ORIGINAL ARTICLE

A New Classication Scheme for Open Fractures

Orthopaedic Trauma Association: Open Fracture Study Group

Objectives: The purpose of this study was to propose a new

classication system for open fractures developed by the Classication Committee of the Orthopaedic Trauma Association.

Setting: Academic and Level I trauma center. Patients/Participants: Consecutive patients with open fractures. Main Outcome Measurements: Open fracture classication scale. Results: Evaluation of 34 factors identied through systematic
literature review and ranking in order of importance by our panel resulted in consensus on ve essential categories of open fracture severity assessment: skin injury, muscle injury, arterial injury, contamination, and bone loss. These categories were chosen with particular attention paid to avoiding redundancy with existing tools for assessment of fracture conguration. Evaluation of the system through prospective data collection revealed that the ve categories were widely applicable to open fractures, but the subcategories of open fracture characteristics required alteration to reect clinically important parameters for open fracture severity stratication. Skin injury was best assessed by its potential for approximation rather than laceration length. Muscle injury required quantitative and qualitative assessment of potential for function. The effects of arterial injury were most accurately assessed through distal limb ischemia. The depth and nature of contamination were important indicators of overall contamination signicance. Bone loss was best assessed quantitatively.

Conclusions: The proposed Orthopaedic Trauma Association

Classication of Open Fractures is a scientically derived assessment tool for determining the severity of open fractures. This tool will require further testing to establish validity characteristics and determine its clinical use compared with existing open fracture classication systems. Key Words: open fractures, classications (J Orthop Trauma 2010;24:457465)

INTRODUCTION
Systematically classifying fractures assists clinicians and researchers in communicating information by grouping
Accepted for publication October 26, 2009. The authors have no nancial disclosures related to their work on this manuscript. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journals Web site (www.jorthotrauma.com). The members of the study group are listed in Appendix 1. Reprints: Andrew R. Evans, MD, Department of Orthopaedics, UPMC Mercy, 1350 Locust Street, Suite 220, Pittsburgh, PA 15219 (e-mail: arevans@ gmail.com). Copyright 2010 by Lippincott Williams & Wilkins

injuries with similar characteristics and separating dissimilar injuries. Open fractures are a unique subset of fractures because of the exposure of bone to contamination from the environment and the disruption of soft tissue integrity, which increases the risk for infection, delayed union, nonunion, and even amputation.16 Contamination and injuries to the integumentary, myofascial, neurologic, vascular, and/or skeletal systems are factors that have historically been used to identify open fracture patterns with similar natural histories or to guide injury-specic treatment strategies.711 Involvement of these diverse structures, however, has made classifying open fractures difcult. Multiple classication systems have been used for open fractures.1,5,8,1214 Of these, the most widely used has been the system of Gustilo and Anderson.1,13 Its simplicity and ability to stratify open fracture severity in a logical order has allowed it to stand the test of time. Despite being broadly applied to most open fractures, the Gustilo system was designed only for open tibial fractures. The denitions of injury characteristics are imprecise, which leaves room for different interpretations of how a fracture should be classied. One important shortcoming of the Gustilo and Anderson system is that it incorporates concepts of treatment into the classication such as the methods for closure of soft tissue injuries. Ideally, the characteristics of the injury should guide treatment rather than treatment determining the classication. This is particularly problematic when concepts of treatment evolve over time, potentially changing how similar injuries are classied. It is not surprising that reliability studies have shown only moderate interobserver agreement for the Gustilo and Anderson classication.15,16 This is the rst report on a project undertaken by the Classication Committee of the Orthopaedic Trauma Association to develop a new classication of open fractures. The global objective of this new classication will be to classify open fractures of the upper extremity, lower extremity, and pelvis in adults and children in a clinically relevant way, which will allow better communication for treatment and research. To develop this classication, the committee has undertaken a systematic scientic process. The purpose of this article is to report the methods and results of that process, propose a new classication of open fractures, and to outline the process to further assess the classication with reliability studies in the clinical setting.

PATIENTS AND METHODS

An extensive search of three electronic databases (PubMed, EMBASE, and Web of Science) was performed by a medical librarian to locate references that identify factors
www.jorthotrauma.com |

J Orthop Trauma  Volume 24, Number 8, August 2010

457

OTA Open Fracture Study Group

J Orthop Trauma  Volume 24, Number 8, August 2010

in the literature that have been used to clinically evaluate open fractures of the upper extremity, lower extremity, and pelvis. The search strategies used for each database are listed in Appendix 2 (see Appendix 2, Supplemental Digital Content, http://links.lww.com/BOT/A29) and include references dating from the inception of each database (PubMed1949, EMBASE1988, Web of Science1955) through June 2008. All identied titles and abstracts were hand-searched to identify salient references specically citing an injury proling system or classication of factors to rate the severity of open fractures. The resulting list of publications was reviewed by one author (A.R.E.), and a comprehensive list of factors describing upper extremity, lower extremity, or pelvic open fracture tissue injury or treatment characteristics was compiled (Fig. 1). For each factor, the relative frequency of use within the selected body of literature was determined. This frequency was used as a simplistic measure of literaturebased importance of each factor. This list of factors was sent by e-mail to a seven-member panel of experienced, fellowship-trained orthopaedic traumatologists, each of whom was instructed to independently examine and prioritize each factor for inclusion or exclusion in the new open fracture classication scheme. Each panel member was asked to rank the factors in order of their relative

importance (134) using their clinical experience and knowledge of the literature. A rank-order mean was calculated for each factor and the mean used as a measurement of relative importance (Fig. 2). In addition, ve of seven traumatologists independently rated the relative necessity of inclusion of each factor into a comprehensive open fracture classication system with designations of must include, may include, and would not include. Responses were recorded, and agreement between evaluators was used to distinguish heavily favored, disfavored, and controversial factors (Fig. 3). The results of this analysis were presented to an open fracture work group for review, discussion, and consensus during a focused 4-hour meeting at the 2008 American Academy of Orthopaedic Surgeons meeting in San Francisco. The work group was a panel of fellowship-trained traumatologists who were members of the Orthopaedic Trauma Association classication committee. The data and a consensus process were used to: 1) identify factors considered essential to the classication of open fractures; 2) determine whether to include controversial factors; 3) clarify the denitions of each factor; 4) determine whether the denitions applied to each factor would be applicable to all anatomic regions; 5) formulate a draft classication; and 6) develop a plan to test the classication on clinical cases.

FIGURE 1. Comprehensive list of factors describing open fracture tissue injury or treatment characteristics from the literature.

458

| www.jorthotrauma.com

q 2010 Lippincott Williams & Wilkins

J Orthop Trauma  Volume 24, Number 8, August 2010

New Classication Scheme for Open Fractures

FIGURE 2. Rank order mean by committee members.

In addition to using data from this process, the panel also worked from a set of general principles in developing a new proposed classication system. First, the classication should be simple rather than complex and should be anatomically allinclusive. Second, it should be based on injury pathoanatomy rather than treatment principles because treatment should follow the classication rather than the classication following the treatment. Third, it should not include system issues (eg, time to debridement). Fourth, it should focus on the anatomic characteristics of the injury. The proposed initial classication was then tested through a process of clinical data collection between June and September 2008. The goal was to determine the feasibility and ease of use of the proposed classication system. After approval from local Institutional Review Boards, data were collected prospectively. The investigators were committee members who were fellowship-trained orthopaedic traumatologists at ve Level I trauma centers. On the data sheet, in
q 2010 Lippincott Williams & Wilkins

addition to classifying the fracture, clinicians were asked to evaluate the classication in each subcategory and record any problems or difculties. The classication was also presented to an independent group of clinicians in Europe at a focused meeting on fracture classication. At a second meeting of the open fracture work group at the 2008 Orthopaedic Trauma Association meeting in Denver, the data and other feedback were reviewed and the classication was revised.

RESULTS Literature Review

The literature search yielded a total of 9283 references: PubMed, 7313 total; EMBASE, 697 unique (821 total, 123 overlap with PubMed); and Web of Science, 1273 unique (1586 total, 314 overlap with PubMed and EMBASE). In addition to these references, three additional references known
www.jorthotrauma.com |

459

OTA Open Fracture Study Group

J Orthop Trauma  Volume 24, Number 8, August 2010

FIGURE 4. Top 10 factors identied for evaluation of open fractures by experts (as determined by rank order mean).

sustained by the muscular envelope, muscle viability at operation, and energy of injury were the most commonly selected and most highly ranked factors in the survey. The importance of each of these factors was also ranked as must include by all panelists who provided must/maybe/not inclusion evaluation. Psychosocial and logistic factors such as patients socioeconomic background, preinjury health status, time from injury to operative treatment, and treatment in a community hospital versus trauma center setting were among the least highly ranked factors in the survey.

Consensus Process
To draft a new open fracture classication, the factors identied as most important through the literature review and rank order process were further analyzed with emphasis on identifying individual factors that, in aggregate, would serve optimally to classify open fractures. The discussion and analysis resulted in the creation of categories of tissue injury that included skin, muscle, vascular, and osseous injury. The top-ranking factors identied by rank order process that pertain to these tissue systems include: 1) skin defect; 2) muscle injury; 3) arterial injury; and 4) bone loss. Contamination was an additional high-ranking factor that was not accounted for by the classication categories, and so it was included as an additional category. Each factor was divided into mild, moderate, and severe subgroups with the objective of achieving parallel construction in the subgrouping. This provided internal consistency across open fracture injury characteristics, allowing one or more increasingly severe injury characteristics to directly inuence the overall assessment of injury severity.

FIGURE 3. Variable inclusion/exclusion consensus.

to the authors but not identied in the search process were selected for inclusion that were not identied through the search strategies used. In total, 21 salient references were identied by hand-searching the titles and abstracts of all references, selecting those that cited an injury proling system or classication of factors to rate the severity of open fractures (Fig. 1). Thirty-four unique factors identied as clinical indices of open fracture severity were extracted through the review of these publications.2,5,712,1727 Of these, the factor most commonly cited was the mechanism of injury as it relates to soft tissue injury kinetics and the degree of injury sustained by the muscular envelope. Fracture pattern, neurologic injury, and arterial injury were also commonly cited in the literature.

Evaluation of Study Methodology

Independent of this consensus process, the original 34 injury-related factors included in this analysis were divided into six categories by one of the authors (A.R.E.) based on how they would likely be grouped if incorporated into a comprehensive classication system: tissue system injury, mechanical characteristics of injury, physiological parameters, psychosocial factors, contamination level, and logistic factors affecting timing of treatment (Fig. 5). The clinician panel was blinded to this categorization before consensus discussions and proposal of the new classication system. Within each category, the factors were listed by rank order mean to identify which factors were ranked as most important through the determination of rank order by each clinician. The highest ranking factors within selected
q 2010 Lippincott Williams & Wilkins

Rank Order Process

The rank order mean for each of the 34 factors was used as a measure of relative importance of the independent clinician rankings. This identied 10 highly favored factors, as presented in Figure 4. Nine of these 10 factors were in the top 10 selected factors for the majority (greater than 50%) of panelists. Mechanism of injury related to soft tissue injury (transection, avulsion, crush) and the degree of injury

460

| www.jorthotrauma.com

J Orthop Trauma  Volume 24, Number 8, August 2010

New Classication Scheme for Open Fractures

FIGURE 5. Grouping of variables.

categories from the rank order process included skin defect, muscle injury, arterial injury, bone loss, and contamination, all of which were included in the proposed classication system. This demonstrated general agreement between the rank order and consensus processes while maintaining the principles of simplicity, concentration on injury characteristics, and exclusion of system issues as originally set forth by the committee. These results lend further merit and validity to the methodology used to generate this classication system.

quality of skin coverage rather than wound length; 2) muscle injury required qualitative as well as quantitative description to capture the extent of and potential for additional necrosis; 3) effects of arterial injury on open fracture severity were more objectively assessed through the identication of ischemic injury rather than the mode of treatment used for the arterial

Clinical Feasibility Study

Data were collected on a series of 99 open fractures. There were 23 upper extremity fractures, 56 lower extremity fractures, four pelvic fractures, and 13 foot fractures. Three fractures had no fracture code. The age range of these patients was 6 to 85 years. Graphs of the range of severity are presented in Figure 6. Comments on the classication were reviewed and discussed by the Study Group at the 2008 Orthopaedic Trauma Association Meeting in Denver, CO. Based on the comments and data, ndings important to the formulation of a more clinically relevant classication system included: 1) skin injury was more accurately reected by the quantity and
q 2010 Lippincott Williams & Wilkins

FIGURE 6. Severity of open fractures (N = 99). www.jorthotrauma.com |

461

OTA Open Fracture Study Group

J Orthop Trauma  Volume 24, Number 8, August 2010

injury; 4) depth and nature of contamination were essential to measurement of the magnitude of contamination; and 5) bone loss was a unique and important factor in the determination of open fracture severity that remains otherwise unaccounted for by other classication systems. The results of this revision are presented as the Orthopaedic Trauma Association Classication for Open Fractures, which is seen in Appendix 3.

DISCUSSION
The overall objective of this project is to create an open fracture classication system that focuses on injury characteristics dened by pathoanatomy. The classication should apply to anatomically diverse skeletal injuries in both adult and pediatric patients and have greater observer reliability than current open fracture classication schemes. This study reports on our systematic review of the literature and selection of injury characteristics that dene an open fractures severity. This was followed by independent ranking of the importance of these characteristics by experienced observers and development of a preliminary classication scheme through consensus based on these data. We then performed a clinical feasibility study and further rened the categories leading to the current Orthopaedic Trauma Association Open Fracture Classication scheme. There is an important need for a new classication of open fractures. The Gustilo and Anderson classication, along with other lesser-known open fracture classication systems, fails to provide simple, reliable, and reproducible measurements of objective clinical ndings that can be used to guide injury assessment and treatment. Existing open fracture classications that display poor interobserver agreement or excessive complexity provide suboptimal use in clinical practice and research and, as a result, require revision through a systematic process capable of providing a simple, reliable, reproducible, and valid clinical and research tool. There are signicant obstacles and challenges to developing this type of classication for open fractures that also correlates with patient prognosis. The challenges include making determinations of open fracture severity based on relevant objective clinical ndings than can be classied in a way that is easy to remember and communicate, also providing meaningful guidance for treatment and prediction of clinical outcome. Additional challenges and obstacles exist in the validation process of any open fracture classication, including potentially elaborate schemes required for the evaluation of inter- or intraobserver reliability and measurement of validity parameters. We hypothesize, however, that a systematic and scientic approach to the creation of an open fracture classication system will yield an objective, relevant, and evidence-based clinical and research tool that improves signicantly on existing open fracture classication systems. Despite the multitude of studies that evaluate various issues related to open fractures, the number of studies that systematically identify and evaluate factors that categorize open fracture severity, treatment complexity, and predict outcome remains limited. The rank order system used in this study was a way to independently weigh the relative importance for each factor without undue inuence of individual

opinion. The consensus process was based on the following principles: 1) create a multicategory classication allowing the separate scoring of important factors; 2) use factors that provide anatomic description of the injury; 3) base injury scores on parallel categories of injury severity; 4) exclude factors that are treatment-based (eg, ap versus no ap); 4) exclude factors related to determining limb salvage potential; and 5) exclude psychosocial and logistic factors. The initial draft classication was then modied by systematic review of data from a prospective review of 99 cases leading to the current classication. During this process, the denition of each subgrouping was carefully considered in an attempt to make it parallel to each of the remaining four subgroupings. Broad categories of skin injury were chosen: skin lacerations that are able to be closed, skin lacerations that are unable to be closed, and skin lacerations associated with extensive dissociation of the skin and subcutaneous tissues from underlying tissues (degloving injuries) that in turn increases the risk of skin necrosis and infection of large dead spaces. Muscle injury was categorized into subgroups in a qualitative fashion based on function, necrosis, or loss. For vascular injury, we considered objective assessments to distinguish categories such as palpation of pulses, assessment of capillary rell, pulse oximetry, arterial indices, or angiography.28 However, because of the difculty and unknown reliability for some methods of assessing vascular injury, we decided to base vascular injury on whether a major vessel is disrupted and whether distal limb ischemia is present. Contamination, dened by its quantity, depth, and nature (eg, organic versus inorganic), was included because it contributes to the overall risk of infection. Incorporating bone loss into this classication was heavily debated because of its potential overlap with fracture classication, which we advocate be assessed separately by the Orthopaedic Trauma Association fracture classication system. However, bone loss was felt to be an indirect measure of periosteal stripping and energy of injury and has signicant implications regarding treatment distinct from the implications of fracture pattern. Subgroupings were agreed on that include no bone loss, bone loss, or devascularization requiring excision with some cortical opposition and segmental bone loss. All subgroupings were considered applicable to both adult and pediatric patients. Compartment syndrome was considered as a potential add-on given that its occurrence is often related to injury severity. However, compartment syndrome is a distinct clinical entity with a complex diagnostic and treatment algorithm, which suggests that it should be considered separately from the open fracture injury. Neurologic injury was also felt to be a separate issue not directly related to grading open fracture severity. Periosteal stripping is difcult to quantify in isolation and was thought to be better assessed through parameters included in the proposed classication system such as myofascial injury and bone loss as well as through the assessment of fracture pattern. Fracture pattern, with its own complexity of classication, is accounted for by separate classication systems such as the Orthopaedic Trauma Association Comprehensive Fracture Classication System.29
q 2010 Lippincott Williams & Wilkins

462

| www.jorthotrauma.com

J Orthop Trauma  Volume 24, Number 8, August 2010

New Classication Scheme for Open Fractures

The timing of when to classify an open injury is important for many reasons. Classifying an open fracture in the emergency room is useful for physician-to-physician communication as well as for deciding management of the initial injury. Accurately assessing the degree of injury to deep tissues such as muscle, neurovascular structures, and bone is not always possible before operative exploration of the zone of injury. Classifying the fracture in later phases of treatment may be valuable for assessing the evolution of injury characteristics through the treatment period but will not enhance initial injury severity assessment or guide early treatment. As a result, the end of the initial debridement was felt to be the most optimal time to apply the classication. Standardizing the timing of classifying the fracture will assist in comparisons between future studies. In summary, we have described a systematic approach for a new classication of open fractures that with further study hopefully will improve on existing systems. We are currently planning observer reliability studies and further modications. ACKNOWLEDGMENT Special thanks to Amy L. Harper, Medical Librarian, for assistance with the systematic literature review. REFERENCES
1. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-ve open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am. 1976;58:453458. 2. Gustilo R, Merkow R, Templeman D. Current concepts review: the management of open fractures. J Bone Joint Surg Am. 1990;72: 299304. 3. Patzakis MJ, Wilkins J, Moore TM. Considerations in reducing the infection rate in open tibial fractures. Clin Orthop Relat Res. 1983;178: 3641. 4. Patzakis MJ, Wilkins J, Moore TM. Use of antibiotics in open tibial fractures. Clin Orthop Relat Res. 1983;178:3135. 5. Byrd H, Spicer T, Cierney G. Management of open tibial fractures. Plast Reconstr Surg. 1985;76:719730. 6. Chapman MW, Mahoney M. The role of early internal xation in the management of open fractures. Clin Orthop Relat Res. 1979;138: 120131. 7. McNamara MG, Heckman JD, Corley FG. Severe open fractures of the lower extremity: a retrospective evaluation of the Mangled Extremity Severity Score (MESS). J Orthop Trauma. 1994;8:8187. 8. Suedkamp NP, Barbey N, Veuskens A, et al. The incidence of osteitis in open fractures: an analysis of 948 open fractures (a review of the Hannover experience). J Orthop Trauma. 1993;7:473482. 9. Russell WL, Sailors DM, Whittle TB, et al. Limb salvage versus traumatic amputation: a decision based on a seven-part predictive index. Ann Surg. 1991;213:473482. 10. Howe HR, Polle GV, Hansen KJ, et al. Salvage of lower extremities following combined orthopedic and vascular trauma. Am Surg. 1987;53: 205208. 11. Gregory RT, Gould RJ, Peclet M, et al. The Mangled Extremity Syndrome (MES): A severity grading system for multisystem injury of the extremity. J Trauma. 1985;25:11471150. 12. Collins DN, Temple SD. Open joint injuries. Clin Orthop Relat Res. 1989; 243:4856. 13. Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classication of type III open fractures. J Trauma. 1984;24:742746. 14. Veliskakis KP. Primary internal xation in open fractures of the tibal shaft; the problem of wound healing. J Bone Joint Surg Br. 1959;41:342354.

15. Horn BD, Rettig ME. Interobserver reliability in the Gustilo and Anderson classication of open fractures. J Orthop Trauma. 1993;7:357360. 16. Brumback RJ, Jones AL. Interobserver agreement in the classication of open fractures of the tibia. The results of a survey of two hundred and fortyve orthopaedic surgeons. J Bone Joint Surg Am. 1994;76:11621166. 17. Johansen K, Daines M, Howey T, et al. Objective criteria accurately predict amputation following lower extremity trauma. J Trauma. 1990;30: 568573. 18. Bosse MJ, MacKenzie EJ, Kellam JF, et al. A prospective evaluation of the clinical utility of the lower extremity injury-severity scores. J Bone Joint Surg Am. 2001;83:314. 19. Bosse MJ, McCarthy ML, Jones AL, et al. The insensate foot following severe lower extremity trauma: an indication for amputation? J Bone Joint Surg Am. 2005;87:26002608. 20. Castillo RC. Impact of smoking on fracture healing and risk of complications in limb-threatening open tibia fractures. J Orthop Trauma. 2005;19:151157. 21. Harrison WJ, Lewis CP, Lavy C. Open fractures of the tibia in HIV positive patients: a prospective controlled single-blind study. Injury. 2004; 35:852856. 22. Lange RH. Limb reconstruction versus amputation decision making in massive lower extremity trauma. Clin Orthop Relat Res. 1989;243:9299. 23. Muller M, Allgower M, Schneider R, et al. Manual of Internal Fixation. London: Springer-Verlag; 1991. 24. Oestern H, Tscherne H, Pathophysiology and classication of soft tissue injuries associated with fractures. In: Tscherne H, Gotzen L, eds. Fraktur und Weichteilschaden. Berlin, Heidelberg, New York, Tokyo: SpringerVerlag; 2004:19. 25. Slauterbeck JR, Britton C, Moneim MS, et al. Mangled Extremity Severity Score: an accurate guide to treatment of the severely injured upper extremity. J Orthop Trauma. 1994;8:282285. 26. Swiontkowski MF, MacKenzie EJ, Bosse MJ, et al. Factors inuencing the decision to amputate or reconstruct after high-energy lower extremity trauma. J Trauma. 2002;52:641649. 27. Togawa S, Yamami N, Nakayama H, et al. The validity of the Mangled Extremity Severity score in the assessment of upper limb injuries. J Bone Joint Surg Br. 2005;87:15161519. 28. Levy BA, Zlowodzki MP, Graves M, et al. Screening for extremity arterial injury with the arterial pressure index. Am J Emerg Med. 2005;23: 689695. 29. Marsh JL, Slongo TF, Agel J, et al. Fracture and dislocation compendium. Orthopaedic Trauma Association Committee for Coding and Classication. J Orthop Trauma. 1996;(Suppl 1):vix, 1154.

APPENDIX 1 Study Group Members

Andrew R. Evans, MD, Department of Orthopaedics, UPMC Mercy, Pittsburgh, PA; Julie Agel, MA, ATC, Department of Orthopaedic Surgery and Sports Medicine, Harborview Medical Center, Seattle, WA; Gregory L. DeSilva, MD, Department of Orthopaedic Surgery, University of Arizona, Tucson, AZ; Thomas A. DeCoster, MD, Department of Orthopaedics, University of New Mexico, Albuquerque, NM; Douglas R. Dirschl, MD, Department of Orthopaedics, University of North Carolina, Chapel Hill, NC; Clifford B. Jones, MD, Department of Orthopaedics, University of Michigan, Grand Rapids, MI; James F. Kellam, MD, Department of Orthopaedic Surgery, Carolinas Medical Center, Charlotte, NC; Douglas W. Lundy, MD, Resurgens Orthopaedics, Marietta, GA; J. Lawrence Marsh, MD, Department of Orthopaedic Surgery, University of Iowa Hospitals and Clinics, Iowa City, IA; Debra L. Sietsema, PhD, Department of Orthopaedics, University of Michigan, Grand Rapids, MI; and Milan K. Sen, MD, University of Texas Medical Science Center at Houston, Houston, TX.
www.jorthotrauma.com |

q 2010 Lippincott Williams & Wilkins

463

OTA Open Fracture Study Group

J Orthop Trauma  Volume 24, Number 8, August 2010

APPENDIX 3 Proposed Classication of Open Fractures Skin

1. Can be approximated 2. Cannot be approximated 3. Extensive degloving 2. Artery injury without ischemia 3. Artery injury with distal ischemia

Contamination
1. None or minimal contamination 2. Surface contamination (easily removed not embedded in bone or deep soft tissues) 3. a. imbedded in bone or deep soft tissues b. high risk environmental conditions (barnyard, fecal, dirty water etc)

Muscle
1. No muscle in area, no appreciable muscle necrosis, some muscle injury with intact muscle function 2. Loss of muscle but the muscle remains functional, some localized necrosis in the zone of injury that requires excision, intact muscle-tendon unit 3. Dead muscle, loss of muscle function, partial or complete compartment excision, complete disruption of a muscletendon unit, muscle defect does not approximate

Bone Loss
1. None 2. Bone missing or devascularized but still some contact between proximal and distal fragments 3. Segmental bone loss.

Arterial
1. No injury

Invited Commentary
The importance of an open fracture classication system cannot be overemphasized. It is essential for providing guidelines for treatment, prognosis, functional outcomes, interphysician communication, and research. In 1959, Veliskakis classied open fractures into three types based on the severity of the wound. Subsequently, the Gustilo-Anderson open fracture classication has been the mainstay for classication of open fractures and is used extensively by trauma and orthopaedic surgeons throughout the world. This system has been extremely useful. However, it is not without some limitations. Initially, when Gustilo-Anderson rst described the classication system in 1976, it was focused on long-bone fracture wounds and did not address contamination. Over time, modications of the classication occurred. Another drawback was demonstrated in the study by Brumback and Jones showing interobserver agreement of 60% among 245 orthopaedic surgeons. This led to the recommendation of denitively classifying open fractures at the time of surgery for treatment of the open fracture wound. The need for improvement and renement of an open fracture classication system is always the goal of the scientic investigator. The Orthopaedic Trauma Association has proposed a new classication system for open fractures focusing on the injury characteristics dened by the pathoanatomy of the open fracture. They have methodically and carefully evaluated 34 factors identied through a systematic literature review, and a panel identied ve essential categories for assessing the severity of the open fracture. These include skin injury, muscle injury, arterial injury, contamination, and bone loss. Besides focusing on the pathoanatomy of the open fracture wound, it is also all inclusive of open fractures regardless of anatomic site and is applicable to both adult and pediatric patients. There is greater emphasis on interobserver reliability and emphasizes that the end of the initial debridement is the ultimate time to apply the classication. The Orthopaedic Trauma Association authors emphasize that the validity of this classication system will be tested and, in the future, observer reliability studies and further modications will occur. The Orthopaedic Trauma Association is proposing a new classication system that will hopefully be an improvement on the existing open fracture classication systems. Michael J. Patzakis, MD
Professor and Chairman, Department of Orthopaedic, The Vincent and Julia Meyer Chair, Keck School of Medicine of USC, Los Angeles, CA

464

| www.jorthotrauma.com

q 2010 Lippincott Williams & Wilkins

J Orthop Trauma  Volume 24, Number 8, August 2010

New Classication Scheme for Open Fractures

Invited Commentary
The effort spent by the authors of this classication system is appreciated. This is a well thought out approach to solving a vexing problem in orthopaedic surgery that has existed far too long. The strengths of this preliminary treatise include a comprehensive content validation process, including thorough literature review, use of clinical experts, group discussions, and feasibility testing. Going forward, however, the authors should address the following questions. 1. During the content validation phase, did the authors conduct interobserver agreement using some version of the kappa coefcient (eg, agreement on which of the 34 factors should be included or not in a simple yes/no fashion or agreement on which factors were rated must include, maybe include, or dont include)? This would provide information about the actual level of agreement with chance factors subtracted out. In their discussion section, the authors mention that they are planning observer reliability studies, but this should ideally take place during the development of the survey itself. 2. How many people were involved in the consensus process to identify individual factors that would serve optimally to classify open fractures as well as to then divide each factor into the mild, moderate, and severe subgroups? This is not clearly specied but probably should be. Finally, the statistical qualities of this preliminary classication system (including interobserver agreement, construct validity, and criterion-related validity) are still untested. Although it is still too early to tell what, if any, problems this newer system will exhibit, we all look forward to a future publication with a large representative sample that would yield a report with statistical results. William G. De Long, Jr, MD
Professor of Orthopaedic Surgery, Anatomy and Cell Biology, Temple School of Medicine, Chief of Orthopaedics SLHN

Jill Stoltzfus, PhD

Director, Research Institute, St. Lukes Hospital and Health Network

q 2010 Lippincott Williams & Wilkins

www.jorthotrauma.com |

465