Designations of fractures according to horizontal beams and vertical buttresses. Beams: superior orbital rims and glabella, H1; inferior orbital rims and zygomatic arches, H2; alveolar process of maxilla, H3; c indicates central; l, lateral. Buttresses: frontonasomaxillary, V1; frontozygomaticomaxillary, V2; pterygomaxillary, V3; s indicates superior; i, inferior. Reprinted with permission from illustrator William Loechel.
Designations of fractures in case 1. Left: V1i, V2i, H2; right: V1i, V2s, V2i, H2, H2l. See legend to Figure 1 and Table 1 for an explanation of the abbreviations. Reprinted with permission from illustrator William Loechel.
Case 1. Computed tomographic (CT) scans of a 28-year-old man who received multiple manual facial assault impacts. A, Coronal CT scan displaying fractures left V1i, H2, and right V2i. B, Coronal CT scan displaying fractures left V1i, H2, V2i, and right V1i, V2s, V2i, H2. C, Coronal CT scan displaying fractures left V1i, H2, V2i, and right V1i, V2s, V2i, H2. D, Axial CT scan displaying fracture left V2s. E, Axial CT scan displaying fracture H2l. F, Coronal CT scan displaying intact V3. See legend to Figure 1 and Table 1 for an explanation of the abbreviations.
Designation of fractures in case 2. Left: V1s, V1i, V2s, V2i, V3, H1c, H2, H2l, H3c; right: V1s, V1i, V2s, V2i, V3, H1, H1c, H2, H2l. See legend to Figure 1 and Table 1 for an explanation of the abbreviations. Reprinted with permission from illustrator William Loechel.
Case 2. Computed tomographic (CT) scans of a 23-year-old male driver who received a high-velocity facial impact from an automobile tire jettisoned from a truck traveling ahead of the patient's vehicle. A, Coronal CT scan displaying fractures left H1c and right H1, H1c. B, Coronal CT scan displaying fractures left V1s, H1c, and right V1s, H1c. C, Coronal CT scan displaying fractures left V1s, V2s, H2, and right V1s, V2s, H2. D, Coronal CT scan displaying fractures left V1i, V2i, V3, H2l, H3c, and right V1i, V2i, H2l. E, Axial CT scan displaying fractures left V1i, V2i, H2l, and right V1i, V2i, H2l. F, Axial CT scan additionally displaying fractures left V3 and right V3. G, Axial CT scan displaying fractures left V1s, V2s, and right V1s, V2s. See legend to Figure 1 and
Table 1 for an explanation of the abbreviations.
Donat TL, Endress C, Mathog RH. Facial Fracture Classification According to Skeletal Support Mechanisms. Arch Otolaryngol Head Neck Surg. 1998;124(12):1306-1314. doi:10.1001/archotol.124.12.1306
THE INCIDENCE, mechanisms, and pathophysiology of facial fractures are well described in the literature, as are the current approaches to fracture reduction and fixation.1- 5 The 3 goals of therapy in treating midfacial fractures are (1) to restore functional occlusion; (2) to stabilize the major facial skeletal supports, thereby restoring the premorbid 3-dimensional contour (height, width, and projection) to the face; and (3) to provide skeletal support for the proper function and appearance of the overlying facial soft tissue structures.
The current approach to facial fracture repair requires the repositioning of the fracture segments into anatomic position, with a focus on the lattice supports in relation to each other and to the cranial base.6- 10 Modern therapy also mandates the rigid stabilization of the vertical and horizontal facial supports to withstand the forces of mastication.11- 16 Such treatment plans are made possible by the diagnostic capability of computed tomographic (CT) scanning technology, a major advance over plain x-rays, for fracture identification and fragment visualization.17- 20
Despite the widespread acceptance of current diagnostic and treatment methods, the most commonly used classification for describing facial fractures remains that classically described by French physician Rene LeFort, which alone yields insufficient information for fracture description and the complete planning of treatment. LeFort's original classification described "the great lines of weakness" according to fracture patterns he experimentally produced.1 The LeFort classification is inadequate in that it does not define the facial skeletal supports or the more severely comminuted, incomplete, or combination maxillary fractures. Moreover, it does not describe the fractures of the part bearing the occlusal segment.4 The LeFort classification thus often underestimates the complexity of the fractures and limits the complete description of the overall facial fracture pattern, which often includes any array of fronto-orbital, zygomatic, and nasoethmoidal fractures in combination with maxillary injury.
The goals of a classification system oriented toward current therapy for midfacial fractures should ideally include information obtained from the clinical, surgical, and CT radiological examinations (1) to accurately represent the anatomic and functional magnitude and complexity of the overall midfacial fracture pattern, (2) to describe the involved functional skeletal supports critical to the proper design for surgical therapy, (3) to provide a meaningful common terminology for communication of the fracture information between the radiologist and surgeon, and (4) to provide specific information sufficient for comparison of treatment outcomes of midfacial fracture treatments.
The scheme for the midfacial fracture classification system in this study was designed by partitioning the 3 pairs of horizontal structural supports (beams) and 3 pairs of vertical structural supports (buttresses). Based on the intersection of these supports and segmental divisions, there were 11 unilateral sites and 22 bilateral sites (Figure 1 and Table 1). The 3 primary, paired horizontal beams from superior to inferior were the superior orbital rims combined with the glabella (H1), the inferior orbital rims combined with the zygomatic arches (H2), and the alveolar processes of the maxilla (H3). The primary, paired vertical buttresses from anterior to posterior were the nasal maxillary (V1), the zygomaticomaxillary (V2), and the pterygomaxillary (V3) buttresses. The beams were further categorized into central (c) and lateral (l) segments, and the buttresses were categorized into superior (s) and inferior (i) segments. Nasal fractures, thin lamina fractures (such as found along the orbital walls and the walls of the maxilla), and the degree of fracture displacement were not included as part of the classification system. The classic completed LeFort and zygomatic fracture patterns as herein represented according to buttress and beam support involvement are listed in Table 2.
Individual fracture locations were denoted laterally (left or right), by involvement of the fractured buttress or beam (vertical or horizontal), and at the site of the fracture line along the buttress, superior or inferior, or along the beam, central or lateral. The fractures of patients were denoted by listing in sequence the location of individual fractures. Examples shown in Figure 2, Figure 3, Figure 4, and Figure 5 demonstrate the methods.
The classification system was evaluated retrospectively by reviewing medical records of 213 adult patients with midfacial fractures treated at Detroit Receiving Hospital, Detroit, Mich, from January 1, 1993, through June 30, 1995. All patients were evaluated by axial and coronal CT scan imaging. The patients ranged in age from 18 to 70 years with the primary mechanisms of injury being blunt trauma caused by altercation or motor vehicle accident in 208 patients (97.6%) or penetrating trauma caused by gunshot wounds in 5 patients (2.4%). Twenty-six patients were women and 187 were men.
To determine whether the fractures could be analyzed and assigned to a specific notation, the CT scans were evaluated for fracture sites and these sites were then diagrammed and transcribed according to the classification system. Evaluations were blinded to any prior medical records or radiology interpretation. Designations were then related to a standard method of description to show the advantages and disadvantages of the system.
To test whether the notations could serve as an efficient and valid means of communication, 12 resident physicians were presented with a series of mock facial fracture diagrams and fracture designations. Each physician was presented first with 5 distinct fracture patterns diagrammed according to the classification scheme and asked to provide notations of the fracture patterns; they were subsequently presented with 5 new distinct notations of fractures and asked to diagram the fracture pattern represented by the notation. The number of the correctly transcribed fracture diagrams and notations was determined. Transcription errors were also determined.
Among the 213 patients, 43 cases involved fractures of the bony lamina of the orbital walls (blowout type) or nasal bones alone and were not included for analysis. The remaining 170 cases were classified and are listed in Table 3, Part A, and Table 3, Part B, and Table 3, Part C. Cases were described according to whether the injury was unilateral left (L), unilateral right (R), or bilateral, and were compared with prior nomenclature systems, where applicable.
The classification scheme was easily applied to all of the fracture patterns determined by axial and coronal CT scan for the patients in this study. No fracture patterns in this study were deemed unassignable, as might occur due to a fracture line passing through a buttress/beam intersection. Using the proposed classification, 40 differing unilateral fracture patterns were identified among 122 patients who presented with unilateral facial fractures. Forty-seven differing fracture patterns were identified among the 48 patients presenting with bilateral facial fractures. Therefore, using the proposed classification, 87 specific and distinct fracture patterns were described among the 170 patient CT scan studies.
Only 25 (28.7%) fracture patterns reviewed met the defined criteria of LeFort fractures in which they had all of the fractures required for 1 of the classic LeFort fracture patterns. Among these, only 11 could be identified as bilateral LeFort fractures of the same level, although often with considerable variation in complexity between sides. Most illustrative of the complexity of the fractures in this series is that current classifications, if applicable at all, do not describe the multitude of distinct fracture patterns, as have herein been identified by the involved facial supports. The specific information critical to the modern approach and method of treatment is therefore seen to be deficient in currently used classifications of midfacial fracture patterns.
The mock facial fracture patterns presented to the 10 participating physicians for transcription are listed in Table 4, with the first 5 having been transcribed from a classification diagram to the fracture pattern notation and the subsequent 5 having been transcribed from the pattern notation to a classification diagram. Of the 120 total patterns posed for evaluation, only 2 were miscommunicated, each by different physicians. The 2 errors that occurred were an incorrectly diagrammed template for laterality (right and left interposed, mock facial fracture pattern 5) in 1 pattern and an incorrectly diagrammed buttress designation (V1i and V2i interposed, mock facial fracture pattern 9) in the other pattern. No errors occurred in providing the mock fracture pattern notation from the mock fracture classification diagrams.
Recent technological advances in radiologic imaging have been beneficial in the overall approach to the diagnosis and treatment of facial fractures. The widespread advent of CT scanning in the United States in the past decade markedly improved the accuracy of fracture imaging to greater than 95%, subsequently enabling the treating surgeon to better determine the requirements for surgery and a plan for approaches and methods of repair.17- 21 In addition, the recent advances in both the areas of internal rigid fixation and the use of autogenous bone grafts for reconstruction have yielded an improved early treatment of fractures with the reestablishment of anatomic form and function before the onset of the sequelae previously seen with external fixation techniques, poor fracture exposure, and stabilization, or a delay in fracture treatment.7,16 It is in accordance with these diagnostic and treatment advances that the need for a classification system providing a means for the informative description and communication of facial fractures characteristics, especially between the radiologist and surgeon, has become obvious.
The LeFort classification system, albeit quite simple and used for many years, does not amply describe multiple sites of fracture now seen with modern imaging techniques. In an attempt to improve interpretations of these images and applications of treatment, Manson et al6 described a method that looked at displacement and forces to create the fracture and noted the contribution of supporting vertical buttresses of the face especially for application and understanding of the role for internal fixation. Approach algorithms were offered by Gruss et al,7,8 using central and lateral midfacial fracture descriptions, as well as Gruss et al stressing the importance of the zygomatic arch in guiding the reestablishment of facial skeletal contours. Other classifications were described to supplement the LeFort description and were based on detailed descriptions of fractures of individual midfacial regions, such as orbitozygomatic fractures classified by Zingg et al,12 and the nasoethmoid classification by Leipzinger and Manson.13
Our system emphasizes the need to analyze the integrity of the vertical and horizontal supports and the need to focus on these supports for reduction and possible rigid fixation. It is assumed that proper reduction of the fractured beams and buttresses will lead to accurate width, length, and projection of the facial skeleton with a concomitant correction of appearance and function. Finer details of reconstruction can be maintained by analyzing and correcting for bone displacement, orbital wall and floor involvement, and comminution. Using the present scheme, the theoretical numbers of unilateral unique fracture pattern combinations can be determined by the combinations (C)[11, n=1, 2, 3. . .1]=2047. Other modifiers such as the presence or absence of fractures of the nasal bones, fractures of orbital lamina, and fractures with displacement and/or comminution can be added to provide a comprehensive description of the injury. These concepts thereby allow for an increased specificity of information in describing the variations in fracture patterns that may occur in patients with midfacial fractures that is not currently possible using other classification schemes.
The advantage of the classification system is that it provides a method to determine the type of injury and degree of severity based on the site and numbers of buttresses and beams involved with the trauma. Attention can then be focused toward a more direct approach that is necessary to deal with these problems. The system also is an easy means of communication between diagnosticians and surgeons who work on these types of cases.
The disadvantages of the classification system is that it is not all-inclusive, negates soft tissue considerations, and does not describe the status of nasal bones, the orbital lamina, or, necessarily, the displacement and comminution of fractures. Such modifers would add more information, but at the same time further add to the complexity of the system.
The classification system also affords a method of quantitation as shown by the mock exercises in which information was communicated from one surgeon to another. The methods were easily learned and found to be extremely accurate in defining the beam and buttress fracture patterns. The system provided an opportunity to quantitate a variety of patterns and with such methods there should be potential application to the analysis treatment outcomes and series in which patients are provided different treatment modalities. The system thus allows for a comparison of number and location of sites of injury as to their treatment and results. Further investigations, however, will be necessary to evaluate the usefulness of this aspect of the classification system.
The proposed classification system is conceptually in anatomic and descriptive accord with the currently practiced methods of facial fracture reduction and rigid fixation. This newly proposed facial fracture classification scheme provides a convenient, succinct, descriptive, and reproducible method of designating beam and buttress fracture patterns. This scheme may be used to accurately communicate and compare, in greater detail than permitted using the LeFort or other independent classification schemes, the essential site and degree-of-severity characteristics of facial fractures critical to their surgical reduction and reconstruction. The usefulness of this classification scheme in determining optimal methods of treatment and subsequent outcomes in dealing with midfacial fracture requires further investigation.
Accepted for publication March 24, 1998.
Corresponding author: Terry L. Donat, MD, Department of Otolaryngology–Head and Neck Surgery, Wayne State University, 41935 W 12 Mile Road, Novi, MI 48377.