Surgical Stabilization Techniques for Thoracolumbar Fractures

Classifications of Spinal Fractures

Historical Review

The classification, management, surgical approaches, and stabilization techniques for thoracolumbar fractures have been undergoing continuous evolution over the past few decades. Appropriate classification of these fractures is the first step toward successful treatment, so a comprehensive classification system should facilitate the selection of the appropriate approach and surgical technique for fractures that require surgical stabilization.

Boehler was the first to attempt to classify thoracolumbar spine fractures in 1929. In 1963 Holdsworth introduced the two-column theory of spinal stability. In 1978 White and Panjabi defined clinical instability as the inability of the spine under physiologic loads to maintain relationships between vertebrae such that there is neither acute nor subsequent neurologic injury, deformity, or pain.

The three-column theory of the spine was introduced by Francis Denis in 1983. He added a third, or middle, column to Holdsworth’s two-column model. According to Denis, the anterior column of the spine included the anterior longitudinal ligament (ALL) and the anterior half of the vertebral body, annulus, and disk. The middle column included the posterior half of the vertebral body, annulus, and disk in addition to the posterior longitudinal ligament (PLL). The posterior column incorporated the neural arch, facets, and the posterior ligamentous complex (PLC), consisting of the supraspinal and interspinal ligaments, ligamentum flavum, and facet capsules ( Fig. 36-1 ). The importance of the middle column had become evident, because it was necessary to disrupt the middle column with the anterior or posterior columns for dislocation to occur. Denis defined stability based on the integrity of two of the three columns. His classification included four groups: 1) compression fractures resulting from failure of the anterior column under compression, 2) burst fractures resulting from failure of the anterior and middle columns secondary to fractures of the vertebral body under axial load, 3) flexion-distraction injuries secondary to failure of the posterior and middle columns, and 4) fracture-dislocations resulting from failure of all three columns. This classification stood the test of time, because it was simple to apply and in a way simplified management, because most anterior column injuries are treated conservatively or nonsurgically with bracing, and almost all three-column injuries are treated with surgical stabilization. Two-column injuries are still a topic of debate; however, in the presence of a neurologic deficit or severe deformity, surgical stabilization is required.

In 1989, Magerl and colleagues introduced the Association for the Study of Internal Fixation— Arbeitsgemeinschaft für Osteosynthesenfragen, or AO—classification system. This was based on a 10-year review of 1445 thoracolumbar fractures that recognized three main fracture types: compression fracture (type A), distraction (type B), and fracture-dislocation (type C). Subdivisions were created according to the severity of the fractures. This detailed and descriptive classification system resulted in 53 fracture patterns, with A1 being the least severe and C3 the most severe.

The Load-Sharing Classification was introduced by McCormack and colleagues in 1994. The classification was derived from the analysis of failures of thoracolumbar spine fractures treated with transpedicular short-segment arthrodesis. According to their classification, fractures were graded according to the degree of comminution of the body, apposition of the fracture fragments, and deformity. A point system was applied to each fracture, from 3 to 9, with a higher number indicative of increased severity. Fractures with a score greater than 7 had a high risk of short-segment fixation failure. This algorithm was intended to aid in deciding whether to use short-segment arthrodesis and/or anterior column graft support. The classification was validated biomechanically in vitro.

In 2005, the Spine Trauma Study Group introduced the Thoracolumbar Injury Severity Score (TLISS) as a new classification system. The system was based on three injury characteristics: 1) mechanism of injury, 2) neurologic status, and 3) the integrity of the PLC. This classification system was eventually modified to become the Thoracolumbar Injury Classification and Severity Scale (TLICSS). As for the morphology of injury, compression injuries are assigned 1 point; burst fractures and compression fractures with coronal plane deformity greater than 15 degrees score 2 points; translational or rotational injuries score 3 points; and distraction injuries, being the most unstable, receive 4 points. Scoring the severity of neurologic injury is based on a five-category system. Patients with an intact neurologic exam receive 0 points. In the presence of nerve root injury or complete spinal cord injury, the fracture scores 2 points. Patients with an incomplete spinal cord injury or cauda equina syndrome are given 3 points.

Assessing the integrity of the PLC can be accomplished clinically by the presence of a palpable interspinous gap, separation of the spinous processes on plain radiographs, or with the use of magnetic resonance imaging (MRI). Patients with an intact PLC receive 0 points. Those in whom the integrity of the PLC is indeterminate receive 2 points, and those with confirmed injury receive 3 points. If the injury involves multiple levels, the most severe level is scored. If more than one mechanism involves the same level, the score should represent the summation of mechanisms. The total TLICSS score reflects the severity of the injury and helps guide treatment. Patients with a score of 3 or less are treated nonoperatively. Those with scores of 5 or above are treated by surgical stabilization. Lastly, patients who score 4 points fall into the indeterminate category, and treatment is according to surgeon preference. Moreover, Lenarz and colleagues showed that the interobserver reliability of the TLICSS score was comparable with the Denis and AO systems.

The Iowa Classification System and Algorithm

Recognizing the schemes and classification systems described above, and based on data collected prospectively on 300 thoracolumbar fractures, a simple classification and algorithm for thoracolumbar spine fractures was developed ( Fig. 36-2 ). This algorithm is based on three criteria: 1) clinical, 2) biomechanical, and 3) radiographic. The clinical criteria address the presence or absence of pain or neurologic deficit. Biomechanical criteria describe the involvement of one, two, or three columns. Radiographic criteria address the degree of kyphosis, canal compromise, and the integrity of the PLC.

Surgical decompression and stabilization is recommended for those patients with neurologic deficit. Surgery for stabilization is also recommended for those who suffer from persistent pain despite bracing, and in spite of a normal neurologic exam, and cannot be mobilized satisfactorily. In patients with a normal neurologic exam, the biomechanical criteria are then used to address the Denis three-column involvement with the use of plain radiographs and computed tomography (CT) scans. In the presence of three-column injuries—such as a fracture–dislocations or flexion-distraction injuries, which are inherently unstable—patients are operated upon acutely. Patients with a single-column fracture (i.e., a wedge compression fracture) are treated nonoperatively with bracing. With two-column injury, such as burst fractures, the integrity of the PLC is assessed with MRI. Disruption of the PLC on MRI would render a burst fracture unstable; hence it would qualify the patient for stabilization. Clinical experience also suggests that older patients (>60 years) are more likely to fail with nonoperative treatment, necessitating surgical intervention.

Stabilization Techniques

Anterolateral Approach to the Thoracolumbar Spine (T11–L4)

The anterior or anterolateral approach is undertaken when the spinal canal is compromised from fracture, tumor, or infection. Under such circumstances it is difficult or impossible to decompress the cord and proceed with anterior grafting or stabilization through a midline posterior approach. Reconstruction of the anterior two columns is achieved by thoracotomy in the thoracic spine and through a flank retroperitoneal approach in the thoracolumbar or lumbar spine.

Patients are placed in the lateral decubitus position with the left side up ( Figs. 36-3 and 36-4 ). The approach is through the left side, on the same side as the abdominal aorta, which is less likely to be injured than the inferior vena cava. The patient is positioned with the flank over the break in the table. The operating table is flexed, opening up the left flank yet keeping the thoracolumbar spine fairly straight. The patient is immobilized on a beanbag supplemented with adhesive tape. The upper knee is slightly bent, and the knees, heels, and ankles are also padded. Wide adhesive tape is used to strap the patient to the operating table, and it is placed across the greater trochanter in a fashion that allows exposure of the iliac crest, should bone harvesting be required. Pneumatic compression stockings are placed to reduce the incidence of deep venous thrombosis (DVT), and a warming blanket is placed over the lower extremities and torso to maintain body heat during the procedure.

The fracture is confirmed using anteroposterior (AP) and lateral fluoroscopy. For T11, T12, and L1, it is necessary to enter the chest cavity. The left lung is deflated while ventilating the right lung through a double-lumen endotracheal tube. Through a 6-inch incision, the chest is entered, and the spine is visualized. Exposure may be facilitated by resection or by shingling of one or two ribs. Periosteal elevators and rib-resection instruments help prevent injury to the intercostal bundle. The incision should stop short of the lateral edge of the paraspinal muscles.

The exposed segmental intercostal arteries and veins may need to be ligated for the decompression and screw application. When necessary, the ligation should be performed as far from the neural foramina as possible; this should allow adequate blood flow to the spinal cord and conus through collaterals. Commonly, the intercostal arteries between T9 and L2 on the left give rise to the artery radiculomedullaris magna, or the artery of Adamkiewicz, the largest spinal branch to supply the lower thoracic cord and lumbar enlargement. To avoid spinal cord ischemia or infarction, ligation of these intercostal vessels should be undertaken sparingly and only when absolutely necessary.

For L2 through L4, the chest cavity can be avoided by making the incision between the costal margin and the iliac crest and by remaining in the retroperitoneal space, below the diaphragm. A 6-inch curvilinear incision is extended posteriorly from the lateral border of the rectus sheath parallel to the rib cage. Lower incisions that do not overlie the rib cage are carried through the layers of the abdominal wall down to the transversalis fascia. The latter is retracted anteriorly, and through it, the surgeon can visualize the kidney and ureter. The spleen is located above the kidney and may be seen or palpated. Blunt dissection is used to define the space behind the transversalis fascia, and it is carried down (posteriorly) to the psoas muscle.

Table-mounted self-retaining retractors can then be used to retract the psoas muscle posteriorly, the diaphragm and ribs superiorly, and the kidney and abdominal contents anteriorly. Extra care must be exercised in retracting the spleen to protect it from laceration. Using electrocautery, the psoas muscle is incised at its attachment to the vertebral bodies and is retracted farther posteriorly to expose the pedicles and neural foramina of the fractured and adjacent vertebrae to be instrumented.

Under magnification, either by loupe or microscope, the intervertebral disks adjacent to the fractured body are then excised. A corpectomy of the fractured body is performed using rongeurs and chisels, and the bone is saved for fusion ( Fig. 36-5 ). An air drill may sometimes be necessary but is less desirable, because the bone dust is hard to retrieve. The decompression of the canal extends across to the opposite side of the canal, until the contralateral pedicle is palpated ( Fig. 36-6 ). The dura can be felt with a Penfield dissector and can sometimes be inspected with a small laryngeal mirror. The corpectomy must be large enough to accommodate a graft with a large footprint to avoid telescoping into the adjacent bodies, and adjacent end plates are kept intact to provide rigid apposition to the graft and to minimize subsidence.