EPIDEMIOLOGY OF SPINAL TRAUMA SURGICALLY TREATED AT THE UNICAMP HOSPITAL

Objective: To report the epidemiological profile of patients operated for treatment of spinal trauma in the UNICAMP Hospital das Clínicas. Methods: Patients older than 14 years with spinal trauma operated in the service from 2012 to 2017 with complete radiological and clinical data were evaluated. Results: A total of 143 patients were included, 120 men and 23 women, with a mean age of 37.8 years. Falls from height (32%), motorcycle (26%) and car accidents (24%) were the most common trauma mechanisms. The most affected vertebral level was C6 (11%) and most patients (55%) had some neurological deficits. The overall postoperative complication rate was 23%. Conclusions: Spinal trauma has a high socioeconomic impact due to its high morbidity and mortality, and it is necessary to study its epidemiology for the development of public policies for prevention and treatment financing. Level of Evidence: III. Type of study: Retrospective study.


INTRODUCTION
Spinal trauma comprises spine lesions, with or without involvement of the spinal cord, as well as associated disc ligament lesions, secondary to any traumatic event, whether or not the result of direct impact to the spine. 1 Because of the relevance of spinal cord injuries in terms of morbidity and mortality, new care protocols are created to facilitate and improve patient treatment. 2n the United States, it is estimated that around 13,000 patients are victims of traumatic spinal cord injury, with a total cumulative cost of $1.1 to $4.6 million per patient, 3 considering hospital expenses, work days lost, and the burden on social security.In Brazil, there are an estimated six to eight thousand new cases per year.However, epidemiological studies are scarce at the national level. 46] Each segment should be evaluated separately, taking into account the differences and particularities between them.
During preadmission treatment, the rescue team should ensure the use of a cervical collar and decompressive mobilization care. 6ollowing admission, all patients with spinal trauma should undergo an emergency evaluation, which cannot be postponed. 7

METHODS
Patients who underwent surgery for spinal trauma in the Neurosurgery unit of Unicamp from 2012 to 2017 were evaluated.All the patients were operated on by the same surgeon (AFJ).Postoperative outpatient follow-up was generally performed at one month, three months, six months, and then annually.Vertebral lesions of any segment were included and all patients were older than 14 years of age.Cases operated outside of the service or with incomplete clinical and radiographic data were excluded from the study.
Through a review of the medical records, age, sex, date of accident, trauma mechanism, injury level, American Spine Injury Association Impairment Score (AIS) 8 (Table 1), and postoperative complications data were collected.For statistical analysis, we used STATA MP version 13 software (StataCorp ® ).
For the literature review, a search of the Pubmed/MEDLINE database was conducted using the MeSH descriptors: "Spine trauma" OR "Spine" OR "trauma" AND "surgery".We found 249,127 articles.Only articles that fit the theme, that were written in English, and that had complete text were evaluated, for a total of 415 articles.The abstracts of these were read and 25 articles were selected to be used in our study discussion.
This study was approved by the Institutional Review Board of Unicamp as CAAE -17337313.7.0000.5404.Because it was a study based in the review of medical records, the ICF requirement was waived.

Results by Segment Craniovertebral Junction (O-C1-C2)
There were 16 patients who underwent surgery for fractures of the craniovertebral junction, 13 men, and three women with an average age of 35.8 years.The primary injury mechanism was automobile accidents in seven cases (43.7%), motorcycle accidents in four cases (25%), and falls from heights in four cases (25%).
There were postoperative complications in only one case (6.2%), in which there was a dural lesion and meningitis occurred, treated successfully with antibiotics.There were no deaths in this segment.

A
Full neurological deficit, including the sacral segment.

B
Full motor deficit, feeling preserved (sacral segments present).

E
No deficit, however presented a temporary deficit, including nerve root deficit.
The main trauma mechanism was falls in 47 cases (32%), followed by motorcycle accidents in 38 cases (26%), and automobile accidents in 36 cases (32%), with these three causes corresponding to of the total traumatisms.Other less common causes were seven cases (4%) of diving in shallow water, seven cases (4%) of being run over, five (3%) of explosions, one (0.7%) of a kick by a domestic animal, and one (0.7%) case of direct physical aggression.
The postoperative complication rate was 23% (33 cases), the main complication being local infections with an overall rate of 11.9%Most of the fractures (53.9%) were in the thoracolumbar region, L1 being the most affected level (9%) in this segment.However, of the total number of fractures, those of C6 were the most predominant at 11%, even though subaxial fractures accounted for 36% of all the fractures.

Subaxial Cervical (C3-C6)
There were 44 patients who underwent surgery for fractures of the subaxial cervical spine, of whom 41 were men (93.1%) and 3 women (6.8%), with an average age of 40.9 years.The main injury mechanisms were automobile accidents with 15 cases (34%), falls from height with 13 cases (29.5%), motorcycle accidents with seven cases (15.9%), and diving in shallow water with six cases (13.6%).
Eleven patients had postoperative complications (25%), the main one being pneumonia or ventilatory difficulties in five cases (11.3%), followed by systemic infectious complications with four patients (9%), two of whom died from sepsis in the postoperative period (4.5%).
Twenty patients had postoperative complications (24%), the main one (13 cases, 15.6%) being infection.Of these, three cases (3.6%) were of sepsis and one case was meningitis (1.2%).Five cases had problems with the instrumentation and reoperation was required (6%).None of the patients with thoracolumbar trauma died during postoperative hospitalization.

Sacral (S1-Coccyx)
No sacral fractures were treated surgically in this series.

DISCUSSION
It is believed that the worldwide prevalence of spinal trauma is between 236 and 4187 cases per million inhabitants, 9 although these data may be underestimated due to a lack of reporting.However, the prevalence of spinal trauma in men is around three to four times greater than in women, 10 similar to the findings in our series (84% vs. 16%).
The principal causal mechanisms of spinal trauma in several studies were falls, land transportation accidents (car, motorbike), recreation (sports and leisure), and aggression. 11Although this agrees with the data obtained in our series, the lack of adequate information about the trauma mechanism in epidemiological studies and the fact that the information available is not always exact (ex. a fall during a recreational activity versus a fall from a height), underestimating or overestimating the data presented.
There are two peaks of incidence, one in young adults from 15 to 29 years of age, more related to traffic accidents, and a second after 65 years of age, usually related to falls from a standing position. 12aradoxically, there is a higher incidence of spinal trauma in developed countries.However, this could be secondary to underreporting of spinal trauma in underdeveloped countries, as well as to the lower number of published epidemiological studies.
The evaluation of a patient with spinal trauma should follow the criteria proposed by the Advanced Trauma Life Support (ATLS). 13Most patients with spinal trauma are victims of high kinetic energy traumas.In our series, land transportation vehicle accidents corresponded to almost 50% of all trauma mechanisms, with a high probability of lesions to other organs.Adequate ventilation and oxygenation support must be ensured and hemodynamic stability achieved before the secondary assessment of the spine surgeon.
During the secondary evaluation of a polytraumatized patient, the exam of choice for assessing the integrity of the spine is computed tomography (CT) with three-dimensional reconstruction.The use of simple radiographs was widely used for a long time.However, we know that for some injuries, especially those of the craniovertebral junction (CVJ), such as condyle fractures, simple radiography has lower sensitivity and specificity than CT in detecting fractures. 14ver time, numerous ways of classifying spine fractures through imaging exams have been proposed to facilitate their treatment and understand their severity.Thus, classification by the morphology of the injury is well accepted.
Fractures can be classified as type A (compression/burst), secondary to compressive axial overload, type B (injuries with lesion of the tension band of the anterior or posterior spine), and type C (rotation/translation), secondary to translational and rotational loads. 15Classically, type A fractures (compression/burst) tend to be stable and types B and C classified as unstable, each with its own particularities. 16] Additionally, fractures can usually be classified as either stable or unstable.The concept of stability was defined by White and Panjabi 20 as the capacity of the spine to maintain alignment under physiological loads, without pain or neurological injury.
However, the previously proposed concept of stability is subjective, causing divergence in the conduct of the spine surgeons.To standardize the conducts, the Spinal Trauma Study Group proposed algorithms to determine objective instability criteria based on the morphology of the fracture, the neurological status of the patient, and on the integrity of the posterior ligament complex.They suggested the SLICS 21 and the TLICS 22 (Tables 2 and 3).In both scores, patients with fewer than four points are treated non-surgically, patients with more than four points are treated surgically, and patients with four points may be treated surgically or non-surgically depending on other variables, such as comorbidities, the experience of the surgeon, bone or metabolic diseases, etc.
The exception is in the craniovertebral junction (CVJ) that has specific classification for each type of lesion, whether of the condyle, of C1, or of C2. [23][24] Joaquim et al. [23][24] proposed instability criteria for the CVJ to guide surgical conduct (Figure 3).Patients with suspected ligament injury (misalignment of the vertebrae, increase in

Neurological status
Intact 0

Partial deficit 3
Persistent compression of the canal* +1 the atlantodental interval, articular luxation, etc.) may be potentially unstable, requiring stabilization.In general, fractures of the CVJ are handled conservatively, however, attention should be given to Anderson and D'Alonzo 25 type two fractures of the odontoid process, which may present late instability and neural compression.

CONCLUSIONS
Spinal cord injury is a complex pathology and its treatment is full of nuances.It should be treated as a neurological emergency  Adapted from Joaquim et al. [23][24] due to the imminent risk of serious and permanent deficit, which causes high morbidity and mortality rates.Epidemiological knowledge is fundamental to planning preventive measures and to directing treatment resources.

Figure 1 .
Figure1.Shows the distribution of the 143 patients in relation to neurological damage according to the AIS.Most patients (45%) had no neurological deficits in the initial evaluation (E), followed by 35% with partial deficits (D, C, and B), and 20% with full sensory and motor deficit (A).

Figure 2 .
Figure2.The figure shows the distribution of the 204 fractures found in these patients.Most of the fractures (53.9%) were in the thoracolumbar region, L1 being the most affected level (9%) in this segment.However, of the total number of fractures, those of C6 were the most predominant at 11%, even though subaxial fractures accounted for 36% of all the fractures.

Figure 3 .
Figure 3. Flowchart for the evaluation of the instability of high cervical lesions.Adapted from Joaquim et al.[23][24]

Table 1 .
8eurological grading according to the AIS (ASIA Impairment Scale).8

Table 3 .
Thoracolumbar Injury Classification System and Injury Severity Score (TLICS).Adapted from Patel et al. 22 Fractures of the condyle and C1 -initially conservative with a Philadelphia cervical collar Fractures of C2 -initially conservative: Surgery if -type 2 odontoid fracture in patient > 50 years of age, deviation >5mm or comminution if -fractures that do not consolidate in control image exams Morphology of the fracture No Evaluate the ligament injury: misalignment, locked facet joints, alantodental interval > 3.5mm