Posterior short segment pedicle screw fixation for the treatment of thoracolumbar fracture

The choice of the type of stabilization device in the osteosynthesis of dorso-lumbar spine fractures remains a subject of controversy. The present study aims to evaluate the efficiency of short segment in patients suffering post-traumatic thoracolumbar fractures. This study was conducted in the Department of Orthopedic Surgery and Traumatology of the Habib Bourguiba University Hospital, Sfax, Tunisia. All our patients had a spinal osteosynthesis via the posterior approach with a short segment pedicle screw fixation. We established a record of the pre and post-operative data, the functional results in the post-operative stage during the follow-up period and in retrospect according to the Denis Pain Scale, as well as the Oswestry score. The correction was evaluated by determining the relative gain and loss at the last period of retrospect: vertebral kyphosis, regional kyphosis, Gardner Segment Kyphotic Deformity (GSKD), and computed tomography (CT) scan in retrospect to check the quality of the arthrodesis. The average Oswestry score was 14%. Twenty-nine patients had an Oswestry score ≤40%. The relative gain obtained postoperatively was 57.3% for vertebral kyphosis, 67.2% for regional kyphosis and 71.3% for Gardner kyphosis deformity; while the loss of correction at the last follow-up was 0.6° for vertebral kyphosis, 1.5° for regional kyphosis and 0.9° for GSKD. No cases of non-union were reported. The short segment fixation makes it possible to limit operating time, the abundance of bleeding and the aggression of the soft tissues.


Introduction
Over the last two decades, numerous technical advances in instrumentation and in the knowledge of spinal biomechanics have modified the surgical strategies for the synthesis of spinal-lumbar spine fractures [1]. The choice of the stabilization device type in the osteosynthesis of dorsolumbar spine fractures remains a subject of controversy. Some authors take into consideration only the type of fracture and the degree of its comminution [2], while some others choose the segmental instrumentations according to the patient´s habit [3]. Through our work, we evaluate the short segment pedicle screws fixation of thoracolumbar fractures.

Methods
We conducted a retrospective, descriptive study over an 8-year period year. We established a record determining the clinical and radiological data of each patient. On the clinical level, we specified the general information about the patient and the circumstances of the accident.
The record also included the characteristics of the fracture and its consequences on the spinal statics, the neurological status, the pre and post-operative data (operating time, transfusion, neurological release, length of hospital stay, post-operative recovery, and rehabilitation), and the postoperative. Retrospective clinical and anatomical results according to the Denis Pain Scale, the Oswestry score (ODI: Oswestry Disability Index), the Schöber Index and return to work were considered in the report. On the radiological level, we estimated the correction by determining the relative gain and losses at the last follow-up vertebral kyphosis (VK), regional kyphosis (RK) and Gardner Segment Kyphotic Deformity (GSKD). We also performed the study of sagittal balance through the sagittal heel in T9 and the sagittal vertical axis (SVA) measured on an x-ray of the entire spine standing front and in the profile made at the last retrospect.
Scannographic analysis was performed to check the quality of the arthrodesis.

Results
Our population included 29 men and only 1 woman. Average age was 32.6 years with a standard deviation of 10.7. The trauma was due to falls from a high place in 50% of the cases (palm tree climbers) and to road accidents in 46.7% of the cases (Figure 1 Twenty-six patients (86.7%) returned to work within 7.9 months and 4 patients (13.3%) stopped working permanently. The average Schöber index was 12.9/10 cm with a standard deviation of 0.6 cm.
All patients initially presenting a partial neurological deficit improved by at least one grade according to the Frankel classification, 50% of which showed total neurological recovery. We also deplored a case of sepsis on early material. However, no case of non-union or implant failure was noted. Regarding the radiological results, the relative gain obtained postoperatively was 57.3% for VK, 67.2% for RK and 71.3% for GSKD (Table 1) and the loss of correction at the last follow-up was 0.6° for VK , 1.5° for RK and 0.9° for GSKD ( Figure 5). As for the study of sagittal balance, 8 patients (33.3%) were well balanced and 16 patients (66.7%) were unbalanced. In our series, all the fractures were consolidated and no case of implant failure was reported ( Figure 6). According to Parker JW et al. [13], this classification, whose total score is between 3 and 9 points, makes it possible to perform a short segment fixation taking one level above and one level below the fractured vertebra if the score is less than 7 points. However, Elzinga M et al. [14] questions this classification and its therapeutic interest.

Discussion
In fact, he brought in three observers for 40 thoracolumbar fractures to determine the load-sharing score (LSS) twice. He used Cohen's Κ (kappa) test, and found great variability between different observers, and even within the same observer. In agreement with Elzinga M et al. [14], the present work highlights the inadequacies of this classification which neglects the importance of load distribution on the vertebral body without taking into account the state of the complex posterior ligament or of the neurological status. Altay M et al. [15] thinks that this classification is not enough to choose between a short segment fixation and a long segment stabilization and proposes an association between the classification of Load Sharing and that of Magerl. In fact, Katonis P et al. [16] argues that a short segment fixation is sufficient in the lumbar region and constitutes a solid device without loss of correction; while Eno J-JT et al. is 14% for our series and is higher than 40% only in 3.3% of cases.
This score is close to that reported by Steib J-P et al.

What this study adds
• The short segment minimizes the extent of the skin incision; • The short segment limits the damage of soft tissues; • Prescribing the short segment fixation makes it possible for consolidation and arthrodesis of thoracolumbar fractures without incurring a significant loss of correction or a high rate of instrumentation failure.

Competing interests
The authors declare no competing interests.