© 2015 John Wiley & Sons A/S.
Scand J Med Sci Sports 2016: 26: 432–440 doi: 10.1111/sms.12446
Published by John Wiley & Sons Ltd
Spinal muscle activity in simulated rugby union scrummaging is affected by different engagement conditions D. Cazzola1*, B. Stone1*, T. P. Holsgrove2, G. Trewartha1, E. Preatoni1 Sport, Health and Exercise Science, Department for Health, University of Bath, Bath, UK, 2Centre for Orthopaedic Biomechanics, Department of Mechanical Engineering, University of Bath, Bath, UK Corresponding author: Dario Cazzola, PhD, Sport, Health and Exercise Science, Department for Health, University of Bath, Applied Biomechanics Suite, 1.305, BA2 7AY Bath, UK. Tel: +44 (0)1225 385466, E-mail: [email protected]
1
Accepted for publication 9 February 2015
Biomechanical studies of rugby union scrummaging have focused on kinetic and kinematic analyses, while muscle activation strategies employed by front-row players during scrummaging are still unknown. The aim of the current study was to investigate the activity of spinal muscles during machine and live scrums. Nine male front-row forwards scrummaged as individuals against a scrum machine under “crouch-touch-set” and “crouchbind-set” conditions, and against a two-player opposition in a simulated live condition. Muscle activities of the sternocleidomastoid, upper trapezius, and erector spinae were measured over the pre-engagement, engagement, and sustained-push phases. The “crouch-bind-set” condition increased muscle activity of the upper trapezius and
sternocleidomastoid before and during the engagement phase in machine scrummaging. During the sustainedpush phase, live scrummaging generated higher activities of the erector spinae than either machine conditions. These results suggest that the pre-bind, prior to engagement, may effectively prepare the cervical spine by stiffening joints before the impact phase. Additionally, machine scrummaging does not replicate the muscular demands of live scrummaging for the erector spinae, and for this reason, we advise rugby union forwards to ensure scrummaging is practiced in live situations to improve the specificity of their neuromuscular activation strategies in relation to resisting external loads.
Rugby union scrummaging involves a dynamic engagement phase followed by a period of sustained pushing (Milburn, 1990; Cazzola et al., 2014). Scrummaging places intense biomechanical demands on players, particularly those playing in the front row (Quarrie & Wilson, 2000). Because of its physical nature, the scrum is associated with approximately 6% to 8% of all rugby injuries (Brooks et al., 2005; Trewartha et al., 2014), 40% of catastrophic injuries (Quarrie et al., 2002; Brown et al., 2013), and may lead to chronic degenerative spinal injuries (Castinel et al., 2007; Delp et al., 2007; Hogan et al., 2010; Pinsault et al., 2010). Machine and live scrummaging biomechanics have been described in terms of the forces generated and motions observed (Cazzola et al., 2014; Preatoni et al., 2014). From these investigations, the scrum has undergone a number of rule changes, most recently from a “crouch-touch-set” (CTS, in 2012/2013) to a “crouchbind-set” (CBS, in 2013/2014) procedure, in an attempt to improve safety by de-emphasizing the initial impact of the scrum engagement (Cazzola et al., 2014; Preatoni et al., 2014). The CBS technique resulted in a significant reduction, approximately a 20% effect, in the biome-
chanical conditions (force, acceleration, and impact speed) experienced by front-row players in live scrummaging (Cazzola et al., 2014). However, the external mechanical load acting on forwards during a scrum is still considerable and the ways in which these loads transmit across the anatomical structures as well as the possible threshold for injury are still to be understood and described. Players’ neck and spinal strength is, therefore, crucial to absorb and control such mechanical stresses. Rugby forwards have been shown to have a specific neck strength profile (Olivier & Du Toit, 2008) that allows them to generate higher isometric force in extension, flexion and rotation motions, compared with backs (Hamilton & Gatherer, 2014). Currently, no study has investigated the effects of rule and technique changes on spinal muscle activity. The analysis of neuromuscular activation patterns under different engagement conditions is fundamental to elucidate the strategies employed by front-row players as they prepare their bodies for the engagement and pushing actions. Such information may allow sports science and medicine practitioners to provide optimal muscle conditioning and more specific rehabilitation programs, and may have a positive impact in terms of quicker and safer return to competition following injury. In fact, cervical
*These authors equally contributed to the study.
432
Spinal muscle EMG in a simulated rugby scrum spine and neck injuries produce an excessively high number of early recurrent injuries in elite rugby union, potentially due in part to unspecific and inappropriate rehabilitation programs (Williams S, unpublished observation). Analysis of muscle activation patterns during scrummaging may also inform training practice, such as in the conditioning of novice front-row players with regard to preparation for live match scrummaging. The aim of this study was to determine the activity of the bilateral upper trapezius, sternocleidomastoid, and erector spinae muscles under three scrummaging conditions: two machine scrummaging conditions, the “crouch-bind-set” and “crouch-touch-set”; and a single live scrummaging condition were investigated. The first hypothesis was that spinal muscle activity is greater for the CTS than the CBS scrummage caused by the higher biomechanical conditions (forces, accelerations, and impact speed) experienced by front-row players during the engagement in the CTS condition. The second hypothesis was that spinal muscle activity would be greater for live scrummaging compared with machine scrummaging caused by a sustained-push phase against a moveable target that decreases scrum stability. The third hypothesis was that spinal muscles activity across scrum phases would have a maximum value during the engagement phase caused by presence of an impact and the related high physical demand of that phase. Methods Study design In a repeated measures design, a group of rugby union front-row forwards performed multiple trials in three different simulated scrummage conditions (within-group factor) and throughout the phases of scrummaging (within-group factor) to assess and compare spinal muscle activity (dependent variable).
Participants Nine male rugby union players (age 20.3 ± 1.3 years, height 1.80 ± 0.10 m, weight 102.36 ± 15 kg), of at least University 1st XV standard with a minimum of 3 years playing experience in the front row and no history of spinal injuries in the 12 months prior to testing, participated in the study. All participants provided written informed consent prior to participation and ethical approval was obtained from the University of Bath Institutional Ethics Committee.
Data collection For electromyography (EMG) collection, six wireless electrodes (Delsys Trigno, Delsys Inc, Boston, Massachusetts, USA), sampling at 2000 Hz, were attached (Delsys adhesive interface) to the sternocleidomastoid, midway between the rostral and sternal attachments; upper trapezius, 1 cm superior to the scapula spine midway between the medial origin of the scapula spine and the acromion; and erector spinae, 3.5 cm from the midline of the spine at the level of L4-5 (Sharp et al., 2014; Fig. 1). Surface EMG signals were collected bilaterally on each participant (Delsys EMGworks 4.1.05, Delsys Inc). Prior to the mounting of electrodes, the skin surface was prepared by shaving, lightly abrading, and cleaning with alcohol wipes.
Fig. 1. Electrode and marker setup. Electrode positions are highlighted with black circles. Sternocleidomastoid (SCM) is shown on the left-hand side of the figure while upper trapezius (UT) and erector spinae (ES) on the right-hand side.
The skin impedance was verified and measurements were carried out having an impedance lower than 5 kΩ. Following a player-led warm-up, each player performed two 4-s isometric maximal voluntary contractions (MVC) of the upper trapezius, sternocleidomastoid, and erector spinae, with a 1-min break between each measurement using the procedures defined in Table 1 (Vera-Garcia et al., 2010; Morimoto et al., 2013). Each participant then performed a number of submaximal scrummaging trials to become familiar with the experimental and environmental conditions (indoor scrummaging on a rubber-based floor). The three different engagement techniques, crouch-touchset (CTS), crouch-bind-set (CBS), and live 2-vs-1 scrummaging (Live; Table 2), were used. The machine scrummaging condition involved a single participant engaging with a static scrum machine (Dictator, Rhino Rugby, Rooksbridge, UK) using the CTS and CBS variants followed by a sustained push (Preatoni et al., 2012). The beams of the scrum machine were instrumented with strain gauges in order to measure the forces generated by front-row players while scrummaging (Preatoni et al., 2012). The live condition involved a single participant passively engaging, for safety reasons, with two other participants prior to the sustained push (Fig. 2). The movement patters across all the phases are essentially identical for all the scrum engagement techniques except for an increased displacement of the body center of mass in the CTS condition. Each participant then completed at least four successful trials in each of the three scrum conditions, which were presented in a blocked random order, up to a maximum of 24 trials in one session. A trial was considered successful when the player engaged timely with the prerecorded referee calls (± 0.5 s from the start of each call). Recovery intervals
433
Cazzola et al. Table 1. Positions and resistances used to measure MVC for the sternocleidomastoid, upper trapezius, and erector spinae
Condition
Description
Sternocleidomastoid (SCM) Participant stands with trunk and hips flexed to 90 degrees, so that the trunk is parallel to the floor with the neck in a neutral position with the forehead placed on the scrum machine pad. The participant attempts to flex the neck against the fixed pad. Upper trapezius (UT) Participant lies in a prone position with both arms abducted at the shoulder (∼45°) and externally rotated with the elbow flexed. The participant attempts to abduct the arms against manual resistance applied to the elbow. Erector spinae (ES) Participant lies in a prone position with the torso on the table, the legs projected horizontally over the end of the table, and arms out of the side. The participant attempts to extend the lower trunk and hip against manual resistance applied to the posterior thigh and shoulders bilaterally.
Table 2. Scrummage conditions
Condition
Description
CTS (2012–2013)
A single participant engaged with a scrum machine following the engagement cues “crouch-touch-set.” On “crouch” (t = 0 s), the participant moved into their normal crouch posture. On “touch” (t = 1.7 s), the participant moved their right arm to touch the pad and then withdrew the arm to assume the crouch posture. On “set” (t = 4 s) the participant dynamically engaged with the scrum machine and then started a sustained push for approximately 3 s. A single participant engaged with a scrum machine following the engagement cues “crouch-bind-set.” On “crouch” (t = 0 s), the participant moved into their normal crouch posture. On “bind” (t = 1.7 s), the participant moved their right and left arm to bind onto the scrum machine pusher arms. On “set” (t = 4 s), the participant dynamically engaged with the scrum machine and then started a sustained push for approximately 3 s. A single participant scrummaged against two other participants when cued by the call “set” (t = 4 s). Prior to the cue, the participant passively engaged with the two opposing participants binding with both arms on to the opposing participants’ backs. The two opposing participants bound together, as would a loose head prop and a hooker in a complete front row. The opposing participants were asked to place their unbound hand on the ground to provide additional stability and to hold the test participant in a relatively static position. On “set,” the participant maximally pushed against the two opposing participants for approximately 3 s.
CBS (2013–present)
Live
Fig. 2. Images of “key instances” in the CBS (left), CTS (central), and live (right) scrummaging conditions. (a) Crouch position; (b) “bind” call; (c) sustained-push phase. Phases a and b are not reported during live scrummaging as muscle activity was only measured during the sustained push caused by the passive engagement. of 1 < t < 2 min between consecutive trials and 7 < t < 10 min between sets were included to avoid fatigue. A 13-camera (12 cameras Oqus 400, 1 Oqus 210c) motion capture system (Qualysis, Goteborg, Sweden) capturing at 250 Hz was used to collect players’ kinematics and to determine the time
434
of actual engagement. A single marker was rigidly attached to the posterior aspect of each of the two central scrum machine pusher-arm pads to detect movement on initial contact. A bespoke control and acquisition system (cRIO-9024, National Instruments, Austin, Texas, USA) was programmed (Labview 2010, National
Spinal muscle EMG in a simulated rugby scrum
Fig. 3. Exemplary compression force and normalized trapezius EMG signal from a CTS scrummaging trial. The three phases (pre-engagement, engagement, and sustained push) considered in the study are highlighted in the graph. Instruments) to synchronously trigger the acquisition hardware (Delsys EMG, Qualysis), collect and store the forces measured through the instrumented scrum machine, and playback prerecorded cues given by the referee (Preatoni et al., 2013). The cues, “crouch-touch-set” and “crouch-bind-set,” were delivered with consistent timing (Table 2) for all the scrummaging conditions. Muscle activity was measured for 10 s from 1 s prior to the “crouch” call.
Data processing Raw electromyograms were filtered by applying a bidirectional second-order Butterworth low-pass and high-pass filter between 20 and 200 Hz. The data were then rectified and smoothed using a moving average over 50-ms windows (Visual 3D v5, C-Motion Inc, Germantown, MD, USA). EMG signals were normalized to the MVCs, which were calculated between 0.2 and 1.2 s after the initiation of maximum isometric muscle contraction, with an average MVC value being calculated over two trials. Average muscle activity (average rectified EMG amplitude) during scrum trials was calculated over three phases of each scrum: the preengagement, the engagement, and the sustained-push phases. For the machine scrummaging conditions (CBS and CTS), the pre-engagement phase was defined as the interval between the “set” cue and the instant of first contact between participant and pusher arm. As in Preatoni et al. (2014), and considering the fast loading rate in the shock-absorption phase (Fig. 3), the engagement phase was defined as the time the participant first contacted the pusher arms until 1 s after the initial contact, the sustainedpush phase extended from the end of engagement for 1 s (Fig. 3). First contact between participant and pusher arms was determined from the horizontal displacement of the markers on the scrum machine pusher arm using the Qualysis analysis software (QTM 2.9). During live scrummaging, players were bound together at the “set” call, which was an invitation for them to start pushing. Therefore, in the Live scrummage, the sustained-push phase was defined as the interval between 1 and 2 s after the “set” cue.
Statistics Separate one-way repeated measure analysis of variance (ANOVA; with scrummage conditions as the within-group factor)
and Bonferroni post-hoc analysis were applied (SPSS software, IBM Corp, New York, USA) to determine if there were any differences (P < 0.05) in muscle activation across scrummage conditions during the sustained-push phase (CBS vs CTS vs Live). A paired t-test was performed to determine the differences in muscle activation in the pre-engagement phase (CBS vs CTS), and the engagement phase (CBS vs CTS), as Live engagement included only the sustained-push phase (Fig. 2). Further one-way repeated measure ANOVAs (with scrum phases as the within-group factor) were used to test possible changes in muscles activation across the different phases of the scrum for the CTS and CBS machine trials, followed by Bonferroni post-hoc comparisons (P < 0.05). Sphericity of the data was assessed using the Greenhouse–Geisser epsilon; if the data were aspherical, a correction was applied to the calculated P-value. Pairwise effect sizes, calculated using Cohen’s d statistic (d) (1988), were also considered (Appendix).
Results The time of contact (tENG) was highly repeatable across all participants (0.55 ± 0.08 s). Also, the time of onset of movement had very low variability (0.04 s), and its average value was 0.03 s. This provided the analysis with a consistent scrum phase duration (e.g., time interval between scrum phases) across all the subjects.
Comparing conditions Left and right muscles in all the analyzed muscle groups exhibited very similar level of activation across the subsequent phases of scrummaging. Therefore, EMG data from left and right side were pooled. In the pre-engagement phase, all the measured muscles in both machine conditions were activated in excess of 25% MVC (Fig. 4). The activity of the sternocleidomastoid and upper trapezius was significantly higher (P < 0.01) in the CBS condition than the CTS
435
Cazzola et al.
Fig. 4. Normalized values (% maximal voluntary contraction) of muscles activation (mean and standard deviation) of sternocleidomastoid (SCM), upper trapezius (TRAP), and erector spinae (ES) during CBS (crouch-bind-set) and CTS (crouch-touch-set) engagements. Muscle activations during CBS and CTS engagements are shown throughout the three scrum phases: pre-engagement, engagement, and sustained push. Live engagement is not included because in that condition, EMG measures were carried out only during sustained-push phase. The dashed and dotted lines show the muscles activation trend in, respectively, CTS and CBS, and are representative of the differences (ANOVA) between (1) pre-engagement and sustained push and (2) engagement and sustained push. ‡ Significant difference between CBS and CTS (P < 0.05 – paired t-test). ¥Significant difference between both (1) pre-engagement and sustained push, and (2) engagement and sustained push in CBS (P < 0.05 – ANOVA). §Significant difference between both (1) pre-engagement and sustained push, and (2) engagement and sustained push in CTS (P < 0.05 – ANOVA).
Fig. 5. Normalized muscle activation (mean and standard deviation) of sternocleidomastoid (SCM), upper trapezius (TRAP), and erector spinae (ES) during CBS, CTS, and Live engagements throughout the sustained-push phase. *Significant difference between CBS and Live (P < 0.01 – ANOVA). †Significant difference between CTS and Live (P < 0.01 – ANOVA).
condition. The activity of the erector spinae tended to be more activated (large effect size, d > 0.8), although not significantly (P > 0.05), in the CTS condition during preengagement than the CBS condition. During the engagement phase, all the measured muscles were more active in the CBS than CTS condition. The activity of the upper trapezius and sternocleidomastoid were significantly higher (P < 0.05 – paired t-test), showing an average increase of 20 ± 12% and 22 ± 20%, respectively (Fig. 4), and large effect size (d > 1). The ES tended to be more activated during the CBS condition (effect size > 0.8), although not significantly.
436
During the sustained-push phase, the activity of the muscles across all three conditions (CBS, CTS, and Live) could be compared (Fig. 5). The activity of the erector spinae was significantly higher during the sustained-push phase of live scrummaging than in either of the CBS or CTS conditions (P < 0.01), with the erector spinae activity approximately 56 ± 26% lower (CBS vs Live) and 62 ± 18% lower (CTS vs Live) and large effect size (d > 0.8). The activity of the upper trapezius tended to be lower in the CTS than the CBS and Live conditions (d > 1.1), whereas the activity of the sternocleidomastoid was similar across conditions.
Spinal muscle EMG in a simulated rugby scrum Comparing across phases The activity of the sternocleidomastoid and erector spinae showed a decreasing trend (P < 0.05) moving from pre-engagement through engagement to the sustained push in both the CBS and CTS conditions (Fig. 4). The activity of the sternocleidomastoid was significantly higher during the pre-engagement and engagement phase than sustained push in CBS (P < 0.05), and in CTS, pre-engagement tended to be higher than sustained push (P = 0.059) showing a large effect size (d > 1.1). The activity of the erector spinae during both CBS and CTS conditions was significantly higher during both the pre-engagement phase (P < 0.05) and engagement phase (P < 0.05) when compared with the sustained push (Fig. 4). There was a significant pattern of decreasing activation from (a) pre-engagement to engagement; and from (b) engagement to sustained push that was also reflected in large effect sizes (d > 1.7). The activity of the upper trapezius in the CBS and CTS conditions peaked, but was not significantly greater, during the engagement phase (Fig. 4). No significant differences were calculated, though large effect sizes (d > 1.2) were found for the upper trapezius between the engagement phase and sustained push in both CBS and CTS conditions. Also, medium effect sizes (d > 0.5) were found for the upper trapezius between the preengagement phase and engagement in both CBS and CTS conditions. Discussion The aim of this study was to gain more insight into the activity of sternocleidomastoid, upper trapezius, and erector spinae muscles during machine (CBS and CTS) and live scrummaging. Compared with the machine conditions, the live condition resulted in significantly higher activation of erector spinae during the sustained-push phase. In contrast with the initial hypothesis, the activity of sternocleidomastoid, upper trapezius, and erector spinae tended to be greater during the CBS condition than the CTS condition throughout the three phases of the scrum. Spinal muscle activity in both machine scrummaging conditions was characterized by a considerable preactivation of all six muscles (>25% MVC) prior to engagement. This pre-activation can functionally lead to an increase in cervical and lumbar spine stiffness (Riemann & Lephart, 2002), which may better prepare the player’s spinal structures for the high biomechanical loads placed upon the spine during engagement. A high level of muscle pre-activation may potentially mitigate the effect of loads on spinal posture (Krajcarski et al., 1999; Choi, 2003) and help maintain optimal neutral spine position (Brooks & Kemp, 2011). However, it must be observed that the stabilization of the spine caused by high level of muscle activation and agonist/antagonist co-contraction (Cheng et al., 2008) may not be enough
to limit cervical spine hyperflexion or buckling mechanisms (Dennison & Macri, 2012; Kuster et al., 2012) during high-dynamic impulsive impacts with misaligned geometry. In fact, in case of a catastrophic injury, the forces and moments acting on the cervical spine are presumably much higher than the forces that can be generated by spinal muscle activation, and further analyses are needed to elucidate the actual contribution that muscles activations can make in certain high-risk loading conditions. The activity of erector spinae has been measured in a similar machine scrummaging study (Sharp et al., 2014), although the engagement condition was not clearly described, and the pre-engagement phase was defined as the 200-ms window prior to engagement, differing from the current study. Sharp et al. (2014) reported a preactivation of 65% of MVC in the erector spinae, which is comparable with our findings of 68% and 75% erector spinae pre-activation in CBS and CTS conditions. Sharp et al. (2014) interpreted erector spinae pre-activation as necessary to maintain an efficient scrummaging position and to overcome gravity forces causing a bending moment and a rotation of the trunk about the pelvis. Our results support this explanation, as the slight reduction in erector spinae activity in the CBS condition when compared with the CTS condition, in the pre-engagement phase, can be attributed to pre-binding, reducing the moment acting on the trunk joints caused by torso weight being partially supported. The engagement phase of a scrum is characterized by a peak in both compression and vertical force, approximately 0.1 s after first contact (Fig. 3), followed by a lower plateau in force traces approximately 0.5 s after engagement (Preatoni et al., 2013, 2014; Cazzola et al., 2014). These changes in force generate the need for a high spinal stiffness and therefore demand compensatory spinal muscle activation. Cholewicki et al. (2000) identified that as the external load placed upon the spine increased, the stiffness of the spine also increased, and most interestingly, this was caused by an increase in spinal muscle activation. Therefore, we propose that the EMG signal of the participant’s spinal musculature can take as an input parameter and respond to the time course of the external load exerted on players’ shoulders. The spinal muscles are maximally activated in response to the peak in external load and then decrease 0.5 s after engagement as the external load reduces (Fig. 3). This stabilization in activation is observed in all six (e.g., left and right) muscles during the sustained-push phase irrespective of condition. We have previously observed a reduction in external loading on players when scrums use the CBS condition when compared with the CTS condition (Cazzola et al., 2014). If these reductions in mechanical loading during the engagement were transferred to the modified scrum condition used in the present study, it can be hypothesized that the spinal muscles would be more active
437
Cazzola et al. during CTS than CBS condition, as the spine is exposed to greater mechanical stresses. However, the activity of the spinal muscles was comparable between the conditions and the activity of the cervical spine muscles (sternocleidomastoid and upper trapezius) was found, in the engagement phase, to be significantly greater in the CBS condition. These differences may be attributed to the differences in binding between the conditions, as the upper trapezius is responsible for the maintenance of position of the cervical spine and also the upper arm (Herrington & Horsley, 2009). Upper trapezius is also a scapula stabilizer, and its higher activation during the pre-engagement and engagement phase in CBS provides more control and stiffness during the binding procedure. In the CBS condition, the players establish a secure pre-bind prior to engagement, while in the CTS condition, the players have to bind during the engagement. Additionally, during the CBS conditions, players set up closer to the scrum machine (Preatoni et al., 2014), and this posture may effect spinal muscle activation, where greater upper trapezius and sternocleidomastoid activation increase cervical spine stiffness and may better maintain cervical spine posture during the engagement. This indicates that a pre-bind procedure makes the upper spine more prepared for the scrum engagement. The activation of the erector spinae was significantly higher (P < 0.05) throughout the sustained-push phase of the live scrummaging condition than either of the machine scrummaging conditions. Live scrummaging is an unstable dynamic condition when compared with machine scrummaging; therefore, the forces applied to the opposition players are not equally matched in direction or magnitude, as they are during a scrum against the machine. These eccentric forces theoretically generate moments at trunk level, causing lumbar flexion, extension or rotation, and placing the participant in a compromised lumbar spinal posture. Thus, the participant needs to maintain optimal lumbar spinal posture via the activation of the stabilizing lumbar muscles (represented by the activation of the erector spinae in this study). These findings demonstrate that although machine and live scrummaging show comparable kinematic and kinetic characteristics, machine scrummaging does not accurately replicate live scrummaging in terms of muscle activation strategies required to maintain the sustained phase. Thus, it can be suggested that front-row players should be required to train in live situations and not only against scrum machines. These findings suggest that although scrum machines are a useful training aid, a relative increase in live scrummaging practice may be beneficial for all front-row players from the perspective of training appropriate and specific neuromuscular activation patterns. This may be particularly important under current rugby practice since the duration of the sustained phase has markedly increased under the CBS engagement procedure.
438
In the current study, each participant scrummaged individually in three conditions; this is a constrained environment. The results observed in such modified conditions cannot be extrapolated directly into a full scrummage scenario. During a complete scrummage, eight players interact and bind to one another generating the forces and movements of the scrummage (Preatoni et al., 2014). Sharp et al. (2014) reported a peak compression force of 3.1 kN in a single participant machine scrummaging, whereas this study recorded an average maximum compression of about 2.8 kN. These magnitudes are about 25% less than the ones found by Preatoni et al. (2014), who measured a peak compression force of 11.7 kN, a vertical force of 1.5 kN, and a lateral force 1.3 kN distributed across the three front-row players in a full scrummage of academy level. The reduction in compression forces and the absence of extra-loading from multiplayer interactions may result in a decrease in muscle activation for one-person scrummaging, suggesting that the results of the current study may underestimate the spinal muscle activity observed throughout a full scrummage. Additionally, front-row and second-row players have loads applied from front to back, which may cause other types of loading patterns around the spine. The current study was undertaken in an indoor lab with an individual scrummaging against a machine or two other participants. The study was one of the first of its kind to measure muscle activity in an impact event similar to match intensity, and is a fundamental first step toward understanding the muscle activations in a full scrummage scenario. A limitation of the current study was that muscle activity in the live condition was only measured during the sustained-push phase. Future research should build on the foundations of the current study, with the objectives of improving its ecological validity through trials on natural turf and in a more complete scrummaging scenario (a full front row or a complete pack) either against a scrum machine or in live conditions. The use of bilateral measurement of muscle activity is advised, as a more complete scrummage is unlikely to match the symmetry of the current binding conditions. In order to provide a full description of the trunk muscles contributing to spine stability, the measurement of abdominal muscle activity should be included. However, soft tissue movement artifact and low signal-to-noise ratio, caused by the physical characteristics of front-row rugby players, are likely to affect EMG signal quality. In conclusion, the activation of selected spinal muscles is greater in the CBS condition than the CTS condition, particularly in the pre-engagement phase. This indicates that the muscles of the cervical spine, in the CBS condition, are better prepared for the forces experienced during the scrum engagement than in the CTS condition as cervical spine stiffness is greater. Furthermore, this research provides evidence that the
Spinal muscle EMG in a simulated rugby scrum erector spinae is significantly more active during live scrummaging than machine scrummaging. This reinforces the requirement for individuals to practice and learn scrummage techniques in a live situation, rather than purely against a machine, as machine scrummaging does not replicate the demands of a live contest. Perspectives The findings of this study may have an impact in sports medicine, especially in the injury prevention and injury biomechanics areas. Firstly, the evidence of spinal muscle pre-activation highlights the presence of a specific trunk-stiffening strategy aiming to prepare the body for the impact, and therefore, its contribution needs to be considered during any real-world injury mechanism analysis in contact sports. Also, the description of the activation patterns of the trunk muscles may provide new insights for the optimization of specific training and rehabilitation programs with a specific view to injury prevention. Finally, this study can provide further evi-
dence on which to inform discussions relating to the scrum laws of rugby union when seeking to improve player welfare. In fact, the resulting information adds the measure of spinal muscle activation to the analysis of the movements and forces involved during different scrum engagements (Preatoni et al., 2013, 2014; Cazzola et al., 2014), providing a better understanding of the biomechanical load experienced by rugby forwards. Key words: Biomechanics, sports injury, sports performance, scrummaging technique, lumbar spine, cervical spine.
Acknowledgements The authors would like to thank Dr Polly McGuigan, Dr Richie Gill, Dr Sabina Gheduzzi, Dr Keith Stokes, and Dr Tony Miles for their involvement in the wider programme of research and their comments on this manuscript. Also, we would like to thank all rugby forward players that took part to the study. This project is funded by the Rugby Football Union (RFU) Injured Players Foundation.
References Brooks JH, Fuller CW, Kemp SP, Reddin DB. Epidemiology of injuries in English professional rugby union: part 1 match injuries. Br J Sports Med 2005: 39: 757–766. Brooks JHM, Kemp SPT. Injury-prevention priorities according to playing position in professional rugby union players. Br J Sports Med 2011: 45: 765–775. Brown JC, Lambert MI, Verhagen E, Readhead C, van Mechelen W, Viljoen W. The incidence of rugby-related catastrophic injuries (including cardiac events) in South Africa from 2008 to 2011: a cohort study. BMJ Open 2013: 3 (2): e002475. doi:10.1136/bmjopen2012-002475. Castinel BH, Adam P, Prat C. A stress fracture of the lumbar spine in a professional rugby player. Br J Sports Med 2007: 41: 337–338. Cazzola D, Preatoni E, Stokes KA, England ME, Trewartha G. A modified prebind engagement process reduces biomechanical loading on front row players during scrummaging: a cross-sectional study of 11 elite teams. Br J Sports Med 2014: doi:10.1136/ bjsports-2013-092904. Cheng CH, Lin KH, Wang JL. Co-contraction of cervical muscles during sagittal and coronal neck motions at different movement speeds. Eur J Appl Physiol 2008: 103: 647–654. Choi H. Quantitative assessment of co-contraction in cervical musculature.
Med Eng Phys 2003: 25: 133– 140. Cholewicki J, Simons APD, Radebold A. Effects of external trunk loads on lumbar spine stability. J Biomech 2000: 33: 1377–1385. Cohen J. Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates, 1988. Delp SL, Anderson FC, Arnold AS, Loan P, Habib A, John CT, Guendelman E, Thelen DG. OpenSim: open-source software to create and analyze dynamic simulations of movement. IEEE Trans Biomed Eng 2007: 54: 1940–1950. Dennison CR, Macri EM, Cripton PA. Mechanisms of cervical spine injury in rugby union: is it premature to abandon hyperflexion as the main mechanism underpinning injury? Br J Sports Med 2012: 46: 545–549. Hamilton D, Gatherer D. Cervical isometric strength and range of motion of elite rugby union players: a cohort study. BMC Sports Sci Med Rehabil 2014: 6: 32. Herrington L, Horsley I. Electromyographic analysis of selected shoulder muscles during a rugby football tackle. Sports Med Arthrosc Rehabil Ther Technol 2009: 1: 10. doi:10.1186/1758-2555-1-10. Hogan BA, Hogan NA, Vos PM, Eustace SJ, Kenny PJ. The cervical spine of professional front-row rugby players: correlation between degenerative changes and symptoms. Ir J Med Sci 2010: 179: 259–263.
Krajcarski SR, Potvin JR, Chiang J. The in vivo dynamic response of the spine to perturbations causing rapid flexion: effects of pre-load and step input magnitude. Clin Biomech (Bristol, Avon) 1999: 14: 54–62. Kuster D, Gibson A, Abboud R, Drew T. Mechanisms of cervical spine injury in Rugby Union: a systematic review of the literature. Br J Sports Med 2012: 46 (8): 550–554. doi: 10.1136/bjsports2011-090360. Milburn PD. The kinetics of rugby union scrummaging. J Sports Sci 1990: 8: 47–60. Morimoto K, Sakamoto M, Fukuhara T, Kato K. Electromyographic study of neck muscle activity according to head position in rugby tackles. J Phys Ther Sci 2013: 25: 563–566. Olivier PE, Du Toit DE. Isokinetic neck strength profile of senior elite rugby union players. J Sci Med Sport 2008: 11: 96–105. Pinsault N, Anxionnaz M, Vuillerme N. Cervical joint position sense in rugby players versus non-rugby players. Phys Ther Sport 2010: 11: 66–70. Preatoni E, Stokes KA, England ME, Trewartha G. The influence of playing level on the biomechanical demands experienced by rugby union forwards during machine scrummaging. Scand J Med Sci Sports 2013: 23: e178–e184. Preatoni E, Stokes KA, England ME, Trewartha G. Engagement techniques and playing level impact the biomechanical demands on rugby forwards during machine-based
439
Cazzola et al. scrummaging. Br J Sports Med 2014: doi:10.1136/bjsports-2013-092938. Preatoni E, Wallbaum A, Gathercole N, Coombes S, Stokes KA, Trewartha G. An integrated measurement system for analysing impact biomechanics in the rugby scrum. Proc Inst Mech Eng Part P J Sports Eng Technol 2012: 226 (3–4): 266–273. Quarrie KL, Cantu RC, Chalmers DJ. Rugby union injuries to the cervical spine and spinal cord. Sports Med 2002: 32: 633–653.
Quarrie KL, Wilson BD. Force production in the rugby union scrum. J Sports Sci 2000: 18: 237–246. Riemann BL, Lephart SM. The sensorimotor system, part I: the physiologic basis of functional joint stability. J Athl Train 2002: 37 (1): 71–79 Sharp T, Halaki M, Greene A, Vanwanseele B. An EMG assessment of front row rugby union scrummaging. Int J Perform Anal Sport 2014: 14: 225–237.
Trewartha G, Preatoni E, England ME, Stokes KA. Injury and biomechanical perspectives on the rugby scrum: a review of the literature. Br J Sports Med 2014: doi:10.1136/bjsports-2013092972. Vera-Garcia FJ, Moreside JM, McGill SM. MVC techniques to normalize trunk muscle EMG in healthy women. J Electromyogr Kinesiol 2010: 20: 10–16.
Appendix Table A1. Pairwise effect sizes calculated between CBS, CTS, and live conditions in pre-engagement (Pre-Eng), engagement (Eng), and sustained push (Sus-Push) scrum phases
Condition
CBS CTS
Phase
Pre-Eng–Eng Eng–Sus-Push Pre-Eng–Sus-Push Pre-Eng–Eng Eng–Sus-Push Pre-Eng–Sus-Push
Muscles SCM
TRAP
ES
0.38 2.70 1.54 0.57 2.50 1.15
0.51 1.46 0.01 0.69 1.38 0.01
1.70 11.46 4.89 4.11 9.78 2.81
Note: Pairwise effect sizes were calculated using Cohen’s (d) values. |d| > 0.8 large effects; |d| > 0.5 moderate effects; |d| > 0.2 small effects.
Table A2. Pairwise effect sizes calculated between pre-engagement (Pre-Eng), engagement (Eng), and sustained push (Sus-Push) scrum phases, in CBS, CTS, and live conditions
Phase
Pre-Eng Eng Sus-Push Sus-Push Sus-Push
Condition
CBS vs CTS CBS vs CTS CBS vs CTS CBS vs Live CTS vs Live
Muscles SCM
TRAP
ES
1.26 2.97 0.57 0.26 1.80
1.46 1.17 1.86 0.88 0.87
0.87 0.91 1.74 0.88 0.87
Note: Pairwise effect sizes were calculated using Cohen’s (d) values. |d| > 0.8 large effects; |d| > 0.5 moderate effects; |d| > 0.2 small effects.
440
Scand J Med Sci Sports 2016: 26: 432–440 doi: 10.1111/sms.12446
Published by John Wiley & Sons Ltd
Spinal muscle activity in simulated rugby union scrummaging is affected by different engagement conditions D. Cazzola1*, B. Stone1*, T. P. Holsgrove2, G. Trewartha1, E. Preatoni1 Sport, Health and Exercise Science, Department for Health, University of Bath, Bath, UK, 2Centre for Orthopaedic Biomechanics, Department of Mechanical Engineering, University of Bath, Bath, UK Corresponding author: Dario Cazzola, PhD, Sport, Health and Exercise Science, Department for Health, University of Bath, Applied Biomechanics Suite, 1.305, BA2 7AY Bath, UK. Tel: +44 (0)1225 385466, E-mail: [email protected]
1
Accepted for publication 9 February 2015
Biomechanical studies of rugby union scrummaging have focused on kinetic and kinematic analyses, while muscle activation strategies employed by front-row players during scrummaging are still unknown. The aim of the current study was to investigate the activity of spinal muscles during machine and live scrums. Nine male front-row forwards scrummaged as individuals against a scrum machine under “crouch-touch-set” and “crouchbind-set” conditions, and against a two-player opposition in a simulated live condition. Muscle activities of the sternocleidomastoid, upper trapezius, and erector spinae were measured over the pre-engagement, engagement, and sustained-push phases. The “crouch-bind-set” condition increased muscle activity of the upper trapezius and
sternocleidomastoid before and during the engagement phase in machine scrummaging. During the sustainedpush phase, live scrummaging generated higher activities of the erector spinae than either machine conditions. These results suggest that the pre-bind, prior to engagement, may effectively prepare the cervical spine by stiffening joints before the impact phase. Additionally, machine scrummaging does not replicate the muscular demands of live scrummaging for the erector spinae, and for this reason, we advise rugby union forwards to ensure scrummaging is practiced in live situations to improve the specificity of their neuromuscular activation strategies in relation to resisting external loads.
Rugby union scrummaging involves a dynamic engagement phase followed by a period of sustained pushing (Milburn, 1990; Cazzola et al., 2014). Scrummaging places intense biomechanical demands on players, particularly those playing in the front row (Quarrie & Wilson, 2000). Because of its physical nature, the scrum is associated with approximately 6% to 8% of all rugby injuries (Brooks et al., 2005; Trewartha et al., 2014), 40% of catastrophic injuries (Quarrie et al., 2002; Brown et al., 2013), and may lead to chronic degenerative spinal injuries (Castinel et al., 2007; Delp et al., 2007; Hogan et al., 2010; Pinsault et al., 2010). Machine and live scrummaging biomechanics have been described in terms of the forces generated and motions observed (Cazzola et al., 2014; Preatoni et al., 2014). From these investigations, the scrum has undergone a number of rule changes, most recently from a “crouch-touch-set” (CTS, in 2012/2013) to a “crouchbind-set” (CBS, in 2013/2014) procedure, in an attempt to improve safety by de-emphasizing the initial impact of the scrum engagement (Cazzola et al., 2014; Preatoni et al., 2014). The CBS technique resulted in a significant reduction, approximately a 20% effect, in the biome-
chanical conditions (force, acceleration, and impact speed) experienced by front-row players in live scrummaging (Cazzola et al., 2014). However, the external mechanical load acting on forwards during a scrum is still considerable and the ways in which these loads transmit across the anatomical structures as well as the possible threshold for injury are still to be understood and described. Players’ neck and spinal strength is, therefore, crucial to absorb and control such mechanical stresses. Rugby forwards have been shown to have a specific neck strength profile (Olivier & Du Toit, 2008) that allows them to generate higher isometric force in extension, flexion and rotation motions, compared with backs (Hamilton & Gatherer, 2014). Currently, no study has investigated the effects of rule and technique changes on spinal muscle activity. The analysis of neuromuscular activation patterns under different engagement conditions is fundamental to elucidate the strategies employed by front-row players as they prepare their bodies for the engagement and pushing actions. Such information may allow sports science and medicine practitioners to provide optimal muscle conditioning and more specific rehabilitation programs, and may have a positive impact in terms of quicker and safer return to competition following injury. In fact, cervical
*These authors equally contributed to the study.
432
Spinal muscle EMG in a simulated rugby scrum spine and neck injuries produce an excessively high number of early recurrent injuries in elite rugby union, potentially due in part to unspecific and inappropriate rehabilitation programs (Williams S, unpublished observation). Analysis of muscle activation patterns during scrummaging may also inform training practice, such as in the conditioning of novice front-row players with regard to preparation for live match scrummaging. The aim of this study was to determine the activity of the bilateral upper trapezius, sternocleidomastoid, and erector spinae muscles under three scrummaging conditions: two machine scrummaging conditions, the “crouch-bind-set” and “crouch-touch-set”; and a single live scrummaging condition were investigated. The first hypothesis was that spinal muscle activity is greater for the CTS than the CBS scrummage caused by the higher biomechanical conditions (forces, accelerations, and impact speed) experienced by front-row players during the engagement in the CTS condition. The second hypothesis was that spinal muscle activity would be greater for live scrummaging compared with machine scrummaging caused by a sustained-push phase against a moveable target that decreases scrum stability. The third hypothesis was that spinal muscles activity across scrum phases would have a maximum value during the engagement phase caused by presence of an impact and the related high physical demand of that phase. Methods Study design In a repeated measures design, a group of rugby union front-row forwards performed multiple trials in three different simulated scrummage conditions (within-group factor) and throughout the phases of scrummaging (within-group factor) to assess and compare spinal muscle activity (dependent variable).
Participants Nine male rugby union players (age 20.3 ± 1.3 years, height 1.80 ± 0.10 m, weight 102.36 ± 15 kg), of at least University 1st XV standard with a minimum of 3 years playing experience in the front row and no history of spinal injuries in the 12 months prior to testing, participated in the study. All participants provided written informed consent prior to participation and ethical approval was obtained from the University of Bath Institutional Ethics Committee.
Data collection For electromyography (EMG) collection, six wireless electrodes (Delsys Trigno, Delsys Inc, Boston, Massachusetts, USA), sampling at 2000 Hz, were attached (Delsys adhesive interface) to the sternocleidomastoid, midway between the rostral and sternal attachments; upper trapezius, 1 cm superior to the scapula spine midway between the medial origin of the scapula spine and the acromion; and erector spinae, 3.5 cm from the midline of the spine at the level of L4-5 (Sharp et al., 2014; Fig. 1). Surface EMG signals were collected bilaterally on each participant (Delsys EMGworks 4.1.05, Delsys Inc). Prior to the mounting of electrodes, the skin surface was prepared by shaving, lightly abrading, and cleaning with alcohol wipes.
Fig. 1. Electrode and marker setup. Electrode positions are highlighted with black circles. Sternocleidomastoid (SCM) is shown on the left-hand side of the figure while upper trapezius (UT) and erector spinae (ES) on the right-hand side.
The skin impedance was verified and measurements were carried out having an impedance lower than 5 kΩ. Following a player-led warm-up, each player performed two 4-s isometric maximal voluntary contractions (MVC) of the upper trapezius, sternocleidomastoid, and erector spinae, with a 1-min break between each measurement using the procedures defined in Table 1 (Vera-Garcia et al., 2010; Morimoto et al., 2013). Each participant then performed a number of submaximal scrummaging trials to become familiar with the experimental and environmental conditions (indoor scrummaging on a rubber-based floor). The three different engagement techniques, crouch-touchset (CTS), crouch-bind-set (CBS), and live 2-vs-1 scrummaging (Live; Table 2), were used. The machine scrummaging condition involved a single participant engaging with a static scrum machine (Dictator, Rhino Rugby, Rooksbridge, UK) using the CTS and CBS variants followed by a sustained push (Preatoni et al., 2012). The beams of the scrum machine were instrumented with strain gauges in order to measure the forces generated by front-row players while scrummaging (Preatoni et al., 2012). The live condition involved a single participant passively engaging, for safety reasons, with two other participants prior to the sustained push (Fig. 2). The movement patters across all the phases are essentially identical for all the scrum engagement techniques except for an increased displacement of the body center of mass in the CTS condition. Each participant then completed at least four successful trials in each of the three scrum conditions, which were presented in a blocked random order, up to a maximum of 24 trials in one session. A trial was considered successful when the player engaged timely with the prerecorded referee calls (± 0.5 s from the start of each call). Recovery intervals
433
Cazzola et al. Table 1. Positions and resistances used to measure MVC for the sternocleidomastoid, upper trapezius, and erector spinae
Condition
Description
Sternocleidomastoid (SCM) Participant stands with trunk and hips flexed to 90 degrees, so that the trunk is parallel to the floor with the neck in a neutral position with the forehead placed on the scrum machine pad. The participant attempts to flex the neck against the fixed pad. Upper trapezius (UT) Participant lies in a prone position with both arms abducted at the shoulder (∼45°) and externally rotated with the elbow flexed. The participant attempts to abduct the arms against manual resistance applied to the elbow. Erector spinae (ES) Participant lies in a prone position with the torso on the table, the legs projected horizontally over the end of the table, and arms out of the side. The participant attempts to extend the lower trunk and hip against manual resistance applied to the posterior thigh and shoulders bilaterally.
Table 2. Scrummage conditions
Condition
Description
CTS (2012–2013)
A single participant engaged with a scrum machine following the engagement cues “crouch-touch-set.” On “crouch” (t = 0 s), the participant moved into their normal crouch posture. On “touch” (t = 1.7 s), the participant moved their right arm to touch the pad and then withdrew the arm to assume the crouch posture. On “set” (t = 4 s) the participant dynamically engaged with the scrum machine and then started a sustained push for approximately 3 s. A single participant engaged with a scrum machine following the engagement cues “crouch-bind-set.” On “crouch” (t = 0 s), the participant moved into their normal crouch posture. On “bind” (t = 1.7 s), the participant moved their right and left arm to bind onto the scrum machine pusher arms. On “set” (t = 4 s), the participant dynamically engaged with the scrum machine and then started a sustained push for approximately 3 s. A single participant scrummaged against two other participants when cued by the call “set” (t = 4 s). Prior to the cue, the participant passively engaged with the two opposing participants binding with both arms on to the opposing participants’ backs. The two opposing participants bound together, as would a loose head prop and a hooker in a complete front row. The opposing participants were asked to place their unbound hand on the ground to provide additional stability and to hold the test participant in a relatively static position. On “set,” the participant maximally pushed against the two opposing participants for approximately 3 s.
CBS (2013–present)
Live
Fig. 2. Images of “key instances” in the CBS (left), CTS (central), and live (right) scrummaging conditions. (a) Crouch position; (b) “bind” call; (c) sustained-push phase. Phases a and b are not reported during live scrummaging as muscle activity was only measured during the sustained push caused by the passive engagement. of 1 < t < 2 min between consecutive trials and 7 < t < 10 min between sets were included to avoid fatigue. A 13-camera (12 cameras Oqus 400, 1 Oqus 210c) motion capture system (Qualysis, Goteborg, Sweden) capturing at 250 Hz was used to collect players’ kinematics and to determine the time
434
of actual engagement. A single marker was rigidly attached to the posterior aspect of each of the two central scrum machine pusher-arm pads to detect movement on initial contact. A bespoke control and acquisition system (cRIO-9024, National Instruments, Austin, Texas, USA) was programmed (Labview 2010, National
Spinal muscle EMG in a simulated rugby scrum
Fig. 3. Exemplary compression force and normalized trapezius EMG signal from a CTS scrummaging trial. The three phases (pre-engagement, engagement, and sustained push) considered in the study are highlighted in the graph. Instruments) to synchronously trigger the acquisition hardware (Delsys EMG, Qualysis), collect and store the forces measured through the instrumented scrum machine, and playback prerecorded cues given by the referee (Preatoni et al., 2013). The cues, “crouch-touch-set” and “crouch-bind-set,” were delivered with consistent timing (Table 2) for all the scrummaging conditions. Muscle activity was measured for 10 s from 1 s prior to the “crouch” call.
Data processing Raw electromyograms were filtered by applying a bidirectional second-order Butterworth low-pass and high-pass filter between 20 and 200 Hz. The data were then rectified and smoothed using a moving average over 50-ms windows (Visual 3D v5, C-Motion Inc, Germantown, MD, USA). EMG signals were normalized to the MVCs, which were calculated between 0.2 and 1.2 s after the initiation of maximum isometric muscle contraction, with an average MVC value being calculated over two trials. Average muscle activity (average rectified EMG amplitude) during scrum trials was calculated over three phases of each scrum: the preengagement, the engagement, and the sustained-push phases. For the machine scrummaging conditions (CBS and CTS), the pre-engagement phase was defined as the interval between the “set” cue and the instant of first contact between participant and pusher arm. As in Preatoni et al. (2014), and considering the fast loading rate in the shock-absorption phase (Fig. 3), the engagement phase was defined as the time the participant first contacted the pusher arms until 1 s after the initial contact, the sustainedpush phase extended from the end of engagement for 1 s (Fig. 3). First contact between participant and pusher arms was determined from the horizontal displacement of the markers on the scrum machine pusher arm using the Qualysis analysis software (QTM 2.9). During live scrummaging, players were bound together at the “set” call, which was an invitation for them to start pushing. Therefore, in the Live scrummage, the sustained-push phase was defined as the interval between 1 and 2 s after the “set” cue.
Statistics Separate one-way repeated measure analysis of variance (ANOVA; with scrummage conditions as the within-group factor)
and Bonferroni post-hoc analysis were applied (SPSS software, IBM Corp, New York, USA) to determine if there were any differences (P < 0.05) in muscle activation across scrummage conditions during the sustained-push phase (CBS vs CTS vs Live). A paired t-test was performed to determine the differences in muscle activation in the pre-engagement phase (CBS vs CTS), and the engagement phase (CBS vs CTS), as Live engagement included only the sustained-push phase (Fig. 2). Further one-way repeated measure ANOVAs (with scrum phases as the within-group factor) were used to test possible changes in muscles activation across the different phases of the scrum for the CTS and CBS machine trials, followed by Bonferroni post-hoc comparisons (P < 0.05). Sphericity of the data was assessed using the Greenhouse–Geisser epsilon; if the data were aspherical, a correction was applied to the calculated P-value. Pairwise effect sizes, calculated using Cohen’s d statistic (d) (1988), were also considered (Appendix).
Results The time of contact (tENG) was highly repeatable across all participants (0.55 ± 0.08 s). Also, the time of onset of movement had very low variability (0.04 s), and its average value was 0.03 s. This provided the analysis with a consistent scrum phase duration (e.g., time interval between scrum phases) across all the subjects.
Comparing conditions Left and right muscles in all the analyzed muscle groups exhibited very similar level of activation across the subsequent phases of scrummaging. Therefore, EMG data from left and right side were pooled. In the pre-engagement phase, all the measured muscles in both machine conditions were activated in excess of 25% MVC (Fig. 4). The activity of the sternocleidomastoid and upper trapezius was significantly higher (P < 0.01) in the CBS condition than the CTS
435
Cazzola et al.
Fig. 4. Normalized values (% maximal voluntary contraction) of muscles activation (mean and standard deviation) of sternocleidomastoid (SCM), upper trapezius (TRAP), and erector spinae (ES) during CBS (crouch-bind-set) and CTS (crouch-touch-set) engagements. Muscle activations during CBS and CTS engagements are shown throughout the three scrum phases: pre-engagement, engagement, and sustained push. Live engagement is not included because in that condition, EMG measures were carried out only during sustained-push phase. The dashed and dotted lines show the muscles activation trend in, respectively, CTS and CBS, and are representative of the differences (ANOVA) between (1) pre-engagement and sustained push and (2) engagement and sustained push. ‡ Significant difference between CBS and CTS (P < 0.05 – paired t-test). ¥Significant difference between both (1) pre-engagement and sustained push, and (2) engagement and sustained push in CBS (P < 0.05 – ANOVA). §Significant difference between both (1) pre-engagement and sustained push, and (2) engagement and sustained push in CTS (P < 0.05 – ANOVA).
Fig. 5. Normalized muscle activation (mean and standard deviation) of sternocleidomastoid (SCM), upper trapezius (TRAP), and erector spinae (ES) during CBS, CTS, and Live engagements throughout the sustained-push phase. *Significant difference between CBS and Live (P < 0.01 – ANOVA). †Significant difference between CTS and Live (P < 0.01 – ANOVA).
condition. The activity of the erector spinae tended to be more activated (large effect size, d > 0.8), although not significantly (P > 0.05), in the CTS condition during preengagement than the CBS condition. During the engagement phase, all the measured muscles were more active in the CBS than CTS condition. The activity of the upper trapezius and sternocleidomastoid were significantly higher (P < 0.05 – paired t-test), showing an average increase of 20 ± 12% and 22 ± 20%, respectively (Fig. 4), and large effect size (d > 1). The ES tended to be more activated during the CBS condition (effect size > 0.8), although not significantly.
436
During the sustained-push phase, the activity of the muscles across all three conditions (CBS, CTS, and Live) could be compared (Fig. 5). The activity of the erector spinae was significantly higher during the sustained-push phase of live scrummaging than in either of the CBS or CTS conditions (P < 0.01), with the erector spinae activity approximately 56 ± 26% lower (CBS vs Live) and 62 ± 18% lower (CTS vs Live) and large effect size (d > 0.8). The activity of the upper trapezius tended to be lower in the CTS than the CBS and Live conditions (d > 1.1), whereas the activity of the sternocleidomastoid was similar across conditions.
Spinal muscle EMG in a simulated rugby scrum Comparing across phases The activity of the sternocleidomastoid and erector spinae showed a decreasing trend (P < 0.05) moving from pre-engagement through engagement to the sustained push in both the CBS and CTS conditions (Fig. 4). The activity of the sternocleidomastoid was significantly higher during the pre-engagement and engagement phase than sustained push in CBS (P < 0.05), and in CTS, pre-engagement tended to be higher than sustained push (P = 0.059) showing a large effect size (d > 1.1). The activity of the erector spinae during both CBS and CTS conditions was significantly higher during both the pre-engagement phase (P < 0.05) and engagement phase (P < 0.05) when compared with the sustained push (Fig. 4). There was a significant pattern of decreasing activation from (a) pre-engagement to engagement; and from (b) engagement to sustained push that was also reflected in large effect sizes (d > 1.7). The activity of the upper trapezius in the CBS and CTS conditions peaked, but was not significantly greater, during the engagement phase (Fig. 4). No significant differences were calculated, though large effect sizes (d > 1.2) were found for the upper trapezius between the engagement phase and sustained push in both CBS and CTS conditions. Also, medium effect sizes (d > 0.5) were found for the upper trapezius between the preengagement phase and engagement in both CBS and CTS conditions. Discussion The aim of this study was to gain more insight into the activity of sternocleidomastoid, upper trapezius, and erector spinae muscles during machine (CBS and CTS) and live scrummaging. Compared with the machine conditions, the live condition resulted in significantly higher activation of erector spinae during the sustained-push phase. In contrast with the initial hypothesis, the activity of sternocleidomastoid, upper trapezius, and erector spinae tended to be greater during the CBS condition than the CTS condition throughout the three phases of the scrum. Spinal muscle activity in both machine scrummaging conditions was characterized by a considerable preactivation of all six muscles (>25% MVC) prior to engagement. This pre-activation can functionally lead to an increase in cervical and lumbar spine stiffness (Riemann & Lephart, 2002), which may better prepare the player’s spinal structures for the high biomechanical loads placed upon the spine during engagement. A high level of muscle pre-activation may potentially mitigate the effect of loads on spinal posture (Krajcarski et al., 1999; Choi, 2003) and help maintain optimal neutral spine position (Brooks & Kemp, 2011). However, it must be observed that the stabilization of the spine caused by high level of muscle activation and agonist/antagonist co-contraction (Cheng et al., 2008) may not be enough
to limit cervical spine hyperflexion or buckling mechanisms (Dennison & Macri, 2012; Kuster et al., 2012) during high-dynamic impulsive impacts with misaligned geometry. In fact, in case of a catastrophic injury, the forces and moments acting on the cervical spine are presumably much higher than the forces that can be generated by spinal muscle activation, and further analyses are needed to elucidate the actual contribution that muscles activations can make in certain high-risk loading conditions. The activity of erector spinae has been measured in a similar machine scrummaging study (Sharp et al., 2014), although the engagement condition was not clearly described, and the pre-engagement phase was defined as the 200-ms window prior to engagement, differing from the current study. Sharp et al. (2014) reported a preactivation of 65% of MVC in the erector spinae, which is comparable with our findings of 68% and 75% erector spinae pre-activation in CBS and CTS conditions. Sharp et al. (2014) interpreted erector spinae pre-activation as necessary to maintain an efficient scrummaging position and to overcome gravity forces causing a bending moment and a rotation of the trunk about the pelvis. Our results support this explanation, as the slight reduction in erector spinae activity in the CBS condition when compared with the CTS condition, in the pre-engagement phase, can be attributed to pre-binding, reducing the moment acting on the trunk joints caused by torso weight being partially supported. The engagement phase of a scrum is characterized by a peak in both compression and vertical force, approximately 0.1 s after first contact (Fig. 3), followed by a lower plateau in force traces approximately 0.5 s after engagement (Preatoni et al., 2013, 2014; Cazzola et al., 2014). These changes in force generate the need for a high spinal stiffness and therefore demand compensatory spinal muscle activation. Cholewicki et al. (2000) identified that as the external load placed upon the spine increased, the stiffness of the spine also increased, and most interestingly, this was caused by an increase in spinal muscle activation. Therefore, we propose that the EMG signal of the participant’s spinal musculature can take as an input parameter and respond to the time course of the external load exerted on players’ shoulders. The spinal muscles are maximally activated in response to the peak in external load and then decrease 0.5 s after engagement as the external load reduces (Fig. 3). This stabilization in activation is observed in all six (e.g., left and right) muscles during the sustained-push phase irrespective of condition. We have previously observed a reduction in external loading on players when scrums use the CBS condition when compared with the CTS condition (Cazzola et al., 2014). If these reductions in mechanical loading during the engagement were transferred to the modified scrum condition used in the present study, it can be hypothesized that the spinal muscles would be more active
437
Cazzola et al. during CTS than CBS condition, as the spine is exposed to greater mechanical stresses. However, the activity of the spinal muscles was comparable between the conditions and the activity of the cervical spine muscles (sternocleidomastoid and upper trapezius) was found, in the engagement phase, to be significantly greater in the CBS condition. These differences may be attributed to the differences in binding between the conditions, as the upper trapezius is responsible for the maintenance of position of the cervical spine and also the upper arm (Herrington & Horsley, 2009). Upper trapezius is also a scapula stabilizer, and its higher activation during the pre-engagement and engagement phase in CBS provides more control and stiffness during the binding procedure. In the CBS condition, the players establish a secure pre-bind prior to engagement, while in the CTS condition, the players have to bind during the engagement. Additionally, during the CBS conditions, players set up closer to the scrum machine (Preatoni et al., 2014), and this posture may effect spinal muscle activation, where greater upper trapezius and sternocleidomastoid activation increase cervical spine stiffness and may better maintain cervical spine posture during the engagement. This indicates that a pre-bind procedure makes the upper spine more prepared for the scrum engagement. The activation of the erector spinae was significantly higher (P < 0.05) throughout the sustained-push phase of the live scrummaging condition than either of the machine scrummaging conditions. Live scrummaging is an unstable dynamic condition when compared with machine scrummaging; therefore, the forces applied to the opposition players are not equally matched in direction or magnitude, as they are during a scrum against the machine. These eccentric forces theoretically generate moments at trunk level, causing lumbar flexion, extension or rotation, and placing the participant in a compromised lumbar spinal posture. Thus, the participant needs to maintain optimal lumbar spinal posture via the activation of the stabilizing lumbar muscles (represented by the activation of the erector spinae in this study). These findings demonstrate that although machine and live scrummaging show comparable kinematic and kinetic characteristics, machine scrummaging does not accurately replicate live scrummaging in terms of muscle activation strategies required to maintain the sustained phase. Thus, it can be suggested that front-row players should be required to train in live situations and not only against scrum machines. These findings suggest that although scrum machines are a useful training aid, a relative increase in live scrummaging practice may be beneficial for all front-row players from the perspective of training appropriate and specific neuromuscular activation patterns. This may be particularly important under current rugby practice since the duration of the sustained phase has markedly increased under the CBS engagement procedure.
438
In the current study, each participant scrummaged individually in three conditions; this is a constrained environment. The results observed in such modified conditions cannot be extrapolated directly into a full scrummage scenario. During a complete scrummage, eight players interact and bind to one another generating the forces and movements of the scrummage (Preatoni et al., 2014). Sharp et al. (2014) reported a peak compression force of 3.1 kN in a single participant machine scrummaging, whereas this study recorded an average maximum compression of about 2.8 kN. These magnitudes are about 25% less than the ones found by Preatoni et al. (2014), who measured a peak compression force of 11.7 kN, a vertical force of 1.5 kN, and a lateral force 1.3 kN distributed across the three front-row players in a full scrummage of academy level. The reduction in compression forces and the absence of extra-loading from multiplayer interactions may result in a decrease in muscle activation for one-person scrummaging, suggesting that the results of the current study may underestimate the spinal muscle activity observed throughout a full scrummage. Additionally, front-row and second-row players have loads applied from front to back, which may cause other types of loading patterns around the spine. The current study was undertaken in an indoor lab with an individual scrummaging against a machine or two other participants. The study was one of the first of its kind to measure muscle activity in an impact event similar to match intensity, and is a fundamental first step toward understanding the muscle activations in a full scrummage scenario. A limitation of the current study was that muscle activity in the live condition was only measured during the sustained-push phase. Future research should build on the foundations of the current study, with the objectives of improving its ecological validity through trials on natural turf and in a more complete scrummaging scenario (a full front row or a complete pack) either against a scrum machine or in live conditions. The use of bilateral measurement of muscle activity is advised, as a more complete scrummage is unlikely to match the symmetry of the current binding conditions. In order to provide a full description of the trunk muscles contributing to spine stability, the measurement of abdominal muscle activity should be included. However, soft tissue movement artifact and low signal-to-noise ratio, caused by the physical characteristics of front-row rugby players, are likely to affect EMG signal quality. In conclusion, the activation of selected spinal muscles is greater in the CBS condition than the CTS condition, particularly in the pre-engagement phase. This indicates that the muscles of the cervical spine, in the CBS condition, are better prepared for the forces experienced during the scrum engagement than in the CTS condition as cervical spine stiffness is greater. Furthermore, this research provides evidence that the
Spinal muscle EMG in a simulated rugby scrum erector spinae is significantly more active during live scrummaging than machine scrummaging. This reinforces the requirement for individuals to practice and learn scrummage techniques in a live situation, rather than purely against a machine, as machine scrummaging does not replicate the demands of a live contest. Perspectives The findings of this study may have an impact in sports medicine, especially in the injury prevention and injury biomechanics areas. Firstly, the evidence of spinal muscle pre-activation highlights the presence of a specific trunk-stiffening strategy aiming to prepare the body for the impact, and therefore, its contribution needs to be considered during any real-world injury mechanism analysis in contact sports. Also, the description of the activation patterns of the trunk muscles may provide new insights for the optimization of specific training and rehabilitation programs with a specific view to injury prevention. Finally, this study can provide further evi-
dence on which to inform discussions relating to the scrum laws of rugby union when seeking to improve player welfare. In fact, the resulting information adds the measure of spinal muscle activation to the analysis of the movements and forces involved during different scrum engagements (Preatoni et al., 2013, 2014; Cazzola et al., 2014), providing a better understanding of the biomechanical load experienced by rugby forwards. Key words: Biomechanics, sports injury, sports performance, scrummaging technique, lumbar spine, cervical spine.
Acknowledgements The authors would like to thank Dr Polly McGuigan, Dr Richie Gill, Dr Sabina Gheduzzi, Dr Keith Stokes, and Dr Tony Miles for their involvement in the wider programme of research and their comments on this manuscript. Also, we would like to thank all rugby forward players that took part to the study. This project is funded by the Rugby Football Union (RFU) Injured Players Foundation.
References Brooks JH, Fuller CW, Kemp SP, Reddin DB. Epidemiology of injuries in English professional rugby union: part 1 match injuries. Br J Sports Med 2005: 39: 757–766. Brooks JHM, Kemp SPT. Injury-prevention priorities according to playing position in professional rugby union players. Br J Sports Med 2011: 45: 765–775. Brown JC, Lambert MI, Verhagen E, Readhead C, van Mechelen W, Viljoen W. The incidence of rugby-related catastrophic injuries (including cardiac events) in South Africa from 2008 to 2011: a cohort study. BMJ Open 2013: 3 (2): e002475. doi:10.1136/bmjopen2012-002475. Castinel BH, Adam P, Prat C. A stress fracture of the lumbar spine in a professional rugby player. Br J Sports Med 2007: 41: 337–338. Cazzola D, Preatoni E, Stokes KA, England ME, Trewartha G. A modified prebind engagement process reduces biomechanical loading on front row players during scrummaging: a cross-sectional study of 11 elite teams. Br J Sports Med 2014: doi:10.1136/ bjsports-2013-092904. Cheng CH, Lin KH, Wang JL. Co-contraction of cervical muscles during sagittal and coronal neck motions at different movement speeds. Eur J Appl Physiol 2008: 103: 647–654. Choi H. Quantitative assessment of co-contraction in cervical musculature.
Med Eng Phys 2003: 25: 133– 140. Cholewicki J, Simons APD, Radebold A. Effects of external trunk loads on lumbar spine stability. J Biomech 2000: 33: 1377–1385. Cohen J. Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates, 1988. Delp SL, Anderson FC, Arnold AS, Loan P, Habib A, John CT, Guendelman E, Thelen DG. OpenSim: open-source software to create and analyze dynamic simulations of movement. IEEE Trans Biomed Eng 2007: 54: 1940–1950. Dennison CR, Macri EM, Cripton PA. Mechanisms of cervical spine injury in rugby union: is it premature to abandon hyperflexion as the main mechanism underpinning injury? Br J Sports Med 2012: 46: 545–549. Hamilton D, Gatherer D. Cervical isometric strength and range of motion of elite rugby union players: a cohort study. BMC Sports Sci Med Rehabil 2014: 6: 32. Herrington L, Horsley I. Electromyographic analysis of selected shoulder muscles during a rugby football tackle. Sports Med Arthrosc Rehabil Ther Technol 2009: 1: 10. doi:10.1186/1758-2555-1-10. Hogan BA, Hogan NA, Vos PM, Eustace SJ, Kenny PJ. The cervical spine of professional front-row rugby players: correlation between degenerative changes and symptoms. Ir J Med Sci 2010: 179: 259–263.
Krajcarski SR, Potvin JR, Chiang J. The in vivo dynamic response of the spine to perturbations causing rapid flexion: effects of pre-load and step input magnitude. Clin Biomech (Bristol, Avon) 1999: 14: 54–62. Kuster D, Gibson A, Abboud R, Drew T. Mechanisms of cervical spine injury in Rugby Union: a systematic review of the literature. Br J Sports Med 2012: 46 (8): 550–554. doi: 10.1136/bjsports2011-090360. Milburn PD. The kinetics of rugby union scrummaging. J Sports Sci 1990: 8: 47–60. Morimoto K, Sakamoto M, Fukuhara T, Kato K. Electromyographic study of neck muscle activity according to head position in rugby tackles. J Phys Ther Sci 2013: 25: 563–566. Olivier PE, Du Toit DE. Isokinetic neck strength profile of senior elite rugby union players. J Sci Med Sport 2008: 11: 96–105. Pinsault N, Anxionnaz M, Vuillerme N. Cervical joint position sense in rugby players versus non-rugby players. Phys Ther Sport 2010: 11: 66–70. Preatoni E, Stokes KA, England ME, Trewartha G. The influence of playing level on the biomechanical demands experienced by rugby union forwards during machine scrummaging. Scand J Med Sci Sports 2013: 23: e178–e184. Preatoni E, Stokes KA, England ME, Trewartha G. Engagement techniques and playing level impact the biomechanical demands on rugby forwards during machine-based
439
Cazzola et al. scrummaging. Br J Sports Med 2014: doi:10.1136/bjsports-2013-092938. Preatoni E, Wallbaum A, Gathercole N, Coombes S, Stokes KA, Trewartha G. An integrated measurement system for analysing impact biomechanics in the rugby scrum. Proc Inst Mech Eng Part P J Sports Eng Technol 2012: 226 (3–4): 266–273. Quarrie KL, Cantu RC, Chalmers DJ. Rugby union injuries to the cervical spine and spinal cord. Sports Med 2002: 32: 633–653.
Quarrie KL, Wilson BD. Force production in the rugby union scrum. J Sports Sci 2000: 18: 237–246. Riemann BL, Lephart SM. The sensorimotor system, part I: the physiologic basis of functional joint stability. J Athl Train 2002: 37 (1): 71–79 Sharp T, Halaki M, Greene A, Vanwanseele B. An EMG assessment of front row rugby union scrummaging. Int J Perform Anal Sport 2014: 14: 225–237.
Trewartha G, Preatoni E, England ME, Stokes KA. Injury and biomechanical perspectives on the rugby scrum: a review of the literature. Br J Sports Med 2014: doi:10.1136/bjsports-2013092972. Vera-Garcia FJ, Moreside JM, McGill SM. MVC techniques to normalize trunk muscle EMG in healthy women. J Electromyogr Kinesiol 2010: 20: 10–16.
Appendix Table A1. Pairwise effect sizes calculated between CBS, CTS, and live conditions in pre-engagement (Pre-Eng), engagement (Eng), and sustained push (Sus-Push) scrum phases
Condition
CBS CTS
Phase
Pre-Eng–Eng Eng–Sus-Push Pre-Eng–Sus-Push Pre-Eng–Eng Eng–Sus-Push Pre-Eng–Sus-Push
Muscles SCM
TRAP
ES
0.38 2.70 1.54 0.57 2.50 1.15
0.51 1.46 0.01 0.69 1.38 0.01
1.70 11.46 4.89 4.11 9.78 2.81
Note: Pairwise effect sizes were calculated using Cohen’s (d) values. |d| > 0.8 large effects; |d| > 0.5 moderate effects; |d| > 0.2 small effects.
Table A2. Pairwise effect sizes calculated between pre-engagement (Pre-Eng), engagement (Eng), and sustained push (Sus-Push) scrum phases, in CBS, CTS, and live conditions
Phase
Pre-Eng Eng Sus-Push Sus-Push Sus-Push
Condition
CBS vs CTS CBS vs CTS CBS vs CTS CBS vs Live CTS vs Live
Muscles SCM
TRAP
ES
1.26 2.97 0.57 0.26 1.80
1.46 1.17 1.86 0.88 0.87
0.87 0.91 1.74 0.88 0.87
Note: Pairwise effect sizes were calculated using Cohen’s (d) values. |d| > 0.8 large effects; |d| > 0.5 moderate effects; |d| > 0.2 small effects.
440
