24 Classifications of Thoracic Spinal Fractures Abstract The thoracolumbar junction (T10–L2) is a unique transition zone between the relatively rigid thoracic spine (T1–T10) and the more mobile lower lumbar spine (L3–L5). At this junction, the spine is subject to great biomechanical stress and, consequently, the thoracolumbar spine is the most common site of failure from blunt trauma. To the authors’ knowledge, there are no classification systems specific to the thoracic spine. Thus, thoracic and lumbar injuries are most often considered together in classification systems. The earliest classification of thoracolumbar fractures dates back to Boehler’s sentinel work in 1929, when he first categorized fractures based on the mechanism of injury and fracture morphology. Denis’ three-column spine model marked an important point in history, as this classification offered insight into the stability of the spine and attempted to guide treatment. Since these publications, additional classifications have been developed, each of which has advanced our understanding in characterizing and managing thoracolumbar trauma. This chapter reviews the common classification schemes for thoracic injuries, specifically detailing their historical context as well as highlighting their advantages and disadvantages. Keywords: thoracic spine, thoracolumbar spine, thoracolumbar fracture, classification, spine trauma, thoracolumbar trauma Clinical Pearls • The thoracolumbar junction (T10–L2) is the most common site of spinal failure from blunt trauma. • Both thoracic and lumbar injuries are considered together in classification systems. • Boehler published the first classification in 1929, organizing injuries into five categories on the basis of the mechanism of injury and fracture morphology. • Watson-Jones simplified injuries into three categories and emphasized the importance of the posterior ligamentous complex (PLC) to the stability of the spinal column. • Nicoll classified injuries into four groups on an anatomical basis and offered treatment guidelines based on the neurological status and spinal stability. • Holdsworth introduced the two-column concept and upheld that the posterior column, and thus the integrity of the PLC, was the determining factor of stability. • Under his three-column principle, Denis proposed that the middle column was integral to spinal stability. The extent of thoracolumbar trauma could be measured by three degrees of instability. • McAfee et al grouped thoracolumbar trauma into six patterns of injuries determined solely by sagittal computed tomography imaging. • According to their mechanistic classification, Ferguson and Allen categorized injuries into six groups depending on the mode on failure of the anterior and posterior spinal elements. • The Gaines load sharing classification was the first to incorporate a point system to quantify injury severity and guide surgical treatment. • The Magerl/AO classification was the most comprehensive system and subcategorized thoracolumbar trauma into 53 distinct injuries. • The thoracolumbar injury classification and severity score (TLICS) assigned a comprehensive injury severity score to every thoracolumbar trauma based on the injury morphology, integrity of the PLC, and neurological status. The TLICS score helped guide clinical decision-making. • The AOSpine classification assigned a thoracolumbar AOSpine injury score (TL AOSIS) score to every thoracolumbar trauma based on the fracture morphology, neurological status, and patient-specific modifiers. Operative versus nonoperative treatment was recommended depending on the TL AOSIS score. The thoracic spine is a complex three-dimensional structure made up of bones, soft tissues (intervertebral discs and ligaments), and the neural elements (spinal cord and nerve roots). It can fail under supraphysiological forces in flexion, compression, extension, distraction, lateral bending, axial rotation, translation, shear, or any combination thereof. The same directional forces can produce different fracture patterns and severity of injury depending on the position of the spine at the instantaneous moment of injury. Injuries can be broadly considered to be mechanically stable or unstable. Either may be associated with a neurological deficit. In distinction from the more mobile and flexible lumbar spine, the thoracic spine is additionally stabilized from the interaction with the ribs. Thus, more energy is required to lead to injury of this more rigid spinal region. Adding to these complexities is the diversity of treatment options for thoracic injuries which can vary widely between institutions and surgeons. In following, there is currently no universally accepted classification that can perfectly characterize these injuries, predict treatment outcomes, or unanimously guide clinical decision-making. Starting with Boehler’s original work and Holdsworth’s two-column concept and progressing to more modern systems such as the thoracolumbar injury classification and severity score (TLICS) and the thoracolumbar AOSpine injury score (TL AOSIS), it is this chapter’s goal to review the milestones along this 90-year journey of thoracic spine fracture classification. Much has been learned about the natural history and behavior of specific injuries with certain types of management. Below, we detail the development of various fracture classifications in chronological order and underscore how each has added to our understanding of injury care ( The first published classification of thoracolumbar injuries was reported by Boehler in 1929. In his seminal work, Boehler reviewed the plain radiographs of patients with spinal injury during World War I and developed his classification scheme based on the mechanism of injury and fracture morphology. He grouped thoracolumbar trauma into five categories: compression, flexion distraction, extension, shear, and rotational injuries.1 In 1938, Watson-Jones was credited as the first to emphasize the importance of the posterior ligamentous complex (PLC) to the stability of the thoracolumbar spinal column.2 The PLC is the capsuloligamentous structures supporting the vertebral arch comprising the facet joint capsule, ligamentum flavum, interspinous ligament, and supraspinous ligament ( Simple wedge fractures were the most common that he observed. He felt these were caused by axial loading. With this fracture pattern, Watson-Jones observed that the intervertebral disc was generally preserved and therefore ankylosis of the fractured segment to an adjacent segment over time was rare. He additionally postulated that the wedging through an intact disc increased the strain on the facet joints which could lead to persistent pain. In contrast, the disc in comminuted fractures was felt to be more generally disrupted, and thus ankylosis was found to occur more commonly. Interestingly, despite being what we would consider to be higher energy injuries, these fractures tend to be less painful because of the “autofusion” that would occur anteriorly. Finally, fracture dislocations were a distinct group that was found to have a higher risk for spinal cord injury (SCI). Such injuries carried a particularly poor prognosis in the high thoracic spine. According to his writings, Watson-Jones insisted on obtaining “perfect reduction” to achieve the best clinical outcomes, similar to burgeoning appendicular (long-bone) fracture recommendations. Fig. 24.1 Posterior ligamentous complex (PLC). Illustrative example of an injury to the PLC, which comprises four capsuloligamentous structures: facet capsule, ligamentum flavum, interspinous ligament, and supraspinous ligament. (Copyright AO Foundation, Switzerland. Reproduced with permission.) Nicoll further advanced the work of Watson-Jones and in 1949 published his thoracolumbar classification system organized on an anatomical basis into four main categories: anterior wedge fractures, lateral wedge fractures, fracture dislocations, and isolated fractures of the neural arch.3 Nicoll also acknowledged the importance of the PLC, most importantly the interspinous ligament, in predicting stability. Like Watson-Jones, he observed that in anterior wedge fractures, the fulcrum of angular deformity lies at the nucleus pulposus, and therefore compromise of the interspinous ligament must take place to produce instability. Radiographically, disruption of the interspinous ligament was implied if there was separation of the spinous processes. Nicoll divided wedge fractures into anterior and lateral types. He believed lateral wedge fractures occurred when the axial skeleton was forced forward and to one side (i.e., a flexion-rotation mechanism). This force would lead to fractures of the transverse processes on the concave side and of the intervertebral joints on the convex side of the spinal column. Nicoll reported that this fracture pattern had a poorer prognosis than its anterior counterpart. Neural arch fractures were thought to be caused by rotational injuries. Nicoll observed that bilateral arch fractures anywhere in the thoracic spine were stable. Like Watson-Jones, he called for perfect reduction, stating that “a good anatomical result is indispensable to a good functional result.” Nicoll offered treatment guidelines based on the neurological status of the patient and spinal stability. In patients with a normal neurological examination, the first step was to establish whether the fracture was stable or unstable. Wedge fractures with intact interspinous ligament and neural arch fractures were considered stable. Stable fractures would undergo so-called “functional treatment,” which does not require reduction or immobilization. Wedge fractures with rupture of the interspinous ligament were considered unstable, being treated with reduction and immobilization to minimize the risk of deformity and disability. Understand that all treatment at this time was closed as there were no internal fixation options for the spine at the time. Holdsworth introduced the two-column concept of spinal stability in 1963.4,5 The anterior column comprises the anterior longitudinal ligament (ALL), vertebral body, disc, and posterior longitudinal ligament (PLL). The posterior column comprises the osseoligamentous structures posterior to the PLL, which include the neural arches, facets, and PLC ( Fig. 24.2 Spinal columns and elements. Holdsworth divided the spinal column into the anterior (from ALL to PLL) and posterior column (comprising the neural arches, facets, and PLC). Denis divided the spine into the anterior (from ALL to anterior half of body/disc), middle (from posterior half of body/disc to PLL), and posterior column (comprising the posterior bony elements and ligamentous complex). Ferguson and Allen divided the spinal column into the anterior (from ALL to anterior two-thirds of body/disc) and posterior element (from posterior one-third of body/disc to PLC). Morphologically, Holdsworth classified thoracolumbar trauma based on clinical and radiographic findings into five categories: simple wedge fractures, rotational fracture dislocations, extension dislocations, vertical compression fractures, and shear fractures. Wedge fractures are caused by pure flexion and are inherently stable. Flexion rotation results in fracture dislocations with disruption of the PLC, which pose high risk of neurological injury. Extension causes failure of the anterior column, specifically rupture of the intervertebral disc and avulsion injury of the ALL. Extension dislocations are stable in flexion, so immobilization in this position was recommended. Holdsworth was the first to introduce the concept of a burst fracture. Of note, he felt that these injuries were generally stable because the ligaments remained intact. Finally, shear force across the posterior column can produce instability and spondylolisthesis and therefore neurological injury was common. In 1983, Denis advanced the two-column concept by dividing Holdsworth’s anterior column into separate anterior and middle columns.6,7 His three-column principle changed the way surgeons considered stability and managed thoracolumbar trauma. The three columns consist of the anterior (ALL to anterior half of the vertebral body and disc), middle (posterior half of the vertebral body and disc to the PLL), and posterior column (posterior bony elements and ligamentous complex) ( In short, Denis denoted that the middle column was the key to predicting neurological injury and stability. Compression of the middle column (or vertical collapse) can cause narrowing of the neuroforaminae and injury to the exiting nerve roots. More importantly, posterior displacement of the vertebral body fragmens (or retropulsion) can lead to compression of the spinal cord ( Denis divided thoracolumbar trauma into minor and major injuries. Minor injuries included fractures of the transverse processes, articular processes, pars, and spinous processes. Major injuries were classified into four groups from most to least stable: compression, burst, seat-belt type (flexion distraction), and fracture dislocation. These four fracture patterns are due to failure of one, two, or all three columns (
24.1 Introduction
Table 24.1).
24.2 Boehler: 1929
24.3 Watson-Jones: 1938
Fig. 24.1). Like Boehler, Watson-Jones’ classification scheme was based on the mode of injury and fracture pattern, specifically of the vertebral body. He simplified thoracolumbar trauma into three categories: simple wedge fractures, comminuted fractures, and fracture dislocations.
24.4 Nicoll: 1949
24.5 Holdsworth: 1963
Fig. 24.2). The major contribution that this system made, which still stands today, is that the posterior column, and thus the integrity of the PLC, was deemed the determining factor of stability.
24.6 Denis: 1983
Fig. 24.2). In contrast to Watson-Jones, Nicoll, and Holdsworth, Denis maintained that disruption of the PLC (or posterior column) alone is insufficient to results in spinal instability. Instead, his definition of instability required two adjacent columns to be compromised. Accordingly, spinal instability develops when the middle column fails either with the anterior or posterior columns (or both).1,6,7,8
Fig. 24.3).
Table 24.2).1,8 Compression fractures occur from axial force causing isolated failure of the anterior column. In minor compression fractures, only the anterior column fails in compression. In severe compression fractures, the middle column serves as a hinge, so the posterior column may fail in tension. Like Nicoll, Denis distinguished two types of compression fractures, anterior and lateral. Burst fractures occur when the anterior and middle columns fail in compression. Denis described five different types of burst fractures. Seat-belt type fractures involve the middle and posterior columns, which fail in tension. The axis lies at the anterior column, which may fail in compression. Fracture dislocations are complex and arise when all three columns fail from a combination of flexion rotation, shear, and flexion distraction forces (
Fig. 24.4). The main drawback of Denis’ classification is it quickly branches from 4 major fracture patterns into 21 different types and subtypes.
Thoracic Key
Fastest Thoracic Insight Engine
