Relationship of Mechanism of Injury to Asymptomatic Cervical Spine Fractures in the Elderly
===========================================================================================

* Christopher Lawrence
* Mahla Radmard
* Armin Tafazolimoghadam
* Akua Afrah Amoah
* Dhairya A. Lakhani
* Javad Azadi
* Arjun Chanmugam
* David M. Yousem

## SUMMARY:

The Canadian Cervical Spine Rule, a clinical decision-making tool for patients post-trauma, is often interpreted as recommending cervical spine CT in patients ≥65 years old, who sustain a dangerous mechanism of injury, and/or have extremity paresthesias. We retrospectively reviewed 6 years’ of emergency department cervical spine CT reports to determine fracture rates in patients ≥65, symptomatic or not, who did and did not have a dangerous mechanism. Of those ≥65 years old, 240 of 13,925 (1.72%) patients had cervical spine fractures. The fracture rate in asymptomatic patients ≥65 was 0.27%. The fracture rate in asymptomatic patients ≥65, who did not have a dangerous mechanism of injury was 0.15%. The rate of unstable fractures requiring surgery was 0.007%. The findings suggest that the algorithm to scan asymptomatic patients ≥65, and/or those ≥65 without a dangerous injury mechanism, should be revisited for appropriateness and overall value.

## ABBREVIATIONS:

CCR
:   Canadian Cervical Spine Rule
CSCT
:   cervical spine CT
ED
:   emergency department
EMR
:   electronic medical record
EMS
:   emergency medical service
MVC
:   motor vehicle collision

The Canadian Cervical Spine Rule (CCR) is a clinical decision-making tool designed to assist health care professionals in evaluating the need for cervical spine imaging in patients who have experienced trauma.1 The CCR is often interpreted as recommending cervical spine imaging after trauma in patients 1) ≥65 years old, 2) who sustain a dangerous mechanism of injury, and/or 3) who have extremity paresthesias.2 Dangerous mechanisms, as defined by the CCR, refer to high-risk events that significantly increase the likelihood of cervical spine injury.

The CCR seminal article, published more than 20 years ago, describes a dangerous mechanism as: a fall ≥1 m/5 stairs, axial load to head, high-speed motor vehicle collision (MVC) >100 kph, rollover or ejection, incident with motorized recreational vehicle, or bicycle collision.1 The authors explored the relationship between traumatic event mechanism and clinically important cervical spine injury through univariable logistic regression analysis and performed supplemental statistical modeling accounting for additional clinical factors such as age to create the final CCR algorithm. MVCs represented the most common mechanism of injury prompting evaluation for cervical spine injury for 67% of all patients. Of these patients in MVCs, there was an even higher likelihood of injury with higher speed collisions, rollovers and ejections, leading to these factors being included as high-risk “dangerous mechanisms.” In isolation, dangerous mechanisms were associated with an odds ratio of 5.2 for a clinically important cervical spine injury, thus, their inclusion into the CCR criteria.

We have previously reported on the low rate of cervical spine fractures in asymptomatic patients ≥65 years of age (28/9455; 0.30%) and suggested more refined decision support criteria for imaging in this select population.3 At that time, we did not look at the other CCR criterion of dangerous mechanism to see how many of the 28 had a dangerous mechanism of trauma that would dictate imaging by the CCR rule. To further clarify the appropriate use for cervical spine CT (CSCT) in trauma patients, we included an additional year of data from our emergency departments (EDs) and measured the rate of fractures in patients ≥65 years of age who were asymptomatic and did or did not have a dangerous mechanism of injury.

## METHODOLOGY

This study was a retrospective, Health Insurance Portability and Accountability Act–compliant project approved by our institutional review board (IRB00392389), with informed consent waived. We retrieved reports of noncontrast CSCT images from the enterprise-level PACS for all patients scanned at the 2 EDs (academic tertiary care and academic community-based) in our health system from July 1, 2018, to May 31, 2024. The rate of fractures was previously reported for a sample obtained from July 1, 2018 to June 30, 2023,3 but another year’s worth of CSCT has been added to the cohort reported herein.

The electronic medical records (EMRs) of all patients whose CSCT reports documented fractures, ligamentous injuries, and/or spinal canal hemorrhage were reviewed to determine the patients’ age, sex, clinical symptoms, physical examination findings, and mechanism of injury, separated into dangerous and nondangerous similar to the definitions provided by the CCR. While the CCR specified dangerous mechanisms as a fall from an elevation of 3 feet or higher, a bicycle collision, an axial load to the head (eg, resulting from a dive into an empty swimming pool), or a motor vehicle collision involving high speed, rollover, or ejection, we included any fall from a height (ie, not a ground level fall) and any MVC (because we rarely found the vehicular speed reported in ED notes). Cervical spine range of motion data were inconsistently reported in the EMR and were not considered. We did not exclude patients with prior cervical spine surgery, while the CCR study excluded patients with prior cervical spine surgery.

Patients were classified as asymptomatic if the medical record documented no history/complaints or physical examination findings suggestive of cervical spine injuries and that the patients were fully capable of giving a reliable history (ie, not impaired, demented, unconscious, under the influence of alcohol or drugs).

Statistical analysis was performed by using SPSS Version 27 software (IBM). For each categoric variable, a χ2 test was conducted to compare the distribution between the asymptomatic and symptomatic groups, as well as between patients with and without dangerous mechanisms of injury. Additionally, we utilized normal approximation for the binomial distribution to calculate a 95% CI of proportion. A significance level of *P* < .05 was considered statistically significant in our analysis.

## RESULTS

For patients <65 years old, 18,476 CSCT studies were obtained in the 2 EDs (Table). Of the 237 (1.3%) cervical spine fractures reported, 25 (0.14%) occurred in asymptomatic competent patients. Of these asymptomatic patients, 10 had a nondangerous mechanism of injury. If these asymptomatic patients without a dangerous mechanism of injury had not been scanned, 10 of 237 (4.2%) of originally identified cervical fractures would have been missed. A 27-year-old asymptomatic patient with a nondangerous mechanism had a fracture requiring surgical fixation.

View this table:
[Table1](http://www.ajnr.org/content/46/4/820/T1)

Fractures in patients <65 and ≥65 years old

Of 13,925 studies performed in patients ≥65 years old, there were 240 (1.7%) fractures (Table). Of these patients with fractures, 52.6% (*n*=126) were women. Thirty-seven fractures (0.27% of 13,925 studies and 15.4% of 240 fractures) occurred in asymptomatic competent patients. Of these 37 patients who were asymptomatic and ≥65 years old, 21 (56.8%) did not have a dangerous mechanism of injury. If in this cohort asymptomatic patients ≥65 who did not have a dangerous mechanism were not scanned, 21 of 13,925 (0.15%) of fractures would be missed. Of these fractures only 1 (1/21 = 4.8%, 1/13,925 = 0.007%) required surgical fixation. This patient was unique in that they had a prior cervical spine fixation (Table).

We found no correlations between sex and 1) rate of asymptomatic fractures or 2) mechanism of injury. The fracture rate in ≥65-year-old cohort (1.7%) was significantly higher than that in <65-year-old cohort (1.3%).

Of the 240 patients with fractures who were ≥65 years old, 203 (84.6%) were symptomatic. Of these 203 symptomatic patients, 73 (36.0%) had a dangerous and 130 (64.0%) a nondangerous mechanism. In the population <65 years old, 212/237 (89.4%) of the patients with fractures were symptomatic and only 135 (63.7%) of the 212 symptomatic had a dangerous mechanism with 77 (36.3%) a nondangerous mechanism.

## DISCUSSION

This follow-up study of ED trauma patients who are ≥65 years old shows a low rate overall of cervical spine fractures (1.7% = 240/13,925). If asymptomatic older patients were not scanned, 37 cervical spine fractures would have been missed. While refraining from scanning asymptomatic patients may be justifiable, further refinement of a proposed recommendation would be to refrain from imaging asymptomatic patients who also do not have a dangerous mechanism as part of their trauma scenario. In that scenario, the rate of missed fractures would decline to 0.15%. Only 1 of these 21 patients had an unstable fracture surgically corrected (0.007% of the 13,925 patients scanned). The cervical spine fracture rates for those <65 years old are 0.05% in asymptomatic nondangerous mechanisms, and 0.005% for unstable fractures. This suggests that other criteria, such as frailty, high burden of comorbidity, or previous fracture history might be useful in helping clinicians determine which asymptomatic patients over 65 years old with minor trauma should undergo CSCT.

Previously, Lomoschitz et al4 looked at patients 65 years and older and their mechanisms of injury. Of their 149 patients with fractures, 70% had a high energy mechanism, in contrast to our 50% rate.4 Such dangerous mechanisms resulted in 36 of 59 (61%) fractures in the 65- to 75-year-old age range, whereas those from falls from a seated or standing height resulted in 36 of 90 (40%) fractures in those >75 years old. Low-energy fractures and older elderly patient fractures were more likely to occur at C1 and C2. It is important to note that Lomoschitz et al4 did not report whether the patients were symptomatic or not.

The study herein represents the radiology results from 2 different facilities of a major academic center with a large volume of case reports reviewed (*n* = 32,401). The application of the CCR in the EDs of this institution is relatively common but not universal; nevertheless, we have high confidence that all patients 65 and older with some degree of reported trauma received a CSCT, symptomatic or not, especially if they arrived to the ED via emergency medical service (EMS) personnel and were wearing C-collar. All patients with a cervical spine fracture had a consultation from either the neurosurgery or orthopedic team. All competent patients were specifically asked about neck symptoms and had a physical examination of their neck from a neurosurgeon or orthopedic surgeon.

In this retrospective study, to the best of our ability, based on the available clinical documentation, we identified which patients were asymptomatic. The mechanism of injury and speed of motor vehicles was also assessed as reliably as possible within the limitations of ED and EMS notes in this retrospective cohort. The strict definition of the CCR for dangerous mechanism for fall from 3 feet or greater and “high-speed” collision was modified to any fall from a height and any MVC in our inclusion criteria due to inadequate documentation of details in the EMR.

In conclusion, appropriate use of CSCT imaging in elderly patients with head trauma in many EDs has largely been guided by the CCR since the 2000s.1,5 Given how often the CCR has been adopted in many EDs, and the increased availability of CSCT, cervical spine imaging has become common for patients ≥ 65 years old with any trauma. A reassessment as to the value of imaging patients 65 and older who are asymptomatic with low energy mechanism injuries seems warranted given that the rate of cervical spine fractures in this patient population is 0.15% and that, of these, only 1 of 21 (4.8% of the fractures and 0.007% of the total scanned population) were unstable fractures. Our findings support tailoring imaging decisions more closely to individual risk factors and clinical presentations, or refining existing decision support tools rather than applying a blanket rule based on age and/or mechanism alone.

## Footnotes

*   Christopher Lawrence and Mahla Radmard contributed equally to this article.

*   [Disclosure forms](https://www.ajnr.org/sites/default/files/additional-assets/Disclosures/April%202025/0774.pdf) provided by the authors are available with the full text and PDF of this article at [www.ajnr.org](http://www.ajnr.org).

## References

1.  1.Stiell IG, Wells GA, Vandemheen KL, et al. The Canadian C-spine rule for radiography in alert and stable trauma patients. JAMA 2001;286:1841–48 doi:10.1001/jama.286.15.1841 pmid:11597285
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.1001/jama.286.15.1841&link_type=DOI) 
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=11597285&link_type=MED&atom=%2Fajnr%2F46%2F4%2F820.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=000171595300024&link_type=ISI) 

2.  2.Stiell IG, Clement CM, McKnight RD, et al. The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med 2003;349:2510–18 doi:10.1056/NEJMoa031375 pmid:14695411
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.1056/NEJMoa031375&link_type=DOI) 
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=14695411&link_type=MED&atom=%2Fajnr%2F46%2F4%2F820.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=000187529500006&link_type=ISI) 

3.  3.Radmard M, Tafazolimoghadam A, Hoseinyazdi M, et al. Is older age an appropriate criterion alone for ordering cervical spine computed tomography after trauma. Acad Emerg Med [Epub ahead of print] June 28, 2024 doi:10.1111/acem.14976
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.1111/acem.14976&link_type=DOI) 
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=38940447&link_type=MED&atom=%2Fajnr%2F46%2F4%2F820.atom) 

4.  4.Lomoschitz FM, Blackmore CC, Mirza SK, et al. Cervical spine injuries in patients 65 years old and older: epidemiologic analysis regarding the effects of age and injury mechanism on distribution, type, and stability of injuries. AJR Am J Roentgenol 2002;178:573–77 doi:10.2214/ajr.178.3.1780573 pmid:11856676
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.2214/ajr.178.3.1780573&link_type=DOI) 
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=11856676&link_type=MED&atom=%2Fajnr%2F46%2F4%2F820.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=000174055400007&link_type=ISI) 

5.  5.Baratloo A, Ahmadzadeh K, Forouzanfar MM, et al. NEXUS vs. Canadian C-Spine Rule (CCR) in predicting cervical spine injuries; a systematic review and meta-analysis. Arch Acad Emerg Med 2023;11:e66 doi:10.22037/aaem.v11i1.2143 pmid:37840870
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.22037/aaem.v11i1.2143&link_type=DOI) 
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=37840870&link_type=MED&atom=%2Fajnr%2F46%2F4%2F820.atom) 

*   Received August 6, 2024.
*   Accepted after revision October 15, 2024.


*   © 2025 by American Journal of Neuroradiology