Prevalence of Cervical Spinal Injury in Trauma

Neurosurg Focus 25 (5):E10, 2008

Prevalence of cervical spinal injury in trauma

An d r e w H. Mi l b y , B.S.,1 Cas e y H. Ha l p e r n , M.D.,1 We n sh e n g Gu o , Ph.D.,2 a n d Sh e r ma n C. St e i n , M.D.1 1Department of Neurosurgery, Hospital of the University of Pennsylvania; and 2Center for Clinical Epidemi- ology and Biostatistics, University of Pennsylvania, School of Medicine, Philadelphia, Pennsylvania

Object. Diagnosis of cervical spinal injury (CSI) is an essential aspect of the trauma evaluation. This task is especially difficult in patients who are not clinically able to be evaluated (unevaluable) because of distracting pain- ful injuries, intoxication, or concomitant head injury. For this population, the appropriate use of advanced imaging techniques for cervical spinal clearance remains undetermined. This study was undertaken to estimate the prevalence of unstable CSI, particularly among patients in whom clinical evaluation is impossible or unreliable. Methods. Estimates of the prevalence of CSI in populations consisting of all trauma patients, alert patients only, and clinically unevaluable patients only were determined by variance-weighted pooling of data from 65 publications (281,864 patients) that met criteria for review. Results. The overall prevalence of CSI among all trauma patients was 3.7%. The prevalence of CSI in alert patients was 2.8%, whereas unevaluable patients were at increased risk of CSI with a prevalence of 7.7% (p = 0.007). Overall, 41.9% of all CSI cases were considered to exhibit instability. Conclusions. Trauma patients who are clinically unevaluable have a higher prevalence of CSI than alert patients. Knowledge of the prevalence and risk of such injuries may help establish an evidence-based approach to the detection and management of clinically occult CSI. (DOI: 10.3171/FOC.2008.25.11.E10)

Ke y Wo r d s • cervical spine injury • prevalence • trauma

uadriplegia due to spinal cord injury is a dev- in 6–71% of critically ill patients during and after trans- astating consequence of trauma to the cervical port.94 Accurate knowledge of the prevalence of CSI in spine, involving numerous functional, psycho- trauma patients is therefore essential for assessing the Q social, and economic ramifications.7,12,13,24,25,27–29,45, need for immobilization and/or further imaging. Scat- 49,61 Identification of unstable CSI is therefore an essential tered studies of CSI in clinical series composed of all aspect of the trauma evaluation in preventing subsequent trauma patients report CSI prevalences ranging from neurological damage.6,22,71,72,75,76 This task is especially 1 to 14%.59,87 However, unevaluable patients require a difficult in patients who are not clinically evaluable (un- higher index of suspicion than the general trauma popu- evaluable group) because of intoxication or concomitant lation,5,46,51,64,80,96 with one patient series estimating that head injury, and has led to the use of advanced imaging a GCS score ≤ 8 incurs an almost 6-fold increase in the techniques such as CT and MR imaging for radiologi- risk of CSI.50 Numerous patient series have examined the cal clearance.1,2,15,20,58,83,90,91 Continued advances in imag- sensitivities of various imaging modalities in the detec- ing quality and sensitivity now raise questions about the tion of CSI, and as such represent a large volume of data practice of clearing even alert, low-risk patients by clini- from which to calculate overall prevalence. However, cal criteria alone,30 and have precluded the establishment considerable variation exists in rates of radiographic evi- of any consensus regarding the appropriate indications dence of CSI, with 1 study reporting CT or MR imaging for the use of imaging studies.58,65,67,85,90 findings in 40% of obtunded patients.90 By systematically Although a lower threshold for the use of advanced pooling data from relevant clinical series, more general- imaging would hypothetically result in the detection and izable estimates of CSI prevalence in all trauma patients, possible prevention of a greater number of CSIs, these alert patients, and unevaluable patients with trauma can benefits must be weighed against the associated risks be determined, along with the proportion of patients and considerable costs of performing such studies and whose unstable injuries confer a risk of quadriplegia. the additional treatments initiated due to false-positive results.18,19,41,77 Indeed, complications have been reported Methods

Abbreviations used in this paper: CSI = cervical spine injury; We performed English-language searches of Med- GCS = Glasgow Coma Scale. line and PubMed for articles published between 1985

Neurosurg. Focus / Volume 25 / November 2008 1

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC A. H. Milby et al. 0.119 0.143 0.017 0.014 0.013 0.017 0.072 0.024 0.081 0.021 0.026 0.032 0.028 0.035 0.038 0.020 0.045 0.020 0.089 0.030 0.050 0.022 0.085 0.090 Injury 0.036 0.048 0.043 0.084 0.044 Incidence 91 410 176 219 179 145 676 106 407 467 775 667 434 480 886 604 860 400 1757 6766 5021 4285 2603 9903 6500 4460 No. of 14,755 34,069 111,219 Patients Patient Population trauma admissions undergoing cervical CT all trauma admissions adult pedestrians injured by vehicles all trauma admissions all trauma admissions blunt trauma admissions blunt trauma admissions blunt trauma admissions “multitrauma” admissions all trauma admissions undergoing CT patients with neck pain, neurological deficit, or intoxication trauma patients undergoing both radiography and CT motor vehicle crash trauma admissions blunt trauma admissions blunt head or neck trauma admissions trauma patients undergoing both radiography and CT patients with neck pain, neurological deficit, or intoxication all trauma admissions trauma patients with nonspinal injuries all trauma admissions trauma patients undergoing flexion-extension radiography all trauma admissions all trauma admissions blunt trauma admissions high-energy trauma admissions blunt trauma admissions blunt trauma admissions all trauma admissions blunt head trauma admissions Case Accrual Method retrospective evaluation of clinical algorithm retrospective observational prospective observational prospective observational prospective observational prospective evaluation of clinical algorithm prospective observational prospective evaluation of clinical algorithm retrospective observational retrospective observational prospective observational prospective observational retrospective observational prospective observational retrospective evaluation of clinical algorithm retrospective observational prospective observational prospective observational prospective evaluation of clinical algorithm retrospective evaluation of clinical algorithm retrospective observational retrospective survey prospective evaluation of clinical algorithm retrospective observational prospective evaluation of clinical algorithm retrospective evaluation of clinical algorithm retrospective evaluation of clinical algorithm retrospective evaluation of clinical algorithm retrospective observational Setting single UK trauma center single US Level I trauma center Trauma registry, US countyTrauma single US Level trauma II center single US university hospital emergency department single US Level I trauma center single US Level I trauma center single US Level I trauma center single Canadian Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center single Canadian Level I trauma center US university21 and community hospitals four US trauma centers single US Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center 106 US Level I–III106 trauma centers single US Level I trauma center single US Level I trauma center single Netherlands Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center Author & Year Spiteri 2006 et al., Williams 1992 et al., Yanar et al., 2007 et al., Yanar Sanchez 2005 et al., Sharma 2007 et al., Roberge & Wears, 1992 Ross 1992 et al., Roberge 1988 et al., Prasad 1999 et al., Ptak 2001 et al., Nguyen & Clark, 2005 McCulloch 2005 et al., MacDonald 1990 et al., Mower 2001 et al., Neifeld 1988 et al., Lee 2001 et al., Mathen 2007 et al., Kreipke 1989 et al., Harris 2000 et al., Hanson 2000 et al., Insko 2002 et al., Grossman 1999 et al., Cox et al., 2001 Cox et al., Demetriades 2000 et al., Edwards 2001 et al., Gale 2005 et al., Borock 1991 et al., Banit 2000 et al., Bayless & Ray, 1989 1989 Bayless & Ray, Table 1: Prevalence 1: Table of CSI in all trauma patients

2 Neurosurg. Focus / Volume 25 / November 2008

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC Prevalence of cervical spine injury in trauma

and January 2008. The search used various combinations

- of the key words “spinal injuries,” “cervical vertebrae,” “instability,” “trauma,” “clearance,” “neck,” “diagnosis,” Inci 0.017 0.017 0.017 0.015 0.020 Injury 0.025 0.026 0.046 0.043 dence “epidemiology,” “prevalence,” and “incidence.” We re- fined the search by eliminating laboratory studies, case reports, editorials, or reviews without newly reported 351 324 286 286 303 353 2176 8924 8283 No. of data, and case series with duplicated or overlapping data. Patients These findings were supplemented by using the “Find Similar” and “Find Citing Articles” features of Medline and “Related Articles” feature of PubMed, as well as the bibliographies of selected articles. Articles were analyzed and compared with reference to the setting, study organi- zation, definitions of clinical criteria, and data collection methods. Studies restricted to children < 15 years of age were excluded. If a study reported prevalence rates from pediatric cases separately from those in an adult population, these data were also excluded from the analysis. Studies meeting our criteria for inclusion were or-

Patient Population ganized into 3 categories: those composed of all trauma patients, alert patients, and unevaluable patients with trauma. Those studies reporting rates of instability upon detection of CSI were also placed into a fourth category, which overlapped in part with the previously listed trauma admissions undergoing cervical CT conscious and oriented blunt trauma admissions awake and alert blunt trauma admissions patients with neck pain following traumatic injury alert, nonintoxicated blunt trauma admissions alert, nonintoxicated blunt trauma admissions patients with acute blunt trauma to head or neck patients with acute blunt trauma to head or neck alert trauma admissions categories. The “all trauma” category contained series in which patients were not further classified by clinical evaluability on presentation. These series were composed of patients with either unrestricted blunt and penetrating trauma or blunt trauma alone, whereas those composed solely of patients with penetrating trauma were excluded. Patients were deemed alert if they had reliable clinical examination findings, consisting minimally of being able to respond to questions regarding neck pain and cooperate with neck movement instructions. Patients were considered unevaluable if impaired consciousness, inebriation,

Case Accrual Method confusion, endotracheal intubation, or distracting injuries rendered the clinical examination of the cervical spine unreliable. An unstable injury was defined as any frac- ture, dislocation, or purely ligamentous injury necessitat- retrospective evaluation of clinical algorithm retrospective observational prospective observational prospective observational retrospective observational prospective cohort prospective evaluation of clinical algorithm prospective evaluation of clinical algorithm retrospective observational ing external stabilization and/or operative fixation. Data concerning unstable injuries were pooled from all series - reporting their prevalence without subclassification on the basis of clinical evaluability. Mean prevalence values for each group were ob- tained using variance-weighted pooling. A mixed-effects logistic regression model was used, using SAS PROC NLMIXED (SAS, Inc.). Data within each study were

Setting considered a cluster and a hierarchical model was used to calculate the average prevalence rate. The binary nature of the outcome allowed the use of summary statistics as a proxy for the entire data set. The effect across studies was assumed to vary as a normal distribution. The overall single US Level I trauma center single university Turkish hospital emer gency department single US Level I trauma center single US Level trauma II center single US Level trauma II center single US military medical center 10 Canadian10 trauma centers 9 Canadian tertiary care hospitals single US Level I trauma center prevalence rate was calculated as the population-average estimate, together with its 95% confidence intervals. Mean prevalence values for alert and unevaluable patients were compared, using a likelihood ratio test for pooled data.60 We considered differences with a probability value < 0.05 to be statistically significant.

Author & Year Results Table 2: PrevalenceTable of CSI in alert trauma patients Barba 2001 et al., Ersoy 1995 et al., Gonzalez 1999 et al., McNamara 1988 et al., McNamara 1990 et al., Roth et al., 1994 Roth et al., Stiell 2001 et al., Stiell 2003 et al., Zabel 1997 et al., Sixty-five studies with a total of 281,864 subjects

Neurosurg. Focus / Volume 25 / November 2008 3

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC A. H. Milby et al. 0.119 0.143 0.017 0.014 0.024 0.072 0.021 0.028 0.020 0.020 0.022 Injury 0.036 0.050 0.085 0.030 0.048 0.084 0.043 0.044 0.090 Incidence 91 176 179 219 145 106 407 775 667 860 604 886 400 1757 4285 6500 4460 No. of 14,755 34,069 111,219 Patients - Patient Population patients with GCS scores on admission < 15 patients with TBI and GCS scores < 8 unconscious trauma patients head with injury or with GCS score GCS scores >12 3-12, abnormalityCT intubated for head or severe multisystem injuries head-injured ICU admissions altered mental status or neurological deficit unconscious or semiconscious trauma patients obtunded trauma patients patients intubated for polytrauma altered mental status or distracting injuries unconscious trauma patients admitted to ICU patients with high-energy mechanisms of injury TBI w/ GCSTBI w/ scores < 8 unconscious with TBI in ICU altered mental status after blunt traumatic injury unevaluable patients undergoing MR imaging after nega tive CT obtunded blunt trauma patients unresponsive or obtunded with GCS scores ≤ 10 obtunded blunt trauma patients Case Accrual Method retrospective observational prospective evaluation of clinical algorithm retrospective observational retrospective observational retrospective observational prospective evaluation of clinical algorithm retrospective observational retrospective observational retrospective observational prospective observational prospective observational retrospective observational retrospective observational retrospective observational prospective observational prospective evaluation of clinical algorithm retrospective observational prospective observational retrospective observational retrospective evaluation of clinical algorithm Setting single US Level I trauma center single US Level I trauma center single UK trauma center 2 US Level I trauma centers 2 US Level I trauma centers single US Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center single UK trauma center single US Level I trauma center single Australian trauma center single US Level I trauma center all Pennsylvania trauma centers single UK trauma center single US Level I trauma center single US Level I trauma center single Canadian Level I trauma center single US military medical center single US Level I trauma center Author & Year ICU = intensive care unit; TBI = traumatic brain injury. Table 3: Prevalence 3: Table of CSI in clinically unevaluable trauma patients* Chiu et al., 2001 Chiu et al., Bolinger 2004 et al., Brooks & Willett, 2001 Holly 2002 et al., D'Alise et al., 1999 D'Alise et al., Davis 2001 et al., Griffen 2003 et al., Griffiths 2002 et al., Hogan 2005 et al., Jelly 2000 et al., Diaz 2003 et al., Freedman 2005 et al., Geck 2001 et al., Piatt 2006 et al., Padayachee 2006 et al., Schenarts 2001 et al., Kihiczak 2001 et al., Widder 2004 et al., * Sees 1998 et al., Stassen 2006 et al.,

4 Neurosurg. Focus / Volume 25 / November 2008

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC Prevalence of cervical spine injury in trauma 0.174 0.171 0.181 0.714 0.612 0.916 0.707 0.219 0.242 1.000 1.000 0.291 0.400 0.250 0.400 Injury 0.302 0.000 0.583 0.348 0.556 0.333 Incidence 3 2 7 3 18 24 70 92 95 60 50 20 43 151 178 818 471 740 165 253 292 No. of Patients - Patient Population pedestrians injured by automobiles trauma admissions undergoing cervical CT obtunded blunt trauma patients altered mental status after blunt traumatic injury obtunded blunt trauma patients head injury with GCS or scoresGCS scores 3–12, > 12 with abnormality CT motor vehicle accident trauma admissions patients with neck pain, neurological deficit, or intoxica tion cohort of patients with known CSIs blunt trauma admissions trauma patients with nonspinal injuries unconscious or semiconscious trauma patients blunt trauma admissions blunt trauma admissions patients with GCS scores on admission < 15 blunt trauma admissions all trauma admissions blunt trauma with intoxication or paralytics all trauma admissions unconscious trauma patients admitted to ICU patients with high-energy mechanisms of injury - - Case Accrual Method retrospective observational retrospective evaluation of clinical algorithm prospective observational prospective evaluation of clinical algo rithm retrospective literature review retrospective observational retrospective observational prospective observational retrospective observational prospective observational prospective evaluation of clinical algo rithm retrospective observational retrospective observational retrospective observational retrospective observational retrospective observational retrospective evaluation of clinical algorithm prospective observational retrospective observational retrospective observational retrospective observational Setting 2 US Level I trauma centers single UK trauma center single Canadian Level I trauma center single US Level I trauma center multiple Level I trauma centers 2 US Level I trauma centers single Canadian Level I trauma center single US Level I trauma center single Canadian Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center 21 US university21 and community hospitals single US Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center single US Level I trauma center 6 US trauma centers single Australian trauma center single US Level I trauma center Author & Year Yanar et al., 2007 et al., Yanar Spiteri 2006 et al., Widder 2004 et al., Schenarts 2001 et al., Sliker 2005 et al., Holly 2002 et al., MacDonald 1990 et al., Mathen 2007 et al., Reid 1987 et al., Ross 1992 et al., Harris 2000 et al., Griffiths 2002 et al., Goldberg 2001 et al., Gerrelts 1991 et al., Table 4: Prevalence 4: Table of unstable CSIs in all trauma patients 2001 Chiu et al., Demetriades 2000 et al., Banit 2000 et al., Berne 1999 et al., Davis 1993 et al., Freedman 2005 et al., Geck 2001 et al.,

Neurosurg. Focus / Volume 25 / November 2008 5

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC A. H. Milby et al.

Table 5: Pooled prevalences of CSIs*

No. of Pooled Patient Population Patients Mean (%) 95% CI all trauma 209,320 3.68 3.64–3.72 alert patients 21,286 2.78 2.74–2.81 unevaluable patients 49,938 7.66 7.59–7.81 proportion of unstable 3,555 41.87 41.46–42.28 injuries

* CI = confidence interval.

use of advanced imaging techniques for cervical spinal clearance is cost effective when compared with prolonged semirigid collar immobilization. A more precise quan- tification of the prevalence of CSI in these populations allows us for the first time to make evidence-based deci- sions in guiding large-scale resource utilization. It is our expectation that these prevalence figures will aid in the calculations needed for indicated cost-effectiveness stud- Fig. 1. Flow chart showing the evidentiary value with regard to prev- ies. alence of CSI in publications fulfilling the initial search criteria. Bold numbers represent the number of publications in each category. There are several important limitations to this study. Many of these limitations are inherent to the technique of meta-analysis, and are the result of variations in the were selected with information on the prevalence of CSI definition of certain clinical terms used by articles that or the proportion of instability in CSI.3,4,8–11,14,16,17,21,23,26,31– met our inclusion criteria. Most notable of these is the 40,42–44,47,48,52–57,62,63,66,68–70,73,74,78,79,81,84,86,88,92,93,95,97,98 Twenty- distinction between alert and unevaluable trauma patient nine of these reports included 209,320 patients who sus- populations. Every attempt was made to categorize the tained nonspecific trauma and were not categorized by included studies systematically, although the existence of level of consciousness (Table 1). Nine series with 21,286 minor disparities between study populations is acknowl- cases pertained specifically to alert patients with reliable edged. Authors also differed as to whether the overall clinical examination findings (Table 2). Twenty series trauma populations included patients with penetrating contained 49,938 unconscious or obtunded patients who trauma or were restricted to those with blunt trauma. The met our criteria for unevaluable (Table 3). Twenty-one se- regional variability in rates of penetrating trauma limits ries composed of 3555 patients with known CSI reported the generalizability of these data to all trauma centers, data on the proportion of these injuries considered to be although these rates were invariably quite low. unstable (Table 4; Fig. 1). Tables 1 through 4 analyze the evidentiary characteristics of these series. The overall prevalence of CSI in all trauma patients Conclusions was 3.7% (Tables 1 and 5). In alert patients only, the Trauma patients who are clinically unevaluable have prevalence of CSI was 2.8% (Tables 2 and 5). Clinically a higher prevalence of CSI than alert patients. Detection unevaluable patients were found to be at increased risk of CSI in this population is especially challenging, and of CSI with a prevalence of 7.7% (Tables 3 and 5). Once places these patients at increased risk for cervical insta- detected, 41.9% of all CSI were subsequently determined bility and quadriplegia. Knowledge of the prevalence and to be unstable (Tables 4 and 5). The difference in preva- risk of such injuries may help establish an evidence-based lence of CSI between the alert and unevaluable groups approach to the detection and management of clinically was statistically significant, with unevaluable patients occult CSI. at a significantly greater risk for CSI than alert patients (p = 0.0072). Disclaimer The authors report no conflict of interest concerning the mate- Discussion rials or methods used in this study or the findings specified in this Our findings demonstrate a higher prevalence of CSI paper. in clinically unevaluable patients with trauma compared with alert patients with trauma. Hence, this high-risk pa- References tient population may be subject to increased occult un- 1. Adams JM, Cockburn MI, Difazio LT, et al: Spinal clearance stable injuries. The potential for quadriplegia following in the difficult trauma patient: a role for screening MRI of the CSI that is undiagnosed underscores the importance of spine. Am Surg 72:101–105, 2006 detecting such injuries, but it is unknown whether the 2. Antevil JL, Sise MJ, Sack DI, et al: Spiral computed tomogra-

6 Neurosurg. Focus / Volume 25 / November 2008

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC Prevalence of cervical spine injury in trauma

phy for the initial evaluation of spine trauma: a new standard etal cervical spine injuries: epidemiology and diagnostic pit- of care? J Trauma 61:382–387, 2006 falls. J Trauma 48:724–727, 2000 3. Banit DM, Grau G, Fisher JR: Evaluation of the acute cervi- 24. DeVivo MJ: Causes and costs of spinal cord injury in the Unit- cal spine: a management algorithm. J Trauma 49:450–456, ed States. Spinal Cord 35:809–813, 1997 2000 25. DeVivo MJ, Rutt RD, Black KJ, et al: Trends in spinal cord 4. Barba CA, Taggert J, Morgan AS, et al: A new cervical spine injury demographics and treatment outcomes between 1973 clearance protocol using computed tomography. J Trauma and 1986. Arch Phys Med Rehabil 73:424–430, 1992 51:652–657, 2001 26. Diaz JJ Jr, Gillman C, Morris JA Jr, et al: Are five-view plain 5. Bayless P, Ray VG: Incidence of cervical spine injuries in as- films of the cervical spine unreliable? A prospective evalu- sociation with blunt head trauma. Am J Emerg Med 7:139– ation in blunt trauma patients with altered mental status. J 142, 1989 Trauma 55:658–654, 2003 6. Bellabarba C, Mirza SK, West GA, et al: Diagnosis and treat- 27. Dryden DM, Saunders LD, Jacobs P, et al: Direct health care ment of craniocervical dislocation in a series of 17 consecu- costs after traumatic spinal cord injury. J Trauma 59:443– tive survivors during an 8-year period. J Neurosurg Spine 449, 2005 4:429–440, 2006 28. Dryden DM, Saunders LD, Rowe BH, et al: Depression fol- 7. Berkowitz M: Assessing the socioeconomic impact of im- lowing traumatic spinal cord injury. Neuroepidemiology proved treatment of head and spinal cord injuries. J Emerg 25:55–61, 2005 Med 11 (Suppl 1):63–67, 1993 29. Dryden DM, Saunders LD, Rowe BH, et al: Utilization of 8. Berne JD, Velmahos GC, El-Tawil Q, et al: Value of complete health services following spinal cord injury: a 6-year follow- cervical helical computed tomographic scanning in identify- up study. Spinal Cord 42:513–525, 2004 ing cervical spine injury in the unevaluable blunt trauma pa- 30. Duane TM, Dechert T, Wolfe LG, et al: Clinical examina- tient with multiple injuries: a prospective study. J Trauma tion and its reliability in identifying cervical spine fractures. 47:896–893, 1999 J Trauma 62:1405–1410, 2007 9. Bolinger B, Shartz M, Marion D: Bedside fluoroscopic flex- 31. Edwards MJ, Frankema SP, Kruit MC, et al: Routine cervical ion and extension cervical spine radiographs for clearance of spine radiography for trauma victims: does everybody need the cervical spine in comatose trauma patients. J Trauma it? J Trauma 50:529–534, 2001 56:132–136, 2004 32. Ersoy G, Karcioglu O, Enginbas Y, et al: Are cervical spine 10. Borock EC, Gabram SG, Jacobs LM, et al: A prospective x-rays mandatory in all blunt trauma patients? Eur J Emerg analysis of a two-year experience using computed tomog- Med 2:191–195, 1995 raphy as an adjunct for cervical spine clearance. J Trauma 33. Freedman I, van Gelderen D, Cooper DJ, et al: Cervical spine 31:1001–1006, 1991 assessment in the unconscious trauma patient: a major trauma 11. Brooks RA, Willett KM: Evaluation of the Oxford protocol service’s experience with passive flexion-extension radiogra- for total spinal clearance in the unconscious trauma patient. J phy. J Trauma 58:1183–1188, 2005 Trauma 50:862–867, 2001 34. Gale SC, Gracias VH, Reilly PM, et al: The inefficiency of 12. Burney RE, Maio RF, Maynard F, et al: Incidence, character- plain radiography to evaluate the cervical spine after blunt istics, and outcome of spinal cord injury at trauma centers in trauma. J Trauma 59:1121–1125, 2005 North America. Arch Surg 128:596–599, 1993 35. Geck MJ, Yoo S, Wang JC: Assessment of cervical ligamen- 13. Cardenas DD, Hoffman JM, Kirshblum S, et al: Etiology and tous injury in trauma patients using MRI. J Spinal Disord incidence of rehospitalization after traumatic spinal cord inju- 14:371–377, 2001 ry: a multicenter analysis. Arch Phys Med Rehabil 85:1757– 36. Gerrelts BD, Petersen EU, Mabry J, et al: Delayed diagnosis of 1763, 2004 cervical spine injuries. J Trauma 31:1622–1626, 1991 14. Chiu WC, Haan JM, Cushing BM, et al: Ligamentous inju- 37. Goldberg W, Mueller C, Panacek E, et al: Distribution and ries of the cervical spine in unreliable blunt trauma patients: patterns of blunt traumatic cervical spine injury. Ann Emerg incidence, evaluation, and outcome. J Trauma 50:457–464, Med 38:17–21, 2001 2001 38. Gonzalez RP, Fried PO, Bukhalo M, et al: Role of clinical 15. Como JJ, Thompson MA, Anderson JS, et al: Is magnetic examination in screening for blunt cervical spine injury. J Am resonance imaging essential in clearing the cervical spine in Coll Surg 189:152–157, 1999 obtunded patients with blunt trauma? J Trauma 63:544–549, 39. Griffen MM, Frykberg ER, Kerwin AJ, et al: Radiographic 2007 clearance of blunt cervical spine injury: plain radiograph or 16. Cox MW, McCarthy M, Lemmon G, et al: Cervical spine in- computed tomography scan? J Trauma 55:222–227, 2003 stability: clearance using dynamic fluoroscopy. Curr Surg 40. Griffiths HJ, Wagner J, Anglen J, et al: The use of forced flex- 58:96–100, 2001 ion/extension views in the obtunded trauma patient. Skeletal 17. D’Alise MD, Benzel EC, Hart BL: Magnetic resonance imag- Radiol 31:587–591, 2002 ing evaluation of the cervical spine in the comatose or ob- 41. Grogan EL, Morris JA Jr, Dittus RS, et al: Cervical spine eval- tunded trauma patient. J Neurosurg 91:54–59, 1999 uation in urban trauma centers: lowering institutional costs 18. Daffner RH: Controversies in cervical spine imaging in trau- and complications through helical CT scan. J Am Coll Surg ma patients. Emerg Radiol 11:2–8, 2004 200:160–165, 2005 19. Daffner RH: Controversies in cervical spine imaging in trau- 42. Grossman MD, Reilly PM, Gillett T, et al: National survey of ma patients. Semin Musculoskelet Radiol 9:105–115, 2005 the incidence of cervical spine injury and approach to cervical 20. Daffner RH, Sciulli RL, Rodriguez A, et al: Imaging for eval- spine clearance in U.S. trauma centers. J Trauma 47:684– uation of suspected cervical spine trauma: a 2-year analysis. 690, 1999 Injury 37:652–658, 2006 43. Hanson JA, Blackmore CC, Mann FA, et al: Cervical spine in- 21. Davis JW, Kaups KL, Cunningham MA, et al: Routine evalu- jury: a clinical decision rule to identify high-risk patients for ation of the cervical spine in head-injured patients with dy- helical CT screening. AJR Am J Roentgenol 174:713–717, namic fluoroscopy: a reappraisal. J Trauma 50:1044–1047, 2000 2001 44. Harris MB, Kronlage SC, Carboni PA, et al: Evaluation of 22. Davis JW, Phreaner DL, Hoyt DB, et al: The etiology of the cervical spine in the polytrauma patient. Spine 25:2884– missed cervical spine injuries. J Trauma 34:342–346, 1993 2892, 2000 23. Demetriades D, Charalambides K, Chahwan S, et al: Nonskel- 45. Harvey LA, Batty J, Jones R, et al: Hand function of C6 and

Neurosurg. Focus / Volume 25 / November 2008 7

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC A. H. Milby et al.

C7 tetraplegics 1 - 16 years following injury. Spinal Cord 66. Mower WR, Hoffman JR, Pollack CV Jr, et al: Use of plain 39:37–43, 2001 radiography to screen for cervical spine injuries. Ann Emerg 46. Hills MW, Deane SA: Head injury and facial injury: is there Med 38:1–7, 2001 an increased risk of cervical spine injury? J Trauma 34:549– 67. Muchow RD, Resnick DK, Abdel MP, et al: Magnetic reso- 544, 1993 nance imaging (MRI) in the clearance of the cervical spine in 47. Hoffman JR, Mower WR, Wolfson AB, et al: Validity of a set blunt trauma: a meta-analysis. J Trauma 64:179–189, 2008 of clinical criteria to rule out injury to the cervical spine in pa- 68. Neifeld GL, Keene JG, Hevesy G, et al: Cervical injury in tients with blunt trauma. National Emergency X-Radiography head trauma. J Emerg Med 6:203–207, 1988 Utilization Study Group. N Engl J Med 343:94–99, 2000 69. Nguyen GK, Clark R: Adequacy of plain radiography in the 48. Hogan GJ, Mirvis SE, Shanmuganathan K, et al: Exclusion of diagnosis of cervical spine injuries. Emerg Radiol 11:158– unstable cervical spine injury in obtunded patients with blunt 161, 2005 trauma: is MR imaging needed when multi-detector row CT 70. Padayachee L, Cooper DJ, Irons S, et al: Cervical spine clear- findings are normal? Radiology 237:106–113, 2005 ance in unconscious traumatic brain injury patients: dynamic 49. Hollingworth W, Relyea-Chew A, Comstock BA, et al: The flexion-extension fluoroscopy versus computed tomography risk of bankruptcy before and after brain or spinal cord injury: with three-dimensional reconstruction. J Trauma 60:341– a glimpse of the iceberg’s tip. Med Care 45:702–711, 2007 345, 2006 50. Holly LT, Kelly DF, Counelis GJ, et al: Cervical spine trauma 71. Piatt JH Jr: Detected and overlooked cervical spine injury associated with moderate and severe head injury: incidence, in comatose victims of trauma: report from the Pennsylva- risk factors, and injury characteristics. J Neurosurg 96:285– nia Trauma Outcomes Study. J Neurosurg Spine 5:210–216, 291, 2002 2006 51. Iida H, Tachibana S, Kitahara T, et al: Association of head 72. Poonnoose PM, Ravichandran G, McClelland MR: Missed trauma with cervical spine injury, spinal cord injury, or both. and mismanaged injuries of the spinal cord. J Trauma J Trauma 46:450–452, 1999 53:314–320, 2002 52. Insko EK, Gracias VH, Gupta R, et al: Utility of flexion and 73. Prasad VS, Schwartz A, Bhutani R, et al: Characteristics of extension radiographs of the cervical spine in the acute evalu- injuries to the cervical spine and spinal cord in polytrauma ation of blunt trauma. J Trauma 53:426–429, 2002 patient population: experience from a regional trauma unit. 53. Jelly LM, Evans DR, Easty MJ, et al: Radiography versus spi- Spinal Cord 37:560–568, 1999 ral CT in the evaluation of cervicothoracic junction injuries in 74. Ptak T, Kihiczak D, Lawrason JN, et al: Screening for cervi- polytrauma patients who have undergone intubation. Radio- cal spine trauma with helical CT: experience with 676 cases. graphics 20:S251-252, 2000 Emerg Radiol 8:315–319, 2001 54. Kihiczak D, Novelline RA, Lawrason JN, et al: Should an MR 75. Ravichandran G, Silver JR: Missed injuries of the spinal cord. scan be performed routinely after a normal clearance CT scan Br Med J (Clin Res Ed) 284:953–956, 1982 in the trauma patient? Experience with 59 cases. Emerg Ra- 76. Reid DC, Henderson R, Saboe L, et al: Etiology and clini- diol 8:276–278, 2001 cal course of missed spine fractures. J Trauma 27:980–986, 55. Kreipke DL, Gillespie KR, McCarthy MC, et al: Reliability 1987 of indications for cervical spine films in trauma patients. J 77. Richards PJ, Summerfield R, George J, et al: Major trauma & Trauma 29:1438–1439, 1989 cervical clearance radiation doses & cancer induction. Injury 56. Lee HJ, Sharma V, Shah K, et al: The role of spiral CT vs 39:347–356, 2008 plain films in acute cervical spine trauma: a comparative 78. Roberge RJ, Wears RC: Evaluation of neck discomfort, neck study. Emerg Radiol 8:311–314, 2001 tenderness, and neurologic deficits as indicators for radiog- 57. MacDonald RL, Schwartz ML, Mirich D, et al: Diagnosis raphy in blunt trauma victims. J Emerg Med 10:539–544, of cervical spine injury in motor vehicle crash victims: how 1992 many x-rays are enough? J Trauma 30:392–397, 1990 79. Roberge RJ, Wears RC, Kelly M, et al: Selective application of 58. Mathen R, Inaba K, Munera F, et al: Prospective evaluation cervical spine radiography in alert victims of blunt trauma: a of multislice computed tomography versus plain radiograph- prospective study. J Trauma 28:784–788, 1988 ic cervical spine clearance in trauma patients. J Trauma 80. Ross SE, O’Malley KF, DeLong WG, et al: Clinical predictors 62:1427–1431, 2007 of unstable cervical spinal injury in multiply injured patients. 59. McCulloch PT, France J, Jones DL, et al: Helical computed Injury 23:317–319, 1992 tomography alone compared with plain radiographs with ad- 81. Roth BJ, Martin RR, Foley K, et al: Roentgenographic evalu- junct computed tomography to evaluate the cervical spine ation of the cervical spine. A selective approach. Arch Surg after high-energy trauma. J Bone Joint Surg Am 87:2388– 129:643–645, 1994 2394, 2005 82. Sanchez B, Waxman K, Jones T, et al: Cervical spine clear- 60. McGee S: Simplifying likelihood ratios. J Gen Intern Med ance in blunt trauma: evaluation of a computed tomography- 17:646–649, 2002 based protocol. J Trauma 59:179–183, 2005 61. McKinley W, Santos K, Meade M, et al: Incidence and out- 83. Sarani B, Waring S, Sonnad S, et al: Magnetic resonance im- comes of spinal cord injury clinical syndromes. J Spinal aging is a useful adjunct in the evaluation of the cervical spine Cord Med 30:215–224, 2007 of injured patients. J Trauma 63:637–640, 2007 62. McNamara RM, Heine E, Esposito B: Cervical spine injury 84. Schenarts PJ, Diaz J, Kaiser C, et al: Prospective comparison and radiography in alert, high-risk patients. J Emerg Med of admission computed tomographic scan and plain films of 8:177–182, 1990 the upper cervical spine in trauma patients with altered men- 63. McNamara RM, O’Brien MC, Davidheiser S: Post-traumatic tal status. J Trauma 51:663–669, 2001 neck pain: a prospective and follow-up study. Ann Emerg 85. Schuster R, Waxman K, Sanchez B, et al: Magnetic resonance Med 17:906–911, 1988 imaging is not needed to clear cervical spines in blunt trauma 64. Michael DB, Guyot DR, Darmody WR: Coincidence of head patients with normal computed tomographic results and no and cervical spine injury. J Neurotrauma 6:177–189, 1989 motor deficits. Arch Surg 140:762–766, 2005 65. Molina DK, Nichols JJ, Dimaio VJ: The sensitivity of com- 86. Sees DW, Rodriguez Cruz LR, Flaherty SF, et al: The use of puted tomography (CT) scans in detecting trauma: are CT bedside fluoroscopy to evaluate the cervical spine in obtunded scans reliable enough for courtroom testimony? J Trauma trauma patients. J Trauma 45:768–771, 1998 63:625–629, 2007 87. Sharma OP, Oswanski MF, Yazdi JS, et al: Assessment for

8 Neurosurg. Focus / Volume 25 / November 2008

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC Prevalence of cervical spine injury in trauma

additional spinal trauma in patients with cervical spine injury. 95. Widder S, Doig C, Burrowes P, et al: Prospective evaluation Am Surg 73:70–74, 2007 of computed tomographic scanning for the spinal clearance 88. Sliker CW, Mirvis SE, Shanmuganathan K: Assessing cervi- of obtunded trauma patients: preliminary results. J Trauma cal spine stability in obtunded blunt trauma patients: review of 56:1179–1184, 2004 medical literature. Radiology 234:733–739, 2005 89. Spiteri V, Kotnis R, Singh P, et al: Cervical dynamic screening 96. Williams J, Jehle D, Cottington E, et al: Head, facial, and in spinal clearance: now redundant. J Trauma 61:1171–1177, clavicular trauma as a predictor of cervical-spine injury. Ann 2006 Emerg Med 21:719–722, 1992 90. Stassen NA, Williams VA, Gestring ML, et al: Magnetic 97. Yanar H, Demetriades D, Hadjizacharia P, et al: Pedestrians resonance imaging in combination with helical computed injured by automobiles: risk factors for cervical spine injuries. tomography provides a safe and efficient method of cervical J Am Coll Surg 205:794–799, 2007 spine clearance in the obtunded trauma patient. J Trauma 98. Zabel DD, Tinkoff G, Wittenborn W, et al: Adequacy and ef- 60:171–177, 2006 91. Stelfox HT, Velmahos GC, Gettings E, et al: Computed to- ficacy of lateral cervical spine radiography in alert, high-risk mography for early and safe discontinuation of cervical spine blunt trauma patient. J Trauma 43:952–958, 1997 immobilization in obtunded multiply injured patients. J Trauma 63:630–636, 2007 92. Stiell IG, Clement CM, McKnight RD, et al: The Canadian C- spine rule versus the NEXUS low-risk criteria in patients with Manuscript submitted July 14, 2008. trauma. N Engl J Med 349:2510–2518, 2003 Accepted July 29, 2008. 93. Stiell IG, Wells GA, Vandemheen KL, et al: The Canadian C- spine rule for radiography in alert and stable trauma patients. Address correspondence to: Sherman C. Stein, M.D., Hospital JAMA 286:1841–1848, 2001 of the University of Pennsylvania, Department of Neurosurgery, 3 94. Waydhas C: Intrahospital transport of critically ill patients. Silverstein, 3400 Spruce Street, Philadelphia, Pennsylvania 19104. Crit Care 3:R83–R89, 1999 email: [email protected].

Neurosurg. Focus / Volume 25 / November 2008 9

Unauthenticated | Downloaded 10/09/21 07:27 AM UTC