Nigerian Journal of Orthopaedics and Trauma

ORIGINAL ARTICLE
Year
: 2021  |  Volume : 20  |  Issue : 1  |  Page : 23--31

Magnetic resonant imaging features of acute traumatic cervical spinal cord injury: Its correlation with american spinal injury association impairment scale and prediction of neurological improvement using a low tesla magnetic resonant imaging in a resource-constrained environment


Jeneral Dumura Alfin1, Ayodeji Salman Yusuf1, Raji Mohammed Mahmud1, Danaan Joseph Shilong2, Gyang Markus Bot2, Simon J Yiltok3,  
1 Department of Surgery, Neurosurgical Unit, National Hospital Abuja, Abuja, Nigeria
2 Department of Surgery, Division of Neurosurgery, Jos University Teaching Hospital, Jos, Plateau State, Nigeria
3 Department of Surgery, Division of Plastic and Reconstruction surgery, Jos University Teaching Hospital, Jos, Plateau State, Nigeria

Correspondence Address:
Dr. Jeneral Dumura Alfin
Department of Surgery, Division of Neurosurgery, Jos University Teaching Hospital, Jos, Plateau State
Nigeria

Abstract

Context: Traumatic spinal cord injury (TSCI) is a devastating disease, hence the need to identify clinical and radiological injury features that can predict neurological improvement. Aims: This study aims to identify magnetic resonant imaging (MRI) features in cervical TSCI that correlates with neurological status at admission, and also predict early neurological improvement. Settings and Design: This was a prospective cohort study. Subjects and Methods: Admission MRI features of 47 patients with cervical TSCI and their neurological assessments at admission and 3 months post-injury were reviewed prospectively over a period of 18 months. Correlational and regression analyses were done using SPSS® version 25 software. P < 0.05 was used as the level of significance. Results: Spinal cord oedema and cord contusion (78.0%) constitute the majority of injury patterns seen on MRI. There was a significant association between spinal cord contusion and cord oedema on MRI and incomplete TSCI. Likewise, spinal cord haemorrhage, compression and transection were associated with complete TSCI. Maximum canal compromise (MCC), maximum spinal cord compression (MSCC) and length of cord lesion significantly correlate with American Spinal Injury Association Impairment Scale at admission (P = 0.033, P = 0.015 and P < 0.001, respectively). Increasing values of these variables were found to be independent predictors of complete TSCI. However, length of cord lesion was the only independent predictor of neurological improvement, 3 months post-injury (P = 0.025). Conclusions: Spinal cord haemorrhage, compression, transection and higher values of MCC, MSCC and increased length of cord lesion were predictive of complete TSCI. However, the length of spinal cord lesion was a better predictor of early neurological improvement.



How to cite this article:
Alfin JD, Yusuf AS, Mahmud RM, Shilong DJ, Bot GM, Yiltok SJ. Magnetic resonant imaging features of acute traumatic cervical spinal cord injury: Its correlation with american spinal injury association impairment scale and prediction of neurological improvement using a low tesla magnetic resonant imaging in a resource-constrained environment.Niger J Orthop Trauma 2021;20:23-31


How to cite this URL:
Alfin JD, Yusuf AS, Mahmud RM, Shilong DJ, Bot GM, Yiltok SJ. Magnetic resonant imaging features of acute traumatic cervical spinal cord injury: Its correlation with american spinal injury association impairment scale and prediction of neurological improvement using a low tesla magnetic resonant imaging in a resource-constrained environment. Niger J Orthop Trauma [serial online] 2021 [cited 2021 Dec 6 ];20:23-31
Available from: https://www.njotonline.org/text.asp?2021/20/1/23/320897


Full Text



 Introduction



Cervical spinal cord is the most frequently injured segment of the spinal cord following trauma to the spine; this injury is often associated with significant morbidity and far-reaching physical, emotional and economic consequences for the patients, their families and the society at large.[1],[2],[3],[4],[5],[6],[7],[8] The most important aspect of caring for these patients is a quick, accurate and comprehensive assessment for possible spinal instability and determination of the extend of neurological deficit. However, the initial neurological examination may be fraught with difficulty, because some of these patients are obtunded by sedation or may be multiply injured resulting in poor cooperation and difficulty in achieving a satisfactory neurological examination, upon which the prognosis of this injury is largely based.[9],[10]

It is often difficult, to accurately predict the extent of neurological recovery at presentation, due to our inability to non-invasively visualise and characterise the type and extent of damage sustained by the spinal cord.[9],[11],[12],[13] Plain radiographs and computed tomography scans are valuable in demonstrating skeletal injuries to the spine but are limited in demonstrating soft-tissue injuries or internal architectural changes in the spinal cord such as spinal cord oedema, cord contusions, transections, disk herniation, epidural haematoma and external compression of the spinal cord.[14],[15],[16],[17] As a result, clinical, therapeutic and prognostic decisions are largely dependent on physical examination and bony pattern of injury.[13],[14],[15] On the contrary, magnetic resonant imaging (MRI) has been shown to demonstrate and characterise different types of injury patterns sustained by the spinal cord, as well as other soft-tissue injuries around the cord following trauma to the spine. This makes MRI a valuable diagnostic tool in the evaluation of patients with traumatic spinal cord injury (TSCI).[13],[15],[17],[18],[19]

The correlations between MRI features of cervical TSCI and neurological outcome have been studied extensively in the literatures using high tesla MRI machines.[8],[14],[15],[17],[20],[21],[22],[23],[24],[25] Moreover, these studies demonstrated that these MRI features of cervical TSCI correlate with the American Spinal Injury Association Impairment Scale (AIS) during follow-up.[15],[17],[18],[19],[21],[25],[26],[27],[28] These MRI features need to be evaluated in a resource-constrained setting with low tesla MRI scans, to see if these features can be used to complement clinical parameters in predicting outcome, especially in patients that clinical assessment may be difficult or unreliable. The goal of this study is to identify MRI features of cervical TSCI that correlates with neurological status at admission, and also predict early neurological improvement using a low tesla MRI machine in a resource-constrained environment.

 Subjects and Methods



This is a prospective cohort study, conducted between May 2016 and December 2017. Approval was obtained from the Research and Ethics Committee of our institution (Research code NHA/EC/005), and a desired sample size was calculated using Fisher's formula.[29] The participants were patients brought to the emergency room of our hospital with clinical and or radiological features of acute cervical TSCI. Patients who met the inclusion criteria after review and gave consent were recruited for the study. However, patients who were unconscious, at the time of presentation and review at the emergency room or those who were unable to follow or obey instructions and those who presented 2 weeks after injury were excluded. Patients who could not do MRI due to financial constraints, patients with implants or foreign bodies that were not MRI compatible and those who leave against medical advice (LAMA) were also excluded from the study.

Information on the biodata of the patients, mechanism of injury and MRI features of TSCI were obtained at admission. Results of neurological examination of the patients at admission, at discharge and at 3 months were documented on an AIS/International Spinal Cord Society or AIS form. A 0.3 tesla MRI machine was used for imaging, and a RadiAnt DICOM Viewer 4.0.3 (64 bits) was used for viewing and measuring of MRI features.

The review of MRI injury features was done by one of the authors in collaboration with a radiologist, with the radiologist completely blinded from the neurological status and outcome of treatment of the patients. Different patterns or types of spinal cord injury as described by Ramon et al.[14] were identified from a midsagittal and axial T2W sequence [Figure 1] and [Figure 2]. The maximum canal compromise (MCC) and maximum spinal cord compression (MSCC) were calculated from a midsagittal T2W sequence, as shown in [Figure 2], using the formula: (i) MCC= (1 – [Di/½ [Da + Db]]) × 100 and (ii) MSCC = (1 – [di/½ [da + db]]) × 100{Figure 1}{Figure 2}

where Di and di are the spinal canal and spinal cord diameters in millimetres, measured at the epicentre of spinal cord injury, respectively; Da and da are the spinal canal and spinal cord diameters in millimetres, measured at a normal level immediately above the level of the spinal cord injury, respectively; and Db and db are the spinal canal and spinal cord diameters in millimetres, measured at a normal level immediately below the level of spinal cord injury, respectively [Figure 2], while the length of spinal cord lesion was measured as the length of spinal cord signal changes.

The patients were managed, based on the neurosurgical unit protocol, used in the management of cervical TSCI; this consist of maintaining blood pressure above 90/60 mmHg within the 1st week of admission, immediate cervical spine stabilisation with Philadelphia cervical collar, analgesia, early physiotherapy and rehabilitation, deep venous thrombosis prophylaxis and pressure ulcer prevention. Cervical spine X-rays or computed tomography (CT) scan and MRI were done as soon as possible. Patients whose cervical X-rays/MRI or CT/MRI revealed cord compression or cervical spine instability were counselled, and had anterior cervical discectomy or corpectomy and fusion (ACDF/ACCF) with or without lateral mass screw fusion, depending on the integrity of posterior ligamentous complex.

The neurological status of these patients was assessed again at discharge and at 3 months post-injury using the AIS forms on the ward and in outpatient clinic, respectively. Patients who were still on admission as at 3 months post-injury were assessed on the ward. However, those patients who could not come to the outpatient department for follow-up were assessed via WhatsApp video call. Eight patients who died before 3 months and nine patients with ASIA E injury at admission were excluded from the analysis for neurological improvement at 3 months post-injury.

Data analysis was done using the IBM SPSS Statistics for Windows, Version 25.0. (Armonk, NY: IBM Corp). Tables and charts were used for descriptive analysis of demographics. Quantitative variables were expressed as mean ± standard deviation. Inferential statistical was done using Fisher's exact test, and Spearman's rank correlation coefficient. Simple and stepwise binary logistic regression was used to determine the predictors of neurological recovery. A two-tailed P < 0.05 was used as the level of significance.

 Results



A total of 60 patients with cervical TSCI were seen within the study period, however, 13 patients were excluded from the study; most of them LAMA on financial ground to seek care from traditional healthcare providers. The remaining 47 patients with cervical TSCI were reviewed and analysed. Most of the patients were male (40, 85.1%). The mean age of the patients was 37.68 ± 13.7, with almost half (22, 46.8%) of this injury occurring in young adults between 20 and 39 years. Motor vehicular crash was the most common cause of injury (39, 83.0%), followed by falls from height (mostly trees) (7, 14.9%). More than half of the injuries occurred in the lower, subaxial cervical spine (C5–7) (27, 57.4%), with axial cervical spine injury being the least (3, 6.4%) [Table 1]. Most (41, 87.2%) of the patients presented within 48 h of injury, with an average time from injury to presentation of 23.71 ± 35.99.{Table 1}

Complete cervical TSCI (ASIA A) was the most common injury seen (19, 40.4%), with ASIA C (5, 10.6%) being the least type of injury at admission. At 3 months post-injury, 41 (87.2%) of the patients survived, whereas 6 (12.8%) of the patients died (all of them had ASIA A and TSCI). Ten (24.4%) of the patients who survived remained as ASIA A, whereas more than half of the patients 25 (61.0%) had good neurological function (ASIA D and E) [Figure 3].

Majority (31, 66.0%) of the patients had their MRI done within 72 h of injury. All patients with spinal cord haemorrhage (3, 100%), spinal cord transection (1, 100%) and 4 (80.0%) of the patients with spinal cord compression type of injury seen on MRI had a complete spinal cord injury (ASIA A). Spinal cord contusion and oedema was the most common (32, 68.1%) type of injury pattern seen on MRI, and more than half (21, 65.63%) of these patients presented with incomplete spinal cord injury. This association was significant (P = 0.01) [Table 2] and [Table 3]. Moreover, the Spearman's correlation coefficient showed a moderately significant correlation (ρ = 0.509, P = 0.000) [Table 4].{Table 2}{Table 3}{Table 4}{Figure 3}

The mean value of MCC was 32.06 ± 25.37%, MSCC was 17.08% ± 18.23% and the mean length of spinal cord lesion was 57.15 ± 41.94 mm. Patients with ASIA A (complete) spinal cord injury had the highest mean values of MCC, MSCC and length of spinal cord lesion (41.53% ± 26.86%, 24.83% ± 22.46% and 93.47% ± 34.57%, respectively), whereas patients with ASIA E had the lowest mean value of MCC, MSCC and length of spinal cord lesion (9.58% ± 9.21%, 7.16% ± 10.86% and 7.74% ± 9.97%, respectively) [Table 2] and [Table 3]. These findings showed a significant correlation between AIS and quantitative MRI features [Table 4]. The type of vertebral column injury and ligamentous injury did not show any significant association or correlation with AIS at admission and type of spinal cord injury [Table 4]

At 3 months post-injury, nine patients with ASIA E injury at admission and six patients who died were excluded from the analysis. Thirty-two patients were analysed for neurological improvement, 19 (59.4%) of the patients showed improvement in their AIS, whereas 13 (40.6%) did not. The mean time from injury to surgical intervention (ACDF/ACCF ± lateral mass screw) was 64.90 ± 54.53 h, only one of the patients had surgery within 48 h of injury and there was no significant relationship between patients who had surgery and their neurological improvement. Likewise, there was no significant correlation (ρ = 0.283, P = 0.054) between the time lapse from injury to presentation and improvement in AIS at 3 months. Spinal cord oedema and cord contusion were observed in most of the patients 18 (94.7%) who improved in their AIS, but none of the patients with spinal cord haemorrhage or cord transection and 3 (75%) of the patients with cord compression improved in their AIS. This association, though significant (P = 0.019) with Fisher's exact test and the Spearman's rank correlation, was not significant (ρ = −0.019, P = 0.917) [Table 4] and [Table 5].{Table 5}

The mean values of MCC, MSCC and length of spinal cord lesion in patients who improved in their AIS were 30.10% ± 23.57%, 12.90% ± 13.32% and 53.78% ± 39%, respectively; these values were lower than the MCC, MSCC and length of spinal cord lesion in patients who did not improve in their AIS (43.27% ± 24.12%, 22.79% ± 21.42% and 85.28% ± 35.31%, respectively). However, this association was only significant for the length of spinal cord lesion (ρ = −0.391, P = 0.027). The type of vertebral injury and ligamentous injury was not significantly associated with improvement in AIS.

A simple binary logistic regression analysis to determine the predictors of the type of spinal cord injury based on severity (complete or incomplete injury) at admission showed that the MCC (P = 0.039, odds ratio [OR]: 1.027, 95% confidence interval [CI]: 1.001, 1.053), MSCC (P = 0.026, OR: 1.045, 95% CI: 1.005, 1.086) and length of spinal cord lesion (P = 0.001, OR: 1.077, 95% CI: 1.032, 1.123) were independent predictors of type of injury, however, the length of spinal cord lesion (P = 0.001, OR: 1.077, 95% CI: 1.032, 1.123) contributed more significantly as a predictor of type of spinal cord injury in a stepwise regression analysis. The length of spinal cord lesion was the only independent predictor of improvement in AIS at 3 months after injury (P = 0.025, OR: 0.974, 95% CI: 0.956–0.999) [Table 6].{Table 6}

 Discussion



Several types of spinal cord signal intensity changes were identified on cervical MRI of majority (41, 87.2%) of these patients with TSCI. These changes were similar to different types of injury patterns described by Bondurant et al. and Ramon et al., using a 1.5 tesla MRI machine.[13],[19] These injury types or patterns observed using a low tesla MRI (0.3 tesla) in our practice are shown in [Figure 1]. Patients who had normal spinal cord (type 0) on MRI had transient neurological deficit (ASIA E injury) at initial assessment. Most (32,78.0%) of the patients with spinal cord signal intensity changes had spinal cord oedema and cord contusion, and two-third of them had an incomplete injury with neurological deficits (ASIA B, C and D). However, all the patients with spinal cord haemorrhage, cord transection and most of the patients with cord compression had complete TSCI, which suggest a severe form of injury.

This observation is similar to publications by several authors who reported that traumatic intramedullary haemorrhage or haematoma and cord transection were severe types of spinal cord injury that correlated significantly with complete TSCI, with these patients having poor chance of neurological recovery during follow-up.[8],[13],[15],[16],[17],[19],[23] Although a higher tesla MRI machine provides a better image quality, the finding from this study suggest that resource-constrained setting with low tesla MRI machine could also use this machine to better define spinal cord injuries in their patients. Most especially, in patients whose AIS cannot be determined in the early phase of TSCI.

This study also found that significantly higher mean values of MCC, MSCC and length of cord lesion were observed in patients with complete spinal cord injury (ASIA A). While patients with incomplete spinal cord injury with neurological deficit (ASIA B, C and D) have slightly lower mean values of these parameters, patients with ASIA E (spinal cord injury without neurological deficits) had the lowest mean value of MCC, MSCC and length of spinal cord lesion. This observation significantly correlated with AIS at admission and type spinal cord injury. These findings were achievable with our low tesla MRI machine, as shown in [Figure 2]. Several authors have reported similar findings using a higher (1.5) tesla MRI machine.[8],[16],[17],[21] The explanation for these findings could be that, severe cord compression or spinal canal compromise is more likely to cause severe and extensive damage to the spinal cord from either ischaemia or disruption of nerve fibres with much more severe consequences than mild form of cord compression or canal compromise.

A simple regression analysis model to determine the predictor of severity of injury showed that the types of spinal cord injury pattern, MCC, MSCC and length of spinal cord lesion were independent predictors of severity of injury (complete versus incomplete injury) at admission. However, in a multistage or stepwise regression model, only the length of spinal cord lesion contributed significantly as a predictor of severity of injury.

At 3 months post-injury, all the patients with cervical spinal cord haemorrhage, cord transection and most 3 (75%) of the patients with cord compression, showed no improvement in their admission AIS. Whereas, 18 (75.0%) out of 24 patients with cervical spinal cord oedema and cord contusion improved by at least one level in their initial AIS. Higher values of MCC, MSCC and length of cord lesion were seen in patients who did not improve in their AIS, however, the only independent predictor of neurological improvement was the length of spinal cord lesion (P = 0.025, OR: 0.974, 95% CI: 0.956–0.999).

AIthough, several studies, published in the literature, showed that, spinal cord haemorrhage is associated with poor neurological recovery, with spinal cord oedema and cord contusion, associated with good recovery.[13],[15],[18],[19] However, there were no studies in which regression analysis model was used to show how quantitative MRI features can be used to predicted neurological improvement, as done in our study. Although some authors reported that the best model that predicted unadjusted follow-up ASIA motor score was MSCC, intramedullary haemorrhage and cord swelling,[17] others found that the length of spinal cord lesion and intramedullary haemorrhage were predictors of Frankel score on admission, with MSCC, length of lesion and intramedullary haemorrhage, predicting Frankel score at follow-up.[8]

The findings in this study and that from other authors show a similar pattern, where the presence of intramedullary haemorrhage, higher mean values of length of spinal cord lesion and MSCC are strong predictors of complete TSCI with poor potential for neurological recovery. However, spinal cord oedema, cord contusion and lower values of length of spinal cord lesion and MSCC are predictors of incomplete injury with good potential for recovery.[8],[13],[15],[16],[17],[21],[30] The possible explanation for these findings could be that traumatic cord compression, often from disc herniation, displaced or dislocated fractured vertebrae or haematoma, results in secondary injury. This further worsens the primary injury caused by the initial energy transfer to the cord at the time of primary injury, leading to permanent spinal cord damage with less likelihood of recovery.[31],[32] This is further supported by the STASIS study where early decompression resulted in improvement in AIS by at least two levels.[33]

Limitation

Delay in carrying out MRI scan was a major challenge, because most of our patients fund their health services; as a result, financial constraint and frequent breakdown of the MRI machine affected the timing of image acquisition and limited our ability to do follow-up of MRI. Although some authors suggested that the best timing for performing a prognostic MRI should be within 24–72 h after trauma,[13] others reported that the ideal timing should be at the time of injury and 2–3 weeks later.[34] However, no substantial evidence exists to support this precise guideline.[35],[36]

Second, our MRI machine is a low (0.3) tesla, spin-echo MR imaging, with limited pathophysiological image characterisation, especially in trauma. More so that, there are higher tesla MRI with advanced imaging technique such as diffusion tensor imaging that has the ability to show axonal injury and additional shearing injuries that are not visible on low tesla and conventional MR images.[17],[37],[38]

 Conclusions



Low tesla MRI can also be used to determine different types of cervical spinal cord injury characteristics. Spinal cord haemorrhage, compression, transection and higher values of MCC, MSCC and increased length of spinal cord lesion were significantly associated with complete TSCI at admission. Whereas, cord contusion and lower values of MCC, MSCC and shorter length of cord lesion are predictive of incomplete TSCI, with length of cord lesion being a better predictor of outcome at 3 months post-injury.

Acknowledgments

A big thanks and sincere appreciation to Dr DC. Dolapo who help with the statistical analysis of the data, and Dr B. Faruk for helping with the review of the MR imaging.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Badhiwala JH, Wilson JR, Fehlings MG. Global burden of traumatic brain and spinal cord injury. Lancet Neurol 2019;18:24-5.
2Lifshutz J, Colohan A. A brief history of therapy for traumatic spinal cord injury. Neurosurg Focus 2004;16:1-8.
3Ihegihu CC, Ugezu AI, Ndukwu CU, Chukwuka NC, Ofiaeli RO, Ihegihu EY. A review of traumatic spinal cord injuries at Nnamdi Azikiwe University Teaching Hospital. Trop J Med Res 2014;17:31-6.
4Bedbrook GM. The development and care of spinal cord paralysis (1918 to 1986) Paraplegia 1987;25:172-84.
5Guttmann L. Spinal Cord Injuries: Comprehensive Management and Research. 2nd ed. Oxford, UK: Blackwell Scientific; 1976. p. 1-76.
6GBD 2016 Traumatic Brain Injury and Spinal Cord Injury Collaborators. Global, regional, and national burden of traumatic brain injury and spinal cord injury, 1990-2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol 2019;18:56-87.
7Nobunaga AI, Go BK, Karunas RB. Recent demographic and injury trends in people served by the Model Spinal Cord Injury Care Systems. Arch Phys Med Rehabil 1999;80:1372-82.
8Haar M, Naidoo SM, Govender S, Parag P, Esterhuizen TM. Acute traumatic cervical spinal cord injuries: Correlating MRI findings with neurological outcome. SA Orthop J 2011;10:35-41.
9Ditunno JF Jr. The John Stanley Coulter Lecture. Predicting recovery after spinal cord injury: A rehabilitation imperative. Arch Phys Med Rehabil 1999;80:361-4.
10La Rosa G, Conti A, Cardali S, Cacciola F, Tomasello F. Does early decompression improve neurological outcome of spinal cord injured patients? Appraisal of the literature using a meta-analytical approach. Spinal Cord 2004;42:503-12.
11Burns AS, Ditunno JF. Establishing prognosis and maximizing functional outcomes after spinal cord injury: A review of current and future directions in rehabilitation management. Spine (Phila Pa 1976) 2001;26:S137-45.
12van Middendorp JJ, Goss B, Urquhart S, Atresh S, Williams RP, Schuetz M. Diagnosis and prognosis of traumatic spinal cord injury. Global Spine J 2011;1:1-8.
13Bondurant FJ, Cotler HB, Kulkarni MV, McArdle CB, Harris JH Jr. Acute spinal cord injury. A study using physical examination and magnetic resonance imaging. Spine (Phila Pa 1976) 1990;15:161-8.
14Singh R, Kumar RR, Setia N, Magu S. A prospective study of neurological outcome in relation to findings of imaging modalities in acute spinal cord injury. Asian J Neurosurg 2015;10:181-9.
15Flanders AE, Schaefer DM, Doan HT, Mishkin MM, Gonzalez CF, Northrup BE. Acute cervical spine trauma: Correlation of MR imaging findings with degree of neurologic deficit. Radiology 1990;177:25-33.
16Magu S, Singh D, Yadav RK, Bala M. Evaluation of traumatic spine by magnetic resonance imaging and correlation with neurological recovery. Asian Spine J 2015;9:748-56.
17Miyanji F, Furlan JC, Aarabi B, Arnold PM, Fehlings MG. Acute cervical traumatic spinal cord injury: MR imaging findings correlated with neurologic outcome – Prospective study with 100 consecutive patients. Radiology 2007;243:820-7.
18Kulkarni MV, Bondurant FJ, Rose SL, Narayana PA. 1.5 tesla magnetic resonance imaging of acute spinal trauma. Radiographics 1988;8:1059-82.
19Ramón S, Domínguez R, Ramírez L, Paraira M, Olona M, Castelló T, et al. Clinical and magnetic resonance imaging correlation in acute spinal cord injury. Spinal Cord 1997;35:664-73.
20Rao KV, Vijaya Saradhi M, Purohit AK. Factors affecting long-term outcome in acute cervical cord injury. Indian J Neurotrauma 2010;7:149-55.
21Gupta R, Mittal P, Sandhu P, Saggar K, Gupta K. Correlation of qualitative and quantitative MRI parameters with neurological status: A prospective study on patients with spinal trauma. J Clin Diagn Res 2014;8:RC13-7.
22Ochie K, Okpala OC, Ohagwu CC, Eze CU. Profile of MRI findings and neurological impairment in spinal cord injury patients at Lagos University Teaching Hospital, Nigeria. World J Med Sci 2013;8:418-23.
23Parashari UC, Khanduri S, Bhadury S, Kohli N, Parihar A, Singh R, et al. Diagnostic and prognostic role of MRI in spinal trauma, its comparison and correlation with clinical profile and neurological outcome, according to ASIA impairment scale. J Craniovertebr Junction Spine 2011;2:17-26.
24Mahmood NS, Kadavigere R, Avinash KR, Rao VR. Magnetic resonance imaging in acute cervical spinal cord injury: A correlative study on spinal cord changes and 1 month motor recovery. Spinal Cord 2008;46:791-7.
25Kulkarni MV, McArdle CB, Kopanicky D, Miner M, Cotler HB, Lee KF, et al. Acute spinal cord injury: MR imaging at 1.5 T. Radiology 1987;164:837-43.
26Rao SC, Fehlings MG. The optimal radiologic method for assessing spinal canal compromise and cord compression in patients with cervical spinal cord injury. Part I: An evidence-based analysis of the published literature. Spine (Phila Pa 1976) 1999;24:598-604.
27Fehlings MG, Rao SC, Tator CH, Skaf G, Arnold P, Benzel E, et al. The optimal radiologic method for assessing spinal canal compromise and cord compression in patients with cervical spinal cord injury. Part II: Results of a multicenter study. Spine 1999;24:605-13.
28Fehlings MG, Furlan JC, Massicotte EM, Arnold P, Aarabi B, Harrop J, et al. Interobserver and intraobserver reliability of maximum canal compromise and spinal cord compression for evaluation of acute traumatic cervical spinal cord injury. Spine (Phila Pa 1976) 2006;31:1719-25.
29Kartz LD, Elmore JG, Wild DM, Lucan SC. Applying statistics to trial design: Sample size randomization and control for multiple hypotheses. In: Kartz LD, Elmore JG, Wild DM, Lucan SC, editors. Jekels Epidemiology, Biostatistics, Preventive Medicine and Public Health. 4th ed.. Philadelphia: Elsevier Saunders; 2014. p. 153-62.
30Martínez-Pérez R, Paredes I, Cepeda S, Ramos A, Castaño-León AM, García-Fuentes C, et al. Spinal cord injury after blunt cervical spine trauma: Correlation of soft-tissue damage and extension of lesion. AJNR Am J Neuroradiol 2014;35:1029-34.
31Talbott JF, Whetstone WD, Readdy WJ, Ferguson AR, Bresnahan JC, Saigal R, et al. The Brain and Spinal Injury Center score: A novel, simple, and reproducible method for assessing the severity of acute cervical spinal cord injury with axial T2-weighted MRI findings. J Neurosurg Spine 2015;23:495-504.
32Fehlings MG, Tator CH. An evidence-based review of decompressive surgeryin acute spinal cord injury: Rationale, indications, and timing based on experimental and clinical studies. J Neurosurg (Spine 1) 1999;91:1-11.
33Fehlings MG, Vaccaro A, Wilson JR, Singh A, Cadotte DW, Harrop JS, et al. Early versus delayed decompression for traumatic cervical spinal cord injury: Results of the Surgical Timing in Acute Spinal Cord Injury Study (STASCIS). PLoS One 2012;7:E32037.
34Shimada K, Tokioka T. Sequential MR studies of cervical cord injury: Correlation with neurological damage and clinical outcome. Spinal Cord 1999;37:410-5.
35Bozzo A, Marcoux J, Radhakrishna M, Pelletier J, Goulet B. The role of magnetic resonance imaging in the management of acute spinal cord injury. J Neurotrauma 2011;28:1401-11.
36Goradia D, Linnau KF, Cohen WA, Mirza S, Hallam DK, Blackmore CC. Correlation of MR imaging findings with intraoperative findings after cervical spine trauma. AJNR Am J Neuroradiol 2007;28:209-15.
37Arfanakis K, Haughton VM, Carew JD, Rogers BP, Dempsey RJ, Meyerand ME. Diffusion tensor MR imaging in diffuse axonal injury. AJNR Am J Neuroradiol 2002;23:794-802.
38Huisman TA. Diffusion-weighted imaging: Basic concepts and application in cerebral stroke and head trauma. Eur Radiol 2003;13:2283-97.