- Email: [email protected]
Effect of rotated head posture on dynamic vertebral artery elongation during simulated rear impact
Clinical Biomechanics 21 (2006) 213–220 www.elsevier.com/locate/clinbiomech
Effect of rotated head posture on dynamic vertebral artery elongation during simulated rear impact Paul C. Ivancic a, Shigeki Ito b, Yasuhiro Tominaga b, Erik J. Carlson a, Wolfgang Rubin a, Manohar M. Panjabi a,* a
Biomechanics Research Laboratory, Department of Orthopaedics and Rehabilitation, Yale University School of Medicine, 333 Cedar St., P.O. Box 208071, New Haven, CT 06520-8071, USA b Department of Orthopaedic Surgery, St. Marianna University School of Medicine, Kanagawa, Japan Received 30 August 2005; accepted 26 October 2005
Abstract Background. Elongation-induced vertebral artery injury has been hypothesized to occur during non-physiological coupled axial rotation and extension of head. No studies have quantified dynamic vertebral artery elongation during head-turned rear impacts. Therefore, we evaluated effect of rotated head posture vs. forward head posture at the time of impact on dynamic vertebral artery elongation during simulated rear impacts. Methods. A whole cervical spine model with surrogate head and muscle force replication underwent either simulated head-turned (n = 6) or head-forward (n = 6) rear impacts of 3.5, 5, 6.5 and 8 g. Continuous dynamic vertebral artery elongation was recorded using custom transducer and compared to physiological values obtained during intact flexibility testing. Findings. Average (SD) peak dynamic vertebral artery elongation of up to 30.5 (2.6) mm during head-turned rear-impact significantly exceeded (P < 0.05) the physiological beginning at 5 g. Highest peak elongation of 5.8 (2.1) mm during head-forward rear impact did not exceed physiological limit. Head-turned rear impact caused earlier occurrence of average peak vertebral artery elongation, 84.5 (4.2) ms vs. 161.0 (43.8) ms, and higher average peak vertebral artery elongation rate, 1336.7 (74.5) mm/s vs. 211.5 (97.4) mm/s, as compared to head-forward rear impact. Interpretation. Elongation-induced vertebral artery injury is more likely to occur in those with rotated head posture at the time of rear impact, as compared to head-forward. 2005 Elsevier Ltd. All rights reserved. Keywords: Vertebral artery; Whiplash biomechanics; Rear impact; Cervical spine
1. Introduction Chronic symptoms of headaches, blurred vision, tinnitus, dizziness, and vertigo have been documented in whiplash patients (Spitzer et al., 1995), however their exact etiology remains poorly understood. Clinical studies have associated these symptoms with altered blood flow rates due to spasm and/or narrowing of vertebral
*
Corresponding author. E-mail address: [email protected] (M.M. Panjabi).
0268-0033/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2005.10.011
arteries (VAs) in victims of high velocity automobile collisions with or without blunt head impact, many of whom sustained severe cranio-cervical injuries or cervical spine fractures (Reddy et al., 2002; Seric et al., 2000). VA injuries resulting in vascular compromise have been documented clinically due to a variety of trauma scenarios from whiplash (Chung and Han, 2002; Taneichi et al., 2005) to blunt-force head injury (Carpenter, 1961; McLean et al., 1985), and also due to activities of daily living such as fitness exercises (DeBehnke and Brady, 1994), painting (Okawara and Nibbelink, 1974) and coughing (Herr et al., 1992), and
214
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
chiropractic spinal therapy (Schellhas et al., 1980). Researchers have hypothesized that coupled cervical spine extension and axial rotation beyond physiological limits may cause elongation-induced VA injury and vascular compromise, particularly at the upper cervical spine (Barton and Margolis, 1975; Chung and Han, 2002; Davis and Zimmerman, 1983; Sherman et al., 1981). VA injury, often a partial tear in the vesselÕs intimal layer, and resulting vascular compromise, may cause symptoms of vertebrobasilar insufficiency in whiplash victims with severe spinal injuries such as craniocervical injuries or bony fractures or dislocations. Previous biomechanical studies have evaluated the potential for elongation-induced VA injury during simulated whiplash. Nibu et al. (1997) and Panjabi et al. (1998b) measured dynamic VA elongation between occiput and C6 using a custom-designed transducer in a whole cervical spine model with surrogate head during simulated rear impacts up to 8.5 g. Statistically significant increases in dynamic VA elongation above physiological limits were observed beginning at 4.5 g, with the highest average peak elongation exceeding 15 mm at 8.5 g. While not able to determine the precise location of VA elongation within the cervical spine, the authors noted that if the elongation occurred over a short length at the upper cervical spine, then VA injury may occur. Previous biomechanical studies have been limited to evaluation of elongation-induced VA injury due to head-forward rear impacts. No previous biomechanical studies have investigated the effects of rotated head posture at the time of rear impact on dynamic VA elongation. The goal of the present study was to evaluate the relative effect of rotated head posture vs. forward head posture at the time of impact on dynamic VA elongation during simulated rear impacts using a biofidelic whole cervical spine model with muscle force replication and surrogate head.
changes, the donors did not suffer from any disease or trauma that could have affected the osteoligamentous structures. A thin nylon-coated steel cable constituted the VA cable and was passed through the right and left VAs of the specimens prepared for head-turned rear impact, and through the right VA of the specimens prepared for head-forward rear impact. The superior ends of the cables were attached to the occipital bone with a screw and the inferior ends to separate vertebral artery transducers (VATs) after existing the C6 transverse foramen. This constituted the whole cervical spine model ready for intact three-plane flexibility testing. 2.2. Vertebral artery transducer design
2. Methods
The VAT (Fig. 1B and C) consisted of two main parts: frame (a) with a horizontal slot containing a rigidly fixed Hall effect sensor (b) (A3506LU, Allegro Microsystems, Worcester, MA, USA), and movable carriage (c) containing two rare earth magnets (d) (13 · 13 · 5 mm, part no. PR28ES4187B, Dexter Magnetic, Billerica, MA, USA). The carriage, connected to VA cable (e) at one end and to a tension spring (f) at the other end, moved smoothly within the horizontal slot of the frame with minimal friction. The magnets, separated by 40 mm, were aligned with the Hall effect sensor and glued to the carriage using cyanoacrylate. The Hall effect sensor provided an output voltage proportional to the magnetic field strength and required no amplification. The excitation voltage, the size of the magnets, and their separation generated a magnetic field gradient such that the range of the Hall effect sensor output was from 0 to 5 V. The VATs were calibrated using a precision micrometer (resolution 0.0001 mm, Model #18010, Oriel Corporation, Stamford, CT, USA). The exact displacement vs. voltage data were fit to sixthorder polynomial calibration curves. The average difference between the exact displacement and the calibration curves over the entire measurement range of 32 mm was 0.00 mm (SD 0.04 mm) for the left and 0.01 mm (SD 0.09 mm) for the right VAT.
2.1. Specimen preparation
2.3. Physiological VA elongation
Twelve fresh-frozen human osteoligamentous whole cervical spine specimens (occiput-T1) were mounted in resin (Polyester Fiberglass Resin, Bondo Corporation, Atlanta, GA) at the occiput and T1 in normal neutral posture with the foramen magnum parallel to the occipital mount, and the T1 vertebra tilted anteriorly by 24 (Fig. 1A). Six of the specimens were prepared for head-turned rear impact (average age 80.2 years; range: 79–93 years), while the other six were prepared for headforward rear impact (average age 70.8 years; range: 52– 84). There were four male and two female donors in each group. Apart from typical age-related degenerative
Three-plane flexibility testing was used to determine physiological VA elongation of the intact specimens. Pure moments were applied to the occipital mount via loading jig in four equal steps up to peak loads of 1.5, 3.0, and 1.5 Nm, in flexion–extension, axial rotation, and lateral bending, respectively, while the T1 mount remained fixed. The weights of the loading jig and occipital mount were counterbalanced during the tests. After two preconditioning cycles, VA elongation data were recorded at each load increment of the third loading cycle. To allow for viscoelastic creep, 30 s break periods were given following each load application.
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
215
Fig. 1. (A) Schematics of the biofidelic whole cervical spine model with surrogate head and muscle force replication used to simulate whiplash, showing position of vertebral artery transducer (VAT) and its cable passing through the VA foramen and attaching to the occiput. (B) Lateral and top schematic views of the VAT: frame (a), Hall effect sensor (b), movable carriage (c) carrying two rare earth magnets (d), VA cable (e), and tension spring (f). (C) Photograph of the VAT with VA cable at upper left.
2.4. The biofidelic whiplash model To prepare the specimen for impact, a custom surrogate head (mass 3.3 kg and sagittal, horizontal, and frontal plane moments of inertia of 0.019, 0.014, 0.015 kg m2, respectively) (Becker, 1972; Walker et al., 1973) was rigidly attached to the occipital mount. The spine and surrogate head were stabilized using the compressive muscle force replication (MFR) system (Fig. 1A) (Ivancic et al., 2005). The MFR system was symmetric about the mid-sagittal plane and consisted of two anterior, two posterior, and eight lateral cables anchored to preloaded springs. The stiffness coefficients of the anterior, lateral and posterior springs were 4.0, 4.0, and 8.0 N/ mm, respectively. The anterior cables originated at the occipital mount, ran through guideposts at C4, through pulleys within the T1 mount and were each connected to a separate spring. The preload in each anterior spring was 15 N. To apply the posterior MFR, wire loops were inserted into the spinal canal through the laminae (C2
through C7) and tightly secured above each vertebral spinous process. The two posterior MFR cables originated from the occipital mount, ran through the wire loops, through a pulley at the T1 mount and each were connected to a spring, preloaded to 15 N. Bilateral MFR cables originated from C0, C2, C4, and C6, passed alternately along lateral guide rods, ran through pulleys at the T1 mount, and were each attached to a spring, preloaded to 30 N. With this MFR arrangement the compressive neutral posture preloads at each intervertebral level were: 120 N (C0–C1, C1–C2); 180 N (C2–C3, C3–C4); 240 N (C4–C5, C5–C6) and 300 N (C6–C7, C7–T1). The whole cervical spine with MFR and surrogate head was used to simulate head-turned and head-forward rear impacts. The model has been validated against in vivo simulated whiplash data (Ivancic et al., 2005). Prior to impact of the head-turned specimens, the head was rotated such that the average head-T1 rotation relative to the neutral posture was 28.4 of left axial rotation, 17.9 of left lateral bending, and 3.5 of flexion.
216
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
2.5. Rear impact and dynamic VA elongation In a bench-top apparatus (Panjabi et al., 1998a), rear impacts were performed using the incremental trauma approach at nominal T1 maximum horizontal accelerations of 3.5, 5, 6.5, and 8 g (Ghole et al., 2004). The major components of the bench-top apparatus included the sled, containing the rigidly attached T1 mount, impacting mass, and acceleration and braking systems. VA elongation data were continuously sampled at 1 kHz using an analog-to-digital converter and a personal computer. 2.6. Data analyses Dynamic VA elongation data were digitally filtered using a third order, dual pass, Butterworth low-pass filter at a cutoff frequency of 30 Hz. Residual and Fourier analyses demonstrated that most of the signal power was contained under 20 Hz. VA elongation rate was computed by numerically differentiating the dynamic VA elongation with respect to time. The peak VA elongation, peak VA elongation rate, and their times of occurrence relative to the onset of T1 horizontal acceleration were determined for each impact. Physiological VA elongation was defined as the maximum elongation obtained during the intact flexibility testing in any of the three planes. 2.6.1. Statistics Single factor, repeated measures ANOVA and Bonferonni post hoc tests were performed to determine increases in peak VA elongation during impact above physiological values. Single factor, non-repeated measures ANOVA and pair-wise Bonferonni post hoc tests were performed to determine differences in peak elongation rate, and times of peak elongation and peak elongation rate among head-forward VA, and head-turned right and left VA. Significance for all statistical tests was set at P < 0.05.
3. Results Representative results were chosen to exemplify dynamic VA elongations during head-forward (specimen #3) and head-turned (specimen #2) rear impacts, both at 8 g impact acceleration. Following the onset of the T1 horizontal acceleration of the head-forward specimen, head extension occurred causing minimal VA elongation, reaching a peak of only 4.1 mm at 186 ms (Fig. 2A). In sharp contrast, rotated head posture at the time of rear impact caused dramatic effects on VA elongation peaks and timing (Fig. 2B). During headturned rear impact, coupled head rotations were observed, consisting of extension, right axial rotation,
and left followed by right lateral bending. Peak left and right VA elongations of 29.1 mm and 26.5 mm, respectively, were achieved between 75 ms and 85 ms following impact, prior to peak head extension. Average maximum physiological VA elongations measured during intact flexibility tests were 7.3 mm (left VA of head-turned specimens), 7.5 mm (right VA of head-turned specimens), and 8.1 mm (head-forward specimens) (Fig. 3). Significant increases (P < 0.05) in dynamic VA elongation above physiological were observed in both left and right VA due to head-turned rear impact beginning at 5 g, while no significant increases due to head-forward rear impact were observed. The highest average dynamic VA elongation peaks of 30.5 mm (left VA) and 27.3 mm (right VA) occurred during the 6.5 g and 8 g head-turned rear impacts, respectively, and were over four times greater than the highest head-forward rear impact peak value. Temporal analyses demonstrated that the average times of peak dynamic left and right VA elongation due to head-turned rear impact occurred significantly earlier than corresponding times during head-forward rear impact for all impact accelerations (Table 1). Peak dynamic VA elongation in the left VA was attained as early as 84.5 ms following 8 g head-turned rear impact. Average peak left and right VA elongation rates were significantly greater with head-turned as compared to head-forward for all impacts with the exception of the right VA at 3.5 g and 5 g (Table 2A). Average peak elongation rate of the left VA significantly exceeded that of the right VA during 5 g and 6.5 g head-turned rear impacts. The highest peak VA elongation rate of 1336.7 mm/s occurred in the left VA during 8 g headturned rear impact. Average times of peak VA elongation rates occurred significantly earlier in both left and right VA with head-turned as compared to head-forward during all impacts except 6.5 g (Table 2B).
4. Discussion Clinical and epidemiological studies have documented chronic symptoms in whiplash patients including headaches, blurred vision, tinnitus, dizziness, and vertigo (Spitzer et al., 1995). Using a custom transducer, the present study has documented dynamic vertebral artery (VA) elongation during simulated head-turned and head-forward rear impacts of a whole cervical spine model with muscle force replication and surrogate head. Dynamic VA elongation of up to 30.5 mm due to headturned rear impact significantly exceeded physiological values beginning at 5 g, while VA elongation during head-forward rear impact remained within physiological limits (Fig. 3). Peak VA elongation occurred as early as 84.5 ms following head-turned rear impact (Table 1). A recent study measured the electromyographic activity of
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
217 70
1. T1 Horizontal Acceleration 2. Head-T1 Flexion-Extension: Rx
25
40 15 30 10
VA Elongation
5
10
1
0
0
-5
-10
2
-20
-10
-30 -15 -40 -20
-50
-25
-60
0
A
50
100
150
200
-70 250
Time (ms) 30
VA Elongation (mm) and T1 Horizontal Acceleration (g)
20
Head-T1 Flexion-Extension: Rx (Degrees)
50
20
-30
25
Left VA Elongation Right VA Elongation
1. T1 Horizontal Acceleration 2. Head-T1 Flexion-Extension: Rx 3. Head-T1 Axial Rotation: Ry 4. Head-T1 Lateral Bending: Rz
20
70 60 50 40
15 30 10
3
20
5
10 1
0
0 -10
2
-5
-20
-10 4
-30
-15 -40 -20
-50
-25
-60
-30 0
B
60
Head-T1 Rotations: Rx, Ry, Rz (Degrees)
VA Elongation (mm) and T1 Horizontal Acceleration (g)
30
50
100
150
200
-70 250
Time (ms)
Fig. 2. Representative examples of dynamic vertebral artery (VA) elongation during (A) head-forward (specimen #3) and (B) head-turned (specimen #2) rear impact, both at 8 g impact acceleration. Head-T1 rotations and T1 horizontal acceleration are also shown. Directions of head-T1 rotation are: +Rx = flexion, Rx = extension, +Ry = left axial rotation, Ry = right axial rotation, +Rz = right lateral bending, Rz = left lateral bending.
the cervical muscles during simulated head-turned rear impacts up to 1.3 g and found that peak muscle activity was achieved no earlier than 192 ms following the impact (Kumar et al., 2005). Thus, VA injury during head-turned rear impact may occur prior to neuromuscular protective mechanisms achieved via peak muscle tension. In contrast, peak VA elongation during headforward rear impact occurred significantly later. Peak VA elongation rates were greater and occurred earlier with head-turned as compared to head-forward rear impact (Table 2A and B). These cumulative results dem-
onstrate that elongation-induced VA injury is more likely to occur in those with rotated head posture at the time of rear impact, as compared to head forward. The limitations of this study must be considered before interpreting the present results. The custom transducers measured the total VA elongations between occiput and C6, thus, it was not possible to determine the VA strain distribution throughout its length. Dynamic VA elongation was calculated relative to neutral posture, assuming no initial VA laxity. Any initial VA laxity would result in overestimation of dynamic
218
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220 35 Head-Forward
Head-Turned Right VA
Head-Turned Left VA
30
* P<0.05 Peak VA Elongation (mm)
* 25
* * *
20
* 15
* 10
5
0 Physiological
3.5 g
5g
6.5 g
8g
Impact Acceleration
Fig. 3. Average peak vertebral artery (VA) elongations (mm) during head-forward and head-turned rear impacts. Significant increases (P < 0.05) in peak VA elongation, relative to physiological are indicated with asterisks.
Table 1 Average (SD) times (ms) at which the dynamic vertebral artery (VA) elongation peaks were attained, relative to onset of T1 horizontal acceleration, during head-forward and head-turned rear impacts
Significantly different times (P < 0.05) from the pair-wise comparisons are indicated with brackets.
Table 2 Average (SD) (A) peak vertebral artery elongation rates (mm/s) and (B) times of occurrence relative to onset of T1 horizontal acceleration during head-forward and head-turned rear impacts
Significant differences (P < 0.05) from the pair-wise comparisons are indicated with brackets.
VA elongation. Average ages of present head-turned and head-forward rear impact specimens were 80.2 and 70.8 years, respectively, and their spinal mechanical properties were stiffer than those of the younger population most likely to suffer whiplash trauma. Thus, peak dynamic VA elongation data of the present study are
conservative, as compared to the younger population. The muscle force replication system provided passive resistance to cervical motion and did not include the active neuromuscular response, thus simulating the response of an unwarned subject. The incremental trauma approach was used to determine the relation
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
between dynamic VA elongation and impact acceleration. A previous comparative study of incremental and single trauma protocols reported equivalent subfailure ligamentous injury severity, thus justifying the incremental trauma approach (Ghole et al., 2004). In previous biomechanical studies, Nibu et al. (1997) and Panjabi et al. (1998b) documented dynamic VA elongation during simulated head-forward rear impacts of the whole cervical spine model with surrogate head, without muscle force replication. Average peak dynamic VA elongation increased with increasing impact acceleration, analogous to the results of the present study; however, there were some differences. The highest average peak VA elongation of 5.8 mm due to head-forward rear impact of the present model was much less than previously reported highest peak, which exceeded 15 mm. There are two reasons for this difference. First, the muscle force replication used in the present model most likely reduced VA elongation by stabilizing the cervical spine. Second, the previous study reported non-zero VA elongation in neutral posture, which served to increase dynamic peak VA elongation. In contrast, the present study reported dynamic VA elongation relative to zero elongation in neutral posture. The exact mechanism by which VA injury occurs in the cervical spine remains unknown. Researchers have hypothesized that non-physiological coupled extension and axial rotation of the head may cause VA injury, particularly at the upper cervical spine due to its high mobility (Barton and Margolis, 1975; Chung and Han, 2002; Davis and Zimmerman, 1983; Sherman et al., 1981). Coupled cervical spine rotations observed during head-turned rear impact of the present study caused dynamic hyper-elongation of the left and right VAs. VA elongation causes a decrease in its diameter due to PoissonÕs effect and may lead to transient vascular compromise (Dobrin, 1978). In addition, while being elongated, the VA may be pinched at a bend of its circuitous anatomical route. The latter mechanism can precipitate VA injury, often manifested as a tear in the intimal layer, which has been clinically documented primarily at C1–C2 (Chung and Han, 2002). Severe VA injury results in acute vascular compromise caused by arterial dissection with possible rupture or thrombosis leading to stroke (Stahmer et al., 1997). The exact relationship between VA mechanical properties and elongation rate is unknown. The VA is a viscoelastic soft tissue structure; the adventitia is composed primarily of collagen fibers and the media consists of collagen as well as more substantial portion of muscular and elastic fibers (Fung, 1993). There are no previous studies that have documented the mechanical characteristics of the VA at elongation rates of the present study. However, a previous biomechanical study of cervical alar and transverse ligaments, viscoelastic structures which are composed primarily of collagen with some
219
elastin (Dvorak et al., 1988), documented a decrease in failure strain at high elongation rate of 920 mm/s as compared to 0.1 mm/s (Panjabi et al., 1998b). In the present study, head-turned rear impact caused highest peak elongation rate in the left VA of 1336.7 mm/s, significantly greater than the corresponding head-forward peak rate of only 211.5 mm/s. Assuming that the VA failure strain decreases with increased elongation rate, the high elongation rate observed due to head-turned rear impact may further accentuate elongation-induced injury of the VA.
5. Conclusions In conclusion, the present study, using a biofidelic whole cervical spine model with muscle force replication and surrogate head in simulated head-forward and head-turned rear impacts, has demonstrated potential elongation-induced VA injury due to head-turned rear impact. In contrast, VA injury during head-forward rear impact is unlikely. VA injury during head-turned rear impact may occur prior to peak muscle tension, indicating that the neuromuscular control system may not be able to protect the VAs from injury. The high VA elongation rate observed during head-turned rear impact may cause a decreased failure strain, as compared to head-forward. Present results cumulatively demonstrate that elongation-induced VA injury is more likely to occur during head-turned rear impact, as compared to forward facing.
Acknowledgment This research was supported by NIH Grant 1 RO1 AR45452 1A2.
References Barton, J.W., Margolis, M.T., 1975. Rotational obstructions of the vertebral artery at the atlantoaxial joint. Neuroradiology 9, 117–120. Becker, E., 1972. Measurement of mass distribution parameters of anatomic segments. Society of Automotive Engineers, Paper No. 720964. Carpenter, S., 1961. Injury of neck as cause of vertebral artery thrombosis. J. Neurosurg. 18, 849–853. Chung, Y.S., Han, D.H., 2002. Vertebrobasilar dissection: a possible role of whiplash injury in its pathogenesis. Neurol. Res. 24, 129–138. Davis, J.M., Zimmerman, R.A., 1983. Injury of the carotid and vertebral arteries. Neuroradiology 25, 55–69. DeBehnke, D.J., Brady, W., 1994. Vertebral artery dissection due to minor neck trauma. J. Emerg. Med. 12, 27–31. Dobrin, P.B., 1978. Mechanical properties of arteries. Physiol. Rev. 58, 397–460. Dvorak, J., Schneider, E., Saldinger, P., Rahn, B., 1988. Biomechanics of the craniocervical region: the alar and transverse ligaments. J. Orthop. Res. 6, 452–461.
220
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
Fung, Y.C., 1993. Mechanical properties and active remodeling of blood vessels. Biomechanics: Mechanical Properties of Living Tissues, second ed. Springer-Verlag, New York, pp. 321–391. Ghole, S.A., Ivancic, P.C., Tominaga, Y., Gimenez, S.E., Panjabi, M.M., 2004. Incremental and single trauma produce equivalent subfailure soft tissue injury of the cervical spine. Clin. Biomech. 19, 784–789. Herr, R.D., Call, G., Banks, D., 1992. Vertebral artery dissection from neck flexion during paroxysmal coughing. Ann. Emerg. Med. 21, 88–91. Ivancic, P.C., Panjabi, M.M., Ito, S., Cripton, P.A., Wang, J.L., 2005. Biofidelic whole cervical spine model with muscle force replication for whiplash simulation. Eur. Spine J. 14, 346–355. Kumar, S., Ferrari, R., Narayan, Y., 2005. Looking away from whiplash: effect of head rotation in rear impacts. Spine 30, 760– 768. McLean, J.M., Wright, R.M., Henderson, J.P., Lister, J.R., 1985. Vertebral artery rupture associated with closed head injury. Report of two cases. J. Neurosurg. 62, 135–138. Nibu, K., Cholewicki, J., Panjabi, M.M., Babat, L.B., Grauer, J.N., Kothe, R., Dvorak, J., 1997. Dynamic elongation of the vertebral artery during an in vitro whiplash simulation. Eur. Spine J. 6, 286– 289. Okawara, S., Nibbelink, D., 1974. Vertebral artery occlusion following hyperextension and rotation of the head. Stroke 5, 640–642. Panjabi, M.M., Cholewicki, J., Nibu, K., Babat, L.B., Dvorak, J., 1998a. Simulation of whiplash trauma using whole cervical spine specimens. Spine 23, 17–24. Panjabi, M.M., Crisco 3rd, J.J., Lydon, C., Dvorak, J., 1998b. The mechanical properties of human alar and transverse
ligaments at slow and fast extension rates. Clin. Biomech. 13, 112–120. Reddy, M., Reddy, B., Schoggl, A., Saringer, W., Matula, C., 2002. The complexity of trauma to the cranio-cervical junction: correlation of clinical presentation with Doppler flow velocities in the V3-segment of the vertebral arteries. Acta Neurochir. (Wien) 144, 575–580, discussion 580. Schellhas, K.P., Latchaw, R.E., Wendling, L.R., Gold, L.H., 1980. Vertebrobasilar injuries following cervical manipulation. JAMA 244, 1450–1453. Seric, V., Blazic-Cop, N., Demarin, V., 2000. Haemodynamic changes in patients with whiplash injury measured by transcranial Doppler sonography (TCD). Coll. Antropol. 24, 197–204. Sherman, D.G., Hart, R.G., Easton, J.D., 1981. Abrupt change in head position and cerebral infarction. Stroke 12, 2–6. Spitzer, W.O., Skovron, M.L., Salmi, L.R., Cassidy, J.D., Duranceau, J., Suissa, S., Zeiss, E., 1995. Scientific monograph of the Quebec Task Force on Whiplash-Associated Disorders: redefining ‘‘whiplash’’ and its management. Spine 20, 1S–73S. Stahmer, S.A., Raps, E.C., Mines, D.I., 1997. Carotid and vertebral artery dissections. Emerg. Med. Clin. North Am. 15, 677–698. Taneichi, H., Suda, K., Kajino, T., Kaneda, K., 2005. Traumatically induced vertebral artery occlusion associated with cervical spine injuries: prospective study using magnetic resonance angiography. Spine 30, 1955–1962. Walker, L., Harris, E., Pontius, U., 1973. Mass, volume, center of mass, and mass moment of inertia of head and neck of human body. Society of Automotive Engineers, Paper No. 730985.
Effect of rotated head posture on dynamic vertebral artery elongation during simulated rear impact Paul C. Ivancic a, Shigeki Ito b, Yasuhiro Tominaga b, Erik J. Carlson a, Wolfgang Rubin a, Manohar M. Panjabi a,* a
Biomechanics Research Laboratory, Department of Orthopaedics and Rehabilitation, Yale University School of Medicine, 333 Cedar St., P.O. Box 208071, New Haven, CT 06520-8071, USA b Department of Orthopaedic Surgery, St. Marianna University School of Medicine, Kanagawa, Japan Received 30 August 2005; accepted 26 October 2005
Abstract Background. Elongation-induced vertebral artery injury has been hypothesized to occur during non-physiological coupled axial rotation and extension of head. No studies have quantified dynamic vertebral artery elongation during head-turned rear impacts. Therefore, we evaluated effect of rotated head posture vs. forward head posture at the time of impact on dynamic vertebral artery elongation during simulated rear impacts. Methods. A whole cervical spine model with surrogate head and muscle force replication underwent either simulated head-turned (n = 6) or head-forward (n = 6) rear impacts of 3.5, 5, 6.5 and 8 g. Continuous dynamic vertebral artery elongation was recorded using custom transducer and compared to physiological values obtained during intact flexibility testing. Findings. Average (SD) peak dynamic vertebral artery elongation of up to 30.5 (2.6) mm during head-turned rear-impact significantly exceeded (P < 0.05) the physiological beginning at 5 g. Highest peak elongation of 5.8 (2.1) mm during head-forward rear impact did not exceed physiological limit. Head-turned rear impact caused earlier occurrence of average peak vertebral artery elongation, 84.5 (4.2) ms vs. 161.0 (43.8) ms, and higher average peak vertebral artery elongation rate, 1336.7 (74.5) mm/s vs. 211.5 (97.4) mm/s, as compared to head-forward rear impact. Interpretation. Elongation-induced vertebral artery injury is more likely to occur in those with rotated head posture at the time of rear impact, as compared to head-forward. 2005 Elsevier Ltd. All rights reserved. Keywords: Vertebral artery; Whiplash biomechanics; Rear impact; Cervical spine
1. Introduction Chronic symptoms of headaches, blurred vision, tinnitus, dizziness, and vertigo have been documented in whiplash patients (Spitzer et al., 1995), however their exact etiology remains poorly understood. Clinical studies have associated these symptoms with altered blood flow rates due to spasm and/or narrowing of vertebral
*
Corresponding author. E-mail address: [email protected] (M.M. Panjabi).
0268-0033/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2005.10.011
arteries (VAs) in victims of high velocity automobile collisions with or without blunt head impact, many of whom sustained severe cranio-cervical injuries or cervical spine fractures (Reddy et al., 2002; Seric et al., 2000). VA injuries resulting in vascular compromise have been documented clinically due to a variety of trauma scenarios from whiplash (Chung and Han, 2002; Taneichi et al., 2005) to blunt-force head injury (Carpenter, 1961; McLean et al., 1985), and also due to activities of daily living such as fitness exercises (DeBehnke and Brady, 1994), painting (Okawara and Nibbelink, 1974) and coughing (Herr et al., 1992), and
214
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
chiropractic spinal therapy (Schellhas et al., 1980). Researchers have hypothesized that coupled cervical spine extension and axial rotation beyond physiological limits may cause elongation-induced VA injury and vascular compromise, particularly at the upper cervical spine (Barton and Margolis, 1975; Chung and Han, 2002; Davis and Zimmerman, 1983; Sherman et al., 1981). VA injury, often a partial tear in the vesselÕs intimal layer, and resulting vascular compromise, may cause symptoms of vertebrobasilar insufficiency in whiplash victims with severe spinal injuries such as craniocervical injuries or bony fractures or dislocations. Previous biomechanical studies have evaluated the potential for elongation-induced VA injury during simulated whiplash. Nibu et al. (1997) and Panjabi et al. (1998b) measured dynamic VA elongation between occiput and C6 using a custom-designed transducer in a whole cervical spine model with surrogate head during simulated rear impacts up to 8.5 g. Statistically significant increases in dynamic VA elongation above physiological limits were observed beginning at 4.5 g, with the highest average peak elongation exceeding 15 mm at 8.5 g. While not able to determine the precise location of VA elongation within the cervical spine, the authors noted that if the elongation occurred over a short length at the upper cervical spine, then VA injury may occur. Previous biomechanical studies have been limited to evaluation of elongation-induced VA injury due to head-forward rear impacts. No previous biomechanical studies have investigated the effects of rotated head posture at the time of rear impact on dynamic VA elongation. The goal of the present study was to evaluate the relative effect of rotated head posture vs. forward head posture at the time of impact on dynamic VA elongation during simulated rear impacts using a biofidelic whole cervical spine model with muscle force replication and surrogate head.
changes, the donors did not suffer from any disease or trauma that could have affected the osteoligamentous structures. A thin nylon-coated steel cable constituted the VA cable and was passed through the right and left VAs of the specimens prepared for head-turned rear impact, and through the right VA of the specimens prepared for head-forward rear impact. The superior ends of the cables were attached to the occipital bone with a screw and the inferior ends to separate vertebral artery transducers (VATs) after existing the C6 transverse foramen. This constituted the whole cervical spine model ready for intact three-plane flexibility testing. 2.2. Vertebral artery transducer design
2. Methods
The VAT (Fig. 1B and C) consisted of two main parts: frame (a) with a horizontal slot containing a rigidly fixed Hall effect sensor (b) (A3506LU, Allegro Microsystems, Worcester, MA, USA), and movable carriage (c) containing two rare earth magnets (d) (13 · 13 · 5 mm, part no. PR28ES4187B, Dexter Magnetic, Billerica, MA, USA). The carriage, connected to VA cable (e) at one end and to a tension spring (f) at the other end, moved smoothly within the horizontal slot of the frame with minimal friction. The magnets, separated by 40 mm, were aligned with the Hall effect sensor and glued to the carriage using cyanoacrylate. The Hall effect sensor provided an output voltage proportional to the magnetic field strength and required no amplification. The excitation voltage, the size of the magnets, and their separation generated a magnetic field gradient such that the range of the Hall effect sensor output was from 0 to 5 V. The VATs were calibrated using a precision micrometer (resolution 0.0001 mm, Model #18010, Oriel Corporation, Stamford, CT, USA). The exact displacement vs. voltage data were fit to sixthorder polynomial calibration curves. The average difference between the exact displacement and the calibration curves over the entire measurement range of 32 mm was 0.00 mm (SD 0.04 mm) for the left and 0.01 mm (SD 0.09 mm) for the right VAT.
2.1. Specimen preparation
2.3. Physiological VA elongation
Twelve fresh-frozen human osteoligamentous whole cervical spine specimens (occiput-T1) were mounted in resin (Polyester Fiberglass Resin, Bondo Corporation, Atlanta, GA) at the occiput and T1 in normal neutral posture with the foramen magnum parallel to the occipital mount, and the T1 vertebra tilted anteriorly by 24 (Fig. 1A). Six of the specimens were prepared for head-turned rear impact (average age 80.2 years; range: 79–93 years), while the other six were prepared for headforward rear impact (average age 70.8 years; range: 52– 84). There were four male and two female donors in each group. Apart from typical age-related degenerative
Three-plane flexibility testing was used to determine physiological VA elongation of the intact specimens. Pure moments were applied to the occipital mount via loading jig in four equal steps up to peak loads of 1.5, 3.0, and 1.5 Nm, in flexion–extension, axial rotation, and lateral bending, respectively, while the T1 mount remained fixed. The weights of the loading jig and occipital mount were counterbalanced during the tests. After two preconditioning cycles, VA elongation data were recorded at each load increment of the third loading cycle. To allow for viscoelastic creep, 30 s break periods were given following each load application.
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
215
Fig. 1. (A) Schematics of the biofidelic whole cervical spine model with surrogate head and muscle force replication used to simulate whiplash, showing position of vertebral artery transducer (VAT) and its cable passing through the VA foramen and attaching to the occiput. (B) Lateral and top schematic views of the VAT: frame (a), Hall effect sensor (b), movable carriage (c) carrying two rare earth magnets (d), VA cable (e), and tension spring (f). (C) Photograph of the VAT with VA cable at upper left.
2.4. The biofidelic whiplash model To prepare the specimen for impact, a custom surrogate head (mass 3.3 kg and sagittal, horizontal, and frontal plane moments of inertia of 0.019, 0.014, 0.015 kg m2, respectively) (Becker, 1972; Walker et al., 1973) was rigidly attached to the occipital mount. The spine and surrogate head were stabilized using the compressive muscle force replication (MFR) system (Fig. 1A) (Ivancic et al., 2005). The MFR system was symmetric about the mid-sagittal plane and consisted of two anterior, two posterior, and eight lateral cables anchored to preloaded springs. The stiffness coefficients of the anterior, lateral and posterior springs were 4.0, 4.0, and 8.0 N/ mm, respectively. The anterior cables originated at the occipital mount, ran through guideposts at C4, through pulleys within the T1 mount and were each connected to a separate spring. The preload in each anterior spring was 15 N. To apply the posterior MFR, wire loops were inserted into the spinal canal through the laminae (C2
through C7) and tightly secured above each vertebral spinous process. The two posterior MFR cables originated from the occipital mount, ran through the wire loops, through a pulley at the T1 mount and each were connected to a spring, preloaded to 15 N. Bilateral MFR cables originated from C0, C2, C4, and C6, passed alternately along lateral guide rods, ran through pulleys at the T1 mount, and were each attached to a spring, preloaded to 30 N. With this MFR arrangement the compressive neutral posture preloads at each intervertebral level were: 120 N (C0–C1, C1–C2); 180 N (C2–C3, C3–C4); 240 N (C4–C5, C5–C6) and 300 N (C6–C7, C7–T1). The whole cervical spine with MFR and surrogate head was used to simulate head-turned and head-forward rear impacts. The model has been validated against in vivo simulated whiplash data (Ivancic et al., 2005). Prior to impact of the head-turned specimens, the head was rotated such that the average head-T1 rotation relative to the neutral posture was 28.4 of left axial rotation, 17.9 of left lateral bending, and 3.5 of flexion.
216
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
2.5. Rear impact and dynamic VA elongation In a bench-top apparatus (Panjabi et al., 1998a), rear impacts were performed using the incremental trauma approach at nominal T1 maximum horizontal accelerations of 3.5, 5, 6.5, and 8 g (Ghole et al., 2004). The major components of the bench-top apparatus included the sled, containing the rigidly attached T1 mount, impacting mass, and acceleration and braking systems. VA elongation data were continuously sampled at 1 kHz using an analog-to-digital converter and a personal computer. 2.6. Data analyses Dynamic VA elongation data were digitally filtered using a third order, dual pass, Butterworth low-pass filter at a cutoff frequency of 30 Hz. Residual and Fourier analyses demonstrated that most of the signal power was contained under 20 Hz. VA elongation rate was computed by numerically differentiating the dynamic VA elongation with respect to time. The peak VA elongation, peak VA elongation rate, and their times of occurrence relative to the onset of T1 horizontal acceleration were determined for each impact. Physiological VA elongation was defined as the maximum elongation obtained during the intact flexibility testing in any of the three planes. 2.6.1. Statistics Single factor, repeated measures ANOVA and Bonferonni post hoc tests were performed to determine increases in peak VA elongation during impact above physiological values. Single factor, non-repeated measures ANOVA and pair-wise Bonferonni post hoc tests were performed to determine differences in peak elongation rate, and times of peak elongation and peak elongation rate among head-forward VA, and head-turned right and left VA. Significance for all statistical tests was set at P < 0.05.
3. Results Representative results were chosen to exemplify dynamic VA elongations during head-forward (specimen #3) and head-turned (specimen #2) rear impacts, both at 8 g impact acceleration. Following the onset of the T1 horizontal acceleration of the head-forward specimen, head extension occurred causing minimal VA elongation, reaching a peak of only 4.1 mm at 186 ms (Fig. 2A). In sharp contrast, rotated head posture at the time of rear impact caused dramatic effects on VA elongation peaks and timing (Fig. 2B). During headturned rear impact, coupled head rotations were observed, consisting of extension, right axial rotation,
and left followed by right lateral bending. Peak left and right VA elongations of 29.1 mm and 26.5 mm, respectively, were achieved between 75 ms and 85 ms following impact, prior to peak head extension. Average maximum physiological VA elongations measured during intact flexibility tests were 7.3 mm (left VA of head-turned specimens), 7.5 mm (right VA of head-turned specimens), and 8.1 mm (head-forward specimens) (Fig. 3). Significant increases (P < 0.05) in dynamic VA elongation above physiological were observed in both left and right VA due to head-turned rear impact beginning at 5 g, while no significant increases due to head-forward rear impact were observed. The highest average dynamic VA elongation peaks of 30.5 mm (left VA) and 27.3 mm (right VA) occurred during the 6.5 g and 8 g head-turned rear impacts, respectively, and were over four times greater than the highest head-forward rear impact peak value. Temporal analyses demonstrated that the average times of peak dynamic left and right VA elongation due to head-turned rear impact occurred significantly earlier than corresponding times during head-forward rear impact for all impact accelerations (Table 1). Peak dynamic VA elongation in the left VA was attained as early as 84.5 ms following 8 g head-turned rear impact. Average peak left and right VA elongation rates were significantly greater with head-turned as compared to head-forward for all impacts with the exception of the right VA at 3.5 g and 5 g (Table 2A). Average peak elongation rate of the left VA significantly exceeded that of the right VA during 5 g and 6.5 g head-turned rear impacts. The highest peak VA elongation rate of 1336.7 mm/s occurred in the left VA during 8 g headturned rear impact. Average times of peak VA elongation rates occurred significantly earlier in both left and right VA with head-turned as compared to head-forward during all impacts except 6.5 g (Table 2B).
4. Discussion Clinical and epidemiological studies have documented chronic symptoms in whiplash patients including headaches, blurred vision, tinnitus, dizziness, and vertigo (Spitzer et al., 1995). Using a custom transducer, the present study has documented dynamic vertebral artery (VA) elongation during simulated head-turned and head-forward rear impacts of a whole cervical spine model with muscle force replication and surrogate head. Dynamic VA elongation of up to 30.5 mm due to headturned rear impact significantly exceeded physiological values beginning at 5 g, while VA elongation during head-forward rear impact remained within physiological limits (Fig. 3). Peak VA elongation occurred as early as 84.5 ms following head-turned rear impact (Table 1). A recent study measured the electromyographic activity of
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
217 70
1. T1 Horizontal Acceleration 2. Head-T1 Flexion-Extension: Rx
25
40 15 30 10
VA Elongation
5
10
1
0
0
-5
-10
2
-20
-10
-30 -15 -40 -20
-50
-25
-60
0
A
50
100
150
200
-70 250
Time (ms) 30
VA Elongation (mm) and T1 Horizontal Acceleration (g)
20
Head-T1 Flexion-Extension: Rx (Degrees)
50
20
-30
25
Left VA Elongation Right VA Elongation
1. T1 Horizontal Acceleration 2. Head-T1 Flexion-Extension: Rx 3. Head-T1 Axial Rotation: Ry 4. Head-T1 Lateral Bending: Rz
20
70 60 50 40
15 30 10
3
20
5
10 1
0
0 -10
2
-5
-20
-10 4
-30
-15 -40 -20
-50
-25
-60
-30 0
B
60
Head-T1 Rotations: Rx, Ry, Rz (Degrees)
VA Elongation (mm) and T1 Horizontal Acceleration (g)
30
50
100
150
200
-70 250
Time (ms)
Fig. 2. Representative examples of dynamic vertebral artery (VA) elongation during (A) head-forward (specimen #3) and (B) head-turned (specimen #2) rear impact, both at 8 g impact acceleration. Head-T1 rotations and T1 horizontal acceleration are also shown. Directions of head-T1 rotation are: +Rx = flexion, Rx = extension, +Ry = left axial rotation, Ry = right axial rotation, +Rz = right lateral bending, Rz = left lateral bending.
the cervical muscles during simulated head-turned rear impacts up to 1.3 g and found that peak muscle activity was achieved no earlier than 192 ms following the impact (Kumar et al., 2005). Thus, VA injury during head-turned rear impact may occur prior to neuromuscular protective mechanisms achieved via peak muscle tension. In contrast, peak VA elongation during headforward rear impact occurred significantly later. Peak VA elongation rates were greater and occurred earlier with head-turned as compared to head-forward rear impact (Table 2A and B). These cumulative results dem-
onstrate that elongation-induced VA injury is more likely to occur in those with rotated head posture at the time of rear impact, as compared to head forward. The limitations of this study must be considered before interpreting the present results. The custom transducers measured the total VA elongations between occiput and C6, thus, it was not possible to determine the VA strain distribution throughout its length. Dynamic VA elongation was calculated relative to neutral posture, assuming no initial VA laxity. Any initial VA laxity would result in overestimation of dynamic
218
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220 35 Head-Forward
Head-Turned Right VA
Head-Turned Left VA
30
* P<0.05 Peak VA Elongation (mm)
* 25
* * *
20
* 15
* 10
5
0 Physiological
3.5 g
5g
6.5 g
8g
Impact Acceleration
Fig. 3. Average peak vertebral artery (VA) elongations (mm) during head-forward and head-turned rear impacts. Significant increases (P < 0.05) in peak VA elongation, relative to physiological are indicated with asterisks.
Table 1 Average (SD) times (ms) at which the dynamic vertebral artery (VA) elongation peaks were attained, relative to onset of T1 horizontal acceleration, during head-forward and head-turned rear impacts
Significantly different times (P < 0.05) from the pair-wise comparisons are indicated with brackets.
Table 2 Average (SD) (A) peak vertebral artery elongation rates (mm/s) and (B) times of occurrence relative to onset of T1 horizontal acceleration during head-forward and head-turned rear impacts
Significant differences (P < 0.05) from the pair-wise comparisons are indicated with brackets.
VA elongation. Average ages of present head-turned and head-forward rear impact specimens were 80.2 and 70.8 years, respectively, and their spinal mechanical properties were stiffer than those of the younger population most likely to suffer whiplash trauma. Thus, peak dynamic VA elongation data of the present study are
conservative, as compared to the younger population. The muscle force replication system provided passive resistance to cervical motion and did not include the active neuromuscular response, thus simulating the response of an unwarned subject. The incremental trauma approach was used to determine the relation
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
between dynamic VA elongation and impact acceleration. A previous comparative study of incremental and single trauma protocols reported equivalent subfailure ligamentous injury severity, thus justifying the incremental trauma approach (Ghole et al., 2004). In previous biomechanical studies, Nibu et al. (1997) and Panjabi et al. (1998b) documented dynamic VA elongation during simulated head-forward rear impacts of the whole cervical spine model with surrogate head, without muscle force replication. Average peak dynamic VA elongation increased with increasing impact acceleration, analogous to the results of the present study; however, there were some differences. The highest average peak VA elongation of 5.8 mm due to head-forward rear impact of the present model was much less than previously reported highest peak, which exceeded 15 mm. There are two reasons for this difference. First, the muscle force replication used in the present model most likely reduced VA elongation by stabilizing the cervical spine. Second, the previous study reported non-zero VA elongation in neutral posture, which served to increase dynamic peak VA elongation. In contrast, the present study reported dynamic VA elongation relative to zero elongation in neutral posture. The exact mechanism by which VA injury occurs in the cervical spine remains unknown. Researchers have hypothesized that non-physiological coupled extension and axial rotation of the head may cause VA injury, particularly at the upper cervical spine due to its high mobility (Barton and Margolis, 1975; Chung and Han, 2002; Davis and Zimmerman, 1983; Sherman et al., 1981). Coupled cervical spine rotations observed during head-turned rear impact of the present study caused dynamic hyper-elongation of the left and right VAs. VA elongation causes a decrease in its diameter due to PoissonÕs effect and may lead to transient vascular compromise (Dobrin, 1978). In addition, while being elongated, the VA may be pinched at a bend of its circuitous anatomical route. The latter mechanism can precipitate VA injury, often manifested as a tear in the intimal layer, which has been clinically documented primarily at C1–C2 (Chung and Han, 2002). Severe VA injury results in acute vascular compromise caused by arterial dissection with possible rupture or thrombosis leading to stroke (Stahmer et al., 1997). The exact relationship between VA mechanical properties and elongation rate is unknown. The VA is a viscoelastic soft tissue structure; the adventitia is composed primarily of collagen fibers and the media consists of collagen as well as more substantial portion of muscular and elastic fibers (Fung, 1993). There are no previous studies that have documented the mechanical characteristics of the VA at elongation rates of the present study. However, a previous biomechanical study of cervical alar and transverse ligaments, viscoelastic structures which are composed primarily of collagen with some
219
elastin (Dvorak et al., 1988), documented a decrease in failure strain at high elongation rate of 920 mm/s as compared to 0.1 mm/s (Panjabi et al., 1998b). In the present study, head-turned rear impact caused highest peak elongation rate in the left VA of 1336.7 mm/s, significantly greater than the corresponding head-forward peak rate of only 211.5 mm/s. Assuming that the VA failure strain decreases with increased elongation rate, the high elongation rate observed due to head-turned rear impact may further accentuate elongation-induced injury of the VA.
5. Conclusions In conclusion, the present study, using a biofidelic whole cervical spine model with muscle force replication and surrogate head in simulated head-forward and head-turned rear impacts, has demonstrated potential elongation-induced VA injury due to head-turned rear impact. In contrast, VA injury during head-forward rear impact is unlikely. VA injury during head-turned rear impact may occur prior to peak muscle tension, indicating that the neuromuscular control system may not be able to protect the VAs from injury. The high VA elongation rate observed during head-turned rear impact may cause a decreased failure strain, as compared to head-forward. Present results cumulatively demonstrate that elongation-induced VA injury is more likely to occur during head-turned rear impact, as compared to forward facing.
Acknowledgment This research was supported by NIH Grant 1 RO1 AR45452 1A2.
References Barton, J.W., Margolis, M.T., 1975. Rotational obstructions of the vertebral artery at the atlantoaxial joint. Neuroradiology 9, 117–120. Becker, E., 1972. Measurement of mass distribution parameters of anatomic segments. Society of Automotive Engineers, Paper No. 720964. Carpenter, S., 1961. Injury of neck as cause of vertebral artery thrombosis. J. Neurosurg. 18, 849–853. Chung, Y.S., Han, D.H., 2002. Vertebrobasilar dissection: a possible role of whiplash injury in its pathogenesis. Neurol. Res. 24, 129–138. Davis, J.M., Zimmerman, R.A., 1983. Injury of the carotid and vertebral arteries. Neuroradiology 25, 55–69. DeBehnke, D.J., Brady, W., 1994. Vertebral artery dissection due to minor neck trauma. J. Emerg. Med. 12, 27–31. Dobrin, P.B., 1978. Mechanical properties of arteries. Physiol. Rev. 58, 397–460. Dvorak, J., Schneider, E., Saldinger, P., Rahn, B., 1988. Biomechanics of the craniocervical region: the alar and transverse ligaments. J. Orthop. Res. 6, 452–461.
220
P.C. Ivancic et al. / Clinical Biomechanics 21 (2006) 213–220
Fung, Y.C., 1993. Mechanical properties and active remodeling of blood vessels. Biomechanics: Mechanical Properties of Living Tissues, second ed. Springer-Verlag, New York, pp. 321–391. Ghole, S.A., Ivancic, P.C., Tominaga, Y., Gimenez, S.E., Panjabi, M.M., 2004. Incremental and single trauma produce equivalent subfailure soft tissue injury of the cervical spine. Clin. Biomech. 19, 784–789. Herr, R.D., Call, G., Banks, D., 1992. Vertebral artery dissection from neck flexion during paroxysmal coughing. Ann. Emerg. Med. 21, 88–91. Ivancic, P.C., Panjabi, M.M., Ito, S., Cripton, P.A., Wang, J.L., 2005. Biofidelic whole cervical spine model with muscle force replication for whiplash simulation. Eur. Spine J. 14, 346–355. Kumar, S., Ferrari, R., Narayan, Y., 2005. Looking away from whiplash: effect of head rotation in rear impacts. Spine 30, 760– 768. McLean, J.M., Wright, R.M., Henderson, J.P., Lister, J.R., 1985. Vertebral artery rupture associated with closed head injury. Report of two cases. J. Neurosurg. 62, 135–138. Nibu, K., Cholewicki, J., Panjabi, M.M., Babat, L.B., Grauer, J.N., Kothe, R., Dvorak, J., 1997. Dynamic elongation of the vertebral artery during an in vitro whiplash simulation. Eur. Spine J. 6, 286– 289. Okawara, S., Nibbelink, D., 1974. Vertebral artery occlusion following hyperextension and rotation of the head. Stroke 5, 640–642. Panjabi, M.M., Cholewicki, J., Nibu, K., Babat, L.B., Dvorak, J., 1998a. Simulation of whiplash trauma using whole cervical spine specimens. Spine 23, 17–24. Panjabi, M.M., Crisco 3rd, J.J., Lydon, C., Dvorak, J., 1998b. The mechanical properties of human alar and transverse
ligaments at slow and fast extension rates. Clin. Biomech. 13, 112–120. Reddy, M., Reddy, B., Schoggl, A., Saringer, W., Matula, C., 2002. The complexity of trauma to the cranio-cervical junction: correlation of clinical presentation with Doppler flow velocities in the V3-segment of the vertebral arteries. Acta Neurochir. (Wien) 144, 575–580, discussion 580. Schellhas, K.P., Latchaw, R.E., Wendling, L.R., Gold, L.H., 1980. Vertebrobasilar injuries following cervical manipulation. JAMA 244, 1450–1453. Seric, V., Blazic-Cop, N., Demarin, V., 2000. Haemodynamic changes in patients with whiplash injury measured by transcranial Doppler sonography (TCD). Coll. Antropol. 24, 197–204. Sherman, D.G., Hart, R.G., Easton, J.D., 1981. Abrupt change in head position and cerebral infarction. Stroke 12, 2–6. Spitzer, W.O., Skovron, M.L., Salmi, L.R., Cassidy, J.D., Duranceau, J., Suissa, S., Zeiss, E., 1995. Scientific monograph of the Quebec Task Force on Whiplash-Associated Disorders: redefining ‘‘whiplash’’ and its management. Spine 20, 1S–73S. Stahmer, S.A., Raps, E.C., Mines, D.I., 1997. Carotid and vertebral artery dissections. Emerg. Med. Clin. North Am. 15, 677–698. Taneichi, H., Suda, K., Kajino, T., Kaneda, K., 2005. Traumatically induced vertebral artery occlusion associated with cervical spine injuries: prospective study using magnetic resonance angiography. Spine 30, 1955–1962. Walker, L., Harris, E., Pontius, U., 1973. Mass, volume, center of mass, and mass moment of inertia of head and neck of human body. Society of Automotive Engineers, Paper No. 730985.