- Email: [email protected]
Comparison of CT and MRI findings for cervical spine clearance in obtunded patients without high impact trauma
Clinical Neurology and Neurosurgery 120 (2014) 23–26
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Comparison of CT and MRI findings for cervical spine clearance in obtunded patients without high impact trauma Lee A. Tan, Manish K. Kasliwal ∗ , Vincent C. Traynelis Department of Neurosurgery, Rush University Medical Center, Chicago, 60612, USA
a r t i c l e
i n f o
Article history: Received 18 July 2013 Received in revised form 12 October 2013 Accepted 17 February 2014 Available online 25 February 2014 Keywords: Cervical spine Computed tomography MRI Obtunded
a b s t r a c t Objective: Cervical spinal injuries occur in 2.0–6.6% of patients after blunt trauma and can have devastating neurological sequelae if left unrecognized. Although there is high quality evidence addressing cervical clearance in asymptomatic and symptomatic awake patients, cervical spine clearance in patients with altered level of alertness (i.e., obtunded patients with Glasgow coma scale (GCS) of 14 or less) following blunt trauma has been a matter of great controversy. Furthermore, there are no data on cervical spine clearance in obtunded patients without high impact trauma and these patients are often treated based on evidence from similar patients with high impact trauma. This retrospective study was conducted on this specific subgroup of patients who were admitted to a neurointensive care unit (NICU) with primary diagnoses of intracranial hemorrhage with history of minor trauma; the objective being to evaluate and compare cervical spinal computed tomography (CT) and magnetic resonance imaging (MRI) findings in this particular group of patients. Methods: Patients with GCS of 14 or less admitted to neruointensive care unit (NICU) at RUSH University Medical Center from 2008 to 2010 with diagnoses of intracranial hemorrhage (surgical or non-surgical) who had reported or presumed fall (i.e., “found down”) were queried from the computer data registry. A group of these patients had cervical spine CT and subsequently MRI for clearing the cervical spine and removal of the cervical collar. Medical records of these patients were reviewed for demographics, GCS score and injury specific data and presence or absence of cervical spine injury. Results: Eighty-three patients were identified from the computer database. Twenty-eight of these patients had positive findings on both CT and MRI (33.73% – Group I); four patients had a negative CT but had positive findings on follow-up MRI (4.82% – Group II); fifty-one patients had both negative CT and MRI (61.44% – Group III). All patients in Group I required either surgical stabilization or continuation of rigid cervical orthosis. All four patients in Group II had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI, but did not have any signs of fracture or ligamentous injury to suggest instability. They eventually underwent surgical decompression of the spinal cord during the same hospital stay. Cervical collars were safely removed in all patients in Group III. In our retrospective study, CT had a sensitivity of 0.875 [0.719–0.950, 95% CI] and a specificity of 1.000 [0.930–1.000, 95% CI] in detecting all cervical spine injuries compared to MRI. However, all patients with missed injuries had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI and were not unstable precluding cervical spine clearance. If only unstable injuries are considered, CT had a sensitivity of 1.00 [0.879–1.000, 95% CI] and a specificity is 1.000 [0.935–1.000, 95% CI] compared to MRI in this particular group of patients. Conclusion: CT is highly sensitive in detecting unstable injuries in obtunded patients with GCS of 14 or less in the absence of high impact trauma. In the absence of high impact trauma, neurosurgeons should be comfortable to discontinue the cervical collar after a negative, high-quality CT in this patient population. In the presence of focal neurological deficits unexplained by associated intracranial injury, an MRI may help diagnose intrinsic spinal cord injuries which necessarily may not be unstable in the presence of a negative CT and does not precludes clearance of cervical spine. © 2014 Elsevier B.V. All rights reserved.
∗ Corresponding author at: Department of Neurosurgery, Rush University Medical Center, Suite 855, Chicago, IL 60612, USA. Tel.: +1 312 942 6644; fax: +1 312 942 2176. E-mail addresses: manish [email protected], m [email protected] (M.K. Kasliwal). http://dx.doi.org/10.1016/j.clineuro.2014.02.006 0303-8467/© 2014 Elsevier B.V. All rights reserved.
24
L.A. Tan et al. / Clinical Neurology and Neurosurgery 120 (2014) 23–26
1. Introduction Cervical spine assessment and clearance is a priority in patients with suspected with cervical spine trauma due to potential devastating neurological deficits that can result from cervical spine injuries. There are well-established guidelines for cervical clearance in patients who are fully awake (Glasgow coma scale [GCS] = 15) with a reliable neurological exam, as described in the National Emergency X-Radiography Utilization Study (NEXUS) criteria and Canadian C-Spine (cervical-spine) Rule (CCR) [1]. Cervical spine clearance can however be challenging in obtunded patients with GCS of 14 or less and is a matter of great conjecture. Obtunded patients with suspected cervical spine injury are usually kept in rigid collars for cervical immobilization. Prolonged use of rigid collar is not only associated with increased incidence of pressure ulcers and intracranial pressure (ICP) elevation [2–4], but also presents challenges for oral hygiene, airway management and daily nursing care. Thus, safe and timely removal of cervical collars in obtunded patients has important clinical implications. With increased availability of high-resolution multi-detector computed tomography (MDCT), there have been reports of using MDCT alone for clearance of cervical spine in patients with blunt trauma who have non-focal examinations. The advantages of it being wider availability of CT, quick scan time and high sensitivity in detecting bony abnormalities such as acute fractures. However, many studies have shown that CT alone can miss clinically significant soft tissue injuries [5–12], especially in patients with high impact trauma. Magnetic resonance imaging (MRI) is highly sensitive in detecting soft tissue and ligamentous injuries but is much more time intensive and is not as readily available compared to CT. Even with technological advances with dramatic improvement in quality and resolution of CT, there is contradictory evidence in the value of CT alone for cervical spine clearance in obtunded patients [5,7–9,12–20]. Hence various studies have evaluated the role of CT alone or in conjunction with single upright radiographs or MRI to clear cervical spine injury in obtunded patients. While negative CT and MRI in obtunded patients have a very high degree of specificity in clearing the cervical spine, the value of a positive MRI is poorly understood and can be redundant leading to retaining the collar for longer periods due to lack of well defined criteria for instability on MRI with its very high sensitivity for detecting soft tissue trauma not necessarily ligamentous with undetermined clinical significance [16]. There is no universally accepted protocol regarding the cervical spine clearance in obtunded patients without history high impact trauma and the data on whether a negative CT alone is enough to clear cervical spine in this subgroup of patients is lacking. This clinical scenario is commonly encountered in many neruointensive care units (NICUs), where patients are admitted with being obtunded following some intracranial pathology and either have a fall secondary to that or are reported as “found down” with not even a witnessed history of trauma. These patients arrive in cervical collar to the NICU where neurosurgeons are often consulted for cervical spine clearance by the neurologists and neurointensivists. Given many of these patients have altered level of alertness with concurrent intracranial pathologies (subarachnoid hemorrhage [SAH], intracerebral hemorrhage [ICH], subdural hemorrhage [SDH]), they often do not cooperative fully for adequate and accurate assessment of strength or sensation or are not even conscious secondarily to the intracranial injury. While the absence of significance trauma makes the probability of cervical spine injury unlikely, theoretically increasing the negative predictive value of a negative CT scan, their intracranial pathologies often produce neurological deficits that cannot be clearly distinguished from possible cervical spine injury. Given above reasons, many of these patients ended up getting MRI for the purpose of cervical spine clearance.
Table 1 Comparison of cervical spine findings on CT vs. MRI.
CT MRI # of pts (%)
Group I
Group II
Group III
Positive Positive 28/83 (33.73%)
Negative Positive 4/83 (4.82%)
Negative Negative 51/83 (61.44%)
At our institution, we receive a significant number of patients from other institutions with the diagnosis of intracranial hemorrhage (primarily SAH, ICH, and SDH). Many of these patients arrive with cervical collars due to history of “fall” or “found down” with no history of high impact trauma and have unreliable neurological exams. It is the routine practice of the neurointensivists at our institution to image the cervical spines of such individuals which often involves CT and MRI. We initiated this study to focus on this subgroup of patients to evaluate and compare the cervical CT and MRI findings. 2. Methods Electronic medical record database at RUSH University Medical Center, Chicago, IL was queried for patients admitted to the NICU with the diagnosis of intracranial hemorrhage and concomitant history of minor cervical spine trauma most commonly ground level fall between January 2008 and December 2010. The study group primarily consisted of patients with SAH, ICH or SDH with history of “ground level fall” or “found down” and have unreliable neurological exams. Those who underwent both CT (Siemens, 64-slice) and MRI (Siemens, 1.5 or 3 T) of cervical spine were identified, which were read both by a board certified attending neuroradiologist and an attending neurosurgeon. Medical records of these patients were reviewed for demographics, admitting diagnosis, cervical CT/MRI findings and cervical spine management. Patients were divided into three groups based on CT and MRI findings. Group I consisted of patients with evidences of cervical injuries on both CT and MRI; Group II had negative CT but positive findings on MRI; Group III had both negative CT and MRI. Cervical spine management in each group was analyzed. Sensitivity and specificity of CT were calculated using MRI as the gold standard, since MRI has a 100% negative predicative value to detect soft tissue injuries as described by Muchow et al. in their meta-analysis [9]. 3. Results Eighty-three patients were identified from the computer database using our search criteria. Twenty-eight (33.73%) patients had pathologies identified on both CT and MRI (Group I); four patients (4.82%) had a negative CT but had positive finding on MRI (Group II); fifty-one patients (61.44%) had both negative CT and MRI (Group III). Results are summarized in Table 1. All patients in Group I required either surgical stabilization or continuation of rigid cervical orthosis. All four patients in Group II had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI, but without any obvious fracture or evidence of instability on CT which was read as negative. They eventually underwent surgical decompression of the spinal cord during the same hospital stay due to presence of spinal stenosis and clinical suspicion of central cord syndrome with no significant recovery once they were clinically stable. Cervical collars were safely removed in all patients in Group III. In our retrospective study, CT had a sensitivity of 0.875 [0.719–0.950, 95% CI] and a specificity of 1.000 [0.930–1.000, 95% CI] in detecting all cervical spine injuries compared to MRI. However, all patients with missed injuries on CT scan had intramedullary T2 hyper intensity consistent with possible central cord syndrome are not unstable
L.A. Tan et al. / Clinical Neurology and Neurosurgery 120 (2014) 23–26
injuries. If only unstable injuries are included, the sensitivity of CT for detecting was 1.00 [0.879–1.000, 95% CI] and a specificity of 1.000 [0.935–1.000, 95% CI] compared to MRI. The positive predictive value (PPV) for CT is calculated to be 100%, the negative predictive value (NPV) for unstable injuries was also 100%, while NPV for all injuries (including spinal cord injury) was 92.7%. The GCS scores of the 4 patients with injuries missed by CT scan but having positive MRI findings ranged from 12 to 14. The mean GCS for the entire study group is 12.09. The mean GCS scores for Groups I (CT+/MR+), Group II (CT−/MR+) and Group III (CT−/MR−) are 13.6, 13.4 and 11.1, respectively. 24% of patients (20/83) had a GCS 3–10, while 76% (63/83) had a GCS11–14 in our cohort. Though all 4 missed injuries occurred in the latter group, there was no statistically significant difference between the likelihood of missed injuries between the two GCS groups (p = 0.5696).
4. Discussion NEXUS and Eastern Association of the Surgery of Trauma (EAST) guidelines have been well-established and are standards for cervical spine clearance in patients who are awake, alert and cooperative with neurological exam. The NEXUS criteria have a sensitivity of 99.9% in detecting significant cervical injuries [1]. However, cervical spine clearance in obtunded patients (GCS of 14 or less) has been a persistent topic of debate. No universally accepted guidelines exist for this group of patients. There is even less data on cervical spine clearance in obtunded patients without history of high impact trauma. Unlike the awake and alert patients, obtunded patients cannot express pain and often lack reliable neurological exam that can help clinicians to identify potential cervical injuries. The management options for such patients include cervical immobilization with collar until reliable clinical exam can be obtained, collar removal after negative CT alone, or collar removal after negative CT and an additional imaging modality such as dynamic fluoroscopy or MRI. Prolonged cervical collar use has been associated with pressure ulcers and ICP elevations [2,4,6,21,22]. It also limits venous access, airway management and general nursing care for these patients which is important for this specific patient population who often have concomitant intracranial injuries/pathology [23,24]. Thus early and safe removal of cervical collar has important clinical implications. CT is the initial imaging modality of choice for cervical spine trauma due to its wide availability, quick scan time and high sensitivity in detecting bony abnormalities such as acute fractures. However, many studies have shown that CT scan alone can miss clinically significant soft tissue injuries [5–12]. Adjuvant modalities such as dynamic fluoroscopy and MRI have been used to further assess ligamentous injuries and cervical stability. The efficacy of dynamic fluoroscopy is limited because of high rate of inadequate visualization of the entire cervical spine, especially in the cervicothoracic junction [25]. MRI is highly sensitive in detecting soft tissue and ligamentous injuries but it is much more time intensive and is not as readily available as CT. With recent technological advances, the imaging quality and resolution of CT have dramatically improved. Many have argued that with modern multi-detector CT scanners, negative CT alone is sufficient for cervical spine clearance in obtunded patients [13–18]. However, many other studies have shown that CT can miss clinically important cervical injuries [5,7–9,12,15,19,20]. There are flaws in various studies precluding drawing firm conclusions. Most of these studies are from trauma centers where the probability of sustaining a cervical spine trauma in the form of fractures or ligamentous injuries is high justifying the aggressiveness in clearing the C spine and value of MRI in patients suspected with high impact trauma. Following such a protocol in subgroup as studied in this study may
25
be of questionable value as these patients do not have any major trauma predisposing them to injuries which may be occult on high resolution CT scans. A recent meta-analysis by Panczykowski et al. [16] reported that following a protocol of clearing cervical spine based on CT alone in obtunded patients, the incidence of unstable cervical spine injury among the studies of negative modern CT approximates to 1 in every 4776 patients. In fact, cervical spine is cleared after a negative MDCT in some trauma centers in the US and other countries (personal communication) in obtunded patients [18]. The most recent guidelines from the American Association of Neurological Surgeons (AANS)/Congress of Neurological Surgeons (CNS) Joint Guidelines Committee, recommends CT imaging as the initial imaging modality of choice in the obtunded patient (Level I). Subsequent to a negative CT, the recommendations becomes weak (Level II) and in patients in whom there is a high clinical suspicion of injury yet have a normal high-quality CT imaging study, the decision for further patient management is left to physicians trained in the diagnosis and management of spinal injuries. A further Level III recommendation following a negative CT was either continue cervical immobilization until asymptomatic, discontinue cervical immobilization following a normal MRI study obtained within 48 h of injury or discontinue cervical immobilization at the discretion of the treating physician [26]. While continuing cervical collars seems pretty straight forward, these patients often have associated intracranial hematomas and may remain comatose/obtunded for a long period of time making this approach less appealing, due to issues related to nursing care and complications due to cervical collar use itself. As related to MRI, the performance of MRI for cervical spine clearance may not be that difficult as these patients often require MRI for assessment of brain injuries and evaluation of stroke, obviating the technical challenge often seen in patients with trauma. Nevertheless, the value of obtaining MRI cervical spine should be carefully considered due to cost issues and benefits analysis, not ignoring the fact that a number of this patients do not even have any mechanism of injury with might really reduce the overall yield of MRI in changing clinical management. Also, while significant number of this patients end up having MRI brain for assessment of intracranial pathology, not all of them require it. In that case, there is value in clearing the cervical spine based on negative CT alone as it is cost effective and quick. Also, in the absence of high impact trauma, the chances of this patients harboring an injury detectable on MRI is very low as compared to patients with high trauma even in whom the chances of detecting a clinically significant injury being missed on CT and detectable per MRI is very low as per the meta-analysis by Panczykowski et al. [16]. As can be seen in this study, there was no patient who had a negative CT who subsequently had positive bony or ligamentous injuries identified on MRI. The only patients who underwent surgery following a negative CT had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI and no unstable injury or any fractures which were missed on CT scan. These can be suspected based on focal neurology which if present is an indication for MRI not for clearing the spine itself but for planning further treatment. Since the subgroup of patients in this study was old, they do have high chances of sustaining central cord syndrome (CCS) after a minor fall and should be watched for this and appropriately evaluated. Also patients with CCS do not really needs collar and C-spine can be cleared even for this subgroup of patients once CT is negative. In our retrospective study, we analyzed cervical CT and MRI of eighty-three patients who were obtunded but had no history high impact trauma. CT did not miss any unstable injuries. MRI identified four patients who had intramedullary T2 hyper intensity consistent with possible spinal cord injury possibly secondary to significant pre-existing degenerative changes, osteophytes or ligamentous hypertrophy in the setting of minor trauma such as falling from standing height. Although clinical exam is unreliable in many
26
L.A. Tan et al. / Clinical Neurology and Neurosurgery 120 (2014) 23–26
obtunded patients, any asymmetric response to the painful stimuli or an exam that cannot be explained by the intracranial pathology should alert the clinician to further evaluate cervical spine with MRI. It is also interesting to note that the overall incidence of cervical spine injury was 38.6% in our study. Thirty-two of the eighty-three patients had positive findings on MRI and required either continuation of cervical collar or surgical decompression and stabilization. This number is much higher than the 2–6% incidence commonly quoted in literature. This discrepancy may be due to the fact that some of the prior studies only used plain X-ray or CT to identify bony fractures [27]. The demographic of our study population may also have contributed to the higher incidence since the patients in our study group tend to be older with pre-existing degenerative changes and are prone to various fractures even after a ground level fall. This limits the generalizability of our study to certain extent but remains very relevant to the patient population admitted to a typical NICU. The drawback of this study is that it is a retrospective and is subject to its biases. Also we do not have longterm follow up on these patients to rule out any delayed instability but considering the fact that all these patients who had a negative CT also had a MRI to rule out ligamentous injury which might manifest in a delayed fashion, a problem in various studies that compared CT with X-rays [16]. 5. Conclusion CT is a very useful imaging modality in evaluation cervical spine in obtunded patients with unreliable neurological exams. A negative high-quality CT scan is very reassuring and in the absence of high impact trauma as is seen in this patient population seems adequate for cervical spine clearance. Performance of routine MRI in this subgroup of patients without history of high impact trauma seems redundant following a negative high quality CT scan and may be avoided for the purpose of cervical spine clearance. In the presence of focal neurological deficits unexplained by associated intracranial injury, an MRI may help diagnose intrinsic spinal cord injuries which necessarily may not be unstable in the presence of a negative CT and does not precludes clearance of cervical spine. Funding None. References [1] Hoffman JR, Mower WR, Wolfson AB, Todd KH, Zucker MI. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. National Emergency X-Radiography Utilization Study Group. N Engl J Med 2000;343:94–9. [2] Mobbs RJ, Stoodley MA, Fuller J. Effect of cervical hard collar on intracranial pressure after head injury. ANZ J Surg 2002;72:389–91. [3] Davies G, Deakin C, Wilson A:. The effect of a rigid collar on intracranial pressure. Injury 1996;27:647–9. [4] Hunt K, Hallworth S, Smith M:. The effects of rigid collar placement on intracranial and cerebral perfusion pressures. Anaesthesia 2001;56:511–3.
[5] Brown CV, Foulkrod KH, Reifsnyder A, Bui E, Lopez A, Hummell M, et al. Computed tomography versus magnetic resonance imaging for evaluation of the cervical spine: how many slices do you need? Am Surg 2010;76: 365–8. [6] Dunham CM, Brocker BP, Collier BD, Gemmel DJ. Risks associated with magnetic resonance imaging and cervical collar in comatose, blunt trauma patients with negative comprehensive cervical spine computed tomography and no apparent spinal deficit. Crit Care 2008;12:R89. [7] Menaker J, Philp A, Boswell S, Scalea TM. Computed tomography alone for cervical spine clearance in the unreliable patient – are we there yet? J Trauma 2008;64:898–903 [discussion 903–4]. [8] Menaker J, Stein DM, Philp AS, Scalea TM. 40-slice multidetector CT: is MRI still necessary for cervical spine clearance after blunt trauma? Am Surg 2010;76:157–63. [9] Muchow RD, Resnick DK, Abdel MP, Munoz A, Anderson PA. Magnetic resonance imaging (MRI) in the clearance of the cervical spine in blunt trauma: a metaanalysis. J Trauma 2008;64:179–89. [10] Hogan GJ, Mirvis SE, Shanmuganathan K, Scalea TM. Exclusion of unstable cervical spine injury in obtunded patients with blunt trauma: is MR imaging needed when multi-detector row CT findings are normal? Radiology 2005;237: 106–13. [11] Spiteri V, Kotnis R, Singh P, Elzein R, Madhu R, Brooks A, et al. Cervical dynamic screening in spinal clearance: now redundant. J Trauma 2006;61:1171–7 [discussion 1177]. [12] Stassen NA, Williams VA, Gestring ML, Cheng JD, Bankey PE. Magnetic resonance imaging in combination with helical computed tomography provides a safe and efficient method of cervical spine clearance in the obtunded trauma patient. J Trauma 2006;60:171–7. [13] Brohi K, Healy M, Fotheringham T, Chan O, Aylwin C, Whitley S, et al. Helical computed tomographic scanning for the evaluation of the cervical spine in the unconscious, intubated trauma patient. J Trauma 2005;58:897–901. [14] Como JJ, Thompson MA, Anderson JS, Shah RR, Claridge JA, Yowler CJ, et al. Is magnetic resonance imaging essential in clearing the cervical spine in obtunded patients with blunt trauma? J Trauma 2007;63:544–9. [15] Harris TJ, Blackmore CC, Mirza SK, Jurkovich GJ. Clearing the cervical spine in obtunded patients. Spine (Phila Pa 1976) 2008;33:1547–53. [16] Panczykowski DM, Tomycz ND, Okonkwo DO. Comparative effectiveness of using computed tomography alone to exclude cervical spine injuries in obtunded or intubated patients: meta-analysis of 14,327 patients with blunt trauma. J Neurosurg 2011;115:541–9. [17] Steigelman M, Lopez P, Dent D, Myers J, Corneille M, Stewart R, et al. Screening cervical spine MRI after normal cervical spine CT scans in patients in whom cervical spine injury cannot be excluded by physical examination. Am J Surg 2008;196:857–62 [discussion 862–3]. [18] Tomycz ND, Chew BG, Chang YF, Darby JM, Gunn SR, Nicholas DH, et al. MRI is unnecessary to clear the cervical spine in obtunded/comatose trauma patients: the four-year experience of a level I trauma center. J Trauma 2008;64:1258–63. [19] Simon JB, Schoenfeld AJ, Katz JN, Kamath A, Wood K, Bono CM, et al. Are “normal” multidetector computed tomographic scans sufficient to allow collar removal in the trauma patient? J Trauma 2010;68:103–8. [20] Diaz Jr JJ, Aulino JM, Collier B, Roman C, May AK, Miller RS, et al. The early work-up for isolated ligamentous injury of the cervical spine: does computed tomography scan have a role? J Trauma 2005;59:897–903 [discussion 903–4]. [21] Kolb JC, Summers RL, Galli RL. Cervical collar-induced changes in intracranial pressure. Am J Emerg Med 1999;17:135–7. [22] Raphael JH, Chotai R. Effects of the cervical collar on cerebrospinal fluid pressure. Anaesthesia 1994;49:437–9. [23] Hogan BJ, Blaylock B, Tobian TL. Trauma multidisciplinary QI project: evaluation of cervical spine clearance, collar selection, and skin care. J Trauma Nurs 1997;4:60–7. [24] Ackland HM, Cooper DJ, Malham GM, Kossmann T. Factors predicting cervical collar-related decubitus ulceration in major trauma patients. Spine (Phila Pa 1976) 2007;32:423–8. [25] Khan SN, Erickson G, Sena MJ, Gupta MC. Use of flexion and extension radiographs of the cervical spine to rule out acute instability in patients with negative computed tomography scans. J Orthop Trauma 2011;25:51–6. [26] Ryken TC, Hadley MN, Walters BC, Aarabi B, Dhall SS, Gelb DE, et al. Radiographic assessment. Neurosurgery 2013;72(Suppl. 2):54–72. [27] Demetriades D, Charalambides K, Chahwan S, Hanpeter D, Alo K, Velmahos G, et al. Nonskeletal cervical spine injuries: epidemiology and diagnostic pitfalls. J Trauma 2000;48:724–7.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Comparison of CT and MRI findings for cervical spine clearance in obtunded patients without high impact trauma Lee A. Tan, Manish K. Kasliwal ∗ , Vincent C. Traynelis Department of Neurosurgery, Rush University Medical Center, Chicago, 60612, USA
a r t i c l e
i n f o
Article history: Received 18 July 2013 Received in revised form 12 October 2013 Accepted 17 February 2014 Available online 25 February 2014 Keywords: Cervical spine Computed tomography MRI Obtunded
a b s t r a c t Objective: Cervical spinal injuries occur in 2.0–6.6% of patients after blunt trauma and can have devastating neurological sequelae if left unrecognized. Although there is high quality evidence addressing cervical clearance in asymptomatic and symptomatic awake patients, cervical spine clearance in patients with altered level of alertness (i.e., obtunded patients with Glasgow coma scale (GCS) of 14 or less) following blunt trauma has been a matter of great controversy. Furthermore, there are no data on cervical spine clearance in obtunded patients without high impact trauma and these patients are often treated based on evidence from similar patients with high impact trauma. This retrospective study was conducted on this specific subgroup of patients who were admitted to a neurointensive care unit (NICU) with primary diagnoses of intracranial hemorrhage with history of minor trauma; the objective being to evaluate and compare cervical spinal computed tomography (CT) and magnetic resonance imaging (MRI) findings in this particular group of patients. Methods: Patients with GCS of 14 or less admitted to neruointensive care unit (NICU) at RUSH University Medical Center from 2008 to 2010 with diagnoses of intracranial hemorrhage (surgical or non-surgical) who had reported or presumed fall (i.e., “found down”) were queried from the computer data registry. A group of these patients had cervical spine CT and subsequently MRI for clearing the cervical spine and removal of the cervical collar. Medical records of these patients were reviewed for demographics, GCS score and injury specific data and presence or absence of cervical spine injury. Results: Eighty-three patients were identified from the computer database. Twenty-eight of these patients had positive findings on both CT and MRI (33.73% – Group I); four patients had a negative CT but had positive findings on follow-up MRI (4.82% – Group II); fifty-one patients had both negative CT and MRI (61.44% – Group III). All patients in Group I required either surgical stabilization or continuation of rigid cervical orthosis. All four patients in Group II had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI, but did not have any signs of fracture or ligamentous injury to suggest instability. They eventually underwent surgical decompression of the spinal cord during the same hospital stay. Cervical collars were safely removed in all patients in Group III. In our retrospective study, CT had a sensitivity of 0.875 [0.719–0.950, 95% CI] and a specificity of 1.000 [0.930–1.000, 95% CI] in detecting all cervical spine injuries compared to MRI. However, all patients with missed injuries had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI and were not unstable precluding cervical spine clearance. If only unstable injuries are considered, CT had a sensitivity of 1.00 [0.879–1.000, 95% CI] and a specificity is 1.000 [0.935–1.000, 95% CI] compared to MRI in this particular group of patients. Conclusion: CT is highly sensitive in detecting unstable injuries in obtunded patients with GCS of 14 or less in the absence of high impact trauma. In the absence of high impact trauma, neurosurgeons should be comfortable to discontinue the cervical collar after a negative, high-quality CT in this patient population. In the presence of focal neurological deficits unexplained by associated intracranial injury, an MRI may help diagnose intrinsic spinal cord injuries which necessarily may not be unstable in the presence of a negative CT and does not precludes clearance of cervical spine. © 2014 Elsevier B.V. All rights reserved.
∗ Corresponding author at: Department of Neurosurgery, Rush University Medical Center, Suite 855, Chicago, IL 60612, USA. Tel.: +1 312 942 6644; fax: +1 312 942 2176. E-mail addresses: manish [email protected], m [email protected] (M.K. Kasliwal). http://dx.doi.org/10.1016/j.clineuro.2014.02.006 0303-8467/© 2014 Elsevier B.V. All rights reserved.
24
L.A. Tan et al. / Clinical Neurology and Neurosurgery 120 (2014) 23–26
1. Introduction Cervical spine assessment and clearance is a priority in patients with suspected with cervical spine trauma due to potential devastating neurological deficits that can result from cervical spine injuries. There are well-established guidelines for cervical clearance in patients who are fully awake (Glasgow coma scale [GCS] = 15) with a reliable neurological exam, as described in the National Emergency X-Radiography Utilization Study (NEXUS) criteria and Canadian C-Spine (cervical-spine) Rule (CCR) [1]. Cervical spine clearance can however be challenging in obtunded patients with GCS of 14 or less and is a matter of great conjecture. Obtunded patients with suspected cervical spine injury are usually kept in rigid collars for cervical immobilization. Prolonged use of rigid collar is not only associated with increased incidence of pressure ulcers and intracranial pressure (ICP) elevation [2–4], but also presents challenges for oral hygiene, airway management and daily nursing care. Thus, safe and timely removal of cervical collars in obtunded patients has important clinical implications. With increased availability of high-resolution multi-detector computed tomography (MDCT), there have been reports of using MDCT alone for clearance of cervical spine in patients with blunt trauma who have non-focal examinations. The advantages of it being wider availability of CT, quick scan time and high sensitivity in detecting bony abnormalities such as acute fractures. However, many studies have shown that CT alone can miss clinically significant soft tissue injuries [5–12], especially in patients with high impact trauma. Magnetic resonance imaging (MRI) is highly sensitive in detecting soft tissue and ligamentous injuries but is much more time intensive and is not as readily available compared to CT. Even with technological advances with dramatic improvement in quality and resolution of CT, there is contradictory evidence in the value of CT alone for cervical spine clearance in obtunded patients [5,7–9,12–20]. Hence various studies have evaluated the role of CT alone or in conjunction with single upright radiographs or MRI to clear cervical spine injury in obtunded patients. While negative CT and MRI in obtunded patients have a very high degree of specificity in clearing the cervical spine, the value of a positive MRI is poorly understood and can be redundant leading to retaining the collar for longer periods due to lack of well defined criteria for instability on MRI with its very high sensitivity for detecting soft tissue trauma not necessarily ligamentous with undetermined clinical significance [16]. There is no universally accepted protocol regarding the cervical spine clearance in obtunded patients without history high impact trauma and the data on whether a negative CT alone is enough to clear cervical spine in this subgroup of patients is lacking. This clinical scenario is commonly encountered in many neruointensive care units (NICUs), where patients are admitted with being obtunded following some intracranial pathology and either have a fall secondary to that or are reported as “found down” with not even a witnessed history of trauma. These patients arrive in cervical collar to the NICU where neurosurgeons are often consulted for cervical spine clearance by the neurologists and neurointensivists. Given many of these patients have altered level of alertness with concurrent intracranial pathologies (subarachnoid hemorrhage [SAH], intracerebral hemorrhage [ICH], subdural hemorrhage [SDH]), they often do not cooperative fully for adequate and accurate assessment of strength or sensation or are not even conscious secondarily to the intracranial injury. While the absence of significance trauma makes the probability of cervical spine injury unlikely, theoretically increasing the negative predictive value of a negative CT scan, their intracranial pathologies often produce neurological deficits that cannot be clearly distinguished from possible cervical spine injury. Given above reasons, many of these patients ended up getting MRI for the purpose of cervical spine clearance.
Table 1 Comparison of cervical spine findings on CT vs. MRI.
CT MRI # of pts (%)
Group I
Group II
Group III
Positive Positive 28/83 (33.73%)
Negative Positive 4/83 (4.82%)
Negative Negative 51/83 (61.44%)
At our institution, we receive a significant number of patients from other institutions with the diagnosis of intracranial hemorrhage (primarily SAH, ICH, and SDH). Many of these patients arrive with cervical collars due to history of “fall” or “found down” with no history of high impact trauma and have unreliable neurological exams. It is the routine practice of the neurointensivists at our institution to image the cervical spines of such individuals which often involves CT and MRI. We initiated this study to focus on this subgroup of patients to evaluate and compare the cervical CT and MRI findings. 2. Methods Electronic medical record database at RUSH University Medical Center, Chicago, IL was queried for patients admitted to the NICU with the diagnosis of intracranial hemorrhage and concomitant history of minor cervical spine trauma most commonly ground level fall between January 2008 and December 2010. The study group primarily consisted of patients with SAH, ICH or SDH with history of “ground level fall” or “found down” and have unreliable neurological exams. Those who underwent both CT (Siemens, 64-slice) and MRI (Siemens, 1.5 or 3 T) of cervical spine were identified, which were read both by a board certified attending neuroradiologist and an attending neurosurgeon. Medical records of these patients were reviewed for demographics, admitting diagnosis, cervical CT/MRI findings and cervical spine management. Patients were divided into three groups based on CT and MRI findings. Group I consisted of patients with evidences of cervical injuries on both CT and MRI; Group II had negative CT but positive findings on MRI; Group III had both negative CT and MRI. Cervical spine management in each group was analyzed. Sensitivity and specificity of CT were calculated using MRI as the gold standard, since MRI has a 100% negative predicative value to detect soft tissue injuries as described by Muchow et al. in their meta-analysis [9]. 3. Results Eighty-three patients were identified from the computer database using our search criteria. Twenty-eight (33.73%) patients had pathologies identified on both CT and MRI (Group I); four patients (4.82%) had a negative CT but had positive finding on MRI (Group II); fifty-one patients (61.44%) had both negative CT and MRI (Group III). Results are summarized in Table 1. All patients in Group I required either surgical stabilization or continuation of rigid cervical orthosis. All four patients in Group II had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI, but without any obvious fracture or evidence of instability on CT which was read as negative. They eventually underwent surgical decompression of the spinal cord during the same hospital stay due to presence of spinal stenosis and clinical suspicion of central cord syndrome with no significant recovery once they were clinically stable. Cervical collars were safely removed in all patients in Group III. In our retrospective study, CT had a sensitivity of 0.875 [0.719–0.950, 95% CI] and a specificity of 1.000 [0.930–1.000, 95% CI] in detecting all cervical spine injuries compared to MRI. However, all patients with missed injuries on CT scan had intramedullary T2 hyper intensity consistent with possible central cord syndrome are not unstable
L.A. Tan et al. / Clinical Neurology and Neurosurgery 120 (2014) 23–26
injuries. If only unstable injuries are included, the sensitivity of CT for detecting was 1.00 [0.879–1.000, 95% CI] and a specificity of 1.000 [0.935–1.000, 95% CI] compared to MRI. The positive predictive value (PPV) for CT is calculated to be 100%, the negative predictive value (NPV) for unstable injuries was also 100%, while NPV for all injuries (including spinal cord injury) was 92.7%. The GCS scores of the 4 patients with injuries missed by CT scan but having positive MRI findings ranged from 12 to 14. The mean GCS for the entire study group is 12.09. The mean GCS scores for Groups I (CT+/MR+), Group II (CT−/MR+) and Group III (CT−/MR−) are 13.6, 13.4 and 11.1, respectively. 24% of patients (20/83) had a GCS 3–10, while 76% (63/83) had a GCS11–14 in our cohort. Though all 4 missed injuries occurred in the latter group, there was no statistically significant difference between the likelihood of missed injuries between the two GCS groups (p = 0.5696).
4. Discussion NEXUS and Eastern Association of the Surgery of Trauma (EAST) guidelines have been well-established and are standards for cervical spine clearance in patients who are awake, alert and cooperative with neurological exam. The NEXUS criteria have a sensitivity of 99.9% in detecting significant cervical injuries [1]. However, cervical spine clearance in obtunded patients (GCS of 14 or less) has been a persistent topic of debate. No universally accepted guidelines exist for this group of patients. There is even less data on cervical spine clearance in obtunded patients without history of high impact trauma. Unlike the awake and alert patients, obtunded patients cannot express pain and often lack reliable neurological exam that can help clinicians to identify potential cervical injuries. The management options for such patients include cervical immobilization with collar until reliable clinical exam can be obtained, collar removal after negative CT alone, or collar removal after negative CT and an additional imaging modality such as dynamic fluoroscopy or MRI. Prolonged cervical collar use has been associated with pressure ulcers and ICP elevations [2,4,6,21,22]. It also limits venous access, airway management and general nursing care for these patients which is important for this specific patient population who often have concomitant intracranial injuries/pathology [23,24]. Thus early and safe removal of cervical collar has important clinical implications. CT is the initial imaging modality of choice for cervical spine trauma due to its wide availability, quick scan time and high sensitivity in detecting bony abnormalities such as acute fractures. However, many studies have shown that CT scan alone can miss clinically significant soft tissue injuries [5–12]. Adjuvant modalities such as dynamic fluoroscopy and MRI have been used to further assess ligamentous injuries and cervical stability. The efficacy of dynamic fluoroscopy is limited because of high rate of inadequate visualization of the entire cervical spine, especially in the cervicothoracic junction [25]. MRI is highly sensitive in detecting soft tissue and ligamentous injuries but it is much more time intensive and is not as readily available as CT. With recent technological advances, the imaging quality and resolution of CT have dramatically improved. Many have argued that with modern multi-detector CT scanners, negative CT alone is sufficient for cervical spine clearance in obtunded patients [13–18]. However, many other studies have shown that CT can miss clinically important cervical injuries [5,7–9,12,15,19,20]. There are flaws in various studies precluding drawing firm conclusions. Most of these studies are from trauma centers where the probability of sustaining a cervical spine trauma in the form of fractures or ligamentous injuries is high justifying the aggressiveness in clearing the C spine and value of MRI in patients suspected with high impact trauma. Following such a protocol in subgroup as studied in this study may
25
be of questionable value as these patients do not have any major trauma predisposing them to injuries which may be occult on high resolution CT scans. A recent meta-analysis by Panczykowski et al. [16] reported that following a protocol of clearing cervical spine based on CT alone in obtunded patients, the incidence of unstable cervical spine injury among the studies of negative modern CT approximates to 1 in every 4776 patients. In fact, cervical spine is cleared after a negative MDCT in some trauma centers in the US and other countries (personal communication) in obtunded patients [18]. The most recent guidelines from the American Association of Neurological Surgeons (AANS)/Congress of Neurological Surgeons (CNS) Joint Guidelines Committee, recommends CT imaging as the initial imaging modality of choice in the obtunded patient (Level I). Subsequent to a negative CT, the recommendations becomes weak (Level II) and in patients in whom there is a high clinical suspicion of injury yet have a normal high-quality CT imaging study, the decision for further patient management is left to physicians trained in the diagnosis and management of spinal injuries. A further Level III recommendation following a negative CT was either continue cervical immobilization until asymptomatic, discontinue cervical immobilization following a normal MRI study obtained within 48 h of injury or discontinue cervical immobilization at the discretion of the treating physician [26]. While continuing cervical collars seems pretty straight forward, these patients often have associated intracranial hematomas and may remain comatose/obtunded for a long period of time making this approach less appealing, due to issues related to nursing care and complications due to cervical collar use itself. As related to MRI, the performance of MRI for cervical spine clearance may not be that difficult as these patients often require MRI for assessment of brain injuries and evaluation of stroke, obviating the technical challenge often seen in patients with trauma. Nevertheless, the value of obtaining MRI cervical spine should be carefully considered due to cost issues and benefits analysis, not ignoring the fact that a number of this patients do not even have any mechanism of injury with might really reduce the overall yield of MRI in changing clinical management. Also, while significant number of this patients end up having MRI brain for assessment of intracranial pathology, not all of them require it. In that case, there is value in clearing the cervical spine based on negative CT alone as it is cost effective and quick. Also, in the absence of high impact trauma, the chances of this patients harboring an injury detectable on MRI is very low as compared to patients with high trauma even in whom the chances of detecting a clinically significant injury being missed on CT and detectable per MRI is very low as per the meta-analysis by Panczykowski et al. [16]. As can be seen in this study, there was no patient who had a negative CT who subsequently had positive bony or ligamentous injuries identified on MRI. The only patients who underwent surgery following a negative CT had intramedullary T2 hyper intensity consistent with possible spinal cord injury on MRI and no unstable injury or any fractures which were missed on CT scan. These can be suspected based on focal neurology which if present is an indication for MRI not for clearing the spine itself but for planning further treatment. Since the subgroup of patients in this study was old, they do have high chances of sustaining central cord syndrome (CCS) after a minor fall and should be watched for this and appropriately evaluated. Also patients with CCS do not really needs collar and C-spine can be cleared even for this subgroup of patients once CT is negative. In our retrospective study, we analyzed cervical CT and MRI of eighty-three patients who were obtunded but had no history high impact trauma. CT did not miss any unstable injuries. MRI identified four patients who had intramedullary T2 hyper intensity consistent with possible spinal cord injury possibly secondary to significant pre-existing degenerative changes, osteophytes or ligamentous hypertrophy in the setting of minor trauma such as falling from standing height. Although clinical exam is unreliable in many
26
L.A. Tan et al. / Clinical Neurology and Neurosurgery 120 (2014) 23–26
obtunded patients, any asymmetric response to the painful stimuli or an exam that cannot be explained by the intracranial pathology should alert the clinician to further evaluate cervical spine with MRI. It is also interesting to note that the overall incidence of cervical spine injury was 38.6% in our study. Thirty-two of the eighty-three patients had positive findings on MRI and required either continuation of cervical collar or surgical decompression and stabilization. This number is much higher than the 2–6% incidence commonly quoted in literature. This discrepancy may be due to the fact that some of the prior studies only used plain X-ray or CT to identify bony fractures [27]. The demographic of our study population may also have contributed to the higher incidence since the patients in our study group tend to be older with pre-existing degenerative changes and are prone to various fractures even after a ground level fall. This limits the generalizability of our study to certain extent but remains very relevant to the patient population admitted to a typical NICU. The drawback of this study is that it is a retrospective and is subject to its biases. Also we do not have longterm follow up on these patients to rule out any delayed instability but considering the fact that all these patients who had a negative CT also had a MRI to rule out ligamentous injury which might manifest in a delayed fashion, a problem in various studies that compared CT with X-rays [16]. 5. Conclusion CT is a very useful imaging modality in evaluation cervical spine in obtunded patients with unreliable neurological exams. A negative high-quality CT scan is very reassuring and in the absence of high impact trauma as is seen in this patient population seems adequate for cervical spine clearance. Performance of routine MRI in this subgroup of patients without history of high impact trauma seems redundant following a negative high quality CT scan and may be avoided for the purpose of cervical spine clearance. In the presence of focal neurological deficits unexplained by associated intracranial injury, an MRI may help diagnose intrinsic spinal cord injuries which necessarily may not be unstable in the presence of a negative CT and does not precludes clearance of cervical spine. Funding None. References [1] Hoffman JR, Mower WR, Wolfson AB, Todd KH, Zucker MI. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. National Emergency X-Radiography Utilization Study Group. N Engl J Med 2000;343:94–9. [2] Mobbs RJ, Stoodley MA, Fuller J. Effect of cervical hard collar on intracranial pressure after head injury. ANZ J Surg 2002;72:389–91. [3] Davies G, Deakin C, Wilson A:. The effect of a rigid collar on intracranial pressure. Injury 1996;27:647–9. [4] Hunt K, Hallworth S, Smith M:. The effects of rigid collar placement on intracranial and cerebral perfusion pressures. Anaesthesia 2001;56:511–3.
[5] Brown CV, Foulkrod KH, Reifsnyder A, Bui E, Lopez A, Hummell M, et al. Computed tomography versus magnetic resonance imaging for evaluation of the cervical spine: how many slices do you need? Am Surg 2010;76: 365–8. [6] Dunham CM, Brocker BP, Collier BD, Gemmel DJ. Risks associated with magnetic resonance imaging and cervical collar in comatose, blunt trauma patients with negative comprehensive cervical spine computed tomography and no apparent spinal deficit. Crit Care 2008;12:R89. [7] Menaker J, Philp A, Boswell S, Scalea TM. Computed tomography alone for cervical spine clearance in the unreliable patient – are we there yet? J Trauma 2008;64:898–903 [discussion 903–4]. [8] Menaker J, Stein DM, Philp AS, Scalea TM. 40-slice multidetector CT: is MRI still necessary for cervical spine clearance after blunt trauma? Am Surg 2010;76:157–63. [9] Muchow RD, Resnick DK, Abdel MP, Munoz A, Anderson PA. Magnetic resonance imaging (MRI) in the clearance of the cervical spine in blunt trauma: a metaanalysis. J Trauma 2008;64:179–89. [10] Hogan GJ, Mirvis SE, Shanmuganathan K, Scalea TM. Exclusion of unstable cervical spine injury in obtunded patients with blunt trauma: is MR imaging needed when multi-detector row CT findings are normal? Radiology 2005;237: 106–13. [11] Spiteri V, Kotnis R, Singh P, Elzein R, Madhu R, Brooks A, et al. Cervical dynamic screening in spinal clearance: now redundant. J Trauma 2006;61:1171–7 [discussion 1177]. [12] Stassen NA, Williams VA, Gestring ML, Cheng JD, Bankey PE. Magnetic resonance imaging in combination with helical computed tomography provides a safe and efficient method of cervical spine clearance in the obtunded trauma patient. J Trauma 2006;60:171–7. [13] Brohi K, Healy M, Fotheringham T, Chan O, Aylwin C, Whitley S, et al. Helical computed tomographic scanning for the evaluation of the cervical spine in the unconscious, intubated trauma patient. J Trauma 2005;58:897–901. [14] Como JJ, Thompson MA, Anderson JS, Shah RR, Claridge JA, Yowler CJ, et al. Is magnetic resonance imaging essential in clearing the cervical spine in obtunded patients with blunt trauma? J Trauma 2007;63:544–9. [15] Harris TJ, Blackmore CC, Mirza SK, Jurkovich GJ. Clearing the cervical spine in obtunded patients. Spine (Phila Pa 1976) 2008;33:1547–53. [16] Panczykowski DM, Tomycz ND, Okonkwo DO. Comparative effectiveness of using computed tomography alone to exclude cervical spine injuries in obtunded or intubated patients: meta-analysis of 14,327 patients with blunt trauma. J Neurosurg 2011;115:541–9. [17] Steigelman M, Lopez P, Dent D, Myers J, Corneille M, Stewart R, et al. Screening cervical spine MRI after normal cervical spine CT scans in patients in whom cervical spine injury cannot be excluded by physical examination. Am J Surg 2008;196:857–62 [discussion 862–3]. [18] Tomycz ND, Chew BG, Chang YF, Darby JM, Gunn SR, Nicholas DH, et al. MRI is unnecessary to clear the cervical spine in obtunded/comatose trauma patients: the four-year experience of a level I trauma center. J Trauma 2008;64:1258–63. [19] Simon JB, Schoenfeld AJ, Katz JN, Kamath A, Wood K, Bono CM, et al. Are “normal” multidetector computed tomographic scans sufficient to allow collar removal in the trauma patient? J Trauma 2010;68:103–8. [20] Diaz Jr JJ, Aulino JM, Collier B, Roman C, May AK, Miller RS, et al. The early work-up for isolated ligamentous injury of the cervical spine: does computed tomography scan have a role? J Trauma 2005;59:897–903 [discussion 903–4]. [21] Kolb JC, Summers RL, Galli RL. Cervical collar-induced changes in intracranial pressure. Am J Emerg Med 1999;17:135–7. [22] Raphael JH, Chotai R. Effects of the cervical collar on cerebrospinal fluid pressure. Anaesthesia 1994;49:437–9. [23] Hogan BJ, Blaylock B, Tobian TL. Trauma multidisciplinary QI project: evaluation of cervical spine clearance, collar selection, and skin care. J Trauma Nurs 1997;4:60–7. [24] Ackland HM, Cooper DJ, Malham GM, Kossmann T. Factors predicting cervical collar-related decubitus ulceration in major trauma patients. Spine (Phila Pa 1976) 2007;32:423–8. [25] Khan SN, Erickson G, Sena MJ, Gupta MC. Use of flexion and extension radiographs of the cervical spine to rule out acute instability in patients with negative computed tomography scans. J Orthop Trauma 2011;25:51–6. [26] Ryken TC, Hadley MN, Walters BC, Aarabi B, Dhall SS, Gelb DE, et al. Radiographic assessment. Neurosurgery 2013;72(Suppl. 2):54–72. [27] Demetriades D, Charalambides K, Chahwan S, Hanpeter D, Alo K, Velmahos G, et al. Nonskeletal cervical spine injuries: epidemiology and diagnostic pitfalls. J Trauma 2000;48:724–7.