Comparative strengths and structural properties of the upper and lower cervical spine in flexion and extension

https://doi.org/10.1016/S0021-9290(02)00037-4Get rights and content

Abstract

The purpose of this study is to test the hypothesis that the upper cervical spine is weaker than the lower cervical spine in pure flexion and extension bending, which may explain the propensity for upper cervical spine injuries in airbag deployments. An additional objective is to evaluate the relative strength and flexibility of the upper and lower cervical spine in an effort to better understand injury mechanisms, and to provide quantitative data on bending responses and failure modes. Pure moment flexibility and failure testing was conducted on 52 female spinal segments in a pure-moment test frame. The average moment at failure for the O-C2 segments was 23.7±3.4 N m for flexion and 43.3±9.3 N m for extension. The ligamentous upper cervical spine was significantly stronger in extension than in flexion (p=0.001). The upper cervical spine was significantly stronger than the lower cervical spine in extension. The relatively high strength of the upper cervical spine in tension and in extension is paradoxical given the large number of upper cervical spine injuries in out-of-position airbag deployments. This discrepancy is most likely due to load sharing by the active musculature.

Introduction

Restraint devices for motor vehicle occupants have become increasingly more advanced and complex over the past decade. However, as injury prevention technologies have progressed, the parameters used to assess injury risk in new motor vehicles have remained relatively unchanged. In order to keep pace with new technologies, and to advance the state of the art in injury prevention, the United States National Highway Traffic Safety Administration (NHTSA) is proposing new injury standards for the 30 mph barrier impact test (FMVSS 208). These include new neck injury criteria that will utilize both axial and bending loads in the formulation of injury reference values for Anthropomorphic Test Devices (ATDs).

One of the objectives of the new neck injury standard is to prevent airbag-related injuries, particularly to the upper cervical spine (O-C2). The US National Center for Statistics and Analysis (NCSA) has been conducting Special Crash Investigations (SCIs) since 1991 on all airbag injuries in low to moderate severity crashes [http://www.nhtsa.dot.gov/people/ncsa/sci3.html]. Most of these injuries occur when the occupant is out-of-position and close to the airbag module. The data from the SCIs show that 78% of the cervical spine injuries in adults are occurring between the occiput and C2. This is considerably higher than upper cervical spine injury rates in the general population (Hadley et al., 1986; Levine and Edwards, 1986; Myers and Winkelstein, 1995). The specific mechanisms by which these injuries occur are still unclear. It is generally assumed that neck injury in airbag deployments is caused by tension and extension secondary to direct loading of the head and mandible (Blacksin, 1993; Maxeiner and Hahn, 1997). Studies using ATDs have demonstrated large tensile forces and extension moments when the dummies are close to deploying airbags (Pintar et al., 1999; Tylko and Dalmotas, 2000).

The new neck injury criteria employ a linear combination of normalized neck axial force (FZ) and neck moment at the occipital condyles (MY). The formulation is Nij=FNZ+MNY, where FNZ=FZ/FZCRIT and MNY=MY/MYCRIT. The axial force and the moment are measured at the same time point, and the critical values are the intercepts for axial load (tension or compression) and moment (flexion or extension). The Nij cannot exceed 1.0 at any point in time during a crash test. The most important parameters in this criterion are the critical values for tension and extension. There have been a number of studies on the strength of the neck in tension (Cheng et al., 1982; Sances et al., 1981; Shea et al., 1992; Van Ed et al., 2000; Yoganandan et al., 1996); however, there are no studies on the strength of the ligamentous cervical spine in bending. As a result, the proposed bending tolerance values for the human upper cervical spine are highly inferential (Mertz and Patrick, 1971; Mertz and Prasad, 2000). The lack of data on the strength of the cervical spine in pure bending has been an impediment to the development of new neck injury standards for crash testing.

The purpose of this study is to test the hypothesis that the upper cervical spine is weaker than the lower cervical spine in flexion/extension bending, which may explain the propensity for upper cervical spine injuries in airbag deployments. The hypothesis was tested by using pure bending moments to produce injuries in human cadaver spinal segments. A secondary goal of this study is to provide previously unavailable biomechanical data on the bending responses and the bending strength of the human cervical spine. These data will assist in the development of injury criteria for combined tension and bending of the neck. They will also assist in the development of new physical and computational models of the neck.

Section snippets

Methods

Testing was performed on 52 unembalmed spinal segments from 16 cervical spines. Donor age ranged from 33 to 66 years (50.8±8.8, mean±standard deviation). To minimize variance, only female cervical spines were used (Nightingale et al., 1997). The muscular tissues were removed while keeping all the ligamentous structures intact (with the exception of the ligamentum nuchae). All specimen handling was performed in compliance with CDC guidelines (Cavanaugh and King, 1990). The cervical spines were

Results

The correlation coefficients for the individual flexibility functions were all >0.98, and the correlation coefficients for the averaged flexibility functions were all >0.99. The coefficients for the functions describing the average response of all the spinal segments are given in Table 1. Plots of the averaged flexibility functions for all the segments are shown in Fig. 2.

The tolerance data for all the motion segments are summarized in Table 2, Table 3. Two C6-C7, and one C4-C5 motion segments

Discussion

The results for the flexibility tests are in good agreement with previously published studies, despite the fact that previous studies used mixed genders (Goel et al., 1988; Panjabi et al (1991), Panjabi et al (1994); Voo et al., 1998) (Table 6). The flexibility results are based on a larger sample size than in prior studies and, with the exception of the work by Voo et al., 1998, the moments applied are of considerably larger magnitude (±3.5 N m). Most previous studies have focused on the

Acknowledgements

NHTSA Cooperative Agreement No. DTNH22-94-Y-07133.

References (31)

  • Cheng, R., Yang, K.H., Levine, R.S., King, A.I., Morgan, R., 1982. Injuries to the cervical spine caused by a...
  • B.J. Doherty et al.

    A biomechanical study of odontoid fracture and fracture fixtion

    Spine

    (1993)
  • Y.C.B. Fung

    Stress–strain-history relations of soft tissues in simple elongation

  • M.N. Hadley et al.

    Axis fractures resulting from motor vehicle accidents. The need for occupant restraints

    Spine

    (1986)
  • H. Maxeiner et al.

    Airbag-induced lethal cervical trauma

    Journal of Trauma

    (1997)
  • Cited by (127)

    View all citing articles on Scopus
    View full text