Article Text
Abstract
Aim To determine the potential of the Canadian Cervical Spine Rule (CCR) to safely reduce the number of cervical spine (c-spine) radiographs performed in the UK emergency department setting.
Methods The study was conducted in two UK emergency departments with a combined annual attendance of >150 000 adult patients. Over the 24 month trial period, 148 doctors were provided training in the use of the CCR and instructed to assess eligible patients presenting with potential c-spine injury. Doctors were instructed to manage patients according to existing practice and not according to the decision obtained from the rule. A subsample of patients was reassessed by a second doctor to test interobserver reliability.
Results A total of 1420 patients were enrolled in the study (50.4% male). 987 (69.5%) had c-spine radiography performed, with 8 (0.6%) having a c-spine injury. If the decision for radiography had been made according to the outcome of the CCR, only 815 (57.4%) would have had c-spine radiography and all 8 abnormal cases would have undergone imaging. Doctors were comfortable using the rule in 91% of cases. Interobserver reliability was good (κ=0.75 95% CI 0.44 to 1.06).
Conclusion The CCR can be applied successfully in the UK. Had the CCR been in use during the study period, a 17.4% reduction in radiography could have been achieved without compromising patient care.
- Cervical-spine injuries
- medical-decision-making
- clinical-protocol
- cohort-study
- imaging, x-ray
- paramedics
- guidelines
- research
- cost effectivenss
- trauma
- spine and pelvis
- trauma
Statistics from Altmetric.com
- Cervical-spine injuries
- medical-decision-making
- clinical-protocol
- cohort-study
- imaging, x-ray
- paramedics
- guidelines
- research
- cost effectivenss
- trauma
- spine and pelvis
- trauma
Introduction
Blunt trauma affecting the neck is a common presentation to emergency departments (EDs) in the UK. The incidence of fracture or cervical spine (c-spine) injury is very low, particularly in alert patients with intact neurological status. Despite this, UK clinicians refer a majority of patients with neck injuries for radiography ‘just in case’.1 This overly cautious approach is understandable in view of the potentially devastating consequences of missing a spinal injury. Large numbers of unnecessary c-spine radiographs add significantly to National Health Service (NHS) health costs and create unnecessary work for ED staff. They increase patient time in the ED and lead to prolonged and often unnecessary immobilisation. Overdependence on investigations may also block the development of clinical judgement and compromise rather than improve patient care.2 3
In the past, the lack of specific evidence-based guidelines for radiography referral contributed to considerable variation in practice.4 The NEXUS rule (box 1) and the Canadian Cervical Spine Rule (CCR) (figure 1) are two clinical decision rules that have been developed to permit more selective and consistent ordering of c-spine radiography.5 6
The NEXUS low-risk criteria5
Cervical-spine radiography is indicated for patients with neck trauma unless they meet all the following criteria;
No posterior midline cervical –spine tenderness
No evidence of intoxication
A normal level of alertness
No focal neurological deficit
No painful distracting injuries
Both rules have been shown to decrease the use of radiography without adversely affecting patient outcomes, with the CCR displaying greater sensitivity and specificity.7 Despite evidence of its efficacy in the North American setting, the CCR is still not universally known or applied in the UK. The CCR was developed by the team that produced the Ottawa Ankle and Knee Rules8 9 and was validated within the Canadian emergency healthcare system in a large multicentre study.6 It is recognised that clinical decision rules should be validated with different cohorts of patients in different healthcare systems.10 Prior to this study there has been no large-scale validation of the CCR in the UK.
The aim of this study was to determine the potential of the CCR to reduce unnecessary referrals for radiography of patients presenting with acute blunt neck trauma.
Methods
Study design/setting
This prospective observational study was conducted over 2 years in the EDs of two UK hospitals with a combined annual attendance of >150 000 adult patients.
Medical staff
A total of 148 doctors of all grades from both sites who agreed to take part in the study were given a training session where the CCR was introduced and explained. The proportions from each grade who recruited patients into the study are given in table 1.
Inclusion and exclusion criteria
The inclusion and exclusion criteria for participation in the study are outlined in box 2.
Inclusion and exclusion criteria
Inclusion criteria
Neck pain following acute blunt trauma to the head and/or neck
No neck pain, non ambulatory and evidence of injury above the clavicle
Alert and stable (Glasgow Coma Score >15 with normal vital signs)
Aged >16 years
Injury sustained within the previous 48 h
Exclusion criteria
Patients <16 years
No trauma to head and neck
Ambulatory patients with no neck pain, minor head/facial injury and a low risk mechanism
Major trauma
Glasgow Coma Score <15
Injury occurred >48 h previously
Penetrating trauma
Acute paralysis/paresis
Vertebral disease
Returned for assessment
Pregnant
Procedure
Once agreement to participate had been given, the decision rule algorithm was appended to the recruited patient's notes by the triage nurse. Doctors were asked to record their findings and their judgement according to the CCR after each assessment of an eligible patient. They were also asked to rate their level of confidence in the rule. During the study, doctors were instructed to order radiographs as they normally would, irrespective of the decision rule.
All radiographs were subsequently reported by a radiologist. Patients who did not undergo radiography were followed up by telephone by a study nurse using a validated proxy outcome tool.11 They were recalled for re-assessment by an emergency medicine consultant if any of the following were present: (1) ‘moderate’ or ‘severe’ neck pain; (2) ‘moderate’ or ‘severe’ restriction of neck movement; (3) ongoing use of a neck collar; or (4) the neck injury had prevented a return to their usual preaccident activity. If the re-assessment suggested the possibility of a significant cervical injury, further imaging was performed.
Results
A total of 1420 patients were enrolled in the study, 716 male (50.4%) and 704 female. The recruitment of patients is detailed in figure 2. The proportion of males and the proportion of serious c-spine injuries were similar for the enrolled and non-enrolled patients (male 50.4% vs 46.9%, 95% CI for the difference −0.2% to 7.2%; proportion of clinically important injury 0.6% vs 0.5%, 95% CI for the difference −0.4% to 0.6%; eligible but not enrolled n=1375). The mechanisms of injury are detailed in table 2.
C-spine radiography was performed on 987 patients (69.5%), with 8 (0.6%) having a c-spine injury (5 vertebral fractures and 3 fracture dislocations). If the decision for radiography had been made according to the outcome of the CCR, 815 (57.4%) would have had c-spine radiography and no spinal injuries would have been missed. In 202 cases, doctors did not evaluate range of motion as required by the CCR. These patients were classified as ‘indeterminate’. The characteristics of these patients were similar to those of the other enrolled patients.
The sensitivity of the CCR (excluding the indeterminates) was 100% (95% CI 56% to 100%) and the specificity was 43% (95% CI 39% to 45%). A secondary analysis was performed to include the indeterminate cases. When all the indeterminate cases were assumed to be negative for the CCR, the sensitivity remained 100% (95% CI 56% to 100%) with a specificity of 36% (95% CI 34% to 39%). When all the indeterminate cases were assumed to be positive the sensitivity remained 100% (95% CI 56% to 100%) with a specificity of 50% (95% CI 48% to 53%).
Most of the doctors reported being ‘very comfortable’ or ‘comfortable’ using the rule (81.9%). Interobserver reliability was 0.75 (unweighted κ, n=16), which represents a ‘good’ level of agreement.
Discussion
This study is the first large-scale validation study of the CCR in the UK ED setting. The results demonstrate that the rule could be applied effectively and with confidence by the majority of doctors who participated in the study. If the CCR had been used as a decision aid during the study period it would have resulted in a reduction of radiographs by 17.4% without compromising patient safety.
The high sensitivity (100%), reasonable specificity (43%) and good reliability are similar to results obtained in other validation and implementation studies.6 12 The difference in the proportion of cervical injuries in this study compared with the original Canadian validation study6 suggests a significantly different patient population at the two UK sites (0.06% vs 2%; 95% CI for the difference 0.08% to 1.9%). This gives further credence to the necessity for local validation.
Although the use of evidence-based ED clinical decision rules such as the CCR is increasing, they are still not consistently used in the UK. Our experience has been that most junior doctors are not familiar with the CCR despite the fact that it was developed nearly a decade ago. Its influence on the National Institute for Health and Clinical Excellence (NICE) guidelines for neck and head injuries13 has undoubtedly increased knowledge of elements of the rule. Despite this, neck radiographs continue to be ordered unnecessarily. This inappropriate use of radiography cannot be justified when there are clinical decision rules available to assist clinicians to make more informed decisions.14 We believe that senior emergency medicine clinicians have a responsibility to disseminate the CCR and similar evidence-based decision rules to their juniors, medical students and to peers in other specialities.
The CCR has been demonstrated to be safe in practice.6 7 The majority of doctors in this study were ‘very comfortable’ or ‘comfortable’ using this rule. Nevertheless a number of doctors (9.02%) expressed concern regarding the confidence they had in the outcome. Further research is required to explore the nature of these concerns. Research is also needed to determine other barriers to the acculturation of clinical decision rules such as the CCR, so that these barriers can be overcome.12 15 A recent trial in Canadian EDs evaluated the effectiveness of active strategies to implement the CCR.15 These strategies included education, policy and real-time reminders on radiology requests. Imaging rates were lower at intervention sites than at control sites where no specific interventions were introduced. The authors suggested that issues relating to doctors' compliance and misinterpretation of the rule identified in the study could be addressed by training which emphasised in particular the correct interpretation of high and low risk criteria (figure 2).
Study limitations
There are a number of limitations associated with this study. Although a large number of patients were recruited, the number of c-spine injuries was small and not sufficient to validate the rule statistically in this setting. The results, however, add considerable weight to the existing body of evidence demonstrating the value, safety and applicability of the CCR. Although doctors were instructed to follow their usual practice in ordering radiographs, it is possible that the CCR recommendation could, on occasion, have influenced their decision. Although ‘good’ interobserver reliability was found relating to the rule, the sample size for interobserver reliability in this study was small.
Conclusion
This study shows that the CCR can be applied accurately by emergency doctors in the UK setting. During the 2 year study period, the use of the rule in two UK EDs would have reduced unnecessary radiography by 172 cases out of 987 (a reduction of 17.4% (95% CI 15% to 19%) without adversely affecting patient care. Further research is required to determine factors that cause doctors to lack confidence in applying the rule in clinical practice.
Acknowledgments
We would like to thank the doctors and nurses from the two study sites, The Queen's Medical Centre, Nottingham and the Derbyshire Royal Infirmary, for their cooperation with the study.
References
Footnotes
Funding The study was partially funded by the Special Trustees Fund of the University Hospital Nottingham.
Competing interests None.
Ethics approval This study was conducted with the approval of the Nottingham Research Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Primary survey