Traumatic spine injuries in Eastern India: A retrospective observational study.

BACKGROUND: Trauma is the leading cause of hospitalization globally, and trauma-induced spinal injuries can be devastating and permanent. The objective of this study was to describe the pattern, association, and outcome in patients with traumatic spine injury (TSI).
METHODS: A retrospective cross-sectional study was undertaken on patients with TSI who presented to the trauma and emergency department of a level 1 trauma center in eastern India between August 15, 2018, and August 14, 2019, by including 103 patients. Information pertaining to demography, mode of injury (MOI), fracture morphology, neurological grading, and associated spinal or other regional injuries was obtained. Correlation among injury severity score (ISS), neurological damage as per American Spinal Injury Association (ASIA), and morphological patterns was determined.
RESULTS: The median age was 39 years, and the gender ratio was 5.87:1. Fall from height (43.7%) was the most common MOI. The median ISS was 21, and the percentage of patients with polytrauma was 73% (ISS > 15). The cervical region (n = 30) was the most common site of injury, and multiple vertebral involvement (n = 32) was more common than isolated involvement. Type A pattern (53.4%) was the predominant type, followed by types C and B (29.1% and 15.5%, respectively) for primary spine injury, and type A was the predominant type for secondary spinal injury. Severe neurological damage (ASIA A-C) was noticed in 69 patients. The correlation between ISS and ASIA scores (Spearman's ρ = 0.561, P < 0.001) and between morphology type and ASIA score (Pearson's χ 2= 69.7, P < 0.001) was statistically significant. In total, 53 patients were managed surgically and 24 patients were managed by conservative measures.
CONCLUSION: Our study found a predominantly younger population, multilevel involvement, significant neurological damage, multiple associated injuries, and higher ISS among the patients of TSI. The pattern in eastern India is different from previous reports from other parts of the country. Copyright:

INTRODUCTION

Globally, trauma is the leading cause of hospitalization, and spinal injuries due to trauma can be devastating and permanent.[12] Several studies have reported a high prevalence of trauma-associated spinal injuries in the young and economically productive age group, wherein the patients require long-term intensive rehabilitation endeavors.[35678] The increasing prevalence has been a cause of a significant burden on the family and state in terms of health-care costs. A few epidemiological studies on spinal trauma have been performed in India and in other countries.[34679101112] However, findings pertaining to demographic profile and injury region are disparate in these studies because of varied geographical or cultural differences. Moreover, the morphological pattern has not been given due importance in most of the epidemiological studies.[3] This pattern is closely related to injury severity, neurological damage, and eventually in the long-term functional outcome. Although studies have elaborated on associated injuries (AIs), most of them have mainly focused on extraspinal regions.[378] However, most of the trauma centers perform computed tomography of the entire spine, which has facilitated the identification of multilevel involvement more frequently than expected.

The present study aimed to primarily identify the pattern of injury and presence of associated spinal and extraspinal injuries and to determine the correlation between severity, morphology, and grades of neurological damage in addition to the demographical characteristics of patients with traumatic spine injury (TSI) who presented to the emergency department of a level 1 trauma care center in eastern India.

METHODS

This retrospective cross-sectional study was performed at the trauma and emergency department (TED) of a level 1 trauma care center in eastern India. The study sample included patients with TSI admitted to the TED between August 15, 2018, and August 14, 2019. The Investigational Review Board of All India Institute of Medical Sciences, Bhubaneswar, approved the study (IEC number T/IM-NF/Ortho/19/34). The study is compliant with the Strengthening the Reporting of Observational Studies in Epidemiology statement guidelines for reporting observational studies. The inclusion criteria were as follows: (1) traumatic spinal cord injury; (2) hospital admission through a TED; and (3) adult patients of all age groups. The exclusion criteria were as follows: (1) presentation to an outpatient clinic rather than a TED; (2) nontraumatic (i.e., pathological) spinal fracture; and (3) incomplete data. The patients' paper-based records were retrieved from the medical record department, and the data were compiled on a structured pro forma in an MS Excel worksheet. Data pertaining to patient demographics (age and sex), mode of injury (MOI), region of spinal injury (cervical, thoracic, lumbar, and sacral), and injury morphology were collected. The collected data were classified according to Muller AO (Arbeitsgemeinschaftfür Osteosynthesefragen) classifications,[13] and injury severity score (ISS),[14] surgical outcome, and 7-day mortality rate were determined. All patients underwent spinal imaging that included plain radiographs and whole spine computed tomography (CT) scans (GE LightSpeed 64 slice CT scanner). The involvement of segments was noted as mono (1 segment), multicontinuous (≥2 segments), or multi-noncontiguous (≥2 segments, skip lesions).

Spinal cord injuries were graded according to the American Spinal Injury Association (ASIA) classification.[15] Grades A to C were grouped together as useless functional power, and Grades D to E were grouped together as useful functional power.[16] The ISS was calculated on the basis of AIs classified as per the Advanced Trauma Life Support guidelines as follows: head injury (HI), maxillofacial injury, chest injury, abdominal injury, extremity injury, and external injuries.[17] The outcome after injury was classified on the basis of the selected management approach, surgical or conservative. The number of patients referred was also noted along with early death (within 7 days).

Statistical analysis

Statistical analysis was performed using R version 3.6.1©(The R foundation, Vienna, Austria). Categorical variables are expressed as frequency or percentages. Data were analyzed for normality according to the Shapiro–Wilk test. Bivariate analysis of categorical variables was performed using Chi-square test. Numerical variables are expressed as median ± interquartile range (IQR). Spearman's correlation was used to determine the correlation between ISS and ASIA. P < 0.05 was considered statistically significant.

RESULTS

A total of 153 patients were considered eligible for the study. Of the total, 103 met the inclusion criteria and their complete data were available. The median age of the patients was 39 years, with the IQR and age range of 28–49 years and 15–75 years, respectively. The male: female (M: F) ratio was 5.87:1 (88:15). Fall from height (43.7%) was the predominant MOI, followed by road traffic accidents (RTAs) (37.9%), and trauma in a few cases was due to other reasons [Figure 1]. In the other reasons group, we had included 2 patients who had been electrocuted. The most commonly affected age group was 31–40 years, followed by 21–30-year and 41–50-year age groups. The mode of transport was government ambulance for the majority of patients (50.5%), followed by private ambulance for 28.2% of the patients and private vehicles for 20.4% of the patients.

Figure 1

Bar diagram displaying the mechanism of injury in different age groups

The median ISS was 21, 14–27 (4–75), reflecting that 73% of our patients had polytrauma (ISS > 15). A higher mean ISS in patients with multilevel involvement (31%) than those with thoracic (23%) or cervical (20%) spine injury was observed. Among the isolated spine injuries, cervical spine was involved in 30 patients, dorsal in 24, lumbar in 13, junctional (dorsolumbar) in 3 and one patient had sacral fracture. Continuous multilevel and skipped multilevel were found to be 14 (13.6%) and 18 (17.5%), respectively.

The major fracture morphology type was type A (A1–4) (n = 55, 53.4%), followed by type C (n = 30, 29.1%) and type B (n = 16, 15.5%) in case of primary spinal fractures [Figure 2]. Three patients not included in the above morphological categories are shown as type D in the same figure who exhibited Jefferson's fracture, odontoid fracture, and sacral fracture each. Severe injury type C was more common in males than in females (29:1), with the C5–6 vertebrae as the most frequently dislocated region. Associated spinal (secondary) injuries were mostly A1 and A2 fractures. Three skipped multilevel patients had A0 fracture morphology; 2 patients exhibited the spinous process fracture and 1 patient exhibited transverse process fracture.

Figure 2

Bar diagram showing the morphological pattern in a different region of spinal injury

Complete neurological damage or ASIA A was observed in most of the patients (n = 30, 29.3%), and ASIA E (neurologically intact) was observed only in 14 patients [Figure 3]. Among incomplete neurology, useless functional power was observed in 69 patients, whereas useful functional power was observed in 34 patients. Table 1 summarizes the AIs. Some of the fracture patterns are illustrated in Figures 4–6.

Figure 3

Stacked bar diagram showing the distribution of various regions of spinal injury in different neurological grade

Table 1

The pattern of associated injuries with traumatic spine injury (n=103)

Type	Grade/Class	Region	Frequency (%)
Associated spinal injuries (n=32)	Multilevel continuous (n=14)	Cervical	8 (7.7)
		Dorsal	3 (2.9)
		Lumbar	2 (1.9)
		Junctional (dorsolumbar)	3 (2.9)
	Multilevel discontinuous (n=18)	Cervical and dorsal	3 (2.9)
		Cervical and lumbar	1 (1)
		Dorsal and lumbar	4 (3.8)
		Lumbar and lumbar	3 (2.9)
		Dorsal and dorsal	5 (4.8)
Associated nonspinal injuries (n=95)	Type of injuries	Head injury	25 (24)
		Maxillofacial injury	12 (12)
		Blunt trauma chest	19 (18)
		Blunt trauma abdomen	4 (4)
		Soft tissue injury	60 (58)
		Extremity injury	28 (27)
	Associated fracture injuries	Skull	6 (6)
		Ribs	12 (12)
		Pelvic	2 (2)
		Shoulder girdle	3 (3)
		Upper extremity	16 (16)
		Lower extremity	12 (12)

Figure 4

Various types of isolated type of vertebral injury

Figure 6

Various types of skip injuries are shown with an arrow marking at the levels of injury

Various types of associated continuous spinal injury. The number of arrows showing the number of vertebral involvement

We found a significant correlation between the ISS and ASIA scores (Spearman's ρ = 0.561, P < 0.001), and the numbers are shown in Figure 7. Chi-square association between neurological grade (ASIA) and morphology to grade was also found to be statistically significant (Pearson's χ2= 69.7, P < 0.001).

Figure 7

Box and whisker plot showing the dispersion of injury severity score in various neurological grades

Twenty four TSI patients (23.3%) were managed conservatively at our centre while 53 patients (51.5%) underwent surgery and we had to refer 26 patients (25.2%). Two hospitalized patients died early due to polytrauma. One had a fatal HI along with a cervical spine injury. The other was a dorsal spine fracture with chest and HI. Both died before any surgical intervention.

DISCUSSION

Our study included patients with TSI who presented to the TED of a level 1 trauma center based in eastern India during a 1-year period. Young age group (20–30 years) and males were predominately affected, and this finding is similar to that of most studies in India and abroad.[3812] Conversely, Aleem et al. reported the mean age of the predominantly affected patients to be 51.2 years in their cohort.[6] Interestingly, Singh et al. noted a decreasing trend of sex skewness among injury survivors in several studies performed chronologically, with a M: F ratio of 2.86:1.[12] We reported a greater M: F ratio of 5.9:1, which contradicts the finding of Aleem et al. who reported the lowest sex ratio of 1.52:1.[6] Our patients predominantly belonged to rural areas, where males were breadwinners; the population included in a study performed by Aleem et al. belonged predominantly to the metropolitan cities where the skewness was reported to be less. Similar to the observations of Leucht et al., we found a higher number of males with more severe injuries.[3]

We found that fall from height (roof, trees, and electric poles) is the most common MOI, and the finding is similar to that of other Indian studies, except a study by Singh et al. who found RTA as the more common cause as reported in the western studies.[37] However, our percentage of RTA was higher than that reported in other Indian studies, and this accounts for higher AIs in our study. We also had 2 cases of electrocution, leading to fall from height similar to observations by Mathur et al.[18] Leucht et al. found that thoracic injury was predominantly caused by RTA because a considerably strong force is required to disrupt the dorsal spine protected by a rigid rib cage;[3] however, we did not observe this in our study, although the mean ISS was more for thoracic injury than for cervical injury, which corroborates the findings of den Ouden et al.[11]

Singh et al. found that prehospital services are available to approximately 79% of the patients because they have an organizational setup.[7] Approximately one-fifth of our patients were transported by private vehicles and not by ambulance. The trend in the use of ambulance services has changed in our country after the implementation of toll-free ambulance services throughout the nation; according to a study, ambulance services were available to only 23% of the patients 15 years ago.[12] In more challenging terrains such as the north-eastern part of our country, Birua et al. reported that patients still use private vehicles and that more than 90% of the patients are not accompanied by trained personnel.[8]

The regional pattern of injury has been reported to vary considerably across various Indian studies. Birua et al. found the cervical region as the most common site of injury,[8] whereas Singh et al. reported the thoracic region as the most common site of injury.[7] Thoracolumbar region and lumbar spine were reported as the most common injury site by Singh et al.[12] and in a multicenter study performed by Aleem et al.,[6] respectively. We observed that the cervical spine was the single most commonly involved region, but then multilevel vertebral injury dominated the overall percentage. Studies conducted in China and Germany have reported higher lumbar involvement,[319] whereas a study in The Netherlands reported high thoracic injury and multilevel involvement,[11] which was also observed in our study. None of the Indian studies have focused on the morphological pattern of injury. We found that the morphological pattern was strongly correlated with the ISS and neurological damage at the time of hospitalization. In a study performed by Leucht et al., more than half of the patients were type A (54%), followed by type C (18.5%) and type B (16.9%), and approximately 10% had C1–C2 injuries.[3] In our study, a similar pattern was observed; 53% were type A (A1–A4) but higher type C (29%) and only 2 cases of C1–C2 fractures and one sacral fracture. Unlike the observations reported by Leucht et al., we did not find any fracture bias toward the junctional area in the dorsal spine and noticed only 2 dorsolumbar injuries.[3]

Our study demonstrated a higher percentage of patients with neurological damage at the time of hospitalization than that reported in other Indian studies.[718] On the other hand, Leucht et al. found that 75% of the patients exhibited intact neurology.[3] Van et al. and Kiwerski reported the incidence of complete tetraplegia and paraplegia (ASIA A) to be 4.8%–50.6% and 16.0%–85.1%, respectively.[2021] In our study, 67% of the patients had useless motor power (ASIA A–C) compared to useful motor (ASIA D–E), of which only 13.6% of the patients had ASIA E. Extensive neurological damage implies more devastating consequences for the patient, family, and state in terms of economic burden. A large number of type A fractures reported by Leucht et al. were osteoporotic spine injuries in female patients. This observation could be responsible for balanced sex distribution and probably the intact neurology reported by the authors.[3] Type B and type C fractures increase the neurological impairment compared with that caused by type A fractures, as reported in a study.[22] We found a positive correlation between the injury severity and depth of neurological damage and between morphology and ASIA-based grading severities. Leucht et al. also observed that more than 60% of the patients with multilevel injuries were completely paralyzed.[3] However, in our study, more patients with cervical spine injury were classified in the ASIA A category than multilevel fractures.

Several studies have highlighted the involvement of more than one vertebra in spinal injuries.[2324] However, none of the Indian studies have highlighted any trends. Calenoff et al. described 3 patterns of skipped involvement decades ago.[23] Choi et al. observed noncontiguous cervicothoracic junction or upper thoracic spinal injuries in 28% of the patients with cervical spinal injury.[24] We found continuous multilevel involvement in the cervical region but noncontinuous involvement in the dorsal and lumbar spine. Secondary vertebral fractures mainly showed type A (compression type) pattern. This morphology of secondary injuries has not been reported earlier. The exact etiology and mechanism of this complex type of injury require further biomechanical evaluation. However, RTA as a mechanism of injury, young age, higher neurological damage (ASIA C and above), greater ISS, and AI were observed in the majority of patients in our study.

Extraspinal AIs were observed in 25.7% of the patients in the study performed by Mathur et al.[18] We observed a higher percentage of up to 57.3% in our study (excluding sexually transmitted infections) but lower than that reported by den Ouden et al.[11] Aleem et al. and Singh et al. found HI as the most common associated injury.[67] Lida et al. reported that one-third of the patients with spinal cord injury were associated with moderate or severe HI.[21] We observed more extremity injury (27.3%) compared to HI (24.3%). Leucht et al. also reported higher extremity injuries and a high association between multilevel spinal and associated injury.[3] As also observed by Mathur et al.,we found rib injury to be more common in thoracic spine injury.[18]

More than 50% of our patients underwent surgery, unlike the treatment received by patients included in other Indian studies wherein they were managed conservatively.[812] Our study cohort comprised more patients with unstable morphology and neurological damage requiring surgical intervention.

Although with a limited dataset, we attempted to determine an association between various spinal and extraspinal regions and the morphological and neurological patterns in multiple areas. However, we had to exclude some patients because the study design was retrospective and data were insufficient. Bhubaneswar, being the capital city, has 3 more medical colleges, and there is patient segregation that accounts for low numbers in our study group. Moreover, we did not include prehospital management information and whether any trained personnel accompanied the patient at the time of hospitalization. Long-term outcomes in terms of functional status were not evaluated because many patients exhibited neurological damage. Prospective multicenter studies should be performed to validate and clear picture about the various associations with TSI.

CONCLUSION

The epidemiology of spinal injuries in eastern India is different from that reported in other parts of the country. We found more than one vertebral level injury to be predominant than any regional isolated injury. Injury severity, AIs, and neurological damage were more prevalent in our study cohort. Because the younger age group is predominantly affected, the risk of long-term consequences is considerable. Hence, government health-care policies should be directed toward the treatment and rehabilitation of these patients considering the dearth of proper spinal rehabilitation centers that are crucial for making such patients physically independent to some extent. Furthermore, simultaneous preventive measures should be stringently implemented.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Research quality and ethics statement

The authors of this manuscript declare that this scientific work complies with reporting quality, formatting, and reproducibility guidelines set forth by the EQUATOR Network. The authors also attest that this clinical investigation was determined to require the Institutional Review Board/Ethics Committee review, and the corresponding protocol/approval number is 1212092-1. We also certify that we have not plagiarized the contents in this submission and have done a Plagiarism Check.