Juntendo Medical Journal

Objectives: To examine the effects of (1) trunk constraint and (2) the entry angle on the cervical spine in response to a headfirst impact. Materials: The AM50 Total HUman Model for Safety (THUMS) v4.02, pedestrian finite element model, was subjected to head-first impacts. Methods: The impact speed was 3.2 m/s. The following patterns were simulated: entry angle (0°, 15°to the sagittal plane) and trunk constraint (constraint, unconstraint). Results: As a result of head-first impact, the upper cervical spine was extended and the lower cervical spine was markedly flexed when the trunk was constrained. The mean stress applied to the cervical spine was significantly increased when the trunk was constrained, as indicated by the Mann-Whitney U test. Conclusions: In a head-first impact, the mean stress on the cervical spine increases significantly when the trunk is constrained. In order to reduce the risk of cervical spine injuries, it is desirable not to bind with teammates before a head-first impact.


Introduction
In rugby, players are exposed to a possible risk of head and neck injury during competition 1) 2) . The cause of catastrophic head and neck injury includes several factors, such as misjudgment, improper posture, and education about collision. In the past, the scrum is the most frequent mechanism of these injuries 3) because its collapse directly leads to the head hitting the ground, and binding with teammates further increases the impact force. To decrease damage to the head and neck, the World Rugby revised the scrum regulations in 2013 (https: //rugbyreferee.net/2013/05/08/breaking-news-ne w-scrum-engagement-process-approved-for-glob al-trial-rugbyrefs-scrum-rugbyunited/). The point of the revised scrum process is that props, front row players, use their outside arm to grip the opponent players before engagement. These scrum law variations contributed to a decrease in severe injury 4) 5) . These changes have improved the safety of rugby competitions. However, there are still some mechanisms that may cause severe injuries that have not yet been discussed.
Rucks and mauls are crowded situations where players are competing for the ball. Unlike set plays such as scrums, they occur in post-tackle sequences and there are many variations in the collisions. Ball contest in such situations, the so-called"breakdown" , is one of the alternative mechanisms of head and neck trauma. Breakdowns are the third most frequent cause of severe injuries, after tackles and scrums 6)-8) . Players usually collide with their opponents from the head or shoulder, and they may also bind with teammates just before an impact. These actions are considered high risks for head and neck injuries. Previous studies have reported that trunk constraint and preflexion affect the cervical load and cervical momentum 9) , but none have investigated the stresses at the sites in detail, at the point of collisions. Therefore, the aim of the present study was to examine the stress placed upon the cervical spine by the effects of (1) trunk constraint, and (2) the entry angle of impact. We investigated these issues by using the finite element method (FEM). We hypothesized that trunk constraint and entry angle of impact affected head-first impacts. Findings from this study may contribute to the safety of rugby players.

Materials and Methods
The AM50 Total HUman Model for Safety (THUMS ® ) v4.02 (Toyota Central R&D Labs., Inc. Aichi, Japan), pedestrian finite element model, was subjected to head-first impacts at selected direction and trunk constraint. The THUMS cervical spine has previously been validated against multiple cadaver tests 10)-13) . Impact conditions were set using Hyper-Mesh and Hyper-Crash (Altair Engineering Inc. Michigan, USA). All simulations were performed with LS-DYNA nonlinear explicit finite element solver version 971 (Livermore Software Technology Corporation. California, USA) in Supercomputer Tsubame 3.0 (Tokyo Institute of Tech-nology. Tokyo, Japan). Post-processing was carried out using LS-PrePost-4.3 (Livermore Software Technology Corporation. California, USA).

Data collection
We adopted C1 as the upper cervical spine, C4 as the middle cervical spine, and C7 as the lower cervical spine. C1 was divided into 2 parts (anterior column and posterior column). C4 and C7 were divided into 3 parts (anterior, middle, and posterior columns) (Figure-1).
For each part, 20 elements were randomly selected, and the mean stress was calculated. For the above four simulation patterns, the mean stress of each column was recorded every 5 ms, up to 35 ms.
We also measured the OC2 angle for the upper cervical angle, C2C7 angle for the middle cervical angle, and C6Th1 angle for the lower cervical angle (Figure-2). Each angle was recorded every 5 ms, up to 30 ms.

Statistical analysis
The outcome measure was the mean stress of each column in C1, C4, and C7. Comparisons of the outcome measurement described as earlier among the tasks were analyzed using the Mann-Whitney U test with 2 × 2 factors: entry angle (0°, 15°of sagittal plane) and condition of the trunk constraint (constraint, unconstraint). P value of < 0.05 was considered statistically significant. All tests were two-sided. The data analyses were conducted using R for Windows v. 3.5.2 (The R Project).

Results
In this study, the effects of the entry angle of impact and trunk constraint on the cervical spine was measured using FEM. Regarding the alignment of the cervical spine at the time of collision, the upper cervical spine was extended and the lower cervical spine was markedly flexed when the trunk was constrained. When the trunk was not constraint, the upper cervical spine was also extended, but the lower cervical spine showed normal to mild flexion, and the middle cervical spine was markedly extended (Figures-3, 4).
The maximum stress and site for each condition were as follows: 250 MPa at anterior column of C7 with entry angle 0°and trunk constraint; 87 MPa at posterior column of C4 with entry angle 0°and no trunk constraint; 239 MPa at anterior column of C7 with entry angle 15°and trunk constraint; and 67 MPa at posterior column of C4 with entry angle 15°a nd no trunk constraint (Figure-5).
The stress applied to the cervical spine was significantly increased with trunk constraint as indicated by the Mann-Whitney U test. In contrast, there was no significant increase in stress with the tested entry angles of impact.

Discussion
In this study, the influence of the entry angle of impact and trunk constraint on the cervical spine was measured using the finite element method. In general, the stresses occurring in the cervical spine were significantly greater when the trunk was constrained, but there was no significant difference in cervical spine stress at the tested angles of impact. The site and magnitude of the maximum stress was approximately 250 MPa in the anterior column of C7 with trunk constraint, and about 70 MPa in the posterior column of C4 without trunk constraint. Both the site of onset and magnitude of Comparing the results of the previous cadaver experiment 15) and the trunk constraint of the current study, the cervical posture (extension in the upper cervical spine and flexion in the lower cervical spine) and the site of fracture (C7) were similar (Figure-3). In the cadaveric experiment, the cervical spine of the corpse was rigidly fixed to the mass as the trunk, simulating the constrained trunk. We also considered trunk constraint in this simulation to reproduce the results of the cadaveric experiment. When the trunk was constrained, the stress was significantly increased in the entire cervical spine. This suggests that trunk constraint may be a significant risk factor for cervical spine injuries. In actual play, binding with teammates is considered a trunk constraint and should be avoided before a head-first impact.
We hypothesized that the head down position would also significantly increase stress on the cervical spine, but the results did not demonstrate this. A previous FEM study 9) showed a significant increase in load and momentum with cervical flexion, but the current simulation may be influenced by the fact that the simulation was tested with the head down but without cervical flexion, meaning without cervical spine straightening. The reason why the head down position is said to be dangerous is not the head position in itself, but the straightening of the cervical spine due to cervical flexion. If that is so, players with a straight neck Actual cervical spine injuries in rugby 3) 16) are reported to occur mainly in the posterior part of the middle cervical spine (C4 and C5), and fracture and dislocation are common. In the present simulation, the maximum stress was generated at the posterior column of C4, and the middle cervical spine was markedly extended when the trunk was not constrained, suggesting that the actual cervical spine injury may have occurred without trunk constraint. As mentioned above, the cadaveric experiment was performed with trunk constraint; and the FEM is useful for simulations without a trunk constraint, which may enable a more realistic simulation to be performed.
The present study had some limitations. Although the model was a general male individual, there are individual differences in the arrangement and posture of the cervical spine, and it is unknown whether these results can be generalized. Also, this experiment does not consider the effect of muscle tone. Naturally, changes in stress and cervical spinal alignment may change with muscle tension, so it is unclear whether it accurately approximates the collision situation in rugby competitions. In addition, the trunk was constrained at Th1 in this study, but it is not clear whether binding occurs at Th1 during a scrum or breakdown in rugby in reality. Moreover, although the entry angle in this study was based on the sagittal plane only, it is necessary to consider the three-dimensional angle including the coronal plane in actual collisions. Further simulations under detailed conditions, are required.

Conclusion
In a head-first impact, the mean stress on the cervical spine increases significantly when the trunk is constrained. In order to reduce the risk of injury to the cervical spine, it is desirable to avoid binding with teammates prior to a head-first impact.