To the Editor:

I greatly appreciate Dr. Amaravati's interest in our article, “Diagnosis and Management of Thoracolumbar Spine Fractures" which was intended to be an educational overview of the contemporary management of thoracolumbar spine fractures in the adult patient.

It is clear that we are in need of a comprehensive classification system for thoracolumbar spine fractures that will allow us to easily communicate the characteristics of a fracture, confer an understanding of the mechanism of injury,and assist in formulating principles of treatment.

The McCormack load sharing classification is a theoretical classification system that stresses the importance of the anterior spinal column in regards to fracture stability and predicts the efficacy of a solitary posterior surgical approach in relationship to the degree of fracture comminution (1). This system is not comprehensive. The Gertzbein description of the AO fracture classification is a three part system with multiple subtypes (2). Due to the complexity of the subtypes of this system, a complete and thorough description of a fracture is rarely performed beyond the major description of the injury. A less than optimal intra and interobserver reliability has been determined for this system when it was used to subclassify thoracolumbar fractures.

The Denis Anatomic Classification System is still popular among North American surgeons because it includes the major types of thoracolumbar fractures and generally assists in guiding treatment(3). Most clinicians understand that if the middle column is injured, great vigilance is required to determine if significant instability exists. We have come to understand that if the posterior osteoligamentous complex is significantly disrupted, especially at the thoracolumabr junction, this is often an indicator of instability and more aggressive treatment i.e. surgery may be necessary.

The Ferguson and Allen Classification System is a mechanistic classification system and is also very useful in describing a thoracolumbar fracture.(4) It is less popular than the Denis classification system in North America but does have merit if one simply wants to describe a fracture and its proposed mechanism to a colleague.

Your second question on how to manage a fracture dislocation at the L5-S1 level is also interesting because it is often unnecessary and not optimal to attempt to rigidly stabilize this injury multiple segments above or below the lumbosacral junction. In this fracture a posterior open reduction is first performed followed by placement of an interbody spacer at the L5-S1 disk space for interbody support. Internal fixation is usually placed from L5 to S1 or possibly from L4 to S1. Some surgeons have recommended two points of fixation in the sacrum including a combination sacral ala screw and a S1 pedicle screw or a S2 ala screw and a S1 pedicle screw. We do not recommend iliac screw placement due to the fact that this would span the uninjured sacral iliac joint.

Your final question is in regards to contemporary thinking regarding short segment fixation. We, again, do not recommend short segment fixation at the thoracolumbar junction in the setting of significant anterior column comminution. Some surgeons recommend this as only a temporary stabilization strategy until healing occurs. Following healing, the internal fixation can be removed to avoid potential implant failure. This, unfortunately, requires a second operation. In a comminuted fracture at the thoracolumbar junction, we often recommend fixation two levels above and two levels below the fracture if a posterior approach is chosen. In fractures of the lower lumbar region such as a burst fracture of L4 or L5 that requires surgical intervention, short segment fixation is routinely used due to the anatomy of this region, i.e. lordosis. The lordotic alignment of this spinal region places less stress on the anterior column and multiple fixation points segments above and below a lower lumbar fracture would significantly decrease the amount of remaining lumbar spinal motion. The potential for loss of lumbar lordosis is also present with multiple levels of fixation in this region of the lumbar spine.

Sincerely,

Alexander R. Vaccaro, M.D. AV/lmy

References

(1)McCormack T.: Karaicovic, E, Gaines R.W.: The Load Sharing Classification of Spine Fractures. Spine 1994. 19:1741-1744.

(2)Gerzbein SD: Spine Update: Classification of thoracic and lumbar fractrures. Spine 1994, 79:626-628.

(3)Denis F: the three column spine and its significance in the classification of actute thoracolumbar spinal injuries. Spine. 1983;8:817- 31.

(4)Ferguson RL, Allen BL Jr.. A mechanistic classification of thoracolumbar spine fractures. Clin. Orthop. 1984;189:77-88.

To the Editor:

I am writing with regard to the Instructional course lecture “Diagnosis & management of Thoracolumbar Spine Fractures”, by Alexander R. Vaccaro et al. This is a good article, however there are few points which I feel the authors could clarify.

1. The classification of spinal injuries proposed are by Gertzbein (mechanism of injury and morphology )(1) , McCormack (load sharing ) (2) or Denis (anatomical) and Ferguson and Allen (mechanical). Which of these will be more useful practically in assessing and managing the patients.

The authors have mentioned that fracture dislocations are secondary to complex shearing forces and can occur anywhere along the thoracolumbar spine. They are unstable and need extensive fusion procedures spanning three levels cephalad to the injury level and two or three levels caudad to it. I would pose to them a hypothetical question: if a fracture dislocation occurs at L5-S1, how would they propose to stabilize and fix it.

The authors have mentioned that use of short segment posterior fusion with pedicle screw fixation is associated with a high rate of early failure of hardware and late loss of sagittal plane correction. In the literature, the best candidates for short segment fusion and instrumentation are young patients who need more mobility, have good general health and who can tolerate two stage procedures. In one report,no implant failed, the prevalence of pseudoarthrosis was not recorded, the sagittal index and compression percentage showed a slight loss, but the clinical outcome was not affected(3).

Thank you.

Yours sincerely,

Dr. Rajkumar S Amaravati.

References :

1. Gertzbein SD: Spine update: Classification of thoracic and lumbar fractures. Spine. 1994; 79:626-628

2. McCormack T, Karaikovic E, Gaines RW: The load sharing classification of spine fractures. Spine. 1994; 19:1741- 1744.

3. Aligizakis AC, Kotonis PG, Sapkas G, Papgelopoulos PJ, Galanakis I, Hadjipavlou A : Gertzbein and Load Sharing Classifications for Unstable Thoracolumbar Fractures. Clin Orthop. 2003; 411:77-85.

4. Ditunno Jr JF, Young W, Donovan WH, Creasey G: The international standard booklets for nuerological and functional classification of spinal cord injury: American Spinal Injury Association. Paraplegia. 1994; 32: 70- 80.