty • Reverse total shoulder arthroplasty Case Presentation AP is a 67-year-old female who presents to the emergency department via EMS complaining of severe shoulder pain after a high-speed motor-vehicle accident in which she was a restrained passenger. She denies any loss of consciousness. On primary survey, she demonstrates a GCS of 15, a patent airway, and is hemodynamically stable. On secondary survey, she demonstrates severe pain with passive range of motion of the right shoulder. There is no motor or sensory de fi cit, and skin is intact. The remaining physical exam as well as her past medical history are unremarkable. Radiographs of the right shoulder are demonstrated in Fig. 7.1 . D. J. Stinner, M.D. Department of Orthopedics and Rehabilitation , San Antonio Military Medical Center , 3551 Roger Brooke Drive Fort Sam , Houston , TX 78234 , USA P. N. Streubel, M.D. • W.T. Obremskey, M.D., M.P.H. (�) Department of Orthopedic Surgery and Rehabilitation , Vanderbilt Medical Center , Suite 4200, South Tower Medical Center East – 1215 21st Avenue South , Nashville , TN 37232 , USA e-mail: william.obremskey@vanderbilt.edu Chapter 7 Proximal Humerus Fracture Daniel J. Stinner , Philipp N. Streubel , and William T. Obremskey 104 D.J. Stinner et al. Interpretation of Clinical Presentation The patient’s fi ndings and symptoms are consistent with an isolated proximal humerus fracture. This injury commonly occurs in the elderly in low-energy situations, such as a fall from standing [ 1 ] . Despite the fact that this injury was sustained in a high-speed motor-vehicle collision, the physical exam fi ndings are consistent with those com-monly seen with a low-energy proximal humerus fracture. Open fractures are rare, but closed fractures can be displaced enough to cause severe tenting and pressure necro- sis on the skin, necessitating an attempt at closed reduction. Vascular injuries are also uncommon, as are nerve injuries. However, when nerve injuries occur, they are typi- cally a traction injury to the axillary nerve or a direct injury to the brachial plexus. These injuries are usually clinically observed due to high rates of recovery [ 2 ] . The radiographs demonstrate an impacted fracture of the surgical neck of the humerus, as well as a minimally displaced fracture between the greater tuberosity and the articular segment. The fracture also exhibits valgus angulation of the proxi- mal segment, with shortening of more than 1 cm. Some orthopedic surgeons would refer to this fracture pattern as a Neer three-part fracture. However, according to Neer’s classi fi cation of proximal humerus fractures, this fracture should be classi fi ed as a two-part surgical neck fracture with a minimally displaced greater tuberosity fracture [ 3 ] . Declaration of Speci fi c Diagnosis AP is a 67-year-old female who presents with a high-energy injury that consists of a two-part surgical neck with a minimally displaced greater tuberosity proximal humerus fracture. Fig. 7.1 ( a ) AP radiograph shoulder, ( b ) Lateral Radiograph Shoulder 1057 Proximal Humerus Fracture Brainstorming: What Are the Treatment Goals and Therapeutic Options? Treatment goals consist of the following objectives: 1. Stabilization of the proximal humerus fracture 2. Maximize shoulder range of motion 3. Maintenance of muscle strength 4. Minimize shoulder pain 5. Return to normal life activities Treatment options include the following: Conservative/nonoperative treatment. 1. Sling with early range of motion Surgical. 1. Closed reduction and percutaneous fi xation 2. Open reduction and internal fi xation (a) Tension band fi xation (b) Plate fi xation Standard • Locking • 3. Antegrade intramedullary nail fi xation 4. Arthroplasty (a) Hemiarthroplasty (b) Reverse total shoulder arthroplasty Evaluation of the Literature To identify relevant publications on the treatment of proximal humerus fractures, Medline and PubMed searches were performed. Keywords included the following: “proximal humerus fracture.” Limits were set to include only articles in the English language and involve only humans, published between 1970 and present. A total of 950 abstracts were identi fi ed and reviewed. From this search, 67 of the most relevant articles were reviewed to identify treatment methods, outcomes, and level of evidence. 106 D.J. Stinner et al. Detailed Review of Pertinent Articles Several treatment options have been studied for the management of proximal humerus fractures. However, the quality of available literature continues to be inad- equate to determine whether operative fi xation of displaced proximal humerus frac- tures will provide better long-term outcomes [ 4 ] . The following review recognizes the controversy associated with the treatment of proximal humerus fractures and evaluates the relevant current literature to deter- mine the most appropriate management for AP. Nonoperative Treatment Nonoperative treatment is considered the standard of treatment for most proximal humerus fractures, especially when minimally displaced. Koval et al. evaluated 104 patients with an average age 63 years (range 24–94) who underwent nonoperative treatment with a standardized therapy regimen for stable one-part proximal humerus fractures de fi ned as less than 1 cm displaced and less than 45° of angulation between fracture parts, and no gross motion between fragments. After a minimum follow-up of 1 year all fractures united. Functional outcome as de fi ned by a score developed by the authors was good or excellent in 77%, fair in 13%, and poor in 10% of cases (half of which had reinjured). Pain was either absent or mild in 90% of patients at last follow-up [ 5 ] . Keser et al. found similar results in 27 patients with minimally displaced fractures treated nonoperatively. After a minimum follow-up of 1 year, all fractures healed with an average Constant score of 81 (range 51–100). Abduction peak torque was found to be below the opposite side in 26 patients [ 6 ] . Tejwani et al. studied 67 patients with Neer one-part fractures managed with a standardized reha- bilitation program. Fifty four (80%) patients were available at 1 year for assess- ment. All fractures healed, and internal rotation and forward fl exion were found to be similar to the unaffected shoulder. Only external rotation was signi fi cantly different, with 7.6° less on the fractured shoulder compared to the opposite side. The American Shoulder and Elbow Surgeons (ASES) score was 93.7 points and not signi fi cantly different from baseline (99.1 points). Similarly no difference was found in quality of life as determined by SF-36 between baseline (89 points) and 1-year follow-up (87.8 points) [ 7 ] . These results have been recently con fi rmed by Bahrs et al. In a prospective cohort study of 66 minimally displaced fractures in patients with an average age of 59 years, healing was achieved in all cases. The median Constant score was 89 points after a mean follow-up of 51 months. As in previous studies, the authors found a signi fi cant association between the fi nal Constant score and age. Interestingly, an association was also found with the American Society of Anesthesia (ASA) classi fi cation and the AO fracture classi fi cation. Court-Brown et al. studied 131 two-part surgical neck fractures of the proximal humerus. Mean age was 73 years for females and 69 for males. One-hundred and 1077 Proximal Humerus Fracture eight patients completed a minimum follow-up of 1 year. Patients below the age of 50 consistently achieved Neer scores above 90. In the group of patients above the age of 60, based on Neer’s criteria, only 65% of patients achieved a good or excel- lent result. Age correlated signi fi cantly with Neer score at fi nal follow-up and with the ability to return to shopping and housework. In the group of patients with at least 66% of translation, no difference was found between surgical and nonoperative treatment with regard to Neer scores, return to daily activities, and nonunion [ 8 ] . Valgus impacted proximal humerus fractures have been shown to represent the most frequent fracture type in elderly patients [ 9 ] . Court-Brown et al. reviewed the clinical results of 150 patients (mean age 71 years, range 44–88) with such a fracture treated nonoperatively. One hundred and twenty fi ve patients completed 1-year of follow-up, having 81% good or excellent results according to Neer’s criteria. The authors found that overall, patients subjectively rated their function higher than assessed objectively. Function determined by Constant score was similar among patients with minimally displaced fractures (mean score 74), and those with dis- placement of the greater tuberosity (mean score 73) or surgical neck (mean score 72). However, fractures with displacement of the surgical neck and greater tuberosity had a mean Constant score of 67, suggesting a small decrease of function compared to the remaining fractures. While pain perception was equal among fracture types, the main decrease in function was related to fl exion and abduction power [ 10 ] . In a recent prospective study, Foruria et al. studied the impact of fracture con fi guration determined with CT imaging on clinical outcome.Ninety-three patients (average age 71 years, range 26–93 years) with nonoperatively treated frac- tures, were followed over a 1 year period. All except two fractures healed (98%), and excellent or satisfactory results as determined by Neer’s criteria were obtained in 75% of cases. A decrease in shoulder function was observed with an average change in ASES and Disability of the Arm Shoulder and Hand (DASH) scores from 16 and 91 preoperatively to 23 and 78 after surgery, respectively. Of the analyzed fracture patterns, those with valgus impaction and a greater tuberosity fragment similar to patient AP, and those with varus impaction had the highest loss of func- tion as determined by both ASES and DASH scores [ 11 ] . This study stands in con- trast to the previous fi ndings by Jakob et al. [ 9 ] and Court-Brown et al. [ 10 ] , with regard to favorable outcomes of valgus impacted fractures. The authors however used detailed fracture analysis with adequate follow-up using validated outcome measures, establishing a new reference point on which to analyze these proximal humerus fractures. Hanson et al. prospectively enrolled a cohort of 160 patients (mean age 63.3 years, male to female ratio 1:3) with proximal humerus fractures treated nonoperatively: there were 75 one-part fractures, 60 two-part fractures, 23 three-part fractures, and 2 four-part or head splitting fractures. Of the 124 patients (87%) that completed 1-year follow-up, four subsequently required surgery due to displacement, and fi ve required arthroscopic subacromial decompression due to impingement. At fi nal follow-up, injured shoulders had on average an 8.2 point lower Constant score than the opposite shoulder. Similarly, the mean difference in DASH scores between the injured and contralateral side was 10.2 points. No differences were found between 108 D.J. Stinner et al. fracture types, and the highest variability in outcomes was found in the two-part fracture group [ 12 ] . Based on these studies consisting of mostly Level 3 and Level 4 evidence, one can assume certain expectations if nonoperative management of AP’s fracture is chosen as the de fi nitive treatment. First, nonoperative management of proximal humerus fractures reliably results in union but may demonstrate a small (but mea- surable) loss of external rotation compared to the uninjured shoulder [ 7 ] . Second, there will likely be a decrease in abduction strength when compared to the unin- jured shoulder [ 6, 10 ] . Finally, Constant and ASES Scores, while in fl uenced most by patient age [ 7, 8 ] , may not be signi fi cantly different than baseline values after 1 year [ 7 ] . Few studies have compared surgical to nonsurgical treatment of proximal humerus fractures. Kristiansen et al. randomized 31 two-part, three-part, and four- part fractures to either closed treatment or external fi xation [ 13 ] . Eleven patients treated nonoperatively completed 1 year of follow-up. Nonunion occurred in two surgical neck fractures and two greater tuberosity fractures. Additionally, two patients developed AVN of the humeral head. In the group treated with external fi xation, 13 patients completed 1 year follow-up. One deep infection occurred, two fractures did not unite, and AVN was seen in one case. Median Neer score was 79 after external fi xation and 60 after nonoperative treatment. Only four patients in the nonoperative group achieved satisfactory or excellent results, compared to eight in the external fi xation group. While these results may suggest that external fi xation is bene fi cial for the management of displaced proximal humerus fractures, the sample size is too small to yield de fi nitive conclusions. Furthermore, several studies have shown results opposite to those of Kristiansen et al. As mentioned above, Court- Brown et al. did not fi nd an improvement in outcomes in displaced two-part frac- tures treated surgically versus those treated nonoperatively [ 8 ] . Similar results were found by Zyto et al., who randomized 40 patients with displaced fractures (37 three- part and 3 four-part fractures) to either nonoperative treatment or open reduction and tension band fi xation. At 1 year follow-up, the mean Constant score was 60 in the operative group and 65 in the nonoperative group. No differences were found with regard to pain, range of motion, strength, and activities of daily living sub- scales. Complications in the nonoperatively treated group included posttraumatic arthritis in two patients. In the operatively treated group, six patients had seven complications: including infection, AVN, nonunion, posttraumatic arthritis, pulmo- nary embolus, and K-wire penetration [ 14 ] . In a recent matched controlled cohort study, Sanders et al. compared the results of 18 fractures in 17 patients (13 three part) treated with a locking proximal humerus plate to a matched nonsurgical con- trol group (18 patients, 14 three part) [ 15 ] . Signi fi cantly better range of motion was achieved in the nonoperative group. While not statistically signi fi cant, patient satis- faction and American Shoulder and Elbow Surgeons (ASES) self-assessment score also favored nonsurgical treatment. In addition, 56% ( n = 10) of the patients in the operative group required additional treatment for their shoulder, compared to only 11% ( n = 2) in the nonoperative group ( p = 0.005). 1097 Proximal Humerus Fracture Perhaps the best available evidence applicable to our patient with a two-part surgical neck and minimally displaced greater tuberosity fracture is data from a prospective randomized controlled trial by Olerud et al. This study compared out- comes of 60 three-part fractures treated nonoperatively or with locked plating dem- onstrating no difference in overall functional outcomes [ 16 ] . While AP’s fracture does not meet Neer’s strict de fi nition for a three-part fracture, the minimally dis- placed greater tuberosity fracture must be taken into consideration when determin- ing treatment options. In Olerud’s study, complications were common in the operative group and included screw perforation of the articular surface in 17% of patients, with 30% requiring reoperation. However, only 14% of patients in the nonoperative group had fractures that healed in a good position. It must be noted that although patients treated nonoperatively have similar functional outcomes than those treated operatively, only a small percentage actually heal in a good radio- graphic position. This information is important when counseling the patient on treatment and expectations of that treatment. Historically, conservative management has involved extended immobilization, followed by progressive range of motion. A recent prospective randomized con- trolled trial comparing immediate or early supervised physical therapy to sling immobilization for 3 weeks followed by physical therapy determined that early therapy led to less pain with movement, less pain at night, and less frequent sleep disturbances with 2 years of follow-up [ 17 ] . While this study was only evaluating stable, minimally displaced two-part proximal humerus fractures, should nonopera- tive treatment be pursued in AP, she would likely bene fi t from early supervised physical therapy as opposed to prolonged immobilization. If nonoperative management is chosen for AP, several aspects have to be taken into account, including residual malunion and loss of strength and range of motion. On the contrary, while radiographic alignment may be improved with surgical man- agement, this may not translate into improved functional outcome, especially in the lower demand elderly population. Furthermore, risks associated with surgical treat- ment, such as screw or k-wire penetration and infection, have to be weighed against thebene fi ts of improved reduction, which does not necessarily seem to correlate with improvement in function in this injury pattern [ 18 ] . Surgical Treatment Displaced fractures of the proximal humerus are frequently cited as indications for surgery. Several treatment methods have been studied and are presented below. Each of these treatments can be used for a two-part proximal humerus fracture with a minimally displaced greater tuberosity fracture. However, before proceeding with a particular technique, the surgeon must be aware of the unique complications asso- ciated with that technique and the variation in functional outcomes common between different techniques. 110 D.J. Stinner et al. Closed Reduction and Percutaneous Fixation Concerns about fragment viability in multipart fractures due to extensive soft tissue dissection with open techniques have led to the development of less invasive techniques. Many case series have demonstrated good results in patients with minimally displaced two-part and three-part fractures treated with closed reduction and percutaneous fi xation, whether with screws, wires, or hybrid techniques. Less reliable results are described when these techniques are used in displaced or four- part fractures. Resch et al. studied 27 patients, 9 with three-part and 18 with four- part fractures (13 of which were valgus type) treated with percutaneous reduction and screw fi xation. After an average follow-up of 24 months (range 18–47) good to very good functional results were seen in the three-part fracture group, with an aver- age Constant score of 91% (84–100%) adjusted for age and gender, and no signs of avascular necrosis [ 19 ] . A subsequent study by the same senior author evaluated the results of 76 patients of at least 70 years of age with displaced three- or four-part fractures after percutaneous reduction and fi xation using a proximal humerus speci fi c external fi xator. 50 patients with 51 fractures completed a mean follow-up of 34 months (range 6–81). Primary healing was achieved in 90% of cases. Mean Constant scores for the treated shoulder were 85% (three-part fractures) and 69% (four-part fractures) of the Constant score for the contralateral, uninjured shoulder. Complications included displacement of fragments or migration of Kirschner wires in fi ve cases and AVN of the humeral head in four patients. Of these, three patients were revised to arthroplasty [ 20 ] . Calvo et al. reported the results of 50 of 73 patients treated with large diameter Kirschner wire fi xation after fl uoroscopically assisted reduction. There were 27 surgical neck two-part fractures, 17 three-part and 6 four- part or head splitting fractures. After a mean follow-up of 14 months (range 12–26), Constant scores were on average 82%, 89%, and 68% of scores of the contralateral shoulder for two-, three-, and four-part fractures, respectively. Overall the authors reported wire migration in 36% of cases and loss of reduction in 10% of fractures. One patient required revision closed reduction and percutaneous pinning, one patient developed a nonunion, two patients had a super fi cial pin tract infection and four developed AVN of the humeral head. While malunion of at least one of the fracture parts was found in 28% of cases, this did not correlate with decreased function [ 21 ] . In a prospective multicenter study, Keener et al. reported on the outcomes of 27 patients with displaced proximal humerus fractures with at least 1 year of follow-up (mean 35 months, range 12–77). Mean patient age was 60 years (range 41–79) and fractures were classi fi ed as two-part in 7 cases, three-part in 8 cases and four-part in 12 cases. At fi nal follow-up mean VAS pain scores were 0.89, 1.5 and 1.75 (0 = no pain, 10 = worst pain) for two-part, three-part and four-part fractures, respectively. ASES and Constant scores were 87.9, 85.1 and 79.7, and 78.7, 78.6, and 67.2 points, respectively. All fractures healed, but one 4-part valgus impacted fracture did how- ever develop AVN of the humeral head with associated severe pain. Complications included one early pin loosening, one varus malunion and one posttraumatic arthro fi brosis of the shoulder that required surgical release [ 22 ] . Brunner et al. further 1117 Proximal Humerus Fracture studied the outcomes of closed reduction and percutaneous fi xation using a proximal humerus-speci fi c external fi xator in a prospective case series. Fifty-eight consecutive patients with displaced proximal humerus fractures completed a mini- mum of 12 months of follow-up. The mean VAS pain score was 1.1 points, and the mean Constant score representing the percentage compared to the unaffected shoulder was 88%. Secondary Kirschner wire perforation of the humeral head was observed in 22% of cases. Revision surgery was required in fi ve patients [ 23 ] . Comparing displaced fractures treated with closed manipulation versus external fi xation, Kristiansen et al. demonstrated better results obtained in those treated with external fi xation [ 13 ] . However, this study only included 31 patients, making it dif fi cult to derive de fi nitive conclusions based on the data presented. A retrospective analysis of 41 patients with two-, three-, or four-part fractures treated with closed reduction and percutaneous fi xation demonstrated good results in patients with two- part and three-part fractures. Of the four included four-part fractures, three had AVN, and none achieved satisfactory results based on ASES scores. Furthermore, the authors found that the use of smooth Kirschner wires led to a higher failure rate than termi- nally threaded pins [ 24 ] . Complications of treatment included collapse of the humeral head, pin loosening, and pin migration. For these reasons authors advocate that should this technique be used, one should use terminally threaded pins, ensure restoration of medial support, accurately reduce the tuberosities, and have close follow-up to avoid catastrophic complications that can be associated with pin migration. Based on the available literature, this technique may result in satisfactory out- comes for AP to include low VAS pain score and average Constant scores of approx- imately 82–91% of the contralateral, uninjured shoulder. However, the authors would not recommend this treatment for AP due to the high rate of complications associated with it, such as super fi cial pin tract infections, loss of reduction, and pin migration. Tension Band Fixation Penetration of screws and pins into the articular surface of the humeral head is one of the most common complications of plate and percutaneous pin fi xation. Fracture fi xation with the use of tension bands using either wires or suture has therefore been proposed to mitigate this complication, while providing adequate fi xation strength to allow bony healing. Tissue at the tendon bone interface allows for strong fi xation, especially in the setting of osteoporosis where osseous fi xation is unreliable. Suture fi xation has been proposed mainly in two-part fractures involving the tuberosities, however some authors have advocated its use even for four-part fractures [ 25 ] . Park et al. retrospectively reviewed the outcomes of open reduction and fi xation with rotator cuff-incorporating sutures for the treatment of displaced two-part and three- part proximal humerus fractures. Thirteen greater tuberosity, nine surgical neck, and six greater tuberosity-surgical neck three-part fractures were included. After a mini- mum follow-up of 1 year, 89% of patients had an excellent or satisfactory result. The mean ASES score was 87.1 (range 35.0–100.0). All fractures achieved 112 D.J. Stinner et al. radiographic healing without evidence of AVN of the humeral head. No differences inoutcomes were found among fracture types [ 26 ] . Dimakopoulos et al. further support suture fi xation of proximal humerus fractures as an inexpensive alternative to achieve suf fi cient fracture stabilization to allow early passive range of motion, while avoiding extensive soft-tissue dissection. These authors reported the results of suture fi xation performed on 188 patients over an 11 year period. 165 patients (mean age 54 years, range 18–75) with 45 (27%) four-part, fractures valgus impacted, 64 (39%) three-part, and 56 (34%) two-part greater tuberosity fractures completed follow-up. With the exception of two greater tuberosity fractures, all patients healed without requiring further surgery. After a mean of 5.4 years follow-up, malunion had occurred in nine patients (5%), one-third of which were a consequence of sec- ondary displacement. Humeral head AVN was observed in 11 (7%) fractures, four of which had complete subchondral collapse. Absolute mean Constant score was 91 points (range 54–100), while the mean Constant score as a percentage of the oppo- site shoulder was 94% (range 60–100) [ 27 ] . Zyto et al. performed one of the few randomized studies involving surgical treat- ment of proximal humerus fractures. In their study, 37 patients with three-part and 3 with four-part fractures were randomized to either nonoperative treatment or open reduction and tension band fi xation. As described earlier, no differences were found with regard to pain, range of motion, strength, and activities of daily living. Complications in the nonoperatively treated group included posttraumatic arthritis in two patients. In the operatively treated group, six patients had seven complica- tions, including infection, AVN, nonunion, posttraumatic arthritis, pulmonary embolus, and K-wire penetration [ 14 ] . Tension band fi xation avoids extensive dissection, promoting good rates of heal- ing. However, when compared to nonoperative management there is no reported difference in pain, range of motion, strength, and ability to perform activities of daily living. While tension band fi xation may minimize complications such as screw or wire penetration, it cannot eliminate other risks associated with surgery such as infection. For these reasons, coupled with the dif fi culty in achieving appropriate fi xation of the surgical neck fracture with the use of a tension band construct, the authors would not recommend tension band fi xation for our patient, AP, with a two- part surgical neck and minimally displaced greater tuberosity fracture. Plate and Screw Fixation Open reduction and internal fi xation has many theoretical advantages, especially with the advent of modern anatomic locking plates. These constructs have demon- strated superiority in both biomechanical and clinical studies. Despite this, there appears to be a signi fi cant learning curve, as demonstrated by the high rate of com- plications associated with technical errors of screw and plate placement. Various methods of conventional plate fi xation for displaced proximal humeral fractures have been used with relative success in prior studies, including one-third 1137 Proximal Humerus Fracture tubular plates fi xed orthogonally from one another [ 28, 29 ] . Common problems with these techniques include loss of fi xation and malunion as a result of the poor bone quality common in this region in the elderly, and AVN likely due to contribut- ing factors of both the injury and extensive exposure required for these techniques. In an effort to minimize loss of fi xation, some surgeons have augmented their fi xation with bone substitutes to fi ll bony defects in the proximal humerus with good results [ 28 ] . Initial data described the bene fi ts of fi xed angled constructs such as decreased screw cutout and hardware failure [ 30 ] which made it a promising technology for application to the proximal humerus due to the poor bone quality often seen in this region. With the advent of anatomic fi xed angle constructs, speci fi cally the locking proximal humeral plate, their biomechanical superiority over other commonly used implants was quickly realized. While many thought the locking plate was going to be the panacea for the proxi- mal humerus fracture in a patient with poor bone quality, it has failed to reliably produce improved results in this patient population. In one of the earlier reports on use of a locked plate for proximal humerus fractures, Bjorkenheim et al. reported good functional results with an average Constant score of 77 at 1 year and few com- plications consisting of: two cases of implant failure due to technical errors, three cases of avascular necrosis, and two nonunions in a retrospective analysis of 72 patients [ 31 ] . All patients self-reported that they were able to return to their prein- jury activity level, and 18 of 23 patients that had been working prior to their injury returned to their previous occupation. In more recent, larger series, complications and unplanned reoperations are more common than initially reported. In a prospective, observational series by Brunner et al. of 157 patients with 158 proximal humerus fractures (42% were three-part fractures) fi xed with a locking plate, there were 39 unplanned repeat surgeries (25%) [ 32 ] . Screw perforation occurred in 15% of patients, and avascular necrosis of the humeral head occurred in 8%. Despite the high rate of complications, patients had an average Constant score of 72. Another prospective, observational study by Sudkamp et al. of 187 patients had a similarly high rate of unplanned secondary operations ( n = 29, 19%), with 62 complications seen in 52 patients [ 33 ] . Twenty- fi ve (40%) of the complications were related to incorrect surgical technique, with the most common mistake being intra-operative screw perforation of the humeral head ( n = 21, 14%). One important point is that in the study by Brunner et al., each surgeon treated on average less than three patients, which makes this data more generalizable [ 32 ] . This fact, together with the high rate of complications associated with incorrect surgical technique, demonstrates the steep learning curve associ- ated with the treatment of these injuries. A recent systematic review of locking plate fi xation was performed by Sproul et al. [ 34 ] . Only studies with more than 15 patients, patients older than 18 years of age, a minimum 18 month follow-up, at least one functional outcome score, and an overall study quality score of 5/10 or better, were included. Five-hundred and four- teen proximal humerus fractures (34% two part, 45% three part, and 21% four part) were included. The average Constant score was 72 for patients with three-part 114 D.J. Stinner et al. fractures. There was a high overall complication rate (49%), with 14% requiring secondary surgeries, most commonly due to screw perforation of the articular surface. Avascular necrosis of the humeral head occurred in 10% of patients, but only 4 of the 51 af fl icted patients required conversion to hemiarthroplasty. One interesting fi nding in this systematic review was that the complication rate decreased from 49% to 33% when varus malreduction, which is typically caused by a lack of medial support, was excluded. Gardner et al. demonstrated the impact that the absence of medial support has on subsequent reduction loss, increased rate of screw perforation, and overall loss of height by critical retrospective appraisal of radio- graphs of proximal humerus fractures treated with locked plating [ 35 ] . The impor- tance of the medial support was also demonstrated in a study by Yang et al., who showed that in a prospective observational study of 64 consecutive patients treated with a locking proximal humeral plate, thosewith an intact medial support had signi fi cantly better functional outcomes at 1 year (Constant score of 81 (medial support) versus 65 (no medial support), p = 0.002) [ 36 ] . While valgus impacted fractures are the most common in the elderly [ 9 ] , they also tend to have better outcomes than those with initial varus deformities [ 37, 38 ] . In a retrospective analysis of 45 fractures in 44 patients, Lee et al. showed that a delay in rehabilitation due to medical comorbidities and a decreased head-neck shaft angle incured by the lack of medial support were both prognostic for poor outcomes in patients treated with a locked plate [ 39 ] . Although our patient, AP, has an initial valgus deformity, these are important parameters to recognize when considering treatment options in proximal humerus fractures. In a matched controlled cohort study of 18 fractures treated with locked plating compared to 18 matched fractures treated nonoperatively, Sanders et al. showed better range of motion ( fl exion, abduction, and external rotation) in the nonopera- tive group [ 15 ] . ASES scores were better in the nonoperative group, but it did not reach statistical signi fi cance. In addition, patients treated with locked plating required ten secondary surgeries compared to two in the nonoperative group ( p = 0.0005). Despite this study demonstrating superior outcomes in patients treated nonoperatively, the fi ndings are potentially biased in that the study was retrospec- tive and treatment was not randomized. Although the control group was matched, surgery may have been done on more challenging fractures. The fi rst prospective randomized trial speci fi c to displaced three-part fractures comparing operative treatment with a locked plate to nonoperative treatment has recently been reported by Olerud et al. who randomized 60 patients and followed them for 2 years (16). While range of motion and Health Related Quality of Life scores were marginally superior in the operative group, there was not a statistically signi fi cant difference with the numbers available. Despite this trend toward improved outcomes in patients treated operatively, 30% required secondary surgeries (13% major, 17% minor). When comparing intamedullary nail fi xation of displaced two-part, three-part, and four-part proximal humerus fractures to locked plating in a prospective obser- vational study of 152 patients (76 per group) by Gradle et al., both groups had a similar rate of complications (22 in the locking plate group and 17 in the 1157 Proximal Humerus Fracture intamedullary nail group, p = 0.458) [ 40 ] . Although not signi fi cant, major compli- cations requiring revision surgery were more common in those treated with intramedullary nail fi xation (13 vs. 9, p = 0.49). Constant scores tended to be better at 1 year in those treated with intramedullary nail fi xation, 83 ± 15 versus 75 ± 20 ( p = 0.075), but this difference was not signi fi cant. This data is contrary to the results of a prospective randomized trial of two-part fractures treated with either a locking plate or locked intramedullary nail by Zhu et al. [ 41 ] . In that study Zhu et al. demonstrated that those treated with a locking plate had signi fi cantly better average ASES scores, median VAS scores for pain, and average supraspinatus strength at 1 year. No differences in any parameters measured were seen at 3 years. Despite the improvement in functional outcomes seen at 1 year in those treated with locked plates, they did have a signi fi cantly higher rate of complications than those treated with intamedullary nail fi xation (31% vs. 4%), with screw penetra- tion of the articular surface occurring in 5 (19%) of those treated with a locked plate. Although Gradle et al. demonstrated higher Constant scores in the group treated with intramedullary nail fi xation, the authors would not recommend this method of treatment for AP, given the risk of further displacement of her mini- mally displaced greater tuberosity fracture and the higher rate of complications seen with intramedullary nail fi xation. In a prospective nonrandomized comparison between patients treated with open reduction and internal fi xation or hemiarthroplasty by a single surgeon, Bastian et al. demonstrated that patients treated with open reduction and internal fi xation had better Constant scores at latest follow-up (mean 5 years) ( p = 0.02), but no dif- ference was seen when patients subjectively rated their shoulder on a 100 point scale ( p = 0.93) [ 42 ] . It must be emphasized that this study was not randomized, and patient selection for each procedure depended on bone quality and preservation of the blood supply to the humeral head as measured by laser-Doppler fl owmetry intraoperatively. Olerud et al. evaluated the quality of life and functional outcomes of 50 patients with 2-part proximal humerus fractures after treatment with a locked plate [ 43 ] . Patients demonstrated signi fi cant improvement in their Constant score out to 1 year. However, all values obtained for the HRQoL (EQ-5D index score) were signi fi cantly lower than before the fracture ( p < 0.001). This information is important when coun- seling patients on expectations following treatment of these injuries with open reduction and internal fi xation with a locked plate. In summary, unplanned reoperations are common secondary to incorrect surgical technique with open reduction and internal fi xation of proximal humerus fractures. In fact, Bell et al. demonstrated by comparing Medicare data from 2004 to 2005 that after the introduction of locking plates, there has been a higher rate of early (<12 months post injury) reoperation than when conventional plates were used in 1999–2000 (odds ratio = 1.47, p = 0.043) [ 44 ] . The current data suggests that there is no difference in functional outcomes with surgical treatment of proximal humerus fractures that are similar to our patient, AP, when compared to nonoperative management. However, should surgery be considered, the medial support must be restored to minimize the risk of subsequent loss of reduction and worse functional outcomes. 116 D.J. Stinner et al. Intramedullary Nail Fixation Antegrade nailing of displaced proximal humerus fractures has gained popularity due to its minimally invasive nature and satisfactory outcomes reported in many case series. However, concern for further injury to the rotator cuff through nail insertion has left many skeptical of this technique. A retrospective study of displaced three and four-part fractures comparing those treated with antegrade intramedullary nail fi xation to nonoperative treatment dem- onstrated better Constant scores in those treated nonoperatively. Complications were also more frequent in those treated operatively (42%) compared to those treated nonoperatively (25%) [ 45 ] . Several studies have demonstrated satisfactory to excellent outcomes based on Constant scores in up to 80% of patients with proximal humerus fractures treated with intramedullary nail fi xation [ 46– 49 ] . However, many of these studies are ham- pered by the lack of a comparison group. A retrospective study by Kazakos et al. showed no difference between patients with a two or three-part proximal humeral fracture treated with an intramedullary nail [ 49 ] . However, not surprisingly, it appears that with increasing fracture complexity, functional outcomes worsen, as demonstrated in a retrospective study by Adedapo and Ikpene, who showed that patients with three-part fractures performed better than four part fractures at 1 year, using the Neer Shoulder Score (89 vs. 60) [ 50 ] . In another retrospective study by Sosef et al., 28 patients treated with an intramedullary nail for a proximalhumerus were followed for 1 year [ 51 ] . When strength was excluded from the Constant score, their patients had a mean score of 81% of the maximum allowable score of 75 points. However, the exclusion of strength from the Constant score arti fi cially in fl ates this value. Nine patients had complications ranging from one two-part fracture that developed avascular necrosis requiring hemiarthroplasty to three patients with proximal screw migration requir- ing removal. Mittlmeier et al. reported on 221 patients that underwent locked intramedullary nailing of a proximal humerus fracture [ 52 ] . They demonstrated continued func- tional improvements at 1 year, with a fi nal Constant score of 78.8 ± 16.6, which equated to 85.7 ± 15.9% of the unaffected side. Despite these promising results, which are comparable to other studies evaluating intramedullary nail fi xation of these injuries, there were 59 complications in 115 patients, with an average follow- up of 9 months. The most common complication involved backing out of the screws, which was seen in 22.6% of patients. Radiographic signs of AVN developed in nine patients, eight of which had mild clinical symptoms. Only one of the patients with AVN went on to require revision to hemiarthroplasty. Linhart et al. reported the outcomes of 51 patients with a minimum of 1 year follow-up after treatment with an intramedullary nail [ 53 ] . Twenty percent of the patients had complications (10/51), with loosening of the proximal screws being the most common complication. Three patients developed AVN of the humeral head that required subsequent revision to hemiarthroplasty. As is common with 1177 Proximal Humerus Fracture plate fi xation, there was signi fi cant improvement in the Constant score at 1 year, averaging 71.2 ± 16.1, which was 82.1 ± 14.1% of the noninjured side. No differ- ences were seen in functional outcomes between two-part, three-part, and four-part fractures. While no prospective randomized trials have compared antegrade intramedul- lary nail fi xation to other treatment modalities for three-part fractures, Zhu et al. did compare antegrade nailing to plate fi xation in two-part surgical neck fractures [ 41 ] . Fifty-one consecutive patients were included and followed for 3 years post- operatively. All patients healed within 3 months. At 1 year, there was a signi fi cantly higher complication rate in patients treated with locking plates (31% vs. 4%, respectively, p = 0.02). On the other hand, patients treated with locking plates had better average ASES scores, median VAS scores, and rotator cuff strength. At 3 years there was no difference between groups. While the complication rate asso- ciated with plate fi xation appears on par with what is reported in the literature, the improved function in this group at 1 year over the intramedullary nail group may be the result of rotator cuff injury during nail insertion, as proposed by the authors. Many of the studies presented here on intramedullary nail fi xation of proximal humerus fractures are compromised by the lack of a comparison group, which must be taken into account when analyzing the results. Con fl icting data exists regarding the complication rate associated with this method of treatment as some studies report much higher rates of approximately 20% [ 52, 53 ] , compared to others as low as 4% [ 41 ] . Despite this, it must be mentioned that antegrade intramedullary nail fi xation for AP’s proximal humerus fracture will likely result in rotator cuff symp- toms postoperatively [ 41, 54 ] , and may require additional exposure to ensure secure fi xation of the greater tuberosity to prevent tuberosity nonunion. Arthroplasty As the data presented below will detail, the authors feel that arthroplasty should not be considered as a treatment option for AP given her age and fracture pattern–a two- part surgical neck fracture with a minimally displaced greater tuberosity fracture. However, the following treatment options are presented for completeness and under- standing of all potential options available to the surgeon treating proximal humerus fractures. Hemiarthroplasty While preservation of the humeral head is a desirable goal of proximal humeral fracture treatment, it is frequently not possible due to severe comminution or marked osteoporosis. Furthermore, some fracture types, speci fi cally fracture 118 D.J. Stinner et al. dislocations and those involving the anatomic neck have a high risk of AVN of the humeral head, leading to signi fi cantly lower functional outcomes [ 55 ] . Hemiarthroplasty has therefore been proposed in this setting. Shoulder hemiarthroplasty for proximal humerus fractures differs from hemiarthroplasty in osteoarthritis in that standard surgical landmarks are lost due to proximal metaphy- seal fragmentation and poor bone stock. Boileau et al. demonstrated that tuberosity, malposition, or loss of reduction; and inadequate restitution of the relationship between the prosthetic head; and the greater tuberosity negatively impacts outcome after shoulder hemiarthroplasty for fracture [ 56 ] . Suture fi xation around the ten- don-bone interface of the rotator cuff should be performed and tied around the neck of the implant in order to allow secure and stable fi xation of the tuberosities. Moeckel et al. reported on the results of 22 consecutive patients (average age 70 years, range 49–87) with 5 three-part, 13 four-part, and 4 head-splitting frac- tures treated with a modular implant. After a mean follow-up of 36 months and based on the Hospital for Special Surgery shoulder scoring system, there were 13 excellent and 7 good results, and only 2 failures. Moderate to severe pain was found only in the two failures and subsequently required revision to a total shoulder arthroplasty. Results correlated inversely with age [ 57 ] . Demirhan et al. retrospec- tively assessed the clinical outcomes of 32 patients (mean age 58, range 37–83) with 15 four-part, 2 three-part fractures, and 15 fracture dislocations. After a mean follow-up of 35 months (range 8–80 months), excellent or good results were obtained in 24 cases (75%) according to Neer’s criteria. The mean Constant score was 68 (range 19–98) and mean elevation 113° (range 30–180). Only one patient complained of signi fi cant residual pain. Half of the patients had either a nonunion or displaced their tuberosities, thereby adversely affecting their functional outcome [ 58 ] . Similar results were found by Mighell et al., who described their results in 72 shoulders. At fi nal follow-up a pain free shoulder was achieved in 93% of cases, with mean ASES and Simple Shoulder Test scores of 77 and 7.5 respectively. Greater tuberosity malunion and superior migration of the humeral head correlated with worse functional outcomes [ 59 ] . One of the largest series on hemiarthroplasty for proximal humerus fractures was published by Robinson et al. who reported on 138 patients who were followed for at least 1 year after arthroplasty. Prosthetic survival was 97% at 1 year, 95% at 5 years and 94% at 10 years. Pain control was found to be the most reliable outcome, with a median value for pain on the Constant Score of 15 points. However, greater variability in outcome was found for function, range of motion, and strength [ 60 ] . Kontakis et al. reviewed the outcome of 28 fractures (mean age 66 years, range 38–80) after hemiarthroplasty using a fracture dedicated implant. After a mean follow-up of 39 months, the Constant score aver- aged 68 points (range 37–84), which represented 89.5% of the contralateral side. As with other series, anatomic reconstruction correlated with higher functional scores as well as improved range of motion. A total of 18 patients had active ante- riorelevation ³ 150°. Patients were very satis fi ed or satis fi ed in 24 cases, while four were either dissatis fi ed or disappointed, and only two patients had moderate to severe pain [ 61 ] . 1197 Proximal Humerus Fracture Long-term outcomes after proximal humerus fractures treated with hemiar- throplasty have not been amply studied to date, but in general have shown dete- riorating outcomes with time. Antuna et al. included 57 patients (mean age 66 years, range 23–89) that had completed a minimum follow-up of 5 years (mean 10 years). According to Neer’s criteria, results were satisfactory in 27 patients and unsatisfactory in 30. Mean active elevation and external rotation were 100° (range 20–180) and 30° (range 0–90) respectively. Revision surgery was required in two cases, and 9 (16%) reported moderate to severe pain [ 62 ] . Comparative data on hemiarthroplasty and fracture fi xation is very limited. One of the few comparative studies was performed by Bastian and Hertel, who com- pared the outcomes of a cohort of 100 displaced proximal humerus fractures (98 patients) that were assigned to either open reduction internal fi xation with locked plating ( n = 51, median age 54 years, range 21–88) or hemiarthroplasty ( n = 49, median age 66 years, range 38–87) based on intraoperative assessment of humeral head ischemia. Seventy six patients completed a mean follow-up of 5 years (range 3–7). The median Constant score was 77 (range 37–98) after fi xation and 70 (range 39–84) after hemiarthroplasty. The median Subjective Shoulder Value was 92 (range 40–100) after fi xation and 90 (range 40–100) after hemiarthroplasty [ 42 ] . Based on these results similar outcomes could be expected after locked plate fi xation in proximal humerus fractures with nonischemic humeral heads, and hemiarthro- plasty when intraoperative head vascularity is absent. Solberg et al. similarly com- pared locked plate fi xation ( n = 38) and hemiarthroplasty ( n = 48) in patients with three- and four-part fractures. Similar to the previous study, arthroplasty was cho- sen when reconstruction was deemed not to be possible. The mean Constant score was signi fi cantly better for the locked-plate group (68.6 points) compared to the hemiarthroplasty group (60.6 points). Avascular necrosis of the humeral head, which occurred in six patients (19%), led to functional scores that were signi fi cantly lower than fi xed fractures without AVN. However, these results were similar to those after hemiarthroplasty. Strength, pain, and activities of daily living were similar, while range of motion was signi fi cantly higher in the fi xation group. Complications were common in the fi xation group and included AVN in six patients, screw perforation of the humeral head in six patients, loss of fi xation in four patients, and wound infection in three patients. Complications in the hemiarthroplasty group included nonunion of the tuberosity in seven patients and wound infection in three patients [ 37 ] . A recent systematic review of 810 hemiarthroplasties performed acutely for proximal humerus fractures ranging from two-part to head-splitting fractures dem- onstrated that most patients reported minimal pain with a mean Constant score of 56.6 [ 63 ] . Infection was not common, with super fi cial infections reported in 1.6% and deep infection in 0.6%. Complications related to the fi xation and healing of the tuberosities was more common at 11.2%. Heterotopic ossi fi cation was also com- mon, occurring in 8.8% of cases, but did not appear to limit function. Despite the minimal pain at fi nal follow-up and the relatively low rate of complications when compared to open reduction and internal fi xation, unsatisfactory results were 120 D.J. Stinner et al. observed in 42% of patients on subjective examination. One interesting fi nding in this study was the low number of cases performed per surgeon per year (average 2.96, range 0.21–9.6), which may be a contributing factor to the poor patient outcomes [ 63 ] . In summary, data comparing hemiarthroplasty to plate fi xation is often biased due to the lack of randomization as surgeons typically performed hemiarthroplasty in nonreconstructable fractures or in fractures that lacked humeral head vascularity. Despite this, several important points can be taken from the current evidence. First, tuberosity malunion is common and can negatively impact functional outcomes. Second, moderate to severe pain is seen in up to 16% of patients, with approximately 75% of patients in most series reporting good to excellent results. Finally, while long-term studies are lacking, one study has demonstrated 10 year prosthetic survival of 94% [ 60 ] . Reverse Total Shoulder Arthroplasty Due to the high incidence of tuberosity malreduction or secondary displacement and their correlation with poor functional outcomes after hemiarthroplasty of multipart proximal humeral fractures, some authors have proposed reverse total shoulder arthroplasty as an alternative option. Due to its geometry and character- istic biomechanical behavior, reverse shoulder arthroplasty does not rely on a competent rotator cuff, thereby potentially yielding improved outcomes after tuberosity nonunion or malunion and fracture associated rotator cuff tears. Bufquin et al. presented their results of reverse total shoulder arthroplasty in 43 consecu- tive three- or four-part proximal humerus fractures (mean patient age 78 years, range 65–97). After an average follow-up of 22 months, the mean-adjusted Constant score was 66 (range 25–97). Anterior elevation and external rotation were on average 97° (range 35–160) and 30° (range 0–80) respectively. At fi nal follow-up, 10 cases (25%) showed scapular notching [ 64 ] . Cazeneuve et al. reported the outcome of 36 fractures (mean age 75 years, range 58–92) a mean of 6.6 years after having been treated with reverse total shoulder arthroplasty. The mean Constant score was 53 points (range 20–80), which represented 67% of the values for the contralateral shoulder. Complications included two cases of com- plex sympathetic dystrophy, one infection, and four dislocations (one anterior, three superior). Loosening was seen in two glenoid components and one humeral component. The rate of scapular notching was alarmingly high, found in 19 patients (53%) [ 65 ] . Gallinet et al. reported a comparative study of 40 patients with proximal humerus fractures treated with either hemiarthroplasty or reverse total shoulder arthroplasty. Those in the hemiarthroplasty group reported better internal and external rotation. Those in the reverse group demonstrated better shoulder abduc- tion, forward elevation, and Constant scores. No difference was seen between 1217 Proximal Humerus Fracture groups with respect to DASH scores. While three patients in the hemiarthroplasty group had tuberosity malunions, 15 of the patients in the reverse arthroplasty group had scapular notching [ 66 ] . Young et al. published the outcomes in a small comparative study of ten patients (mean age 77) who underwent reverse total shoulder arthroplasty and ten patients (mean age 75) who had undergone hemiar- throplasty. At a minimum follow-up of 22 months, ASES scores were 65 (range 40–88) in the reverse group and 67 (26–100) in the hemiarthroplasty group. Similar values for forward elevation and external rotation were found in both groups [ 67 ] . Given that the majority of poor outcomes following arthroplasty for complex proximal humeral fractures are rotator cuff and tuberosity related, an implant that does not rely on these structures for function has obvious attractions. However, based on the currently available literature, caution should be advised in using reverse total shoulder replacements as the primaryimplant for severe proximal humeral fractures. Long-term outcomes have not yet become available, rotational weakness is common, and the anticipated dramatic improvement in functional results com- pared to conventional prostheses has not materialized. Additionally, the complica- tion rate seems to be intrinsically high, and the incidence of scapular notching, at least with prior implants and techniques, is of concern. However, in the presence of an irreparable rotator cuff tear, extreme osteoporosis with marked tuberosity comminution/displacement, or delayed presentation, reverse total shoulder arthroplasty is a reasonable alternative in an elderly (>70 years of age), low-demand individual. Literature Inconsistencies As noted by the most recent Cochrane Review [ 4 ] , there are not enough prospective randomized controlled trials to de fi nitively determine the superior treatment for each particular fracture pattern: inconsistencies within the available literature were discussed in their respective sections. The trend towards more prospective random- ized studies comparing treatment methods is encouraging, but further study is needed to appropriately guide the decision making process. Evidentiary Table and Selection of Treatment Method The key studies in treating AP are shown in Table 7.1 . Based on the available literature, the authors feel that the best treatment in this case would be nonoperative treatment, as similar functional outcomes can be reliably obtained without the added complications associated with operative fi xation of AP’s fracture. 122 D.J. Stinner et al. Table 7.1 Evidentiary table: a summary of the quality of evidence for the treatment of proximal humeral fractures Author (year) Description Summary of results Level of evidence Sudkamp et al. (2009) Prospective observational study 187 patients with proximal humerus fractures treated with a locking plate. High rate of complications, including unplanned second operations in 19%. 40% of complications were associated with incorrect surgical technique. III Sanders et al. (2011) Matched controlled cohort study 18 fractures treated with locked plating compared to 18 treated nonoperatively demonstrating signi fi cantly more secondary surgeries in the operative group, and, while not signi fi cant, better ASES scores were achieved in the nonoperatively treated group. III Gradl et al. (2009) Prospective observational study Comparison between locked plating and antegrade nailing of proximal humerus fractures (76 per group) resulted in a similar rate of complications, although not signi fi cant, Constant scores tended to be better at 1 year in those treated with intramedullary fi xation. III Zhu et al. (2011) Prospective randomized Signi fi cantly better average ASES scores, median VAS score for pain, and average supraspinatus strength at 1 year in patients treated with locked plating when compared to intramedul- lary fi xation. However, there was a signi fi cantly higher rate of complica- tions seen in the locked plate group (31% vs. 4%). I Olerud et al. (2011) Prospective randomized 60 patients with a three-part proximal humerus fracture randomized to either locked plating or nonoperative treatment and followed for 2 years. No difference in range of motion and Health Related Quality of Life with numbers available, and of those who underwent surgery 30% required reoperation. I Court-Brown et al. (2002) Retrospective cohort study Of valgus-impacted fractures treated nonoperatively, 80% good to excellent results, 1 month to get to independent dressing, fracture displacement, and age were independently predictive of Constant score at 1 year. III (continued) 1237 Proximal Humerus Fracture Table 7.1 (continued) Author (year) Description Summary of results Level of evidence Hanson et al. (2009) Prospective observational 160 patients (124 completed 1-year follow-up) with proximal humerus fractures treated nonoperatively. 10.6% had displacement or loss of reduction, 7% nonunions. Smokers were 5.5 times more likely to have a nonunion and those who were employed, better DASH scores if employed, Constant score decreased with age. III Kristiansen and Kofoed (1998) Prospective randomized 31 displaced proximal humeral fractures were randomized to closed manipula- tion or percutaneous reduction and external fi xation. Of those with 1-year follow-up, 4 of the 10 patients with nonoperative treatment, and 8 of the 11 patients with operative treatment had satisfactory to excellent results according to their Neer score. There was one deep infection in the operative group, necessitating pin removal. I Zyto et al. (1997) Prospective randomized 40 patients (average age 74) with displaced three- and four-part fractures were randomized to nonoperative manage- ment or tension-band fi xation. No difference was seen in functional outcomes between groups at 1 year (including subjective assessment, Constant score, pain, range of motion, power, and ADLs). Major complications were only seen in the operative group. I Dimakopoulos et al. (2007) Retrospective cohort study 165 patients with proximal humerus fractures, including 64 three-part fractures, were treated with open reduction and internal fi xation with transosseous sutures. All fractures healed except two three-part fractures of the greater tuberosity. Mean Constant score was 91. AVN developed in 11 (7%). III Bastian and Hertel (2009) Prospective observational 98 patients underwent ORIF or hemiar- throplasty dependent on fracture morphology, as determined by a single surgeon. No difference was seen between groups in functional outcomes except in measuring force, which was signi fi cantly higher in the group that underwent plate fi xation. III (continued) 124 D.J. Stinner et al. Table 7.1 (continued) Author (year) Description Summary of results Level of evidence Robinson et al. (2003) Retrospective cohort study Arthroplasty was performed in 163 patients where it was found at surgery to have a fracture that was not amenable to ORIF. Prosthetic survival was 94% at 10 years. Constant score was 64 at 1 year. Patient age, need for reopera- tion, and degree of displacement of the tuberosities, among others, were predictive of the 1-year Constant score. III Solberg et al. (2009) Retrospective comparative study 38 patients treated with a locking plate were compared to 48 patients treated with hemiarthroplasty for three- and four-part fractures. Despite a higher rate of complications seen in the locked plating group, they had better outcome scores, with an average Constant score of 68.6 compared to 60.6. Those with an initial varus extension deformity had worse outcomes compared to valgus impacted fractures. III De fi nitive Treatment Plan and Prediction of Outcomes AP has an isolated two-part proximal humerus fracture with a minimally displaced greater tuberosity that is amenable to nonoperative treatment. Initial immobilization should last no longer than 1 week in order to control symptoms and allow for appro- priate pain management. Early rehabilitation should then be started. The available literature is insuf fi cient to allow de fi nitive conclusions as to which treatment offers the best outcome. However, with nonoperative treatment in this patient it would be reasonable to expect union, a pain-free shoulder, and a subjective satisfaction rate in close to 90% of cases.References 1. Kelsey JL, et al. Risk factors for fractures of the distal forearm and proximal humerus. The Study of Osteoporotic Fractures Research Group. Am J Epidemiol. 1992;135(5):477–89. 2. Visser CP, et al. Nerve lesions in proximal humeral fractures. J Shoulder Elbow Surg. 2001;10(5):421–7. 3. Neer 2nd CS. Displaced proximal humeral fractures. I. Classi fi cation and evaluation. J Bone Joint Surg Am. 1970;52(6):1077–89. 4. Handoll HH, Ollivere BJ. Interventions for treating proximal humeral fractures in adults. Cochrane Database Syst Rev. 2010(12): CD000434. 1257 Proximal Humerus Fracture 5. Koval KJ, et al. Functional outcome after minimally displaced fractures of the proximal part of the humerus. J Bone Joint Surg Am. 1997;79(2):203–7. 6. Keser S, et al. Proximal humeral fractures with minimal displacement treated conservatively. Int Orthop. 2004;28(4):231–4. 7. Tejwani NC, et al. Functional outcome following one-part proximal humeral fractures: a pro- spective study. J Shoulder Elbow Surg. 2008;17(2):216–9. 8. Court-Brown CM, Garg A, McQueen MM. The translated two-part fracture of the proximal humerus. Epidemiology and outcome in the older patient. J Bone Joint Surg Br. 2001;83(6):799–804. 9. Jakob RP, et al. Four-part valgus impacted fractures of the proximal humerus. J Bone Joint Surg Br. 1991;73(2):295–8. 10. Court-Brown CM, Cattermole H, McQueen MM. Impacted valgus fractures (B1.1) of the proximal humerus. The results of non-operative treatment. J Bone Joint Surg Br. 2002; 84(4):504–8. 11. Foruria AM, et al. The pattern of the fracture and displacement of the fragments predict the outcome in proximal humeral fractures. J Bone Joint Surg Br. 2011;93(3):378–86. 12. Hanson B, et al. Functional outcomes after nonoperative management of fractures of the proxi- mal humerus. J Shoulder Elbow Surg. 2009;18(4):612–21. 13. Kristiansen B, Kofoed H. Transcutaneous reduction and external fi xation of displaced fractures of the proximal humerus. A controlled clinical trial. J Bone Joint Surg Br. 1988;70(5):821–4. 14. Zyto K, et al. Treatment of displaced proximal humeral fractures in elderly patients. J Bone Joint Surg Br. 1997;79(3):412–7. 15. Chatrchyan S, et al. Search for supersymmetry in pp collisions at radical7 TeV in events with two photons and missing transverse energy. Phys Rev Lett. 2011;106(21):211802. 16. Olerud P, et al. Internal fi xation versus nonoperative treatment of displaced 3-part proximal humeral fractures in elderly patients: a randomized controlled trial. J Shoulder Elbow Surg. 2011. 17. Hodgson SA, et al. Rehabilitation of two-part fractures of the neck of the humerus (two-year follow-up). J Shoulder Elbow Surg. 2007;16(2):143–5. 18. Olerud P, et al. Internal fi xation versus nonoperative treatment of displaced 3-part proximal humeral fractures in elderly patients: a randomized controlled trial. J Shoulder Elbow Surg. 2011;20(5):747–55. 19. Resch H, et al. Percutaneous fi xation of three- and four-part fractures of the proximal humerus. J Bone Joint Surg Br. 1997;79(2):295–300. 20. Bogner R, et al. Minimally-invasive treatment of three- and four-part fractures of the proximal humerus in elderly patients. J Bone Joint Surg Br. 2008;90(12):1602–7. 21. Calvo E, et al. Percutaneous fi xation of displaced proximal humeral fractures: indications based on the correlation between clinical and radiographic results. J Shoulder Elbow Surg. 2007;16(6):774–81. 22. Keener JD, et al. Outcomes after percutaneous reduction and fi xation of proximal humeral fractures. J Shoulder Elbow Surg. 2007;16(3):330–8. 23. Brunner A, et al. Closed reduction and minimally invasive percutaneous fi xation of proximal humerus fractures using the Humerusblock. J Orthop Trauma. 2010;24(7):407–13. 24. Herscovici Jr D, et al. Percutaneous fi xation of proximal humeral fractures. Clin Orthop Relat Res. 2000;375:97–104. 25. Hertel R. Fractures of the proximal humerus in osteoporotic bone. Osteoporos Int. 2005;16 Suppl 2:S65–72. 26. Park MC, et al. Two-part and three-part fractures of the proximal humerus treated with suture fi xation. J Orthop Trauma. 2003;17(5):319–25. 27. Dimakopoulos P, Panagopoulos A, Kasimatis G. Transosseous suture fi xation of proximal humeral fractures. J Bone Joint Surg Am. 2007;89(8):1700–9. 28. Robinson CM, Page RS. Severely impacted valgus proximal humeral fractures. Results of operative treatment. J Bone Joint Surg Am. 2003;85-A(9):1647–55. 126 D.J. Stinner et al. 29. Wanner GA, et al. Internal fi xation of displaced proximal humeral fractures with two one-third tubular plates. J Trauma. 2003;54(3):536–44. 30. Egol KA, et al. Biomechanics of locked plates and screws. J Orthop Trauma. 2004; 18(8):488–93. 31. Bjorkenheim JM, Pajarinen J, Savolainen V. Internal fi xation of proximal humeral fractures with a locking compression plate: a retrospective evaluation of 72 patients followed for a mini- mum of 1 year. Acta Orthop Scand. 2004;75(6):741–5. 32. Brunner F, et al. Open reduction and internal fi xation of proximal humerus fractures using a proximal humeral locked plate: a prospective multicenter analysis. J Orthop Trauma. 2009; 23(3):163–72. 33. Sudkamp N, et al. Open reduction and internal fi xation of proximal humeral fractures with use of the locking proximal humerus plate. Results of a prospective, multicenter, observational study. J Bone Joint Surg Am. 2009;91(6):1320–8. 34. Sproul RC, et al. A systematic review of locking plate fi xation of proximal humerus fractures. Injury. 2011;42(4):408–13. 35. Gardner MJ, et al. The importance of medial support in locked plating of proximal humerus fractures. J Orthop Trauma. 2007;21(3):185–91. 36. Chatrchyan S, et al. Measurement of the t-Channel Single Top Quark Production Cross Section in pp Collisions at sqrt[s] = 7 TeV. Phys Rev Lett. 2011;107(9):091802. 37. Solberg BD, et al. Surgical treatment of three and four-part proximal humeral fractures. J Bone Joint Surg Am. 2009;91(7):1689–97. 38. Solberg BD, et al. Locked plating of 3- and 4-part proximal humerus fractures in older patients: the effect of initial fracture pattern on outcome. J Orthop Trauma. 2009;23(2):113–9. 39. Lee CW, Shin SJ. Prognostic factors for unstable proximal humeral fractures treated with locking-plate fi xation. J Shoulder Elbow Surg. 2009;18(1):83–8. 40. Gradl G, et al. Is locking nailing of humeral head fractures superior to locking plate fi xation? Clin Orthop Relat Res. 2009;467(11):2986–93. 41. Zhu Y, et al. Locking intramedullary nails and locking plates in the treatment of two-part proximal humeral surgical neck fractures: a prospective randomized trial with a minimum of three years of follow-up. J Bone Joint Surg Am. 2011;93(2):159–68. 42. Bastian JD, Hertel R. Osteosynthesis and hemiarthroplasty of fractures of the proximal humerus: outcomes in a consecutive case series. J Shoulder Elbow Surg. 2009;18(2):216–9. 43. Olerud P, et al. Quality of life and functional outcome after a 2-part proximal humeral fracture: a prospective cohort study on 50 patients treated with a locking plate. J Shoulder Elbow Surg. 2011;19(6):814–22. 44. Bell JE, et al. Trends and variation in incidence, surgical treatment, and repeat surgery of proximal humeral fractures in the elderly. J Bone Joint Surg Am. 2011;93(2):121–31. 45. van den Broek CM, et al. Displaced proximal humeral fractures: intramedullary nailing versus conservativetreatment. Arch Orthop Trauma Surg. 2007;127(6):459–63. 46. Rajasekhar C, Ray PS, Bhamra MS. Fixation of proximal humeral fractures with the Polarus nail. J Shoulder Elbow Surg. 2001;10(1):7–10. 47. Sosef N, et al. Minimal invasive fi xation of proximal humeral fractures with an intramedullary nail: good results in elderly patients. Arch Orthop Trauma Surg. 2010;130(5):605–11. 48. Popescu D, et al. Internal fi xation of proximal humerus fractures using the T2-proximal humeral nail. Arch Orthop Trauma Surg. 2009;129(9):1239–44. 49. Kazakos K, et al. Internal fi xation of proximal humerus fractures using the Polarus intramedul- lary nail. Arch Orthop Trauma Surg. 2007;127(7):503–8. 50. Adedapo AO, Ikpeme JO. The results of internal fi xation of three- and four-part proximal humeral fractures with the Polarus nail. Injury. 2001;32(2):115–21. 51. Sosef N, et al. The Polarus intramedullary nail for proximal humeral fractures: outcome in 28 patients followed for 1 year. Acta Orthop. 2007;78(3):436–41. 52. Mittlmeier TW, et al. Stabilization of proximal humeral fractures with an angular and sliding stable antegrade locking nail (Targon PH). J Bone Joint Surg Am. 2003;85-A Suppl 4:136–46. 1277 Proximal Humerus Fracture 53. Linhart W, et al. Antegrade nailing of humeral head fractures with captured interlocking screws. J Orthop Trauma. 2007;21(5):285–94. 54. Nolan BM, et al. Surgical treatment of displaced proximal humerus fractures with a short intramedullary nail. J Shoulder Elbow Surg. 2011. 55. Gerber C, Hersche O, Berberat C. The clinical relevance of posttraumatic avascular necrosis of the humeral head. J Shoulder Elbow Surg. 1998;7(6):586–90. 56. Boileau P, et al. Tuberosity malposition and migration: reasons for poor outcomes after hemi- arthroplasty for displaced fractures of the proximal humerus. J Shoulder Elbow Surg. 2002; 11(5):401–12. 57. Moeckel BH, et al. Modular hemiarthroplasty for fractures of the proximal part of the humerus. J Bone Joint Surg Am. 1992;74(6):884–9. 58. Demirhan M, et al. Prognostic factors in prosthetic replacement for acute proximal humerus fractures. J Orthop Trauma. 2003;17(3):181–8. discussion 188–9. 59. Mighell MA, et al. Outcomes of hemiarthroplasty for fractures of the proximal humerus. J Shoulder Elbow Surg. 2003;12(6):569–77. 60. Robinson CM, et al. Primary hemiarthroplasty for treatment of proximal humeral fractures. J Bone Joint Surg Am. 2003;85-A(7):1215–23. 61. Kontakis GM, et al. Early management of complex proximal humeral fractures using the Aequalis fracture prosthesis: a two- to fi ve-year follow-up report. J Bone Joint Surg Br. 2009; 91(10):1335–40. 62. Antuna SA, Sperling JW, Co fi eld RH. Shoulder hemiarthroplasty for acute fractures of the proxi- mal humerus: a minimum fi ve-year follow-up. J Shoulder Elbow Surg. 2008;17(2):202–9. 63. Kontakis G, et al. Early management of proximal humeral fractures with hemiarthroplasty: a systematic review. J Bone Joint Surg Br. 2008;90(11):1407–13. 64. Bufquin T, et al. Reverse shoulder arthroplasty for the treatment of three- and four-part frac- tures of the proximal humerus in the elderly: a prospective review of 43 cases with a short-term follow-up. J Bone Joint Surg Br. 2007;89(4):516–20. 65. Cazeneuve JF, Cristofari DJ. The reverse shoulder prosthesis in the treatment of fractures of the proximal humerus in the elderly. J Bone Joint Surg Br. 2010;92(4):535–9. 66. Gallinet D, et al. Three or four parts complex proximal humerus fractures: hemiarthroplasty versus reverse prosthesis: a comparative study of 40 cases. Orthop Traumatol Surg Res. 2009; 95(1):48–55. 67. Young SW, et al. Comparison of functional outcomes of reverse shoulder arthroplasty versus hemiarthroplasty in the primary treatment of acute proximal humerus fracture. ANZ J Surg. 2010;80(11):789–93. 129M.K. Sethi et al. (eds.), Orthopedic Traumatology: An Evidence-Based Approach, DOI 10.1007/978-1-4614-3511-2_8, © Springer Science+Business Media New York 2013 Keywords Radial nerve palsy Management of humeral shaft fracture Posterior interosseous nerve palsy Functional bracing Casting SM: 34-Year-Old Female with Arm Pain Case Presentation SM is a 34-year-old female who presents to the emergency department via EMS complaining of severe arm pain after a high speed MVA. She denies any loss of consciousness. On primary survey, she demonstrates a GCS of 15, a patent airway, and is hemodynamically stable. On secondary survey, she demonstrates an obviously deformed left arm. Her past medical history is unremarkable. She takes no medica- tions and has no allergies. On physical examination, she demonstrates a strong radial pulse in the left upper extremity and clear deformity of the left humerus. Her compartments are soft. Neurological examination reveals intact median and ulnar nerve, motor and sensory functions. Examination of the radial nerve reveals decreased sensation over the dorsoradial aspect of the hand, and she is unable to extend her wrist, thumb, or fi ngers. Radiographs of the left humerus are demonstrated in Fig. 8.1 . B. Ristevski , M.D., M.Sc. (*) Division of Orthopedic Surgery, Department of Surgery , Hamilton General Hospital , 1280 Main Street , West Hamilton , ON , Canada L85 4K1 e-mail: bill.ristevski@gmail.com J. Hall , M.D. Department of Surgery/Orthopedic Surgery , St. Michael’s Hospital , 800-55 Queen Street , Toronto , ON , Canada M5C 1R6 Chapter 8 Humeral Shaft Fractures Bill Ristevski and Jeremy Hall 130 B. Ristevski and J. Hall Interpretation of Clinical Presentation The patient’s fi ndings and symptoms are consistent with an isolated left humeral shaft fracture. However, Adili et al. have demonstrated that a humeral shaft fracture in a patient, which has been sustained in a motor vehicle crash, is an independent predictor for other long bone fractures as well as fractures in the hand. In the multiply injured patient, a humeral shaft fracture is also associated with a greater likelihood of intra-abdominal injury [ 1 ] . Therefore, this patient must be carefully evaluated through a focused history and complete physical examination to rule out associated fractures and other injuries, particularly to the head, chest, and abdomen. The physical examination is consistent with an associated radial nerve palsy. A complete neurological examination is critical to determine the level of injury: a high radial nerve palsy or posterior interosseous nerve (PIN) palsy. A PIN palsy will exhibit weak wrist extension with radial wrist deviation, as the extensor carpi radialis will typically be spared. A high radial nerve palsy will lack wrist extension all together. Metacarpophalangeal joint extension should be tested with the wrist in slight extension to defunction the intrinsic muscles of the hand, which can aid in fi nger extension when the wrist is in fl exion. The vascular examination revealed a “strong radial pulse.” Additionally, the ulnar artery should also be palpated. A careful circumferential inspection of the skin is critical to rule out an open fracture. Radiographs show a transverse midshaft humeral shaft fracture, with complete dis- placement and varus angulation. Initial immobilization via a coaptation splint, sugar tong splint, or collar-and-cuff provides some comfort by minimizing fracture motion. Fig. 8.1 ( a ) AP Radiograph of left humerus, ( b ) Lateral radiograph of left humerus 1318 Humeral Shaft FracturesDeclaration of Speci fi c Diagnosis SM is a 34-year-old female who presents with an isolated closed left humeral shaft fracture that is transverse, completely displaced, and in approximately 35° of varus alignment. This fracture also demonstrates a rotational deformity and has slight comminution. She has a high radial nerve palsy with normally functioning ulnar and median nerves and a normal vascular examination. Brainstorming: What Are the Treatment Goals and the Options? Treatment goals consist of the following objectives: 1. Reduction and stabilization of the humeral fracture 2. Early mobilization of upper extremity to avoid stiffness/contractures of the ipsilateral wrist, hand, elbow, and shoulder 3. Restoration of radial nerve function 4. Maintenance of muscle strength 5. Return to normal life activities Treatment options include the following: Nonoperative treatment 1. Functional bracing 2. Casting Surgical 1. Plate fi xation 2. Antegrade intramedullary nail fi xation 3. Retrograde intramedullary nail fi xation 4. External fi xation Evaluation of the Literature Relevant publications related to mid-shaft humerus fractures and associated neuro- logical injuries were located with searches on Medline, PubMed and the Cochrane Database. Keywords included “fracture,” “humerus,” “shaft,” and “Nerve palsy.” Subheadings included nonoperative and operative treatment. Searches were limited to humans, the English language, and publication years from 1948 to the present. Some searches were done with isolated keywords or combined to hone in on rele- vant articles. Searches were reviewed manually and, if applicable, based on their abstracts. Manuscripts were retrieved and reviewed along with their reference lists. 132 B. Ristevski and J. Hall Detailed Review of Pertinent Articles The most appropriate treatment for a displaced, angulated, mid-shaft humerus fracture with an associated neurological injury remains controversial. The follow- ing discussion evaluates the relevant literature in order to derive the most favorable treatment option for SM. Nonoperative Treatment Various casts, splints, and braces have been used in the past for the treatment of humeral shaft fractures. Functional bracing, which allows the shoulder and elbow to remain free from immobilization, avoiding excessive stiffness secondary to treat- ment, has been the most popular form of nonoperative treatment and has been favored over casting [ 2 ] . The largest series of treatment via functional bracing was reported by Sarmiento et al. [ 3 ] . They were able to track 620 patients treated for a humeral shaft fracture by functional bracing. The overall nonunion rate in closed injuries was less than 2%, while open fractures had a nonunion rate of 6%. He noted transverse and short oblique fractures took longer to heal. Malunion was also quanti fi ed, and 87 and 81% of patients healed with less than 16° of anterior and varus angulation, respectively. Radial nerve palsy was present in 67 patients and recovered in all patients except one. It should be noted that gunshot wounds cre- ated by high-velocity weapons and injuries caused by sharp penetration were not included in this study. These results have been replicated by some authors [ 4– 6 ] , while others have not been as successful with nonunion rates in these studies rang- ing from 10% to 20.6% [ 7– 10 ] . Some authors have suggested that short oblique fractures particularly in the middle and proximal third of humeral shaft may be at greater risk to develop nonunion [ 10, 11 ] . Functional bracing has remained the mainstay of treatment for most humeral shaft fractures. SM currently has greater than 20° degrees of malalignment which some authors have reported to be unacceptable [ 12, 13 ] . However, a surprising degree of malalignment can be corrected with the use of a functional brace. This correction may take some time to allow gravity to overcome post-traumatic muscle spasm. It is feasible that all the goals of treatment could be successfully obtained via functional bracing. The literature would support that in most cases, the radial nerve palsy will recover in this setting [ 3 ] . A radial nerve palsy is not an absolute indica- tion for operative treatment. Acceptable alignment, length, and rotation of the humerus could be achieved with functional bracing. A trial of bracing with serial radiographic and clinical examination is a reasonable option. Speci fi c instructions about passive range of motion therapy and an extension splint for SM’s fi ngers and wrist is critical to avoiding stiffness given her radial nerve palsy. If an unacceptable reduction persists at 2–3 weeks post functional bracing, an operative approach could be considered. 1338 Humeral Shaft Fractures Operative Treatment Operative indications for humeral shaft fractures include the following [ 12 ] : Failure to obtain and maintain a closed reduction Shortening > 3 cm – Angulation > 20° – Rotation > 30° – Segmental fractures Intra-articular extension Pathologic fractures Ipsilateral fractures (i.e., forearm) Bilateral humeral fractures External Fixation External fi xation for acute humeral shaft fractures has demonstrated some degree of success, but overall is associated with increased complication rates in terms of malunion, nonunion, iatrogenic nerve injury, infection, and hardware malfunc- tion [ 14– 16 ] . However, the current literature re fl ects the fact that most surgeons employ an external fi xator for extreme injuries, such as open fractures with massive soft tissue stripping or loss, neurovascularly compromised limbs, and fractures of the humerus caused by high-velocity projectiles. Naturally, patients with a greater zone of injury have poorer outcomes and are susceptible to more frequent and devastating complications compared to patients with less severe injuries. The external fi xator has been shown to be useful in dealing with such extreme injuries for multiple reasons: the surgical dissection is minimal; application of an external fi xator can decrease operative time and potential blood loss, and can be applied to allow access for vascular and nerve repairs or soft tissue reconstructions. In these situations, external fi xation has been used as de fi nitive treatment, but also has an established role as temporary fi xation before de fi nitive conversion to internal fi xation [ 17– 19 ] . External fi xator pin insertion should occur through a mini-open approach, to avoid iatrogenic nerve injury [ 20 ] . SM’s fracture does not exhibit the characteristics (type II/III open, soft tissue loss, vascular injury, polytrauma) for external fi xation. As there is no de fi ned bene fi t of an external fi xator over other surgical techniques in this situation, alternate oper- ative modalities would be more appropriate. In this patient with a closed, nonpene- trating injury, the associated radial nerve palsy is not an indication for external fi xation or nerve exploration [ 21 ] . 134 B. Ristevski and J. Hall Plate Osteosynthesis Versus Intramedullary Nailing Plate osteosynthesis and intramedullary nailing (IMN) remain viable options for humeral shaft fracture fi xation. There have been multiple randomized controlled trials (RCTs) and meta-analyses comparing these treatment techniques [ 22– 29 ] . A meta-analysis on the topic in 2006 demonstrated that plate fi xation was superior to IMN with respect to reoperation rate (Relative risk (RR) 0.26 Con fi dence Interval (CI) 0.07–0.88, p = 0.03), andplates had signi fi cantly lower associated shoulder morbidity (RR 0.10 CI 0.03–0.42, p = 0.002). A more recent meta-analysis including Changulani et al., failed to show statistical differences, until a re-update was published including the Putti et al. study that favored plating with a relative risk reduction in total complications of 0.52, CI 0.3–0.91, p = 0.02. Interestingly, a single study can signi fi cantly change the outcome of the meta-analysis, suggesting that overall, the difference in results between IMN and plate osteosynthesis is not marked. Although each of these RCTs had a relatively small number of patients enrolled, the design of these studies was robust and represents the best available evidence. Overall, these RCTs demonstrated that plate osteosynthesis has an advantage over IMN, despite the theoretical advantages of IMN which include smaller incisions, less soft tissue stripping at the fracture site and better biomechanics compared to plates. A recent Cochrane Review demonstrated that intramedullary nails were associated with increased shoulder impingement, decreased shoulder range of motion, and an increased need for removal of implants [ 30 ] . However, the nature of the current literature is such that a single study has the power to change outcomes in a meta-analysis, demonstrating the need for a large randomized controlled study with adequate power to address the limitations inherent in the above studies. However, there are situations where the advantages of a locked IMN outweighs plate osteosynthesis-speci fi cally, multifocal pathological humeral shaft fractures secondary to metastatic disease. In this situation IMN can stabilize the bone, allow for early weight-bearing through the extremity, and avoids placing incisions in the area of potential radiotherapy [ 31 ] . Other indications for IMN can include segmental fractures with a large intervening segment, morbid obesity, and poor soft tissue coverage. A disadvantage of IMN use has been a higher incidence of shoulder pain compared to plate fi xation [ 24, 26 ] . If nonoperative attempts failed to obtain and maintain an acceptable reduction for SM, either plate osteosynthesis or intramedullary nailing would be reasonable approaches. Two techniques exist for nailing for the humerus: antegrade and retrograde. The major differences include patient positioning for the operation as most antegrade nails require entering the shoulder joint versus an approach through the posterior aspect of the elbow joint for retrograde insertion. Operating through the joint, as one could anticipate, can cause complications involving the shoulder and the elbow, such as impingement, stiffness, and septic arthritis. Regardless of technique, distal locking screws should be placed with a mini-open technique as percutaneous insertion similar to that used for the lower extremities, can lead to iatrogenic nerve injuries [ 20 ] . 1358 Humeral Shaft Fractures For humeral shaft plate osteosynthesis, when possible, broad 4.5 plates with staggered holes have been favored historically, as experts have suggested that narrow 4.5 plates or smaller 3.5 plates with longitudinally aligned holes could result in longitudinal splits and fracture through the screw holes. There are instances when early exploration of the radial nerve has been advocated. Palsies associated with penetrating trauma (i.e., machete injury, or high- velocity gunshot wound), open fractures, and high-energy fractures have been noted to have poorer prognosis, due to a high incidence of nerve transection or severe nerve injury with no possibility of recovery [ 21, 32 ] . In these instances, exploring the radial nerve can be helpful for prognosis, and in rare circumstances (such as a machete injury) can allow for primary repair of the nerve. In high-energy fractures, the zone of injury of the nerve is too great to allow useful results from primary repair, but can allow earlier treatment decisions in terms of subsequent nerve grafting or tendon transfers [ 21 ] . Evidentiary Table and Selection of Treatment Method The cornerstone studies related to the treatment of SM are included in Table 8.1 . Based on the literature, the authors recommend that SM be treated with a trial of functional bracing. If an acceptable reduction is not obtained or maintained by 2 weeks post bracing, the authors prefer plate osteosynthesis via an anterolateral approach, which would also allow for a radial nerve exploration if desired to provide prognostic information. De fi nitive Treatment Plan SM’s fracture of her humerus on presentation has an unacceptable degree of malalignment. However, proper use of a functional brace has the ability to improve fracture alignment. After a brief (7–10 days) of splinting, functional bracing is initiated: it is imperative that the functional brace fi t in a snug fashion. As swelling decreases and atrophy occurs, the brace should be adjusted to maintain a tolerably tight fi t. The brace will work by compressing the soft tissues around the fracture site, improving the bony alignment via the principle of incompressibility of fl uids [ 33 ] . Gravity and the alternating forces resulting from active elbow fl exion and extension are also helpful reduction aids. However, active shoulder elevation and abduction has been noted to have a displacing effect on humeral fractures [ 3 ] . Patients should be instructed not to rest their elbow on the arm of a chair or table, as this can cause further malalignment. In Sarmiento’s very large series, 67 patients had an associated radial nerve palsy; of these, 66 patients had recovery of their nerve. Wrist drop splints were not employed, as rapidly regaining elbow extension intrinsically brought the wrist into a neutral position avoiding fl exion contractures 136 B. Ristevski and J. Hall Table 8.1 Evidentiary table: a summary of the quality of evidence for functional bracing and operative fi xation via plates or IMN of humeral shaft fractures Author (year) Description Summary of results Quality of evidence Sarmiento et al. (2000) Retrospective case series Followed 620 of 922 patients with a humeral shaft fracture treated by functional bracing. Overall, nonunion rate of 2.6% (<2% for closed and 6% for open fractures). Overwhelming majority of patients obtaining functional range of motion and acceptable alignment. Level IV Bhandari et al. (2006) Meta-analysis 3 RCTs pooling of 155 patients comparing compression plating versus IMN of humeral shaft fractures. RR of reopera- tion favoring plates of 0.26, 95% CI 0.007–0.9, p = 0.03. Also reduced risk of shoulder morbidity favoring plates, RR = 0.10, 95% CI 0.03–0.4, p = 0.002. Level I Heineman et al. (2010) Updated meta- analysis (from 2006 above) 4 RCTs pooling 203 patients comparing compression plating versus intramedullary nailing of humeral shaft fractures. No signi fi cant differences between implants for total complications, nonunion, infection, nerve-palsy or the need for reoperation. Level I Heineman et al. (2010) (Correspondence) Meta-analysis re-updated (from 2010 above) 5 RCTs pooling 237 patients comparing compression plating versus IMN of humeral shaft fractures. Total complication rate signi fi cantly lower with plates versus IMN, RR 0.52, CI 0.30–0.91, p = 0.02. Level I Chapman et al. (2000) Prospective randomized controlled trial 84 patients comparing compression plating versus IMN of humeral shaft fractures. IMN had signi fi cantly more shoulder pain and loss of range of motion, p = 0.007. Plate fi xation had signi fi cant range of motion loss at the elbow,p = 0.03. Level I McCormack et al. (2000) Prospective randomized controlled trial 44 patients comparing compression plating versus IMN of humeral shaft fractures. No signi fi cant differences in pain or function as gauged by ASES score. A signi fi cant difference in the need for reoperation with seven patients requiring secondary surgery in the IMN group versus 1 in the plating group, p = 0.016. Level I Changulani et al. (2007) Prospective randomized controlled trial 47 (2 loss to follow-up) patients comparing compression plating versus IMN of humeral shaft fractures. No signi fi cant differences in function as gauged by ASES score. A shorter time to union was found for IMN group (6.3 weeks vs. 8.9 weeks, p = 0.001). Higher infection rate with plating was observed while patients with IMN have more shortening and loss of shoulder motion. Level I (continued) 1378 Humeral Shaft Fractures Table 8.1 (continued) Author (year) Description Summary of results Quality of evidence Putti et al. (2009) Prospective randomized controlled trial 34 patients comparing compression plating versus IMN of humeral shaft fractures. No signi fi cant differences in function as gauged by ASES score. The total complication rate in IMN was signi fi cantly higher, than in the plating group (50% vs. 17% p = 0.038). Level I Kurup et al. (2011) Meta-analysis Dynamic compression plating versus IMN of humeral shaft fractures. IMN associated with shoulder impingement, decreased shoulder range of motion, and the need for hardware removal. Level I RCT randomized controlled trial, IMN intramedullary nailing, RR relative risk, CI con fi dence interval, ASES American Shoulder and Elbow Surgeons of the wrist [ 3 ] . In SM’s case, if a 2–3-week trial of functional bracing results in an unacceptable amount of malalignment, operative fi xation is reasonable. If osteosynthesis is required, our preferred treatment of choice is plate fi xation. For plate osteosynthesis, the patient is placed in a semisitting or supine position and appropriately prepared and draped in a sterile fashion. Tourniquet application is not usually possible. Distal fractures of the shaft can be approached posteriorly; however, an anterolateral approach would be appropriate for SM’s fracture. The incision centers on the fracture and is approximately 15 cm long. The fascia is incised in line with the skin incision. The biceps is retracted medially, being mindful of the musculocutaneous nerve which lies beneath. Any traumatic injury to the muscle typically can be utilized in the surgical approach. The brachialis is split centrally in a longitudinal fashion. Typically, this muscle has dual innervation by the radial and musculocutaneous nerve, and is not rendered functionless by this technique. The radial nerve should be identi fi ed in the distal aspect of the wound (as it lies between the brachioradialis and the brachialis muscles), inspected, and protected. It can then be traced back proximally and inspected through the fracture site: this provides prognostic information. With a closed injury, it is rare for the nerve to be transected; however, knowledge of the status of the nerve (intact, transected, etc.) will affect future care. Thus, primary repair of a transected radial nerve caused by nonpenetrating trauma has poor results, secondary to the large zone of injury which is fundamen- tally different compared to an injury that is caused a sharp instrument [ 21 ] . The radial nerve is vulnerable at the fracture site (the probable site of injury in this case) as well as at the distal aspect of the humerus, where the nerve pierces the intermus- cular septum and becomes an anterior structure. It is here that the nerve can also be inadvertently damaged by surgical dissection, retraction, or hardware placement. Once exposure is completed the fracture ends can be reduced and comminuted pieces large enough to play a role in fracture stability should be reduced and fi xed with lag screws. Temporary fi xation with Kirschner wires may assist in this process. The authors prefer to use a broad 4.5 low contact dynamic compression plate. 138 B. Ristevski and J. Hall Preferably, four screws above and below the fracture for fi xation are used, with a minimum of three. Compression mode should be used for the screws if the fracture pattern is stable and amenable. It is imperative to examine the distal edge of the plate to ensure that the radial nerve has not been trapped between the plate and the bone prior to securing the plate. Fluoroscopy may be used to gauge adequacy of reduction, plate placement, and screw lengths. If there is any question about overall alignment or rotation, fl at plate X-rays can be taken intraoperatively, which are superior to the view offered by fl uoroscopy. A standard closure is performed. Postoperatively, patients fi xed with plates and screws with good cortical contact are allowed to weight-bear as tolerated through the extremity. Weight-bearing on the upper extremity after humeral plating for patients with concomitant lower extremity injuries has been shown to be safe [ 34 ] . Minimally invasive plate osteosynthesis (MIPO) techniques for humeral shaft fractures have been evaluated [ 35– 39 ] . The major perceived danger with this technique is the potential injury to neurological structures, particularly the radial and musculocutaneous nerves. Overall, most authors feel this technique is demand- ing, should be performed by surgeons with experience in the area and that overall alignment is likely harder to achieve with MIPO techniques. Predicting Long-Term Outcomes In the largest series of patients treated with functional bracing after a closed injury, less than 2% of patients had a nonunion [ 3 ] . However, some studies have failed to replicate this degree of success. In Sarmiento’s series of closed and open fractures, the majority of patients (87 and 81%) had less than 16° of varus and anterior angula- tion, respectively [ 3 ] . In addition 88.6 and 92.0% had less than 10° of range of motion loss at the shoulder and elbow, respectively [ 3 ] . The overwhelming majority of patients returned to activities of daily living. Plate fi xation also affords very good long-term results, with good range of motion of the shoulder and elbow, as typically these joints have not been violated for the fi xation of shaft fractures, and early range of motion exercises can be initiated [ 27 ] . Conversely, elbow motion can be affected in plating of distal shaft fractures [ 25 ] . The overall nonunion rate in the most recent meta-analysis was 7.5% [ 22 ] . The overall complication rate as gauged by Heineman et al. was signi fi cantly decreased when plates were compared to IMN [ 26 ] . References 1. Adili A, Bhandari M, Sprague S, et al. Humeral shaft fractures as predictors of intra-abdominal injury in motor vehicle collision victims. Arch Orthop Trauma Surg. 2002;122:5–9. 2. Sharma VK, Jain AK, Gupta RK, et al. Non-operative treatment of fractures of the humeral shaft: a comparative study. J Indian Med Assoc. 1991;89:157–60. 1398 Humeral Shaft Fractures 3. Sarmiento A, Zagorski JB, Zych GA, et al. Functional bracing for the treatment of fractures of the humeral diaphysis. J Bone Joint Surg Am. 2000;82:478–86. 4. Ricciardi-Pollini PT, Falez F. The treatment of diaphyseal fractures by functional bracing. Results in 36 cases. Ital J Orthop Traumatol. 1985;11:199–205. 5. Camden P, Nade S. Fracture bracing the humerus. Injury. 1992;23:245–8. 6. Zagorski JB, Latta LL, Zych GA,
Mais conteúdos dessa disciplina
43 9 8486-9985 - MAPA - FISIOLOGIA HUMANA - 54_2025- 43 9 8486-9985 - MAPA - ENF - ENFERMAGEM NA ASSISTENCIA AO PACIENTE CRITICO - 54_2025
- Doença Pulmonar Obstrutiva Crônica (DPOC) - Aula de Pneumologia SanarFlix - Resumo
- Doença Pulmonar Obstrutiva Crônica
- Exercícios (AIES, Imunossupressores e Manejo de Asma e DPOC)
- Impactos da Poluição na Saúde
- Destaques Diretrizes 2025
- 43 9 8486-9985 - CONTEXTUALIZACAO LEIA O RELATO DE CASO PUBLICADO NO PERIODICO CIENTIFICO NEW ENGLAND JOURN
- 12-POP--Fisioterapia - Resumo
- onsidere uma situação em que um indivíduo realiza uma expiração forçada máxima após uma inspiração igualmente forçada. O volume total de ar expirado é
- 2- A analise dedutiva é a base fundamental na interpretação do eletrocardiograma e requer o conhecimento do significado do processo de ativação do cor
- 2- A analise dedutiva é a base fundamental na interpretação do eletrocardiograma e requer o conhecimento do significado do processo de ativação do cor
- O tempo em que um paciente passa imóvel no leito da Unidade de Terapia Intensiva (UTI) está proporcionalmente ligado as complicações graves que venham
- Em relação à fisioterapia do sistema cardiovascular, assinale a opção correta. a. A resistência oferecida à ejeção dos ventrículos a cada sítole é den
- A escala de RASS, ou escala de agitação-sedação de Richmond, é amplamente utilizada para avaliar o nivel de consciência do paciente critico, Inclus...
- Analise as questões e marque a resposta CORRETA. A. O ritmo e a profundidade da respiração são características importantes na avaliação do desco...
- unip A Ventilação Mecânica Não Invasiva (VMNI) é o método de suporte ventilatório que utiliza a pressão positiva sem o uso de tubos traqueais. Atua...
- Analise e marque a resposta CORRETA. A. A saída de ar dos pulmões acontece pela expiração e ocorre pelo relaxamento da musculatura do diafragma ...
- Frente a um caso de doença pulmonar obstrutiva crônica, assinale os possíveis desfechos desfavoráveis (que não queremos que aconteçam) para estes paci
- A desobstrução rinofaríngea retrógrada (DRR) é uma manobra que visa remover secreções. Qual é a principal diferença entre a DRR e a DRR+I?
- A vibrocompressão é uma técnica utilizada na fisioterapia respiratória que envolve a aplicação de vibrações. Essa técnica é realizada em sinergia c...
- A Expiração Abreviada é uma técnica que visa aumentar a capacidade pulmonar. Qual é o principal objetivo dessa técnica?
- Massoterapia, pompage, mobilizações articulares
- avaliação paciente critico
