Radiography
Volume 17, Issue 1 , Pages 39-42, February 2011

Cervical spine imaging in trauma: Does the use of grid and filter combination improve visualisation of the cervicothoracic junction?

University Hospital of Wales, Heath Park, Cardiff, CF14 4XW, UK

Received 24 July 2009; received in revised form 24 April 2010; accepted 30 April 2010. published online 07 June 2010.

Article Outline

Abstract 

Purpose

To evaluate the usefulness of filter and anti-scatter grid combination in demonstrating the cervicothoracic junction in lateral cervical spine radiographs performed for trauma patients.

Methods

Following a change in departmental protocol in our hospital, anti-scatter grid and filter are routinely used for lateral cervical spine radiograph in all trauma patients with immobilised cervical spine. A retrospective study was done to compare the efficacy of lateral cervical spine radiographs in demonstrating the cervicothoracic junction for a period of three months before and after the implementation of the change. All images were independently evaluated by two observers.

Results

253 trauma patients had a lateral cervical spine radiograph done in January to March 2003 without using the anti-scatter grid and filter while 309 patients in January to March 2007, using filter and grid. Inter-observer variability between the two observers was calculated using Cohen’s Kappa which showed good and very good agreement for 2003 and 2007 respectively. 126 (49.8%) images adequately demonstrated the cervicothoracic junction without using filter and grid while 189 (61.1%) were adequate following their use. This was statistically significant (Fischer exact test, p value = 0.0081).

Conclusion

The use of filter and anti-scatter grids improves the visualisation of cervicothoracic junction in lateral cervical spine imaging and reduces the need to repeat exposure.

Keywords: Radiography, Trauma, Spine

 

Back to Article Outline

Introduction 

Up to six percent of major trauma patients suffer cervical spine injuries.1, 2 Such injuries may cause significant morbidity and mortality and appropriate diagnosis is therefore important to prevent complications. A three view cervical spine series, including anteroposterior, lateral and dedicated odontoid projections are widely accepted as standard.2, 3, 4 Adequate visualisation of the cervicothoracic junction in the lateral cervical spine film is also considered essential.5 Between 9% and 18% of all Cervical spine injuries occur at the cervicothoracic junction.5, 6 It has been suggested that the use of anti-scatter grids improves visualisation of the cervicothoracic junction. We present a study evaluating the role of grids and filter combination in lateral cervical spine radiographs. We discuss other methods suggested to improve the visualisation of cervicothoracic junction and increasing role of CT scan in cervical spine injuries.

Back to Article Outline

Materials and methods 

Our departmental policy was changed in 2004 and use of anti-scatter grid and filter combination was adopted to help improve the visualisation of the cervicothoracic junction and decrease the number of repeat radiographs. We studied the radiographs of all patients presenting to the accident and emergency department with possible C-spine injury from January to March 2003, when they were performed without using grid and filter, and similarly in corresponding months of 2007 when this combination was used. The patients were identified using the departmental Radiology Information System (RIS) database and all trauma patients in the study period who presented to accident and emergency department and had a Cervical spine radiograph were selected.

Agfa CR Impax system was used in both study periods for obtaining radiographs. A 24 × 30 cm parallel grid with 177 lines/inch (grid ratio 8:1) was used. The filter used was a Ferlic filter (Rothband, UK). The source to image distance (SID) was 180 cm. All images were viewed on the Agfa PACS station with a resolution of 2 mega pixels with dynamic windowing and magnification.

An adequate lateral C-spine radiograph has been defined as an unrotated view demonstrating the spine from the occiput to the upper border of T1.4 However in our study, we assessed only the visualisation of the cervicothoracic junction and if there was doubt regarding its visualisation, the radiographs were classified as inadequate. All radiographs were assessed for adequacy independently by two radiologists: a musculoskeletal radiology consultant and a specialist registrar. All demographic data was removed using the demographic toggle function of the PACS and thus the observer was blinded to whether the radiographs were from 2003 or 2007. The data was analysed using Microsoft Excel XP and SPSS version 12 for Windows 2000 software.

Back to Article Outline

Results 

Some 253 patients had a cervical spine radiograph series performed following presentation with trauma in the period of January to March 2003 and 309 patients in 2007. Inter-observer variability between the two observers was calculated using Cohen’s Kappa for both 2003 and 2007 (Fig. 1). The degrees of concordance were assessed using the scale proposed by Landis and Koch.7 The calculated Kappa value for 2003 was 0.763 and for 2007 was 0.812. These values indicate good and very good agreement for 2003 and 2007 respectively between the two observers. Further results were analysed using the findings of the more senior observer 1. In 2003, 126 lateral Cervical spine radiographs (49.8%) were adequate. The number of adequate films in 2007 was 189 (61.1%). This difference was statistically significant (Fischer exact test, p value = 0.0081) (Fig. 2).

Of the 127 inadequate lateral cervical spine images obtained without using grids in 2003, 109 had a repeat film (33 lateral view, 60 swimmer’s view and 16 had both).With the use of grids and filters, there were 120 inadequate films and 113 of these were repeated (31 lateral, 53 swimmer’s, 11 both and 18 had CT scan of the Cervical spine). None of the suspected cervical spine injury patients in 2003 study group had a CT scan performed.

In 2003, 34 patients had further repeat films after an initial repeat film, while only 18 had further repeat radiographs taken with grid in 2007. This was statistically significant (p value = 0.044). However these results could be skewed because of use of CT scans in 2007 patient group following failure of initial cervical spine image to demonstrate the C7/T1 interface.

Back to Article Outline

Discussion 

The National Emergency X-Radiography Utilization Study (NEXUS) and the Canadian C-spine rule for radiography in alert and stable trauma cases are two widely used criteria in deciding which patients undergo imaging to rule out cervical spine injury.8, 9 An initial screening radiograph series usually includes a lateral, anterior-posterior and odontoid views of the cervical spine.3, 10

Missed cervical spine injuries are associated with varying degrees of neurological impairment. Up to 29% of all the patients with delayed diagnosis of cervical spine injury may develop permanent neurological deficit.11 The common reasons for this were suboptimal views, incorrect interpretation and lack of radiographic signs.11, 12 The legal and monetary implications of missed cervical spine injury leading to permanent neurological deficit, has led to liberal use of imaging to rule out such injuries.13 Such cases have been associated with some of the highest injury awards and monetary settlements in medical malpractice litigation.14

Various techniques have been used to adequately visualise the cervicothoracic junction. These include applying downward traction on the arms to pull the shoulders out of the view to gain exposure of the lower cervical vertebrae.4 Ohiorenoya et al15 showed that unless an initial cervical spine radiograph includes the upper one third of the body of the C7 vertebra, the probability of attaining the C7/T1 level with arm traction is less than 15%. Alternative radiographs of the cervical spine such as a swimmer’s or supine oblique views may be necessary in an attempt to adequately visualise the cervical spine.16, 17, 18, 19 The swimmer’s view may not be possible in patients with upper extremity injuries because upper extremity trauma may prohibit the positioning of the patient’s arms for this view. A subluxation of the unstable cervical spine injury during positioning for swimmer’s view has also been reported.20 Ireland et al18 concluded that when initial three view cervical spine series fails to demonstrate the cervicothoracic junction, swimmer’s and supine oblique views show the alignment of the vertebral bodies with equal frequency. However, supine oblique films are safer, expose patients to less radiation, and are more often successful in demonstrating posterior elements. But, radiologists and emergency department staff are less familiar with these views, which make interpretation difficult.

Anti-scatter grids and filters can be used to improve the visualisation of the cervicothoracic junction on the initial radiograph; however their use is often at radiographer’s discretion.21 Anti-scatter grid consists of lead strips separated by bands of a low attenuation material and are positioned between the image receptor and patient. The larger surface area permits modern grids to reduce scatter by up to 90%.22 This improves the signal to noise ratio and thus the image quality. Filters are designed to compensate for the change in densities between adjacent body areas. Grid and filter combination requires a higher mAs and hence the dose area product (DAP) meter reading is bound to be higher. Roberts et al23 calculated the effective dose (ED) in their series and found a small increase in the ED following use of grids in lateral cervical spine radiographs was reported. Their largest calculated mean ED was 3.51 μSv which is less than one day of background radiation in the UK.24

No study has been reported to assess the efficacy of their use. In their paper, Roberts et al23 were unable to recommend clearly whether grids were needed or not in imaging Cervical spine due to reduced inter-rater reliability. One of the main areas of disagreement between the two radiologists was DIMOND criterion F25 which states that the upper border of 1st thoracic vertebra must be visualised.

In our study, the only significant change over the study periods has been the use of filters and grids. Imaging was undertaken by qualified radiographers trained in the appropriate use of filters and grids. The same X-ray tube was used in both groups to obtain the radiographs. We acknowledge that the grids used might have improved between the two study periods. But, this again would have contributed to improved image quality. Though the software used to view the images changed between the two study groups, the current PACS software was used to assess all the radiographs during our study, thus allowing uniformity in the interpretation of the images. We have shown that the use of anti-scatter grids significantly increases the diagnostic quality of the lateral cervical spine images (Fig. 3).

  • View full-size image.
  • Figure 3 

    Lateral X Ray of the Cervical Spine of the same patient performed a) without using filter and grid and b) with use of grid and filter combination (the second radiograph was performed to visualise the C7/T1 junction and therefore upper cervical spine was not included).

A CT scan of the cervical spine is another imaging modality that may be used to clear the cervical spine in difficult cases. Multislice CT scans with multiplanar reformats provide excellent quality images and allow visualisation of the cervical spine in all planes. In the UK, guidelines for management of head injury patients were issued in 20032 by the National Institute of Clinical Excellence suggesting that CT scans should include the cervical spine where it is not possible to achieve the desired cervical spine views with a radiograph. CT is also indicated if the plain film series is technically inadequate, suspicious or definitely abnormal or if there is continued clinical suspicion of injury despite a normal study. CT imaging of the cervical spine should be considered if the patient is having other body areas scanned for head injury/multi-region trauma, and a definitive diagnosis of cervical spine injury is required urgently. CT scan is likely to replace further views where availability permits as it is quick and gives visualisation in almost 100% of patients.26

This was a retrospective study and therefore one of the limitations was that we were assessing the end product available to us on the PACS. It has been departmental policy during both study periods to archive all radiographs, even if they were inadequate. Whether this was adhered to was difficult to gauge and we may have missed some of the rejected inadequate views that were not archived. Perhaps with the current software, it is possible to track all radiographs, even those not archived. This would enable a better assessment of the true value of grids and filters in imaging the cervical spine. However, even if this was the case it would affect both study periods and therefore not likely to affect the results. Also we did not calculate the increased radiation burden to the patient as a consequence of using filters and grids. Certain other factors such as patient’s BMI, shoulder position and severity of injuries may have had an impact on cervical spine radiographs but were not looked into in our study.

Back to Article Outline

Conclusion 

The use of anti-scatter grid and filter combination provides a definite and significant advantage in visualisation of cervicothoracic junction, thus reducing the need for repeat views.

Back to Article Outline

References 

  1. Crim JR, Moore K, Brodke D. Clearance of the cervical spine in multitrauma patients: the role of advanced imaging. Semin Ultrasound CT MR. 2001;22:283–305
  2. National Collaborating Centre for Acute Care . Guideline commissioned by the National Institute for Clinical Excellence. Head injury: triage, assessment, investigation and early management of head injury in infants, children and adults. In: National Institute for clinical Excellence. Jun 2003;
  3. Ross SE, Schwab CW, David ET, et al. Clearing the cervical spine: initial radiologic evaluation. J Trauma. 1987;27:1055–1060
  4. Richards PJ. Cervical spine clearance: a review. Injury. 2005;36:248–269discussion 70
  5. Nichols CG, Young DH, Schiller WR. Evaluation of cervicothoracic junction injury. Ann Emerg Med. 1987;16:640–642
  6. Miller MD, Gehweiler JA, Martinez S, et al. Significant new observations on cervical spine trauma. AJR Am J Roentgenol. 1978;130:659–663
  7. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics. 1977;33:363–374
  8. Hoffman JR, Mower WR, Wolfson AB, et al. National Emergency X-Radiography Utilization Study Group Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. N Engl J Med. 2000;343:94–99
  9. Stiell IG, Wells GA, Vandemheen KL, et al. The Canadian C-spine rule for radiography in alert and stable trauma patients. JAMA. 2001;286:1841–1848
  10. American College of Surgeons . Committed on trauma. Advanced trauma life support instructor manual. Chicago, IL: American College of Surgeons; 1997;
  11. Davis JW, Phreaner DL, Hoyt DB, et al. The etiology of missed cervical spine injuries. J Trauma. 1993;34:342–346
  12. Tins BJ, Cassar-Pullicino VN. Imaging of acute cervical spine injuries: review and outlook. Clin Radiol. 2004;59:865–880
  13. Berlin L. CT versus radiography for initial evaluation of cervical spine trauma: what is the standard of care?. AJR Am J Roentgenol. 2003;180:911–915
  14. Brennan TA, Sox CM, Burstin HR. Relation between negligent adverse events and the outcomes of medical-malpractice litigation. N Engl J Med. 1996;335:1963–1967
  15. Ohiorenoya D, Hilton M, Oakland CD, et al. Cervical spine imaging in trauma patients: a simple scheme of rationalising arm traction using zonal divisions of the vertebral bodies. J Accid Emerg Med. 1996;13:175–176
  16. MacDonald RL, Schwartz ML, Mirich D, et al. Diagnosis of cervical spine injury in motor vehicle crash victims: how many X-rays are enough?. J Trauma. 1990;30:392–397
  17. Woodring JH, Lee C. Limitations of cervical radiography in the evaluation of acute cervical trauma. J Trauma. 1993;34:32–39
  18. Ireland AJ, Britton I, Forrester AW. Do supine oblique views provide better imaging of the cervicothoracic junction than swimmer’s views?. J Accid Emerg Med. 1998;15:151–154
  19. Jenkins MG, Curran P, Rocke LG. Where do we go after the three standard cervical spine views in the conscious trauma patient? A survey. Eur J Emerg Med. 1999;6:215–217
  20. Davis JW. Cervical injuries-perils of the swimmer’s view: case report. J Trauma. 1989;29:891–893
  21. Bell N, Erskine M, Warren-Forward H. Lateral cervical spine examinations: an evaluation of dose for grid and non-grid techniques. Radiography. 2003;9:43–52
  22. Dendy PP, Heaton B. Physics for diagnostic radiology. 2nd ed.. Bristol: Bristol Institute of Physics Publishing; 1999;
  23. Roberts JA, Evans SC, Rees M. Optimisation of imaging technique used in direct digital radiography. J Radiol Prot. 2006;26:287–299
  24. NRPB (National Radiological Protection Board) . Living with radiation. 5th ed.. Chilton: Chilton: NRPB; 1998;
  25. Busch HP. Diagnostic requirements for digital projection radiography image quality and dose management for digital radiography chapter III (DIMOND III), 2004.
  26. Jelly LM, Evans DR, Easty MJ, et al. Radiography versus spiral CT in the evaluation of cervicothoracic junction injuries in polytrauma patients who have undergone intubation. Radiographics. 2000;20:S251–S259discussion S60–2

PII: S1078-8174(10)00049-0

doi:10.1016/j.radi.2010.04.005

Radiography
Volume 17, Issue 1 , Pages 39-42, February 2011

Article Tools

* Image Usage & Resolution