Author information ►Article notes ►Copyright and License information ►
Chronic radicular symptoms have been documented in whiplash patients, potentially caused by cervical neural tissue compression during an automobile rear crash. Our goals were to determine neural space narrowing of the lower cervical spine during simulated rear crashes with whiplash protection system (WHIPS) and active head restraint (AHR) and to compare these data to those obtained with no head restraint (NHR). We extrapolated our results to determine the potential for cord, ganglion, and nerve root compression.
Our model, consisting of a human neck specimen within a BioRID II crash dummy, was subjected to simulated rear crashes in a WHIPS seat (n = 6, peak 12.0 g and ΔV 11.4 kph) or AHR seat and subsequently with NHR (n = 6, peak 11.0 g and ΔV 10.2 kph with AHR; peak 11.5 g and ΔV 10.7 kph with NHR). Cervical canal and foraminal narrowing were computed and average peak values statistically compared (P < 0.05) between WHIPS, AHR, and NHR.
Average peak canal and foramen narrowing could not be statistically differentiated between WHIPS, AHR, or NHR. Peak narrowing with WHIPS or AHR was 2.7 mm for canal diameter and 1.6 mm, 2.7 mm, and 5.9 mm2 for foraminal width, height and area, respectively.
While lower cervical spine cord compression during a rear crash is unlikely in those with normal canal diameters, our results demonstrated foraminal kinematics sufficient to compress spinal ganglia and nerve roots. Future anti-whiplash systems designed to reduce cervical neural space narrowing may lead to reduced radicular symptoms in whiplash patients.
Rear automobile crashes may injure cervical neural tissues leading to clinically documented radicular symptoms of muscle weakness and paresthesias of the neck, shoulders, upper back, and arms in whiplash patients . Sensory hypersensitivity and hypoesthesia have been observed in both chronic whiplash and radiculopathy patients, which may indicate similar injury mechanisms .
Radiculopathy has been associated with cervical nerve root and ganglion compression injury due to foraminal spondylosis, most commonly caused by uncovertebral osteophytes in the intervertebral foramen . Cervical cord injury without vertebral fracture resulting in the central cord syndrome is most often caused by traumatic neck extension with greater injury severity in those with cervical stenosis . Intervertebral extension and posterior translation cause reduction in the cervical foraminal area and canal diameter and volume, which may exacerbate pain in those with cervical stenosis [5, 6]. These motions may also cause dynamic neural tissue compression during whiplash. An imaging study of cervical specimens demonstrated a 12% decrease in the C5/6 foraminal area with each 1-mm incremental posterior translation of C5 relative to C6 . Another cadaveric study demonstrated that physiologic extension can decrease the cervical foraminal area by as much as 20% . Rear crashes simulated using osteoligamentous neck specimens have demonstrated nonphysiologic intervertebral extension and posterior translation of the lower cervical spine [9–11]. These motions may potentially cause transient neural tissue compression leading to injury and chronic radicular symptoms in whiplash patients.
Biomechanical studies have evaluated the potential for transient cervical neural tissue compression during simulated rear crashes with no head restraint or injury prevention system. Abrupt head extension and posterior shear were applied to an in vivo porcine model and the observed spinal fluid pressure increases were hypothesized to cause cervical spinal ganglia injury, documented histologically . Other studies have determined neural tissue injury potential due to simulated rear crashes of human head–neck specimens. Using custom transducers to determine spinal canal and foraminal area narrowing, the greatest foraminal area reductions occurred at C5/6 and C6/7, but were of insufficient magnitude to cause injury . Another study found that spinal cord injury during rear impact was unlikely in subjects with normal cervical canal diameters; however, those with severe cervical spinal canal stenosis may be at risk . Nonphysiologic foraminal width narrowing was observed at the lower cervical spine, which indicated potential ganglia compression injury .
The goals of this study were to determine neural space narrowing of the lower cervical spine during simulated rear crashes with two distinct anti-whiplash systems, whiplash protection system (WHIPS) and active head restraint (AHR), and to compare these data to those obtained with no head restraint (NHR). We also extrapolated the present results to determine the potential for cord, ganglion, and nerve root compression injuries in individuals with and without cervical stenosis.
Twelve fresh-frozen human osteoligamentous whole cervical spine specimens (occiput-T1) were mounted in resin at the occiput and T1 in neutral posture. Six of the specimens were prepared for rear crashes using WHIPS (2 M, 4 F; average age 86 years; range 79–90 years), while the other six were prepared for rear crashes using AHR and subsequently with NHR (4 M, 2 F; average age 83 years, range 79–90 years). Apart from typical age-related degenerative changes, the specimens did not suffer from any disease that could have affected the osteoligamentous structures. Lightweight motion-tracking flags with 9.5-mm diameter markers were rigidly attached to the anterior aspect of each cervical vertebra. The average weight of each vertebral flag, including marker mass, was 14.5 g.
Simulated rear crashes
The human model of the neck (Fig. 1), described in detail elsewhere , consisted of the neck specimen with its T1 mount rigidly connected to the torso of a rear impact dummy, BioRID II (Humanetics Innovative Solutions, Plymouth, MI, USA), and carrying a custom anthropometric surrogate head (4.6 kg mass; 0.0214 kg m2 sagittal moment of inertia). The head and neck were stabilized using the compressive muscle force replication system, which provides static postural neck stability and passive resistance to intervertebral motions during rear impact and produces a high-speed kinematic response similar to in vivo data . The simulated muscle forces did not model dynamic muscle activation. The muscle force replication was used to maintain the cervical spine posture immediately prior to the crash. The cervical posture with AHR was consistent with in vivo neutral neck alignment , while with WHIPS it consisted of anterior head translation relative to the in vivo neutral posture, thus representing an out-of-position forward posture. The average initial gap between the head and head restraint was 3.5 cm (SD 0.5 cm; range 2.5–3.8 cm) with AHR and 11.9 cm (SD 2.1 cm; range 8.9–15.2 cm) with WHIPS.
Photographs of the human model of the neck and schematic drawings of a whiplash protection system (WHIPS) and b active head restraint (AHR). WHIPS and AHR were activated by torso momentum pressing into the seatback during the crash. WHIPS motions consisted ...
Rear crashes were simulated at average maximum measured horizontal sled accelerations of 12.0 g (ΔV11.4 kph) with WHIPS, 11.0 g (ΔV 10.2 kph) with AHR, and 11.5 g (ΔV 10.7 kph) with NHR. The crash apparatus consisted of an automobile seat mounted on a custom sled in which the model was seated in normal driving posture and secured with a seatbelt (Fig. 1). The WHIPS seat was from a 2005 Volvo XC90 minivan (Volvo Car Corporation, Göteborg, Sweden), while the AHR seat was from a 2006 Kia Sedona minivan (Kia Motors America, Inc, Irvine, CA, USA). The AHR was initially set to midrange gap and height. The crashes with NHR were simulated using the Kia seat with the AHR removed. WHIPS and AHR were activated by torso momentum pressing into the seatback during the crash. WHIPS motions consisted of rearward translation and rotation of the seatback relative to the seat base and plastic deformation of the bilateral WHIPS energy-absorbing components (Fig. 1a). The energy-absorbing components were replaced prior to each subsequent crash (Service Kit, Part #31250443, Volvo Car Corporation, Göteborg, Sweden). The AHR rotated forward via a pivoting mechanism between it and the seatback (Fig. 1b).
A high-speed digital camera recorded the sagittal motions of the flag markers at 500 frames/s (Motion Pro, Redlake, MSAD Inc., San Diego, CA). A contact switch mounted to the back of the head determined the time of contact of the head with the head restraint. A bi-axial accelerometer (50 g capacity; part no. ADXL250JQC, Analog Devices, Norwood, MA, USA) fixed to the sled was continuously sampled at 1 kHz using an analog-to-digital converter and a personal computer. A braking system consisting of air cylinders in parallel gradually decelerated the model following the crash.
Spinal canal and intervertebral foramen geometry and kinematics
The coordinates of the vertebral flag markers were computed with sub-pixel accuracy in the global coordinate system using custom Matlab programs (Matlab, MathWorks, Natick, MA, USA). A radiograph of the lower cervical spine, C4 through C7, which most closely matched average vertebral dimensions [18, 19] and clearly showed anatomical landmarks, was scaled and aligned accordingly and digitally superimposed on the first frame of the high-speed movie (Adobe Photoshop CS2, Version 9, San Jose, CA). Two spinal canal points (Fig. 2a) and six foraminal points (Fig. 2b) were mathematically reconstructed for each functional spinal unit, C4/5 through C6/7, using the radiograph and average quantitative cervical anatomy . The canal diameter was defined as the sagittal distance between the posteroinferior corner of the upper vertebral body and the superior end of the spinolaminar line of the lower vertebra. Points at the anterior, middle, and posterior regions of the foraminal medial zone were used to define three foramen parameters of width, height, and area. The width was the distance between the anterior medial zone of the superior vertebra and the posterior medial zone of the inferior vertebra, the height was the distance between the middle medial zone of adjacent vertebrae, and the area was the total area encompassed by the six foraminal points. We established geometrical rigid body relationships between the flag markers and the canal and foraminal points. For each subsequent movie frame, canal and foramen narrowing and widening were computed using as input vertebral rotation data, flag marker translations, and the geometric rigid body relationships. For subsequent evaluation of the potential for nerve root and ganglia compression during the crashes, the sagittal foraminal width and area narrowing were transformed to determine the width and area narrowing at 45° to the midsagittal plane (Fig. 2b). No transformations of the foraminal height or canal narrowing were performed. The average (SD) system errors, as determined in separate studies, were 0.3 mm (0.2 mm) for canal and foramen narrowing and widening  and 0.00° (0.34°) for vertebral rotation .
Schematic drawings of spinal canal and intervertebral foramen points used to determine narrowing and widening during the rear crashes. aSagittal view of the spinal canal. The spinal canal diameter was defined as the distance from the posteroinferior ...