Journal Club January 2018 - Cervical Spine Fractures in the Elderly


More than five million patients require spinal immobilization following blunt trauma each year in the US.   The distinction between which patients may have their cervical spine “cleared clinically” versus those whom require additional imaging is an essential skill for emergency physicians.  Two clinical decision rules are often used to guide decision-making; the National Emergency X-Radiography Utilization Study (NEXUS) or the Canadian Cervical Spine Rule.

Both decision rules have sensitivities greater than 99% for identifying cervical spine injury.  Using the NEXUS criteria, no imaging studies are required if:

  • There is no posterior midline cervical spine tenderness
  • There is no evidence of intoxication
  • The patient exhibits a normal level of alertness
  • There are no focal neurological deficits
  • There is no painful distracting injury.
NEXUS Criteria (

NEXUS Criteria (


Alternatively, the Canadian Cervical Spine Rule considers patients very low risk for cervical spine fracture if:

  • The patient is fully alert with a Glasgow Coma scale of 15
  • The patient has no high-risk factors (ie, age >65 years, dangerous mechanism of injury, fall greater than five stairs, axial load to the head, high-speed vehicular crash, bicycle or motorcycle crash, or the presence of paresthesias in the extremities)
  • The patient has low-risk factors (eg, simple vehicle crash, sitting position in the ED, ambulatory at any time, delayed onset of neck pain, and the absence of midline cervical tenderness)
  • The patient can actively rotate his or her neck 45 degrees to the left and to the right. 2


Geriatric patients (> age 65) account for almost 1 in 5 of all cervical spine fractures, with 50% of cervical spine fractures occurring at the C1/C2 level.  Recent literature has questioned whether these clinical decision rules may be applied in the geriatric population, where over half of identified cervical spine fractures result from ground level falls.  Furthermore the reliability of physical exam, specifically midline tenderness, has been questioned in elderly patients.  This month’s journal club examined three articles that addressed this specific population of blunt trauma patients at risk for cervical spine injury. 




1.  Asymptomatic cervical spine fractures: current guidelines can fail older patients. Healy CD, Spilman SK, King BD, Sherrill JE 2nd, Palaez CA. J Trauma Acute Care Surgery. 2017 Jul;83(1):119-125. [Pubmed]

2.  Cervical Spine Fractures in geriatric blunt trauma patients with low-energy mechanism: are clinical predictors adequate. Sherwin P. Schrag et al. The American Journal of Surgery; 2008. [Pubmed]

3.  Tran, J., Jeanmonod, D., Agresti, D., Hamden, K., & Jeanmonod, R. K. (2016). Prospective Validation of Modified NEXUS Cervical Spine Injury Criteria in Low-risk Elderly Fall Patients. The Western Journal of Emergency Medicine, 17(3), 252–7. [Full Text]



This retrospective study of 1071 patients age 55 and older with cervical spine imaging assessed the association of neck pain and cervical spine fractures in patients presenting to a level 1 trauma center following blunt trauma.  Patients were into categorized two groups: 1) SYMPTOMATIC -  if they had pain documented in the history, review of systems, or on physical exam. If documentation was unclear, patients were considered symptomatic; 2) ASYMPTOMATIC -  if they denied pain and had no tenderness on exam.  36 of the 173 patients (21%) were asymptomatic. This was associated with a 79.2% sensitivity (CI 72.4-85) and 59.6% specificity (56.3-62.8). Neck pain had a positive predictive value of 27.6% (CI 25.5-29.9).

Nearly one fifth (21%) of patients > 55 with a c-spine fracture reported no pain on initial presentation and denied tenderness to palpation on examination.  22% of symptomatic patients and 19% of asymptomatic patients required surgical intervention, which was not a statistically significant difference.


Bottom Line:  This study is limited in its retrospective design and needs to be duplicated in a prospective manner to be widely applicable.  Furthermore, the patients who were asymptomatic with c-spine fractures had higher overall Injury Severity Scores and higher incidence of multi-level fractures, which limits the generalizability to a demographic of lower risk mechanism patients such those with ground level falls.



This retrospective case controlled study addressed whether clinical predictors are adequate in the identification of cervical spine fractures in geriatric blunt trauma patients with low energy mechanisms.  1264 patients were enrolled and 40 (3.17%) had c-spine fractures.  These patients were compared to 64 patients without c-spine fractures identified on CT and the presence of 8 clinical and radiographic predictors was assessed. 

  • CT evaluation of the head
  • C-spine tenderness
  • Cephalohematoma
  • Confusion
  • Focal neuologic deficit
  • Facial fractures
  • Other fractures
  • Ethanol consumption (excluded as this was only present in 2 patients)

C-spine tenderness was only present in 45.5% of the patients with c-spine fractures and no other predictors reached statistical significance.

Bottom Line:  This study was limited by a small sample size and its retrospective design at a single institution.  The Study included patients with GCS ranging from 8-15 and therefore it is unclear how accurate the clinical predictors are in a group of alert elderly patients with lower energy mechanisms of trauma.  Despite this, the 45.5% of patients with no c-spine tenderness on exam is concerning and again raises the possibility of a less reliable clinical exam in elderly trauma patients at risk for c-spine injury.



The objective of this prospective observational cohort study was to validate modified NEXUS criteria in a low-risk elderly fall population. Two modifications to Nexus were: a modified definition for distracting injury (signs of trauma to head and neck only) and the definition of normal mentation (being at patient’s baseline mental status).  Using the patient’s personal baseline mental status rather than GCS and using signs of trauma to the head and neck as the only distracting injury, these modified NEXUS criteria performed well in this study population with a sensitivity of 100%, negative predictive value of 100%  and specificity of 47.7.   Similar to the NEXUS cohort, this population had a low rate (1.4%) of cervical spine injury (CSI).  This report’s incidence is likely lower than NEXUS (which was 4.6%) due to the inclusion criteria including only those with low risk mechanism who were not triaged to the trauma bay. 

Bottom line:  The elderly population enrolled in this study included only injury due to falls that was not triaged to a trauma bay based on the facility’s trauma alert criteria.  This study was limited to a single tertiary care trauma center and the geriatric population may not reflect the experience of other facilities which suggests need for a prospective, randomized control, multi-center clinical trial of above modified NEXUS before it can applied clinically.




download article summaries

healey et al. 

healey et al. 

sherwin ET AL, 2015

sherwin ET AL, 2015

tran ET AL

tran ET AL


Written by Christine Hein, MD

Edited and Posted by Jeffrey A. Holmes, MD