JBJS | Experimental Impact Injury to the Cervical Spine: Relating Motion of the Head and the Mechanism of Injury*

The Journal of Bone & Joint Surgery, Volume 78, Issue 3

Articles | March 01, 1996

Experimental Impact Injury to the Cervical Spine: Relating Motion of the Head and the Mechanism of Injury*

ROGER W. NIGHTINGALE, PH.D.†; JAMES H. MCELHANEY, PH.D.†; WILLIAM J. RICHARDSON, M.D.†; THOMAS M. BEST, M.D., PH.D.†; BARRY S. MYERS, M.D.PH.D.†, DURHAM, NORTH CAROLINA

View Disclosures and Other Information

Investigation performed at the Department of Biomedical Engineering, Duke University, Durham

The Journal of Bone & Joint Surgery. 1996; 78:412-21

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

The purpose of this study was to analyze, with use of an impact model, the relationships among motion of the head, local deformations of the cervical spine, and the mechanisms of injury; the model consisted of the head and neck of a cadaver. Traditionally, the mechanisms of injury to the cervical spine have been associated with flexion and extension motions of the head and neck. However, the classification of the mechanisms is not always in agreement with the patient's account of the injury or with lacerations and contusions of the scalp, which indicate the site of the impact of the head. Eleven specimens were dropped in an inverted posture with the head and neck in an anatomically neutral position. Forces, moments, and accelerations were recorded, and the impacts were imaged at 1000 frames per second. The velocity at the time of impact was on the order of 3.2 meters per second. The angle and the padding of the impact surface varied.Observable motion of the head did not correspond to the mechanism of the injury to the cervical spine. Injury occurred 2.2 to 18.8 milliseconds after impact and before noticeable motion of the head. However, the classification of the mechanism of the injuries was descriptive of the local deformations of the cervical spine at the time of the injury. Accordingly, it is a useful tool in describing the local mechanism of injury. Buckling of the cervical spine, involving extension between the third and sixth cervical vertebrae and flexion between the seventh and eighth cervical vertebrae, was observed. Other, more complex, buckling deformations were also seen, suggesting that the deformations that occur during impact are so complex that they can give rise to a number of different mechanisms of injury.CLINICAL RELEVANCE: Classic concepts of flexion and extension of the head as a mechanism of injury do not apply to a vertical impact of the head. Motions of the head, which often are used to classify the injury, are not a reliable indicator of the mechanism of injury. The mechanism of injury is descriptive of local deformations of the cervical spine and forces at the instant of injury. Although it is a useful tool for describing local mechanisms of injury, care should be taken not to confuse the mechanism of injury at the level of the motion segment with the mechanism as it applies to loads on (and resulting motions of) the head.The complex buckling of the cervical spine that results from a vertical impact of the head may cause concomitant flexion and extension in different regions of the cervical spine. Treatment should be based on the local mechanism, with the understanding that this type of impact may involve multiple, sometimes non-contiguous, mechanisms of injury.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

The characterization of the mechanism of injury to the cervical spine continues to be of interest to the orthopaedic community. Classifications of these mechanisms have been offered by Babcock³, White and Panjabi³², Harris et al.¹⁰, Portnoy et al.²⁴, Allen et al.², and Whitley and Forsyth³². The classifications have been based on reviews of patient data, experiments performed on whole cadavera and cervical spines isolated from cadavera^{4,11,13,16,19-21,23,24,33}, and analysis of videotapes of injuries that occurred during sporting events³⁰. Despite the importance of consistent classification of injury, there is still a lack of consensus on both the mechanisms and the classifications of injuries to the neck. As a result, a particular type of injury is sometimes attributed to different or contradictory mechanisms that may suggest different treatment. For instance, the definition of the so-called teardrop fracture has been very unclear in the literature. Some investigators have used the term to describe a specific injury^4,5,31, whereas others have used it to describe any triangular fragment that has avulsed from the anterior portion of the vertebral body^9,24. Hence, this fracture has been attributed to flexion, vertical compression, and extension.

The difficulty in establishing a generally accepted classification of the mechanism of injury to the cervical spine is not surprising. In clinical studies, mechanisms have been postulated on the basis of reviews of radiographs and magnetic resonance images, studies of patients' histories, and reconstructions of accidents; these studies are, by definition, retrospective. In order to determine the mechanism conclusively, it is necessary to know the actual motions of the cervical spine that produce injury rather than just the fractures and residual displacements that are seen on radiographs. Numerous experimental studies of isolated segments of the cervical spine have provided such information^{4,16,19,21,23,25}; however, they have not related injuries to the cervical spine to traumatic events that occur during actual accidents.

It has been hypothesized that the mechanisms of injury to the cervical spine correspond to forces and moments applied to the head and with the resulting motions of the head^3,5,31. It has been thought that flexion is the mechanism of injury in accidents that produce flexion of the head and that extension is the mechanism in accidents that produce extension of the head. Static testing of spines from cadavera has supported this hypothesis^3,4,19; however, the hypothesis has not been tested for injuries involving impact of the head. Impact of the head occurs in falls, dives into shallow water, and motor-vehicle accidents, causing more than 70 per cent of injuries to the cervical spinal cord¹⁴. Accordingly, the purpose of the present study was to evaluate this hypothesis in an impact model (consisting of the head and neck of a human cadaver) by monitoring the traumatic deformations and forces to the cervical spine. An important secondary goal was to evaluate the clinical relevance of the injuries that were produced.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were Grant R49/CCR402396-08 of the Centers for Disease Control and the Virginia Flowers Baker Chair.

†Department of Biomedical Engineering, Box 90281 (R. W. N., J. H. McE., T. M. B., and B. S. M.) and Division of Orthopaedic Surgery, Department of Surgery, Box 3077 Medical Center (W. J. R.), Duke University, Durham, North Carolina 27708.

View Large | Download Slide(.ppt)
Add to My JBJS

Diagram of the test apparatus, showing the accelerometer on the torso mass (A), the optical velocity sensor (B), the carriage and torso mass (C), the six-axis load-cell at the first thoracic level (D), the accelerometers on the head (E), and the impact surface and three-axis load-cell (F). The angle of the impact surface was varied according to the sign convention shown.

View Large | Download Slide(.ppt)
Add to My JBJS

Graph of the magnitude of the resultant forces on the head and neck, with the time-interval between the onset of force to the head and injury to the cervical spine. Injury is defined by the decrease in resultant load with continued compression by the torso mass. T1 = first thoracic vertebra.

View Large | Download Slide(.ppt)
Add to My JBJS

A typical impact sequence. The initial position is shown at four milliseconds after impact. At eighteen milliseconds, the cervical spine buckled and a vertical compression injury to the atlas occurred, quickly followed by distractive extension injuries to the fourth and fifth cervical vertebrae. Subsequent flexion and extension motions of the head are shown at thirty-two and ninety milliseconds.

View Large | Download Slide(.ppt)
Add to My JBJS

Time line showing the relative occurrences of injury to the neck and noticeable (more than 20-degree) motions of the head with respect to the duration of motion of the head and cervical spine.

View Large | Download Slide(.ppt)
Add to My JBJS

Four frames of video data (specimen E), obtained one millisecond apart, showing the initial serpentine deformation (at three milliseconds) followed by a rapid transition to buckling (at six milliseconds). Buckling involved flexion of the seventh and eighth cervical motion segments and extension of the third through sixth cervical motion segments.

View Large | Download Slide(.ppt)
Add to My JBJS

Illustration of the initial posture of the cervical spine and the buckling deformation. Extension was observed at the third through sixth cervical motion segments and flexion, at the seventh and eighth cervical motion segments. There was a strong association between the mechanism of injury and the local deformation at each motion segment. The number of injuries that were produced is shown, in parentheses, by level and classification. CF = compressive flexion, CE = compressive extension, VC = vertical compression, DE = distractive extension, and DF = distractive flexion.

View Large | Download Slide(.ppt)
Add to My JBJS

Radiographs of specimens F (Fig. 7-A) and B (Fig. 7-B), showing the injuries that were produced, the most evident of which are the bilateral dislocated facets at the sixth and seventh cervical levels (Fig. 7-A) and a hangman's fracture (Fig. 7-B).

View Large | Download Slide(.ppt)
Add to My JBJS

Graph of the positions of the pins in the pars interarticularis in specimen D. Each point represents sequential spatial positions of the pins in the transverse processes of the third through seventh cervical vertebrae before the impact (t = 0) and at six, nine, and twelve milliseconds. The normal lordotic curve can be seen before the impact. The serpentine deformation of the spine is demonstrated by the anterior displacement of the seventh, sixth, and third cervical vertebrae and the posterior displacement of the fourth cervical vertebra at six milliseconds after impact. Illustration of a more complicated buckling, seen two to eight milliseconds after impact of the head.

View Large | Download Slide(.ppt)
Add to My JBJS

Illustration of a more complicated buckling, seen two to eight milliseconds after impact of the head.

View Large | Download Slide(.ppt)
Add to My JBJS

Illustrated example of the variation in local forces at different vertebral levels. The compressive force (P) is shown as components of compression (Pc, perpendicular to the end plate) and shear (Ps, parallel to the end plate) on the end plates of the third and seventh cervical vertebrae (C3 and C7). In the buckled cervical spine, axial compression gives rise to local compression with posterior shear at the third cervical vertebra and local compression with anterior shear at the seventh cervical vertebra. Similarly, the bending moment at the third cervical vertebra is extension, and the moment at the seventh cervical vertebra is flexion.

View Larger

TABLE I DATA ON THE SUBJECTS AND INJURIES AND CLASSIFICATION OF INJURIES

*The magnitude of force that produced the injury.†The elapsed time between impact and injury.‡The injuries to the lower cervical spine were classified according to the system of Allen et al.², and the injuries to the upper cervical spine were classified as described in the text. CE = compressive extension, DE = distractive extension, VC = vertical compression, DF = distractive flexion, and CF = compressive flexion. S1, S2, and S3 are severity indices², specific to each mechanism.

Specimen	Sex, Age	Velocity	Angle of Impact Surface	Force*	Time to Injury†	Injuries to Cervical Spine‡	Maximum Displacement of Head
	(Yrs.)	(m/Sec.)	(Degrees)	(N)	(msec)
Rigid impact surface
A	M, 65	2.43	0	—	—	None	Flexion
B	M, 62	3.20	0	-1839	2.2	C1 (2-part fract., post. ring) (CE) and C2 (hangman's fract.) (DE)	Flexion
C	M, 71	3.26	0	-1955	6.5	C1 (3-part commin. fract.) (VC)	Flexion
D	M, 55	3.14	-15	—	—	None	Flexion
E	F, 35	3.28	-15	—	—	None	Flexion
F	M,—	3.26	+15	-1863	6.4	C1 (fract., lateral mass) (VC), C2 (hangman's fract.) (DE), C2-C3 disc and ant. longitudinal lig. (DES1), and C6-C7 (bilat. dislocation perching, facets) (DFS3)	Extension
G	M, 36	3.23	+30	-1552	8.3	C3 (burst fract.) (VCS3), C3-C4 disc and ant. longitudinal lig. (DES1), and C4-C5 ant. longitudinal lig. (DES1)	Extension

Padded impact surface
H	M, 61	3.13	+30	-1632	14.8	C1 (2-part fract., ant. ring) (CF); C4 (fract., spinous process (CES1); C5 (fract., spinous process) (CES1); and C5-C6 disc, L capsular lig., and ant. longitudinal lig. (DES2)	Extension
I	M, 46	3.03	+30	—	—	None	Extension
J	M, 46	3.51	+30	-2240	18.7	C1 (2-part fract., R aspect) (VC), C3-C4 disc and ant. longitudinal lig. (DES1); and C4-C5 disc, ant. longitudinal lig. , and chip of sup. part of body (DES1)	Extension
K	F, 42	3.07	-15	-1011	18.8	C2-C3 disc, ant. longitudinal lig., and chip, C2 (DES1) and C3-C4 disc, ant. longitudinal lig., and C3 chip (DES1)	Flexion

Add to My JBJS

TABLE I DATA ON THE SUBJECTS AND INJURIES AND CLASSIFICATION OF INJURIES

Specimen	Sex, Age	Velocity	Angle of Impact Surface	Force*	Time to Injury†	Injuries to Cervical Spine‡	Maximum Displacement of Head
	(Yrs.)	(m/Sec.)	(Degrees)	(N)	(msec)
Rigid impact surface
A	M, 65	2.43	0	—	—	None	Flexion
B	M, 62	3.20	0	-1839	2.2	C1 (2-part fract., post. ring) (CE) and C2 (hangman's fract.) (DE)	Flexion
C	M, 71	3.26	0	-1955	6.5	C1 (3-part commin. fract.) (VC)	Flexion
D	M, 55	3.14	-15	—	—	None	Flexion
E	F, 35	3.28	-15	—	—	None	Flexion
F	M,—	3.26	+15	-1863	6.4	C1 (fract., lateral mass) (VC), C2 (hangman's fract.) (DE), C2-C3 disc and ant. longitudinal lig. (DES1), and C6-C7 (bilat. dislocation perching, facets) (DFS3)	Extension
G	M, 36	3.23	+30	-1552	8.3	C3 (burst fract.) (VCS3), C3-C4 disc and ant. longitudinal lig. (DES1), and C4-C5 ant. longitudinal lig. (DES1)	Extension

Padded impact surface
H	M, 61	3.13	+30	-1632	14.8	C1 (2-part fract., ant. ring) (CF); C4 (fract., spinous process (CES1); C5 (fract., spinous process) (CES1); and C5-C6 disc, L capsular lig., and ant. longitudinal lig. (DES2)	Extension
I	M, 46	3.03	+30	—	—	None	Extension
J	M, 46	3.51	+30	-2240	18.7	C1 (2-part fract., R aspect) (VC), C3-C4 disc and ant. longitudinal lig. (DES1); and C4-C5 disc, ant. longitudinal lig. , and chip of sup. part of body (DES1)	Extension
K	F, 42	3.07	-15	-1011	18.8	C2-C3 disc, ant. longitudinal lig., and chip, C2 (DES1) and C3-C4 disc, ant. longitudinal lig., and C3 chip (DES1)	Flexion

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

Most injuries to the cervical spine occur when the head hits an object and the neck is forced to stop the torso⁷. Accordingly, the risk of injury depends on the shape and weight of the head, the impact surface, the energy of the moving torso, and the strength of the cervical spine. Many studies^{4,11,14-16,19-21,23,25} have been performed on cadavera to aid in the understanding of neck injury, but we know of no investigation that has employed all of these effects and has provided sufficient instrumentation to describe fully the mechanism of injury. In addition, most injuries occur dynamically, and it has been established that responses of the cervical spine are sensitive to loading rates^15,16,18; therefore, experiments should be conducted at realistic loading rates.

Experimental Apparatus

The cadaver-based impact model that was developed included the entire head and the spine through the second thoracic vertebra. The apparatus was designed to simulate a vertical impact of the head with a following torso. The apparatus consisted of a drop-track and a steel carriage mounted on two linear bearing sliders (Fig. 1). Weight was added to the carriage to simulate a torso mass of sixteen kilograms. This weight is an approximation of the mass of the upper part of the torso for a man whose weight is in the fiftieth percentile, and it represents the portion of the inertia of the torso that acts on the neck during injury. To quantify the forces and moments fully, two load-transducers were used: a three-axis load-cell (model 9067; Kistler Instrument, Amherst, New York) under the impact surface measured forces on the head, and a six-axis load-cell (model 6607-00; GSE, Farmington Hills, Michigan) at the first thoracic vertebra measured forces and moments on the neck. Three accelerometers (PCB Piezotronics, Buffalo, New York) were used: a uniaxial accelerometer (model 302A02) was attached to the torso to measure decelerations of the torso, and two triaxial accelerometers (model 306A06) were mounted on the head to measure linear and angular accelerations of the head in the sagittal plane. A computer-based sixteen-channel digital data-acquisition system (RC-Electronics, Santa Barbara, California) recorded data from the transducers, and an optical sensor (MTS, Eden Prairie, Minnesota) recorded the impact velocity. The motion of the head and the vertebrae during injury were recorded with a high-speed digital-imaging system (model EM-2; Eastman Kodak, Charlotte, North Carolina). This imaging tool acquires 1000 digitized frames per second and was interfaced with a workstation (Sun Microsystems, Mountain View, California) for analysis of the kinematics of the head and neck. A flat steel plate, 15.25 centimeters in diameter and 4.0 centimeters thick, was used as an impact surface. It was mounted on a locking clevis that allowed variation of the surface angle about the y axis (normal to the sagittal plane); thus, impacts could be delivered to the anterior or posterior portion of the head.

Preparation of the Specimens

The tests were performed with the heads and intact ligamentous cervical spines from unembalmed human cadavera. Previous investigations have shown that the mechanical properties of hydrated bone and ligament from cadavera are similar to those of living tissue; however, the muscular tissues no longer support load^22,28. Accordingly, we removed the muscles to visualize the motions of the cervical spine and the mechanisms of injury better. Medical records and radiographs of the specimens that had been made before testing were examined to ensure that there were no unrecognized abnormalities of the spine that may have degraded their structural integrity. All preparations were carried out in a humidified chamber to prevent dehydration. The caudad two vertebrae (the first and second thoracic vertebrae) were cleaned of muscle, defatted, and mounted in an aluminum cup with reinforced polyester resin. Extreme care was taken to ensure that the motion segment between the seventh cervical and first thoracic vertebrae was free of the resin, to permit uninhibited motion. Otherwise, a stress concentration would be created at the joint, which would predispose the specimen to fail at the cup. The cup was cooled in a water bath to prevent degradation of the specimen as a result of the heat of polymerization. The triaxial accelerometers were attached to exposed parietal bone with the use of dental acrylic and bone screws. A jig was used to ensure that the accelerometers were parallel to the sagittal plane. Finally, photographic target pins (4.0 millimeters in diameter) were inserted into the anterior part of the vertebral body, the spinous processes, and the pars interarticularis of the second through seventh cervical vertebrae. The pins were used for photogrammetric analysis of the motions of the vertebrae.

Test Protocol

All of the specimens were positioned in an inverted posture and in an anatomically neutral configuration (preservation of the resting lordosis and no anteroposterior rotation of the head). The initial position was maintained with the use of sutures passed through the ear lobules and the nasal septum (Fig. 1). The neutral angle for the intervertebral disc between the seventh cervical and first thoracic vertebrae (25 degrees) was taken from radiographic studies of normal volunteers.

Each specimen was mounted to the carriage, raised into position, and preconditioned by manually exercising the head and neck through 60 degrees of combined flexion and extension for fifty cycles¹¹. The specimen was dropped from a height of 0.54 to 0.61 meter, which was sufficient to cause injury to the cervical spine in most of our tests but is less than that which causes fracture of the skull. After the impact, anteroposterior and lateral radiographs were made of the entire specimen, the head and the cervical spine were dissected, and all injuries were documented.

The occurrence of injury was determined by a temporary decrease in the axial component of force measured at the first thoracic vertebra (neck force) with a continued downward translation of the torso mass (Fig. 2). This decrease in load with increasing compressive deformation is the traditional definition of failure. Data from the high-speed images and load-cells were also compared to ensure a temporal association between local increases in deformation of the cervical spine and decreases in the axial component of force.

Testing was performed on eleven specimens (Table I). The angle of the impact surface was varied in 15-degree increments from +30 to -15 degrees. A positive angle indicated an anterior impact; a negative angle, a posterior impact; and 0 degrees, an impact to the top of the head (a flat surface). For seven specimens, the steel-plate impact surface was covered with a sheet of lubricated Teflon (polytetrafluoroethylene). This was designated as the unconstrained head end condition. For the remaining four specimens, the surface was padded with five centimeters of open-cell polyurethane foam (Young's modulus, 158.6 kilopascals; yield stress, 7.0 kilopascals; and density, 0.0277 gram per cubic centimeter). The foam causes the head to pocket and resists its sliding along the impact surface. This was designated as the constrained end condition.

Results

Introduction | Materials and Methods | Results | Discussion | References

Injuries to the cervical spine occurred in seven of the eleven specimens (Fig. 7-A and 7-B; Table I). In six specimens, there was injury at multiple levels, yielding a total of nineteen injured vertebrae or vertebral motion segments. In five of the injured specimens (B, C, F, H, and J), fracture of the first cervical vertebra occurred; in three (F, H, and J), there were also multiple non-contiguous injuries. The system of Allen et al.² was used to classify the injuries in the lower cervical spine. Injuries in the upper cervical spine were classified as follows. Fractures of the posterior ring of the first cervical vertebra were classified as compressive extension injuries and fractures of the anterior ring, as compressive flexion injuries. Comminuted fractures of the first cervical vertebra or through its facets were attributed to vertical compression and bipedicular fracture at the second cervical level (hangman's fractures), to distractive extension. No attempt was made to subclassify the injuries to the upper cervical spine into stages.

All of the injuries were associated with very little concomitant motion of the head. Injury occurred 2.2 to 8.3 milliseconds after impact with the rigid surface and 14.8 to 18.8 milliseconds after impact with the foam-padded surface (Fig. 3 Table I). Noticeable flexion or extension motions of the head (motion of more than 20 degrees) occurred after injury, twenty to 100 milliseconds after impact (Fig. 4). Maximum displacement of the head, on the order of 90 degrees in flexion or extension, occurred during the latter 150 milliseconds of the test, with gross motions completed after 155 to 300 milliseconds. During the latter portion of the impact of the anterior part of the head (surface angle of +30 degrees), the head extended and translated posteriorly. For all other surface orientations, the head flexed and translated anteriorly.

Motion of the head did not correspond to the mechanism of injury to the cervical spine. Flexion of the head was produced in six of the eleven specimens, with a total of five injuries (one vertical compression, four extension, and no flexion injuries), and extension of the head was produced in five, with a total of fourteen injuries (three vertical compression, nine extension, and two flexion injuries) (Table I).

In each of the specimens, regardless of the orientation and padding of the impact surface, the cervical spine buckled and a distinct pattern of deformation developed that explained the distribution of the injuries (Figs. 5 and 6). The pattern was characterized by local extension of the third through sixth cervical motion segments and local flexion of the seventh and eighth cervical motion segments. This mode of deformation persisted for the duration of the injury interval, resulting in a characteristic pattern. Twelve of the thirteen injuries to the second through sixth cervical vertebrae and the inclusive ligamentous complexes were attributed to extension. The one exception (specimen G) was a vertical compression fracture of the third cervical vertebra. Only one specimen (F) had injury (bilateral dislocation of the facets) of the sixth and seventh cervical vertebrae (Fig. 7-A). The injuries to the atlas were attributed to vertical compression (three injuries), compressive flexion (one injury), or compressive extension (one injury).

The presence of concomitant flexion and extension mechanisms of injury, consistent with the pattern of a buckling deformation, was noted in some specimens. For example, in one specimen (F), there was a bipedicular fracture at the second cervical level (hangman's fracture), which occurred as a result of local extension of the cervical spine, and a concomitant bilateral dislocation of the facets between the sixth and seventh cervical vertebrae, which occurred as a result of local flexion (Fig. 7-A). Similarly, in another specimen (H), a two-part fracture of the anterior ring of the first cervical vertebra was caused by local flexion and concomitant fractures of the spinous processes of the fourth and fifth cervical vertebrae were caused by extension.

In four of the specimens that were subjected to an impact on a rigid surface, a second, more complicated, buckling deformation, lasting two to eight milliseconds, was observed before the flexion-extension mode that has already been described (Figs. 5, Fig. 8, Fig. 9). Analysis of the force or videotape data during the period of this type of deformation showed no evidence of injury in any of these specimens.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Numerous clinical and radiographic classification schemes have been developed to assist in the diagnosis and treatment of injuries to the cervical spine. The mechanisms of injury have also been determined by reconstruction of accidents, analysis of the patient's history, and study of injury in human cadavera. Despite these efforts, clear relationships have yet to be defined among the mechanism of injury, motion of the head, and classification. The analysis and experiments presented here were designed to help to clarify this complex problem.

The clinical relevance of an injury model must be evaluated in order to assess its utility. The tests presented here produced injury in a reliable and repeatable fashion in specimens in which the lordosis of the cervical spine was preserved. All of the injuries that were produced have been observed clinically, and the distribution of injuries is consistent with reported epidemiology^1,6,8,27. Four of the five specimens that had injury to the atlas had additional injury to the cervical spine. This finding is consistent with the experience of Levine and Edwards¹², who reported that additional injury to the cervical spine occurred in nine of seventeen surviving patients who had fracture of the arch of the atlas. Multiple non-contiguous injuries were observed in three of our specimens. In a study by Shear et al.²⁷, six of sixty-six consecutive patients who had traumatic injury to the cervical spine had multiple non-contiguous injuries. Four of these patients had fracture of the first cervical vertebra, and three had fracture of the second cervical vertebra. Shear et al. suggested that the prevalence of multiple non-contiguous injuries may be higher than reported because the suspicion of injury may be reduced after one injury has been identified radiographically. As a result, injuries in the lower cervical spine may be identified, while additional stable fractures of the first or second cervical vertebra remain undetected. In the series of Shear et al., five of the six patients who had multiple non-contiguous injuries had been involved in a motor-vehicle accident. Because multiple non-contiguous injuries are not nearly as common in accidents that occur during athletic and recreational activities, this suggests an association between multiple non-contiguous injuries and impact loading. The proportion of injuries to the upper cervical spine in our study was greater than that seen in the clinical setting. This was because we limited the tests to vertical impacts and because some of the impacts were non-survivable. Vertical impacts and injuries to the upper cervical spine are more common during motor-vehicle accidents. Indeed, Fife and Kraus⁸ reported that injuries to the upper cervical spine occurred in 43 per cent (fifty-nine) of 138 survivors and non-survivors of motor-vehicle accidents. The prevalence in fatal motor-vehicle accidents is even greater. In an epidemiological study by Alker et al.¹, the first or second cervical vertebra was injured in seventeen (55 per cent) of thirty-one victims of fatal motor-vehicle accidents. In a similar study by Bucholz et al.⁶, this prevalence was 50 per cent (twelve of twenty-four victims). These findings may indicate that the vertical impacts in our study represent some of the most severe, and potentially fatal, types of traumatic loading.

The primary limitation of the present study is the inherent lack of active muscles of the neck in cadaver models. Muscles undoubtedly play a role in stabilization and absorption of energy during traumatic events; however, their importance is minimized during compressive loading. This was the rationale for limiting the tests to vertical impacts with the head, neck, and torso aligned in an anatomically neutral position. An additional reason for vertical impacts is that, under these conditions, injuries occur two to three times more quickly than the muscles of the cervical spine react. Previous investigators^9,26 have reported reflex times for these muscles of fifty to sixty-five milliseconds, which is considerably longer than the twenty milliseconds that was needed for the injuries to be produced in the current experiment. It is possible, however, that the muscles play a greater role in impacts involving less compression and more flexion or extension bending (as might occur during impacts to the face or to the occipital region of the head). Also, we do not contend that the results would be the same under non-accidental circumstances, such as when a conditioned athlete uses the head to spear-tackle an opponent³⁰.

The injuries produced in our experiments were not related to the observed motions of the head, as the injuries occurred within twenty milliseconds after impact, which is insufficient time for substantial translation or rotation of the head. Maximum displacement of the head, which has been previously associated with the mechanism of injury^3,5,31, did not occur until at least 130 milliseconds after impact. Therefore, it is not surprising that there was no association between the mechanism of injury and motion of the head. In contrast, the classification of injury was very descriptive of the local deformations in the cervical spine at the instant of injury and of the mechanism of injury. For thirteen of the fourteen injuries that were produced in the second through seventh cervical vertebrae, the mechanism of injury was the extension of the middle portion of the cervical spine and the flexion of the caudad portion that was characteristic of the buckling deformation (Fig. 6). Furthermore, the deformations are consistent with the classification scheme of Allen et al.².

In accidental injury to the neck resulting from impact of the head, the deformations of the cervical spine are so complex that they can give rise to a number of different mechanisms of injury. A purely compressive force on the head can resolve itself into compression with posterior shear at one vertebral level and compression with anterior shear at another (Fig. 10). Consequently, flexion and extension moments can occur simultaneously at different vertebral levels. These moments can result in distractive (tensile) stresses in the anterior or posterior elements of the cervical spine. Hence, a distractive mechanism of injury can be present in a compressive impact of the head. Therefore, these experiments suggest that one should not be biased in favor of classifying a second injury the same as the first; flexion and extension or distractive and compressive mechanisms may occur concomitantly after a single impact.

Buckling of the cervical spine has been postulated to play an important role in injury in well documented clinical studies of football injuries²⁹ and has also been observed in static studies of isolated cervical spines from cadavera^19,23; however, its role in injuries to the cervical spine as a result of impact of the head has not been demonstrated experimentally. The buckling that was observed in the present experiment (Fig. 6) indicates that the neutrally positioned cervical spine assumed a curved shape before injury, which predisposed the third through sixth cervical motion segments to extension-type injuries and the seventh and eighth cervical motion segments, to flexion injuries. Clearly, flexion injuries at the third through sixth cervical levels have been observed in the clinical setting. As suggested clinically by Torg et al.²⁹, and experimentally by others^15,19,20,23, straightening of the cervical spine by flexion before the injury may be a necessary element of the compressive flexion mechanism of injury in this region. These experiments suggest that injuries to the middle portion of the cervical spine may also be produced in a neutrally positioned spine at greater impact velocities. We observed a more complicated buckling shape that produced flexion of this portion of the cervical spine (Figs. 9 and 10); however, this occurred early in the impact interval, when the forces on the spine were small. We postulate that, with greater impact velocities, injurious levels of force may develop before the cervical spine can move from this shape to the characteristic flexion-extension shape that produced the injuries that were observed here.

NOTE: The authors thank the National Disease Resource Interchange.

References

Introduction | Materials and Methods | Results | Discussion | References

Alker, G. J.; Oh, Y. S.; Leslie, E. V.; Lehotay, J.; Panaro, V. A.; and |and |Eschner, E. G.: Postmortem radiology of head and neck injuries in fatal traffic accidents. Radiology,114: 611-617, 1975.114611 1975 [PubMed]

Allen, B. L., Jr.; Ferguson, R. L.; Lehmann, T. R.; and |and |O'Brien, R. P.: A mechanistic classification of closed indirect fractures and dislocations of the lower cervical spine. Spine,7: 1-27, 1982.71 1982 [PubMed][CrossRef]

Babcock, J. L.: Cervical spine injuries. Diagnosis and classification. Arch. Surg.,111: 646-651, 1976.111646 1976 [PubMed]

Bauze, R. J., and |and |Ardran, G. M.: Experimental production of forward dislocation in the human cervical spine. J. Bone and Joint Surg.,60-B(2): 239-245, 1978.60-B(2)239 1978

Braakman, R., and Penning, L.: Injuries of the Cervical Spine, Amsterdam, Excerpta Medica Foundation, 1971.

Bucholz, R. W.; Burkhead, W. Z.; Graham, W.; and |and |Petty, C.: Occult cervical spine injuries in fatal traffic accidents. J. Trauma,19: 768-771, 1979.19768 1979 [PubMed][CrossRef]

Burstein, A. H., Otis, J. C.; and Torg, J. S.: Mechanisms and pathomechanics of athletic injuries to the cervical spine. In Athletic Injuries to the Head, Neck, and Face, pp. 139-154. Edited by J. S. Torg. Philadelphia, Lea and Febiger, 1982.

Fife, D., and |and |Kraus, J.: Anatomic location of spinal cord injury. Relationship to the cause of injury. Spine,11: 2-5, 1986.112 1986 [PubMed][CrossRef]

Foust, D. R.; Chaffin, D. B.; Snyder, R. G.; and Baum, J. K.: Cervical range of motion and dynamic response and strength of cervical muscles. In Proceedings of the 17th Stapp Car Crash Conference, p. 285. Warrendale, Pennsylvania, Society of Automotive Engineers, 1971.

Harris, J. H., Jr.; Edeiken-Monroe, B.; and |and |Kopaniky, D. R.: A practical classification of acute cervical spine injuries. Orthop. Clin. North America,17: 15-30, 1986.1715 1986

Hodgson, V. R., and Thomas, L. M.: Mechanisms of cervical spine injury during impact to the protected head. In Proceedings of the 24th Stapp Car Crash Conference, pp. 17-42. Warrendale, Pennsylvania, Society of Automotive Engineers, 1980.

Levine, A. M., and |and |Edwards, C. C.: Fractures of the atlas. J. Bone and Joint Surg.,73-A: 680-691, June 1991.73-A680 1991

Liu, Y. K., Krieger, K. W., Njus, G., Ueno, N., Connors, M. P., Wakano, K., and Thies, D.: Cervical spine stiffness and geometry of the young human male. Technical Report 80-138. Dayton, Ohio, Air Force Aerospace Medical Research Laboratory, 1982.

McElhaney, J. H., and Myers, B. S.: Biomechanical aspects of cervical trauma. In Accidental Injury, pp. 311-361. Edited by A. M. Nahum and J. W. Melvin. New York, Springer, 1993.

McElhaney, J. H.; Paver, J. G.; McCrackin, H. J.; and Maxwell, G. M.: Cervical spine compression responses. In Proceedings of the 27th Stapp Car Crash Conference, pp. 163-177. Warrendale, Pennsylvania, Society of Automotive Engineers, 1983.

McElhaney, J. H. Doherty, B. J. Paver, J. G. Myers, B. S. and Gray, L.: Combined bending and axial loading responses of the cervical spine. In Proceedings of the 32nd Stapp Car Crash Conference, pp. 21-28. Warrendale, Pennsylvania, Society of Automotive Engineers, 1988.

Melvin, J. W.; Mohan, D.; and |and |Stalnaker, R. L.: Occupant injury mechanisms and impact tolerance. Transport. Res. Rec.,586: 11-22, 1976.58611 1976

Myers, B. S.; McElhaney, J. H.; and |and |Doherty, B. J.: The viscoelastic responses of the human cervical spine in torsion: experimental limitation of quasi-linear theory, and a method for reducing these effects. J. Biomech.,24: 811-817, 1991.24811 1991 [PubMed][CrossRef]

Myers, B. S.; Richardson, W. J.; Doherty, B. J.; Nightingale, R. W.; and McElhaney, J. H.: The role of head constraint in the development of lower cervical compression flexion injury. In Proceedings of the 35th Stapp Car Crash Conference, pp. 391-399. Warrendale, Pennsylvania, Society of Automotive Engineers, 1991.

Nusholtz, G. S.; Huelke, D. E.; Lux, P.; Alem, N. M.; and Montalvo, F.: Cervical spine injury mechanisms. In Proceedings of the 27th Stapp Car Crash Conference, pp. 179-197. Warrendale, Pennsylvania, Society of Automotive Engineers, 1983.

Panjabi, M. M.; White, A. A., III; and |and |Johnson, R. M.: Cervical spine mechanics as a function of transection of components. J. Biomech.,8: 327-336, 1975.8327 1975 [PubMed][CrossRef]

Panjabi, M. M.; Krag, M. H.; Summers, D.; and |and |Videman, T.: Biomechanical time-tolerance of fresh cadaveric human spine specimens. J. Orthop. Res.,3: 292-300, 1985.3292 1985 [PubMed][CrossRef]

Pintar, F. A.; Sances, A., Jr.; Yoganandan, N.; Reinartz, J.; Maiman, D. J.; Suh, J. R.; Unger, G.; Cusick, J. F.; and Larson, S. J.: Biodynamics of the total human cadaveric cervical spine. In Proceedings of the 34th Stapp Car Crash Conference, pp. 55-72. Warrendale, Pennsylvania, Society of Automotive Engineers, 1990.

Portnoy, H. D. McElhaney, J. H. Melvin, J. W. and Croissant, P. D.: Mechanism of cervical spine injury in auto accidents. In Proceedings of the 15th Conference of the American Association for Automotive Medicine, pp. 58-83. Des Plaines, Illinois, American Association for Automotive Medicine, 1972.

Roaf, R.: A study of the mechanics of spinal injuries. J. Bone and Joint Surg.,42-B(4): 810-823, 1960.42-B(4)810 1960

Schneider, L. W.; Foust, D. R.; Bowman, B. M.; Snyder, R. G.; Chaffin, D. B.; Abdelnovr, T. A.; and Baum, J. K.: Biomechanical properties of the human neck in lateral flexion. In Proceedings of the 19th Stapp Car Crash Conference, p. 3212. Warrendale, Pennsylvania, Society of Automotive Engineers, 1975.

Shear, P.; Hugenholtz, H.; Richard, M. T.; Russell, N. A.; Peterson, E. W.; Benoit, B. G.; and |and |da Silva, V. F.: Multiple noncontiguous fractures of the cervical spine. J. Trauma,28: 655-659, 1988.28655 1988 [PubMed][CrossRef]

Smeathers, J. E., and |and |Joanes, D. N.: Dynamic compressive properties of human lumbar intervertebral joints: a comparison between fresh and thawed specimens. J. Biomech.,21: 425-433, 1988.21425 1988 [PubMed][CrossRef]

Torg, J. S.; Vegso, J. J.; O'Neill, M. J.; and |and |Sennett, B.: The epidemiologic, pathologic, biomechanical, and cinematographic analysis of football-induced cervical spine trauma. Am. J. Sports Med.,18: 50-57, 1990.1850 1990 [PubMed][CrossRef]

Torg, J. S.; Pavlov, H.; O'Neill, M. J.; Nichols, C. E., Jr.; and |and |Sennett, B.: The axial load teardrop fracture. A biomechanical, clinical and roentgenographic analysis. Am. J. Sports Med.,19: 355-364, 1991.19355 1991 [PubMed][CrossRef]

White, A. A., III, and Panjabi, M. M.: Clinical Biomechanics of the Spine. Philadelphia, J. B. Lippincott, 1978.

Whitley, J. E., and |and |Forsyth, H. F.: The classification of cervical spine injuries. Am. J. Roentgenol.,83: 633-644, 1960.83633 1960

Yoganandan, N.; Sances, A., Jr.; Maiman, D. J.; Myklebust, J. B.; Pech, P.; and |and |Larson, S. J.: Experimental spinal injuries with vertical impact. Spine,11: 855-860, 1986.11855 1986 [PubMed][CrossRef]

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

Time line showing the relative occurrences of injury to the neck and noticeable (more than 20-degree) motions of the head with respect to the duration of motion of the head and cervical spine.

View Large | Download Slide(.ppt)
Add to My JBJS

Illustration of a more complicated buckling, seen two to eight milliseconds after impact of the head.

View Large | Download Slide(.ppt)
Add to My JBJS

View Larger

TABLE I DATA ON THE SUBJECTS AND INJURIES AND CLASSIFICATION OF INJURIES

Specimen	Sex, Age	Velocity	Angle of Impact Surface	Force*	Time to Injury†	Injuries to Cervical Spine‡	Maximum Displacement of Head
	(Yrs.)	(m/Sec.)	(Degrees)	(N)	(msec)
Rigid impact surface
A	M, 65	2.43	0	—	—	None	Flexion
B	M, 62	3.20	0	-1839	2.2	C1 (2-part fract., post. ring) (CE) and C2 (hangman's fract.) (DE)	Flexion
C	M, 71	3.26	0	-1955	6.5	C1 (3-part commin. fract.) (VC)	Flexion
D	M, 55	3.14	-15	—	—	None	Flexion
E	F, 35	3.28	-15	—	—	None	Flexion
F	M,—	3.26	+15	-1863	6.4	C1 (fract., lateral mass) (VC), C2 (hangman's fract.) (DE), C2-C3 disc and ant. longitudinal lig. (DES1), and C6-C7 (bilat. dislocation perching, facets) (DFS3)	Extension
G	M, 36	3.23	+30	-1552	8.3	C3 (burst fract.) (VCS3), C3-C4 disc and ant. longitudinal lig. (DES1), and C4-C5 ant. longitudinal lig. (DES1)	Extension

Padded impact surface
H	M, 61	3.13	+30	-1632	14.8	C1 (2-part fract., ant. ring) (CF); C4 (fract., spinous process (CES1); C5 (fract., spinous process) (CES1); and C5-C6 disc, L capsular lig., and ant. longitudinal lig. (DES2)	Extension
I	M, 46	3.03	+30	—	—	None	Extension
J	M, 46	3.51	+30	-2240	18.7	C1 (2-part fract., R aspect) (VC), C3-C4 disc and ant. longitudinal lig. (DES1); and C4-C5 disc, ant. longitudinal lig. , and chip of sup. part of body (DES1)	Extension
K	F, 42	3.07	-15	-1011	18.8	C2-C3 disc, ant. longitudinal lig., and chip, C2 (DES1) and C3-C4 disc, ant. longitudinal lig., and C3 chip (DES1)	Flexion

Add to My JBJS

TABLE I DATA ON THE SUBJECTS AND INJURIES AND CLASSIFICATION OF INJURIES

Specimen	Sex, Age	Velocity	Angle of Impact Surface	Force*	Time to Injury†	Injuries to Cervical Spine‡	Maximum Displacement of Head
	(Yrs.)	(m/Sec.)	(Degrees)	(N)	(msec)
Rigid impact surface
A	M, 65	2.43	0	—	—	None	Flexion
B	M, 62	3.20	0	-1839	2.2	C1 (2-part fract., post. ring) (CE) and C2 (hangman's fract.) (DE)	Flexion
C	M, 71	3.26	0	-1955	6.5	C1 (3-part commin. fract.) (VC)	Flexion
D	M, 55	3.14	-15	—	—	None	Flexion
E	F, 35	3.28	-15	—	—	None	Flexion
F	M,—	3.26	+15	-1863	6.4	C1 (fract., lateral mass) (VC), C2 (hangman's fract.) (DE), C2-C3 disc and ant. longitudinal lig. (DES1), and C6-C7 (bilat. dislocation perching, facets) (DFS3)	Extension
G	M, 36	3.23	+30	-1552	8.3	C3 (burst fract.) (VCS3), C3-C4 disc and ant. longitudinal lig. (DES1), and C4-C5 ant. longitudinal lig. (DES1)	Extension

Padded impact surface
H	M, 61	3.13	+30	-1632	14.8	C1 (2-part fract., ant. ring) (CF); C4 (fract., spinous process (CES1); C5 (fract., spinous process) (CES1); and C5-C6 disc, L capsular lig., and ant. longitudinal lig. (DES2)	Extension
I	M, 46	3.03	+30	—	—	None	Extension
J	M, 46	3.51	+30	-2240	18.7	C1 (2-part fract., R aspect) (VC), C3-C4 disc and ant. longitudinal lig. (DES1); and C4-C5 disc, ant. longitudinal lig. , and chip of sup. part of body (DES1)	Extension
K	F, 42	3.07	-15	-1011	18.8	C2-C3 disc, ant. longitudinal lig., and chip, C2 (DES1) and C3-C4 disc, ant. longitudinal lig., and C3 chip (DES1)	Flexion