Trunk Muscle Activation and Estimating Spinal Compressive Force in Rope and Harness Vertical Dance Margaret Wilson, Ph.D., Boyi Dai, Ph.D., Qin Zhu, Ph.D., and Neil Humphrey, Ph.D.
Abstract Rope and harness vertical dance takes place off the floor with the dancer suspended from his or her center of mass in a harness attached to a rope from a point overhead. Vertical dance represents a novel environment for training and performing in which expected stresses on the dancer’s body are different from those that take place during dance on the floor. Two male and eleven female dancers with training in vertical dance performed six typical vertical dance movements with electromyography (EMG) electrodes placed bilaterally on rectus abdominus, external oblique, erector spinae, and latissimus dorsi. EMG data were expressed as a percentage of maximum voluntary isometric contraction (MVIC). A simplified musculoskeletal model based on muscle activation for these four muscle groups was used to estimate the compressive force on the spine. The greatest muscle activation for erector spinae and latissimus dorsi and the greatest trunk compressive forces were seen in vertical axis positions where the dancer was moving the trunk into a hyper-extended position. The greatest muscle activation for rectus abdominus and external oblique and the second highest compressive force were seen in a supine position with the arms and legs extended
away from the center of mass (COM). The least muscle activation occurred in positions where the limbs were hanging below the torso. These movements also showed relatively low muscle activation compression forces. Post-test survey results revealed that dancers felt comfortable in these positions; however, observation of some positions indicated insufficient muscular control. Computing the relative contribution of muscles, expressed as muscle activation and estimated spinal compression, provided a measure of how much the muscle groups were working to support the spine and the rest of the dancer’s body in the different movements tested. Additionally, identifying typical muscle recruitment patterns in each movement will help identify key exercises for training that should promote injury prevention.
B
iomechanics has long been used to study specific dance movements and can offer insight into the movement often unnoticed by dance educators.1,2 Some of the dance forms studied to date include ballet, modern, tap, and Irish Dance. 3-6 Utilizing tools, such as motion capture, electromyography
Margaret Wilson, Ph.D., Department of Theatre & Dance, Boyi Dai, Ph.D., Department of Kinesiology and Health, Qin Zhu, Ph.D., Department of Kinesiology and Health, and Neil Humphrey, Ph.D., Department of Geology and Geophysics, and Department of Theatre & Dance, University of Wyoming, Laramie, Wyoming. Correspondence: Margaret Wilson, Ph.D., Department of Theatre & Dance, University of Wyoming, 1000 East University Avenue, Dept. 3951, Laramie, Wyoming 82071; [email protected]. Copyright © 2015 J. Michael Ryan Publishing, Inc. http://dx.doi.org/10.12678/1089-313X.19.4.163
(EMG), ground reaction force, and muscular modeling, movements can be isolated and observed relative to the forces of gravity and torque to understand the relationship that dancers have with their environment, as well as the internal forces acting on the body. Traditional methodology for biomechanical analysis assumes that the dancer is standing on the floor or weightbearing through the upper body. The question driving this research was whether biomechanics could help researchers understand a new dance form, rope and harness vertical dance—i.e., dance that takes place off the floor with some apparatus suspending the body or from which the body hangs. Vertical dance is often classified as a form of aerial dance, an outgrowth of circus tradition that includes trapeze, silks, or fabrics; however, rope and harness vertical dance uses traditional climbing and rescue equipment. Early work in vertical dance was experimental and often inspired by aesthetic climbing.7 The dancer wears a harness around the hips, and the harness is suspended by a rope, which is secured overhead. Thus, the dancer is hanging in midair with support coming only from the pelvis, not from the feet on the ground. In this configuration, the dancer can interact with a vertical wall or hang in “free space” and can reorient the body so that he or she is upright, parallel to the floor, upside down, and many dif163
164
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
ferent angles in-between. The vertical dancer uses familiar movements from traditional dance styles and explores these actions in new relationships to gravity, with a different base of support. The dancer is afforded almost 360° of motion, depending on how the harness is attached to the rope. Suspended in air, the dancer’s trunk and hip muscles must create leverage to control and execute movement without an external source of torque, such as would be found when performing these movements on the floor. Hanging in the harness creates an upward force on the dancer’s pelvis, while at the same time the weight of the dancer’s body creates a downward tensional force. With no other source of support, trunk and hip muscle activation must create the desired orientation of the trunk in the harness and also control and support the movement of the upper body and legs. This is seen in all orientations to gravity, such as when the dancer is upside down or perpendicular to the rope and parallel to the floor. The dancer is creating similar movements to those performed in a traditional dance form, with a different relationship of the weights of the body parts (trunk, arms, and legs) to gravity. The typical use pattern of muscles to create movement is altered with this different orientation to gravity. For example, when standing upright, the line of gravity passes through the length of the body in the vertical axis, and most muscular activity is directed to cocontraction for balance and stability. When the dancer is suspended and the body is perpendicular to the rope and parallel to the floor, all anterior muscles, primarily flexors, must contract in order to keep the body in the horizontal plane, as gravity is creating an extensor force (Fig. 1). The initial concern in this investigation was to look at the potential stresses on the spine in vertical dance. There are two aspects of this stress. First, with the feet off the ground, the dancer’s legs hang down from the pelvis, rather than serving to support the pelvis and spine as they do in a standing position. Second, although
Figure 1 Plank.
Figure 2 Spider.
Figure 3 Side-lying.
the harness acts as suspension for the weight of the body, its fabric construction does not provide structural support. In positions with the limbs acting as long resistance arms, such as Plank (Fig. 1), with the spine bearing weight perpendicular to the spinal column (not a typical stress), are there undue stresses being created? A literature review generated no information on calculating forces acting on the dancer in an airborne environment. Comparisons were sought in the disciplines of gymnastics, swimming, yoga, and circus arts; however, to date, no
evidence of similar movement studies has been identified. Additionally, specific vertical dance movements generate different questions. For Spider (Fig. 2) and Side-lying attitude (Fig. 3), the dancers could approximate these positions with minimal muscular effort. This means that in these positions the dancer’s body is hanging in the harness, yet due to the fabric construction of the harness, there can be little control of the angular relation between body segments (i.e., too much spine hyperextension or
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
Figure 4 High Release.
Figure 5 Inversion.
lateral flexion). This is not the way the movements are taught, but one researcher’s observation of the dancer’s performance, as well as comments from the dancers, indicated that the body passively assumes the position due to gravity, thus the abdominals
are not active. The profound impact of gravity on the suspended body in the vertical dance movement of Spider stretches the muscles to full ranges of elongation. This may override the sensory structures for control, thus transferring the support to
165
the passive tissues and spine. Whereas this movement can be compared to a back bend in gymnastics or camel pose (Ustrasana) or upward facing bow pose (Urdhva Khanurasana) in yoga,8 those movements are supported by the arms and legs on the ground. Performing the same movement with the dancer suspended in the air means that the full weight of the body is transferred through the lumbar spine. As the weight of the dancer in this position is transferred horizontally through the body of the lumbar vertebrae (rather than vertically, as when standing), the dancer may well be experiencing different stresses to the body of the vertebrae. To place this in context, in a standing posture the interlocking facets provide stability and also absorb 30% of the compression force in hyperextension.9 The facets and body of the vertebrae should be sufficient to withstand this pressure, but this has not been tested in vertical dance or in other activities. This is one of the key questions in the research. In the movement Spider, particularly, the dancer can use little muscle control and simply “hangs in the harness.” As this dance form places the spine in a novel weightbearing position, we were concerned that the effect of gravity combined with lack of muscular activity creates a unique stress on the dancer’s spine. In other movements, the harness and rope configuration forces the dancer to overcome gravitational resistance to the movement. With the dancer’s body aligned in the vertical axis, such as High Release (Fig. 4), Inversion (Fig. 5), and where the body is moving toward the vertical axis as in V-sit (Fig. 6), the harness supports the pelvis and also provides a means of leverage for the trunk and limbs. In some positions, the geometry of the harness actually inhibits the desired position of the body and requires additional muscle activation. The harness attaches to the rope in the front via waist and leg loops, which displace the center of mass posteriorly. When in the vertical axis, either inverted or when
166
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
Figure 6 V-Sit.
Figure 7 Prone MVIC test.
working toward the High Release position, the dancer must actively press the pelvis anteriorly to achieve a vertical orientation. As vertical dance is a relatively new form of movement, there was little research to use as a guide in setting up this investigation. Initial attempts at using motion capture were set aside due to the difficulty of capturing all of the reflective markers in the horizontal positions or the actions that changed levels (which obscured the reflective markers) and the lack of a means to measure forces applied by the harness. Specifically, we hoped to provide an analysis of resultant joint forces and resultant joint moments acting on (and within) the body with the dancer hanging in the air. Without the ability to quantify the forces provided by the harness at different contact points, traditional biomechanical modeling was not possible in the current study. Instead, a theoretical approach was adopted based on research modeling compressive forces in the spine.10,11 A protocol was established to mea-
sure muscle activation for bilateral rectus abdominus, external oblique, erector spinae, and latissimus dorsi as percentages of maximal voluntary isometric contraction. Based on the percentages of muscle activation, an estimation of spinal compression could be made. Therefore, the primary purpose of the research was to quantify the trunk muscle activation for selected static postures in rope and harness vertical dance. The secondary purpose was to estimate the spinal compressive force for selected static postures in vertical dance using a musculoskeletal model. Although it is difficult to measure spinal loading directly without invasive measurements, an indirect estimate of trunk compression can provide insight to the internal forces acting on the spine.
Methods The current study was approved by the University of Wyoming’s Institutional Review Board. Participants signed informed consent forms prior to testing. Two male and eleven female
dancers (Table 1) participated in the data collection. The dancers were all at an intermediate level of dance training and vertical dance training and were familiar with the movements to be included in the testing. All testing took place in the biomechanics laboratory. Measures of each dancer’s height and weight were obtained, the length of the spine was measured in a sitting position on a firm surface (base of the spine to the acromion process), and the width and depth of the torso at L5 were measured in supine (Marras and Sommerich).10,11 The dancers were asked to warm up as they would for a vertical dance class. Then EMG electrodes were placed bilaterally on the rectus abdominus (RA: 3 cm lateral to midline of abdomen and 2 cm above umbilicus), external oblique (EO: 15 cm lateral to the umbilicus at a 45º angle above the anterior superior iliac spine at the level of the umbilicus), erector spinae (ES: 4 cm lateral to L3 spinous process), and latissimus dorsi (LD: most lateral portion of the muscle at the T9, 4 cm below the inferior angle of the scapula).12-14 A reference electrode was placed on the manubrium. An EMG transmitter (MyoMonitor®, Delsys, Inc., Boston, MA) was contained in a harness above the subject’s pelvis. The EMG data were captured at a sampling frequency of 2,000 Hz (EMGworks®, Delsys, Inc., Boston, MA). Prior to the vertical dance testing, the dancers performed movements of trunk flexion, extension, and rotation to make sure that a signal was being obtained for the EMG leads. Literature has suggested that surface EMG measurement is a validated and reliable tool to measure surface muscle activation and may provide insight into muscle forces.15 For the data collection, the dancers climbed a ladder to attach their harness to a rope using a gri-gri (self-belay device). The harness was suspended approximately 180 cm above the ground. The dancer was stabilized to stop any swinging motion and asked to assume the position to be tested. When the dancer demonstrated the correct position, the recording of the
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
167
Table 1 Participant Characteristics N = 13
Height (cm)
Mass (kg)
Torso Depth (cm)
Torso Width (cm)
Torso Length (cm)
Vertical dance experience (yrs)
Age
Average
169
61.0
18.0
28.4
57.8
2.8
26.3
Range
159-186.5
45.8-78.9
13.5-19
25.5-35
54-63
1-9
19-60
EMG began, and the test lasted 5 seconds. If the test was successful (the position was maintained for the required time), the dancer only performed the movement once. Visual inspection of EMG signals was performed after each trial to make sure that no noise was observed; otherwise, the trial was repeated. EMG data were collected for the eight muscles in each position. Six positions were tested: V-sit, Plank, High Release, Spider, Inversion, and Side-lying. These positions represent the core movement vocabulary used in rope and harness vertical dance training and are representative of the different orientations to gravity experienced in vertical dance. Specific descriptions of the movements follow. Plank In this position, the goal is to create a body line that is parallel to the floor. The abdominals, trunk and hip flexors, and spine extensors must be active. The abilty of the dancer to achieve a parallel orientaiton to the floor depends on strength and control, relative length and mass of upper body to leg length, and balancing at the attachment point of the harness to the rope (Fig.1). Spider The upward facing dancer allows the arms and legs to hang toward the floor, reaching behind the trunk and below the pelvis, creating full spinal hyperextension. Dancers are instructed to maintain an eccentric contraction of the abdominals throughout (Fig. 2). Side-Lying The dancer rotates in the harness so that the body is oriented to one side. The goal of the exercise is to have a horizontal orientation of the body with the top leg hyperextended at the hip and knee and the lower leg
flexed with a straight knee. The low arm extends to the floor, and the top arm extends to the ceiling, parallel with the rope. The head should be parallel with the shoulders (ideally parallel to the floor). The dancer often will touch the fingers of the top arm to the rope to help balance and support the upper body (Fig. 3). For this study, the dancers were tested on one side only, rotated to the right with the right leg in flexion and the left in hyperextension. High Release The dancer tries to bring the upper body parallel to the rope. He or she is instructed to cross the wrists over the rope rather than grasp it with the hands for leverage. Lumbar hyperextension, hip extension and external rotation, and knee flexion allows the dancers to take the legs behind the torso to the desired position (Fig. 4). Inversion The dancer is upside down with the legs straight, crossed at the ankles, with the back foot on the front of the rope and the front leg crossed over as in a classical fifth position. This position of the legs provides leverage to press the pelvis anteriorly. Inversion has the added stylistic variation of arching the head and shoulders into hyperextension, creating the illusion
of a bow in the body (Fig. 5). V-sit This movement usually starts from a sitting position, but the dancer can move to the V-sit position from Plank or from Spider. The head should remain in line with the spine, with the focus up and over the feet. The torso should be as straight as possible, with the arms and legs each extended 90º relative to the trunk (Fig. 6). Following the vertical dance movements, a test of maximal voluntary isometric contraction (MVIC) was completed. The dancer started supine on the floor, with arms crossed over the chest and legs extended along the floor. The tester applied body weight pressing down on the shoulders of the dancer, who was asked to attempt to flex the spine, bringing the shoulders up off the floor. The dancer was monitored for accessory motion in the legs during the trial. For back extension, the dancer was face down on the floor with the legs extended and arms alongside the legs but off the floor. The tester applied body weight pressing down on the shoulders of the prone dancer who was asked to attempt to take the spine into hyperextension off the floor. The dancer was monitored for accessory motion of the legs during this trial (Fig. 7). MVIC trials were performed for 10 seconds
Table 2 Participant Survey Results* Pre-test
Post-test
How long can you hold this position?
V-sit
4
3.45
6.9 sec
Plank
4.63
5.18
4.18 sec
Spider
2.63
2.72
11.8 sec
3
2.9
8.81 sec
High Release
4.9
4.36
5.72 sec
Side-lying
1.90
1.90
12.45 sec
Inversion
*Participant Survey: Rank the following in terms of difficulty for performing (1- easiest, 6- most difficult). Nine of 13 respondents reported.
168
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
to ensure a true 1-second MVIC was achieved. Following data collection the participants completed a post-test survey (Table 2).
Data Reduction EMG data were band-pass filtered from 20 Hz to 450 Hz.16 Filtered EMG data were rectified and then filtered at a low-pass cut-off frequency of 10 Hz to obtain the linear envelopes of EMG.16 The average of the MVIC data was calculated with an interval of 1 second and an increment of 1 frame (0.0005 second), for example, the average between 0 and 1 second, the average between 0.0005 and 1.0005 second, and the average between 0.001 and 1.001 second until the average between 9 and 10 seconds. The maximum average was extracted as the value for MVIC.16 EMG data during dancing trials were expressed as percentages of MVIC.16 EMG data during the middle 3 seconds of each dancing trial were averaged to represent muscle activation for the trial. A simplified musculoskeletal model was used to estimate muscle forces and L5/S1 compressive forces.12 Muscle forces were modeled as the product of the muscle activation (normalized EMG), muscle cross-sectional areas, and muscle stress. Muscle crosssectional areas were estimated as the product of the trunk length, trunk width, and specific coefficients for different muscles. The coefficients for the EO, RA, LD, and ES were 0.0148, 0.006, 0.0351, and 0.0389, respectively.10 Muscle stress was set at a constant of 47.4 N/cm2.10 RA and ES were modeled as parallel to lumbar spine, and EO and LD were modeled as 45º from lumbar spine.11 Lumbar compressive force was estimated by summing the vertical components of muscle forces and the vertical component of the body weight above the navel (45.76% of the total body weight for females and 49.11% for males).17 The calculations were performed using customized subroutines developed in MATLAB 2009a (MathWorks, Inc., Natick, MA). F = F left rectus abdominus + F right rectus
abdominus + F left erector spinae + F right erector spinae + F left external oblique * cos (45º) +F right external oblique * cos (45º) + F left latissimus dorsi * cos (45º) + F right latissimus dorsi * cos (45º) +F body weight
For the Plank, Spider, and Sidelying conditions, F body weight was zero. For the High Release condition, F body weight equaled positive body weight above the navel. For the Inversion condition, F body weight equaled negative body weight above the navel. For the V-sit condition, F body weight equaled positive body weight above the navel multiplied by cos (45º). Data Analysis One-way repeated measure analyses of variance (ANOVA) for each dance position as a within-subject factor were performed on the EMG and estimated compressive forces to examine the position effect. A Type-I error rate was set at 0.05 for significantly different ANOVAs. Paired comparisons with 95% confidence interval were performed if an ANOVA showed a significant main effect. Statistical analyses were performed using SPSS 16.0 (SPSS, IL, USA).
Results ANOVAs showed significant main effects for all muscle EMG and estimated compressive forces (p < 0.001, Table 3). Post hoc analysis showed that both left and right EO and left and right RA demonstrated the greatest activation for Plank and the second greatest activation for V-sit. Both left and right LD demonstrated the greatest activation for Inversion and the second greatest activation for High Release. Both left and right ES showed the greatest activation for High Release and Inversion. Estimated compressive forces were the greatest for High Release, followed by Inversion, Plank, V-sit, Side-lying, and Spider.
Discussion This study was a first attempt at measuring muscular activity in positions commonly used in vertical dance. The greatest estimated compressive forces
were seen in vertical axis positions where the dancer was moving the trunk into a hyper-extended position. With High Release and Inversion, the dancer is trying to bring the body into a vertical axis, parallel to the rope. The aesthetic requirement is to create the illusion of a straight line up and down. However, the movement becomes isometric as the dancer nears the vertical axis. At the same time, the body rarely achieves a true vertical position due to the configuration of the harness waist and leg loops and the attachment of the harness to the rope. Therefore, the high activation of the muscles had to do with establishing and maintaining the required orientation of the body— upside down in the Inversion and behind the center of gravity for High Release. The high muscle activation may not be expected by simply examining the motion during these conditions. In High Release and Inversion positions, where the upper spine is taken into hyperextension, the dancer’s upper body is brought relatively close to the L5/S1, creating a small moment arm and a small external torque, which are normally associated with low muscle activation. However, because the dancers were hanging on a rope and were in a novel weightbearing environment, they co-activated the abdominal and low back muscles to increase trunk stability in order to maintain the required posture. As McGill and colleagues have described,18 virtually all muscles work together to create the balance in stiffness needed to ensure sufficient stability in all degrees of freedom or to maintain the appropriate level of potential energy of the spine. Vera Garcia and colleagues looking at steady state trunk pre-activation, found increased abdominal co-activation significantly increased spine stability at the cost of increasing spinal compression.19 Stokes and colleagues20 note that antagonistic activation of the abdominal muscles is observed in 3-dimensional movement, not only to stabilize the spine but is associated with intra-abdominal pressurization. In the current study, dancers had to increase the trunk stability to maintain their postures in a relatively unstable weightbearing situation.
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
169
Table 3 Mean and Standard Deviation of Muscle EMG and L5/S1 Compressive Forces, P-values for ANOVAs, and Significant Pairwise Comparisons during Different Dance Conditions HR
IN
PL
SL
SP
VS
LEO (%MVIC)
13.6 ± 13.0
12.8 ± 9.6
86.5 ± 54.8 > HR, IN, SL, SP, VS
36.7 ± 30.3 > HR, IN
17.5 ± 16.2
44.2 ± 27.3 p < 0.0001 > HR, IN, SP
REO (%MVIC)
18.4 ± 17.1
15.9 ± 11.4
100.8 ± 55.8 > HR, IN, SL, SP, VS
21.6 ± 19.1
32.9 ± 41.1
54.4 ± 39.9 p < 0.0001 > HR, IN, SL
LRA (%MVIC)
7.4 ± 4.1
5.6 ± 6.4
65.7 ± 35.6 > HR, IN, SL, SP, VS
23.4 ± 34.4 > IN
24.8 ± 44.1
23.8 ± 14.4 > HR, IN
p < 0.0001
RRA (%MVIC)
5.3 ± 3.7
5.0 ± 6.0
53.8 ± 25.5 > HR, IN, SL, SP, VS
7.3 ± 4.8 > HR
10.2 ± 9.5
21.6 ± 15.0 > HR, IN, SL, SP
p < 0.0001
LLD (%MVIC)
21.7 ± 17.9 > SL, VS
37.4 ± 21.9 > HR, PL, SL, SP, VS
14.3 ± 9.9
10.5 ± 4.2
12.2 ± 7.0
9.2 ± 8.2
p < 0.0001
RLD (%MVIC)
22.4 ± 12.9 > SL, VS
44.9 ± 29.3 > HR, PL, SL, SP, VS
17.5 ± 27.7
14.2 ± 9.6
15.0 ± 12.8
10.8 ± 11.3
p < 0.0001
LES (%MVIC)
48.5 ± 25.8 43.9 ± 19.7 6.5 ± 3.2 > PL, SL, SP, VS > PL, SL, SP, VS
14.7 ± 19.5
5.7 ± 3.6
6.5 ± 3.1
p < 0.0001
RES (%MVIC)
43.6 ± 20.3 40.4 ± 18.2 6.5 ± 2.9 > PL, SL, SP, VS > PL, SL, SP, VS
7.3 ± 10.2
5.1 ± 4.3
7.0 ± 4.6) > SP
p < 0.0001
Compressive 1519.4 ± 570.0 1142.4 ± 512.5 force (N) > IN, PL, SL, SP, > SL, SP, VS VS
965.9 ± 309.3 > SL, SP, VS
P-values
541.4 ± 284.4 447.6 ± 224.6 762.0 ± 242.0 p < 0.0001 > SL, SP
HR: High Release; IN: Inversion; PL: Plank; SL: Side-lying; SP: Spider; VS: V-sit; LEO: left external oblique; REO: right external oblique; LRA: left rectus abdominus; RRA: right rectus abdominus; LLD: left latissimus dorsi; RLD: right latissimus dorsi; LES: left erector spinae; RES: right erector spinae; MVIC: maximum voluntary isometric contraction; Significant pair wise comparisons were indicated using “>” at a p-value level of 0.05.
Plank position showed the next highest level of estimated compressive force. High activation of the rectus abdominus and external oblique would be expected in this position because the upper and lower body segments were relatively far away from the L5/S1 (as there was no flexion or extension of the upper or lower body toward the spine) and created large external torques. This increase in external torque required the abdominals to generate internal trunk and hip flexion torques. However, compared to low back muscles, the cross-sectional areas of the abdominals are smaller.21 This is confirmed by the greater coefficients for cross-sectional areas of low back muscles compared to abdominal muscles (EO, RA, LD, and ES were 0.0148, 0.006, 0.0351, and 0.0389, respectively). With the same level of activation, low back muscles would generate more force compared
to abdominal muscles. Therefore, the estimated compressive force in Plank was lower than that in High Release and Inversion. V-sit was another position that generated high demands for the abdominal muscles. Compared to the Plank, the upper and lower body segments in V-sit were closer to the L5/S1 in the vertical axis and therefore imposed smaller external torques on L5/S1.22 The decreased EMG in V-sit compared to Plank supports this notion. In all positions tested, the amount of muscular work had two purposes: to maintain the position and to stabilize the body in the position. In all of the movements tested, co-contraction of the agonists and antagonists would have contributed to the muscular work and compression. The lowest combined means were seen with Side-lying and full hyperex-
tension with the limbs hanging below the pelvis (Spider). The Side-lying is similar to Plank, but the external loading is mainly in the coronal plane compared to sagittal plane because of the rotation of the body in the harness. Also, dancers may have utilized passive tissues (e.g., bone, fascia, and ligaments) to support themselves during Side-lying. In this instance, the lack of muscle activation would correspond to increased stress on the spine. In addition, the dancers were likely utilizing lateral trunk muscles (quadratus lumborum, for example), but these were not measured in this study. For the Spider position, without engaging the abdominal muscles, the dancers were using passive tissues. In this hyperextended position, dancers may have reached the limit of their extension, and passive tissues were required to support the external
170
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
torques, so muscle activations were the lowest. The post-test survey revealed that the dancers felt most comfortable holding the positions Spider and Sidelying attitude. These positions can be held with a minimum of muscular contraction, thus the dancer’s perception of the action was that it could be maintained for longer. However, when cued to use the abdominals in these positions, one of the dancers commented: “When I went into Spider the first time, I let gravity completely take over, but this caused a lot of strain on my lower back where the harness was holding me, and when I tried to get out of the position, it was much more difficult. When [my abdominals were] engaged in the position, I felt much more active in my entire body, and the position felt more dynamic. I was able to support myself much more effectively and could easily move from this position to any other.” The findings of this study provide useful information for performance training and injury prevention in vertical dance. First, vertical dance is a demanding movement form, specifically taxing for muscles as they often are contracting in a different relationship to gravity. Individuals without good low back and abdominal strength and endurance may not be ready to participate in vertical dance. Muscular strength and endurance are needed to maintain postures and spine stability; a lack of strength is likely to cause injuries. Once again, the dancer is working in a novel relationship to gravity and must use the lower trunk to stabilize, control, and help initiate movement.23,24 Specific to vertical dance, control of the trunk and limbs must take place without the feet on the ground, so “normal” compressive forces in the skeleton as seen in standing cannot contribute to posture. Second, individuals are recommended to progress the tasks according to the difficulty of the movement. The results from this study suggest that individuals may start with Side-lying and V-sit and practice high-release and Inversion last. Muscular strength and endurance are needed to support the
dancer in various movements during vertical dance sequences. The dancer is often in a harness for up to 20 minutes in performance but as much as 30 minutes at a time during rehearsals. Third, as previously mentioned, a tradeoff between stability and loading should be considered in vertical dance. A certain amount of muscle co-activation may be needed to maintain stability; however, too much coactivation may generate great loading to the spine and increase the risk of injury. Therefore, while the dancer must have strength in all trunk muscles for balance, excessive co-contraction of the muscles will increase compression loading and limit the dancer’s range of motion.25 For tasks that induce high loading of the spine, instructors should ensure that the dancers have had adequate conditioning, training, and movement knowledge to avoid high loading in extreme positions. Dancers should also pay attention not to overly co-activate muscles and impose unnecessary loading of the spine. For tasks that cause low loading to the spine, attention should be paid to good posture and form as well as muscle co-activation to maintain spine stability. Finally, dancers should endeavor to use skeletal alignment wherever possible in movements to support the loading and avoid taxing the muscles and tendons.
Limitations In this research, we chose to study static postures because of the difficulties involved in collecting EMG data during dynamic dance movement. While considered a valid and reliable research tool, EMG has inherent limitations, such as cross-talk from other muscles and a nonlinear relationship between EMG and muscle forces.15 Each of the muscles has complex origins and attachments, and the placement of the electrodes could substantially affect the EMG signals. To minimize these limitations in the current study, the electrodes were placed by one researcher, according to established literature. In addition, the participant was only tested during one session. There is a margin of error in
EMG lead placement, but as the current study was a within-subject design, potential errors in placing electrodes should have had a mininal effect on the comparisons within each subject. Additionally, as the participant began to sweat, the adhesive lost contact with the skin, and while the movements tested were static, transitioning to another movement might affect adhesion of the EMG sensors. In this case, the EMG sensors were held in place by one of the researchers while the participant changed position to ensure that there was no movement between the electrode and skin. A modified or instrumented harness may be necessary for further research in vertical dance. The model used in the current study to estimate the spinal compressive force had several limitations. We assumed a linear relationship between muscle activation and muscle forces and a constant muscle stress without considering the muscle force-length relationship. We only included eight surface muscles of the trunk, excluding other surface muscles and deep muscles, so we likely underestimated the spinal loading. The purpose of measuring estimated force was to compare the change in spinal loading across different postures, but the exact value of the estimated compressive force should be interpreted with an awareness of these limitations. The vertical dance ability of the participants was varied, as the dancers were not currently training for vertical dance during the data collection time period. There was no comparison to performing the movements while in a weightbearing position, as this would not have been possible for the side lying, High Release, or Inversion positions. However, this information could be valuable in assessing the effects of a non-weightbearing environment on muscle activation and used for comparison. Thus, future studies in this area are needed. Finally, did the testing conditions replicate performance conditions? There was no music provided for the testing, and the dancers were required
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
to hold the positions for 5 seconds, not something usually measured in the rehearsal process. However, gradually increasing the length of time for each position in training is important for developing strength and endurance in the movements. One participant’s comments in the post-test survey also revealed: “In terms of physical exertion, I do think it [the testing] is similar to class or rehearsal. The positions used in this research are commonly used in rehearsal or class. Staying on the rope and constantly doing something in one of the positions is a lot like being in class. Although we aren’t focusing on the aesthetic of the position or transition in the research, we spend a lot of time holding positions in class.”
Conclusion In this study, muscle activation (RA, EO, ES, and LD) was measured with EMG in typical vertical dance movements. The EMG data were expressed as a percentage of MVIC and were used to estimate the compressive force on the spine. The greatest muscle activation for low back muscles and estimated compressive forces were seen in movements in the vertical axis, with the trunk in a hyper-extended position. The greatest muscle activation for abdominal muscles and compressive forces were seen in a supine position with the arms and legs extended away from the COM. Although this study has helped analyze the stresses on the spine in a range of static vertical dance positions, it has also pointed to the need for future research on dynamic motions. In particular, positions such as Spider, when not performed with muscular activation have the potential to place stress on the spine when entering and exiting the position. These transitional movements are important both for the aesthetics of the dance form and should be monitored for potential stress on the dancer’s body. Vertical dance represents a novel environment for dance training, and the findings of the current study suggest that it is a demanding movement form. Individuals are recommended
to progress the tasks according to the difficulty and physical demands of the movement. Systematic training leads to the development of co-contraction of the trunk muscles; therefore, loading the positions in testing should be done incrementally so that the dancer can develop the strength to sustain the position for 5 seconds or longer. Measuring the relative contribution of muscles, expressed as muscle activation and estimated spinal compression, provides a measure of how much the muscle groups were working to support the spine and the rest of the dancer’s body in the different movements tested. The four muscle groups tested are only a portion of the trunk stabilizers used in vertical dance, but identifying typical muscle recruitment patterns in these groups will help identify key exercises for training. Exercises that target awareness and utilization of all trunk muscles to support the spine and limbs are being developed, and an intervention protocol is being developed that is specific to vertical dance. As vertical dance is increasing in popularity, identification of the stresses involved and development of conditioning programs to ensure the safety of the dancers is warranted.
References 1. Wilson M, Kwon YH. The role of biomechanics in understanding dance movement: a review. J Dance Med Sci. 2008;12(3):109-16. 2. Krasnow D, Wilmerding MV, Stecyk S, et al. Biomechanical research in dance: a literature review. Med Probl Perform Art. 2011 Mar;26(1):3-23. 3. Simpson KJ, Kanter L. Jump distance of dance landings influencing internal joint forces: I. Axial forces. Med Sci Sports Exerc.1997 Jul;29(7):91627. 4. Kulig K, Fietzer AL, Popovich JM Jr. Ground reaction forces and knee mechanics in the weight acceptance phase of a dance leap take-off and landing. J Sports Sci. 2011 Jan;29(2):125-31. 5. Shippen J, May B. Calculation of muscle loading and joint contact forces during the rock step in Irish dance. J Dance Med Sci. 2010;14(1):11-8.
171
6. Mayers L, Bronner S, Agraharasamakulam S, Ojofeitimi. Lower extremity kinetics in tap dance. J Dance Med Sci. 2010;14(1):3-10. 7. Personal communication, Wanda Moretti, Il Posto Dance Company, 4/19/2013. 8. Long R. Yoga Mat Companion 3: Anatomy for Backbends and Twists. Plattsburgh, NY: Bandha Yoga Publications LLC, 2010. 9. Clippinger K. Dance Anatomy and Kinesiology. Champaign, IL: Human Kinetics, 2007. 10. Marras WS, Sommerich CM. A three-dimensional motion model of loads on the lumbar spine: II. Model validation. Hum Factors. 1991 Apr;33(2):139-49. 11. Marras WS, Granta KP. The development of an EMG-assisted model to assess spine loading during wholebody free-dynamic lifting. J Electromyogr Kinesiol. 1997 Dec;7(4):25968. 12. Escamilla RA, Babb E, DeWitt R, et al. Nontraditional abdominal exercises: implications for rehabilitation and training. Phys Ther. 2006 May;86(5):656-71. 13. Beim GM, Giraldo JL, Pincivero DM, et al. Abdominal strengthening exercises: a comparative EMG Study. J Sports Rehab. 1997;6:11-20. 14. Floyd WF, Silver PHS. Electromyographic study of patterns of activity in the anterior abdominal wall muscles in man. J Anat. 1950 Apr;84(2):132-46. 15. De Luca CJ. The use of surface electromyography in biomechanics. J Appl Biomech. 1997;13:135-63. 16. Dai B, Sorensen CJ, Derrick TR, Gillette JC. The effects of postseason break on knee biomechanics and lower extremity EMG in a stop-jump task: implications for ACL injury. J Appl Biomech. 2012 Dec;28(6):70817. 17. de Leva P. Adjustments to Zatsiorsky-Seluyanov’s segment inertia parameters. J Biomech. 1996 Sep;29(9):1223-30. 18. McGill SM, Grenier S, Kavcic N. Cholewicki J. Coordination of muscle activity to assure stability of the lumbar spine. J Electromyogr Kinesiol. 2003 Aug;13(4):353-9. 19. Vera-Garcia FJ, Brown SHM, Gray JR, McGill SM. Effects of different levels of torso coactivation on trunk muscular and kinematic responses
172
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
to posteriorly applied sudden loads. Clin Biomech (Bristol, Avon). 2006 Jun;21(5):443-55. 20. Stokes IAF, Gardner-Morse MG, Henry SM. Abdominal muscle activation increases lumbar stability: analysis of contributions of different muscle groups. Clin Biomech (Bristol, Avon). 2011 Oct;26(8):797803. 21. Marras WS, Jorgensen MJ. Granata KP, Wiand B. Female and male trunk geometry: size and prediction of the spine loading trunk muscles derived
from MRI. Clin Biomech (Bristol, Avon). 2001 Jan;16(1):38-46. 22. Quint U, Wikle HJ, Shirazi-Adl A, et al. Importance of the intersegmental trunk muscles for the stability of the lumbar spine. A biomechanical study in vitro. Spine (Phila Pa 1976). 1998 Sep;23(18):1937-45. 23. Dias JM, Menacho Mde O, Mazuquin BF, et al. Comparison of the electromyographic activity of the anterior trunk during the execution of two Pilates exercises—teaser
and longspine—for healthy people. J Electromyogr Kinesiol. 2014 Oct;24(5):689-97. 24. Marques NR, Morcelli MH, Hallal CZ, Goncalves M. EMG activity of trunk stabilizer muscles during Centering Principle of Pilates Method. J Bodyw Mov Ther. 2013 Apr:17(2):185-91. 25. Liebenson C, Karpowiz AM, Brown SH, et al. The active straight leg raise test and lumbar spine stability. PM R. 2009 Jun;1(6):530-5.
Abstract Rope and harness vertical dance takes place off the floor with the dancer suspended from his or her center of mass in a harness attached to a rope from a point overhead. Vertical dance represents a novel environment for training and performing in which expected stresses on the dancer’s body are different from those that take place during dance on the floor. Two male and eleven female dancers with training in vertical dance performed six typical vertical dance movements with electromyography (EMG) electrodes placed bilaterally on rectus abdominus, external oblique, erector spinae, and latissimus dorsi. EMG data were expressed as a percentage of maximum voluntary isometric contraction (MVIC). A simplified musculoskeletal model based on muscle activation for these four muscle groups was used to estimate the compressive force on the spine. The greatest muscle activation for erector spinae and latissimus dorsi and the greatest trunk compressive forces were seen in vertical axis positions where the dancer was moving the trunk into a hyper-extended position. The greatest muscle activation for rectus abdominus and external oblique and the second highest compressive force were seen in a supine position with the arms and legs extended
away from the center of mass (COM). The least muscle activation occurred in positions where the limbs were hanging below the torso. These movements also showed relatively low muscle activation compression forces. Post-test survey results revealed that dancers felt comfortable in these positions; however, observation of some positions indicated insufficient muscular control. Computing the relative contribution of muscles, expressed as muscle activation and estimated spinal compression, provided a measure of how much the muscle groups were working to support the spine and the rest of the dancer’s body in the different movements tested. Additionally, identifying typical muscle recruitment patterns in each movement will help identify key exercises for training that should promote injury prevention.
B
iomechanics has long been used to study specific dance movements and can offer insight into the movement often unnoticed by dance educators.1,2 Some of the dance forms studied to date include ballet, modern, tap, and Irish Dance. 3-6 Utilizing tools, such as motion capture, electromyography
Margaret Wilson, Ph.D., Department of Theatre & Dance, Boyi Dai, Ph.D., Department of Kinesiology and Health, Qin Zhu, Ph.D., Department of Kinesiology and Health, and Neil Humphrey, Ph.D., Department of Geology and Geophysics, and Department of Theatre & Dance, University of Wyoming, Laramie, Wyoming. Correspondence: Margaret Wilson, Ph.D., Department of Theatre & Dance, University of Wyoming, 1000 East University Avenue, Dept. 3951, Laramie, Wyoming 82071; [email protected]. Copyright © 2015 J. Michael Ryan Publishing, Inc. http://dx.doi.org/10.12678/1089-313X.19.4.163
(EMG), ground reaction force, and muscular modeling, movements can be isolated and observed relative to the forces of gravity and torque to understand the relationship that dancers have with their environment, as well as the internal forces acting on the body. Traditional methodology for biomechanical analysis assumes that the dancer is standing on the floor or weightbearing through the upper body. The question driving this research was whether biomechanics could help researchers understand a new dance form, rope and harness vertical dance—i.e., dance that takes place off the floor with some apparatus suspending the body or from which the body hangs. Vertical dance is often classified as a form of aerial dance, an outgrowth of circus tradition that includes trapeze, silks, or fabrics; however, rope and harness vertical dance uses traditional climbing and rescue equipment. Early work in vertical dance was experimental and often inspired by aesthetic climbing.7 The dancer wears a harness around the hips, and the harness is suspended by a rope, which is secured overhead. Thus, the dancer is hanging in midair with support coming only from the pelvis, not from the feet on the ground. In this configuration, the dancer can interact with a vertical wall or hang in “free space” and can reorient the body so that he or she is upright, parallel to the floor, upside down, and many dif163
164
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
ferent angles in-between. The vertical dancer uses familiar movements from traditional dance styles and explores these actions in new relationships to gravity, with a different base of support. The dancer is afforded almost 360° of motion, depending on how the harness is attached to the rope. Suspended in air, the dancer’s trunk and hip muscles must create leverage to control and execute movement without an external source of torque, such as would be found when performing these movements on the floor. Hanging in the harness creates an upward force on the dancer’s pelvis, while at the same time the weight of the dancer’s body creates a downward tensional force. With no other source of support, trunk and hip muscle activation must create the desired orientation of the trunk in the harness and also control and support the movement of the upper body and legs. This is seen in all orientations to gravity, such as when the dancer is upside down or perpendicular to the rope and parallel to the floor. The dancer is creating similar movements to those performed in a traditional dance form, with a different relationship of the weights of the body parts (trunk, arms, and legs) to gravity. The typical use pattern of muscles to create movement is altered with this different orientation to gravity. For example, when standing upright, the line of gravity passes through the length of the body in the vertical axis, and most muscular activity is directed to cocontraction for balance and stability. When the dancer is suspended and the body is perpendicular to the rope and parallel to the floor, all anterior muscles, primarily flexors, must contract in order to keep the body in the horizontal plane, as gravity is creating an extensor force (Fig. 1). The initial concern in this investigation was to look at the potential stresses on the spine in vertical dance. There are two aspects of this stress. First, with the feet off the ground, the dancer’s legs hang down from the pelvis, rather than serving to support the pelvis and spine as they do in a standing position. Second, although
Figure 1 Plank.
Figure 2 Spider.
Figure 3 Side-lying.
the harness acts as suspension for the weight of the body, its fabric construction does not provide structural support. In positions with the limbs acting as long resistance arms, such as Plank (Fig. 1), with the spine bearing weight perpendicular to the spinal column (not a typical stress), are there undue stresses being created? A literature review generated no information on calculating forces acting on the dancer in an airborne environment. Comparisons were sought in the disciplines of gymnastics, swimming, yoga, and circus arts; however, to date, no
evidence of similar movement studies has been identified. Additionally, specific vertical dance movements generate different questions. For Spider (Fig. 2) and Side-lying attitude (Fig. 3), the dancers could approximate these positions with minimal muscular effort. This means that in these positions the dancer’s body is hanging in the harness, yet due to the fabric construction of the harness, there can be little control of the angular relation between body segments (i.e., too much spine hyperextension or
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
Figure 4 High Release.
Figure 5 Inversion.
lateral flexion). This is not the way the movements are taught, but one researcher’s observation of the dancer’s performance, as well as comments from the dancers, indicated that the body passively assumes the position due to gravity, thus the abdominals
are not active. The profound impact of gravity on the suspended body in the vertical dance movement of Spider stretches the muscles to full ranges of elongation. This may override the sensory structures for control, thus transferring the support to
165
the passive tissues and spine. Whereas this movement can be compared to a back bend in gymnastics or camel pose (Ustrasana) or upward facing bow pose (Urdhva Khanurasana) in yoga,8 those movements are supported by the arms and legs on the ground. Performing the same movement with the dancer suspended in the air means that the full weight of the body is transferred through the lumbar spine. As the weight of the dancer in this position is transferred horizontally through the body of the lumbar vertebrae (rather than vertically, as when standing), the dancer may well be experiencing different stresses to the body of the vertebrae. To place this in context, in a standing posture the interlocking facets provide stability and also absorb 30% of the compression force in hyperextension.9 The facets and body of the vertebrae should be sufficient to withstand this pressure, but this has not been tested in vertical dance or in other activities. This is one of the key questions in the research. In the movement Spider, particularly, the dancer can use little muscle control and simply “hangs in the harness.” As this dance form places the spine in a novel weightbearing position, we were concerned that the effect of gravity combined with lack of muscular activity creates a unique stress on the dancer’s spine. In other movements, the harness and rope configuration forces the dancer to overcome gravitational resistance to the movement. With the dancer’s body aligned in the vertical axis, such as High Release (Fig. 4), Inversion (Fig. 5), and where the body is moving toward the vertical axis as in V-sit (Fig. 6), the harness supports the pelvis and also provides a means of leverage for the trunk and limbs. In some positions, the geometry of the harness actually inhibits the desired position of the body and requires additional muscle activation. The harness attaches to the rope in the front via waist and leg loops, which displace the center of mass posteriorly. When in the vertical axis, either inverted or when
166
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
Figure 6 V-Sit.
Figure 7 Prone MVIC test.
working toward the High Release position, the dancer must actively press the pelvis anteriorly to achieve a vertical orientation. As vertical dance is a relatively new form of movement, there was little research to use as a guide in setting up this investigation. Initial attempts at using motion capture were set aside due to the difficulty of capturing all of the reflective markers in the horizontal positions or the actions that changed levels (which obscured the reflective markers) and the lack of a means to measure forces applied by the harness. Specifically, we hoped to provide an analysis of resultant joint forces and resultant joint moments acting on (and within) the body with the dancer hanging in the air. Without the ability to quantify the forces provided by the harness at different contact points, traditional biomechanical modeling was not possible in the current study. Instead, a theoretical approach was adopted based on research modeling compressive forces in the spine.10,11 A protocol was established to mea-
sure muscle activation for bilateral rectus abdominus, external oblique, erector spinae, and latissimus dorsi as percentages of maximal voluntary isometric contraction. Based on the percentages of muscle activation, an estimation of spinal compression could be made. Therefore, the primary purpose of the research was to quantify the trunk muscle activation for selected static postures in rope and harness vertical dance. The secondary purpose was to estimate the spinal compressive force for selected static postures in vertical dance using a musculoskeletal model. Although it is difficult to measure spinal loading directly without invasive measurements, an indirect estimate of trunk compression can provide insight to the internal forces acting on the spine.
Methods The current study was approved by the University of Wyoming’s Institutional Review Board. Participants signed informed consent forms prior to testing. Two male and eleven female
dancers (Table 1) participated in the data collection. The dancers were all at an intermediate level of dance training and vertical dance training and were familiar with the movements to be included in the testing. All testing took place in the biomechanics laboratory. Measures of each dancer’s height and weight were obtained, the length of the spine was measured in a sitting position on a firm surface (base of the spine to the acromion process), and the width and depth of the torso at L5 were measured in supine (Marras and Sommerich).10,11 The dancers were asked to warm up as they would for a vertical dance class. Then EMG electrodes were placed bilaterally on the rectus abdominus (RA: 3 cm lateral to midline of abdomen and 2 cm above umbilicus), external oblique (EO: 15 cm lateral to the umbilicus at a 45º angle above the anterior superior iliac spine at the level of the umbilicus), erector spinae (ES: 4 cm lateral to L3 spinous process), and latissimus dorsi (LD: most lateral portion of the muscle at the T9, 4 cm below the inferior angle of the scapula).12-14 A reference electrode was placed on the manubrium. An EMG transmitter (MyoMonitor®, Delsys, Inc., Boston, MA) was contained in a harness above the subject’s pelvis. The EMG data were captured at a sampling frequency of 2,000 Hz (EMGworks®, Delsys, Inc., Boston, MA). Prior to the vertical dance testing, the dancers performed movements of trunk flexion, extension, and rotation to make sure that a signal was being obtained for the EMG leads. Literature has suggested that surface EMG measurement is a validated and reliable tool to measure surface muscle activation and may provide insight into muscle forces.15 For the data collection, the dancers climbed a ladder to attach their harness to a rope using a gri-gri (self-belay device). The harness was suspended approximately 180 cm above the ground. The dancer was stabilized to stop any swinging motion and asked to assume the position to be tested. When the dancer demonstrated the correct position, the recording of the
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
167
Table 1 Participant Characteristics N = 13
Height (cm)
Mass (kg)
Torso Depth (cm)
Torso Width (cm)
Torso Length (cm)
Vertical dance experience (yrs)
Age
Average
169
61.0
18.0
28.4
57.8
2.8
26.3
Range
159-186.5
45.8-78.9
13.5-19
25.5-35
54-63
1-9
19-60
EMG began, and the test lasted 5 seconds. If the test was successful (the position was maintained for the required time), the dancer only performed the movement once. Visual inspection of EMG signals was performed after each trial to make sure that no noise was observed; otherwise, the trial was repeated. EMG data were collected for the eight muscles in each position. Six positions were tested: V-sit, Plank, High Release, Spider, Inversion, and Side-lying. These positions represent the core movement vocabulary used in rope and harness vertical dance training and are representative of the different orientations to gravity experienced in vertical dance. Specific descriptions of the movements follow. Plank In this position, the goal is to create a body line that is parallel to the floor. The abdominals, trunk and hip flexors, and spine extensors must be active. The abilty of the dancer to achieve a parallel orientaiton to the floor depends on strength and control, relative length and mass of upper body to leg length, and balancing at the attachment point of the harness to the rope (Fig.1). Spider The upward facing dancer allows the arms and legs to hang toward the floor, reaching behind the trunk and below the pelvis, creating full spinal hyperextension. Dancers are instructed to maintain an eccentric contraction of the abdominals throughout (Fig. 2). Side-Lying The dancer rotates in the harness so that the body is oriented to one side. The goal of the exercise is to have a horizontal orientation of the body with the top leg hyperextended at the hip and knee and the lower leg
flexed with a straight knee. The low arm extends to the floor, and the top arm extends to the ceiling, parallel with the rope. The head should be parallel with the shoulders (ideally parallel to the floor). The dancer often will touch the fingers of the top arm to the rope to help balance and support the upper body (Fig. 3). For this study, the dancers were tested on one side only, rotated to the right with the right leg in flexion and the left in hyperextension. High Release The dancer tries to bring the upper body parallel to the rope. He or she is instructed to cross the wrists over the rope rather than grasp it with the hands for leverage. Lumbar hyperextension, hip extension and external rotation, and knee flexion allows the dancers to take the legs behind the torso to the desired position (Fig. 4). Inversion The dancer is upside down with the legs straight, crossed at the ankles, with the back foot on the front of the rope and the front leg crossed over as in a classical fifth position. This position of the legs provides leverage to press the pelvis anteriorly. Inversion has the added stylistic variation of arching the head and shoulders into hyperextension, creating the illusion
of a bow in the body (Fig. 5). V-sit This movement usually starts from a sitting position, but the dancer can move to the V-sit position from Plank or from Spider. The head should remain in line with the spine, with the focus up and over the feet. The torso should be as straight as possible, with the arms and legs each extended 90º relative to the trunk (Fig. 6). Following the vertical dance movements, a test of maximal voluntary isometric contraction (MVIC) was completed. The dancer started supine on the floor, with arms crossed over the chest and legs extended along the floor. The tester applied body weight pressing down on the shoulders of the dancer, who was asked to attempt to flex the spine, bringing the shoulders up off the floor. The dancer was monitored for accessory motion in the legs during the trial. For back extension, the dancer was face down on the floor with the legs extended and arms alongside the legs but off the floor. The tester applied body weight pressing down on the shoulders of the prone dancer who was asked to attempt to take the spine into hyperextension off the floor. The dancer was monitored for accessory motion of the legs during this trial (Fig. 7). MVIC trials were performed for 10 seconds
Table 2 Participant Survey Results* Pre-test
Post-test
How long can you hold this position?
V-sit
4
3.45
6.9 sec
Plank
4.63
5.18
4.18 sec
Spider
2.63
2.72
11.8 sec
3
2.9
8.81 sec
High Release
4.9
4.36
5.72 sec
Side-lying
1.90
1.90
12.45 sec
Inversion
*Participant Survey: Rank the following in terms of difficulty for performing (1- easiest, 6- most difficult). Nine of 13 respondents reported.
168
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
to ensure a true 1-second MVIC was achieved. Following data collection the participants completed a post-test survey (Table 2).
Data Reduction EMG data were band-pass filtered from 20 Hz to 450 Hz.16 Filtered EMG data were rectified and then filtered at a low-pass cut-off frequency of 10 Hz to obtain the linear envelopes of EMG.16 The average of the MVIC data was calculated with an interval of 1 second and an increment of 1 frame (0.0005 second), for example, the average between 0 and 1 second, the average between 0.0005 and 1.0005 second, and the average between 0.001 and 1.001 second until the average between 9 and 10 seconds. The maximum average was extracted as the value for MVIC.16 EMG data during dancing trials were expressed as percentages of MVIC.16 EMG data during the middle 3 seconds of each dancing trial were averaged to represent muscle activation for the trial. A simplified musculoskeletal model was used to estimate muscle forces and L5/S1 compressive forces.12 Muscle forces were modeled as the product of the muscle activation (normalized EMG), muscle cross-sectional areas, and muscle stress. Muscle crosssectional areas were estimated as the product of the trunk length, trunk width, and specific coefficients for different muscles. The coefficients for the EO, RA, LD, and ES were 0.0148, 0.006, 0.0351, and 0.0389, respectively.10 Muscle stress was set at a constant of 47.4 N/cm2.10 RA and ES were modeled as parallel to lumbar spine, and EO and LD were modeled as 45º from lumbar spine.11 Lumbar compressive force was estimated by summing the vertical components of muscle forces and the vertical component of the body weight above the navel (45.76% of the total body weight for females and 49.11% for males).17 The calculations were performed using customized subroutines developed in MATLAB 2009a (MathWorks, Inc., Natick, MA). F = F left rectus abdominus + F right rectus
abdominus + F left erector spinae + F right erector spinae + F left external oblique * cos (45º) +F right external oblique * cos (45º) + F left latissimus dorsi * cos (45º) + F right latissimus dorsi * cos (45º) +F body weight
For the Plank, Spider, and Sidelying conditions, F body weight was zero. For the High Release condition, F body weight equaled positive body weight above the navel. For the Inversion condition, F body weight equaled negative body weight above the navel. For the V-sit condition, F body weight equaled positive body weight above the navel multiplied by cos (45º). Data Analysis One-way repeated measure analyses of variance (ANOVA) for each dance position as a within-subject factor were performed on the EMG and estimated compressive forces to examine the position effect. A Type-I error rate was set at 0.05 for significantly different ANOVAs. Paired comparisons with 95% confidence interval were performed if an ANOVA showed a significant main effect. Statistical analyses were performed using SPSS 16.0 (SPSS, IL, USA).
Results ANOVAs showed significant main effects for all muscle EMG and estimated compressive forces (p < 0.001, Table 3). Post hoc analysis showed that both left and right EO and left and right RA demonstrated the greatest activation for Plank and the second greatest activation for V-sit. Both left and right LD demonstrated the greatest activation for Inversion and the second greatest activation for High Release. Both left and right ES showed the greatest activation for High Release and Inversion. Estimated compressive forces were the greatest for High Release, followed by Inversion, Plank, V-sit, Side-lying, and Spider.
Discussion This study was a first attempt at measuring muscular activity in positions commonly used in vertical dance. The greatest estimated compressive forces
were seen in vertical axis positions where the dancer was moving the trunk into a hyper-extended position. With High Release and Inversion, the dancer is trying to bring the body into a vertical axis, parallel to the rope. The aesthetic requirement is to create the illusion of a straight line up and down. However, the movement becomes isometric as the dancer nears the vertical axis. At the same time, the body rarely achieves a true vertical position due to the configuration of the harness waist and leg loops and the attachment of the harness to the rope. Therefore, the high activation of the muscles had to do with establishing and maintaining the required orientation of the body— upside down in the Inversion and behind the center of gravity for High Release. The high muscle activation may not be expected by simply examining the motion during these conditions. In High Release and Inversion positions, where the upper spine is taken into hyperextension, the dancer’s upper body is brought relatively close to the L5/S1, creating a small moment arm and a small external torque, which are normally associated with low muscle activation. However, because the dancers were hanging on a rope and were in a novel weightbearing environment, they co-activated the abdominal and low back muscles to increase trunk stability in order to maintain the required posture. As McGill and colleagues have described,18 virtually all muscles work together to create the balance in stiffness needed to ensure sufficient stability in all degrees of freedom or to maintain the appropriate level of potential energy of the spine. Vera Garcia and colleagues looking at steady state trunk pre-activation, found increased abdominal co-activation significantly increased spine stability at the cost of increasing spinal compression.19 Stokes and colleagues20 note that antagonistic activation of the abdominal muscles is observed in 3-dimensional movement, not only to stabilize the spine but is associated with intra-abdominal pressurization. In the current study, dancers had to increase the trunk stability to maintain their postures in a relatively unstable weightbearing situation.
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
169
Table 3 Mean and Standard Deviation of Muscle EMG and L5/S1 Compressive Forces, P-values for ANOVAs, and Significant Pairwise Comparisons during Different Dance Conditions HR
IN
PL
SL
SP
VS
LEO (%MVIC)
13.6 ± 13.0
12.8 ± 9.6
86.5 ± 54.8 > HR, IN, SL, SP, VS
36.7 ± 30.3 > HR, IN
17.5 ± 16.2
44.2 ± 27.3 p < 0.0001 > HR, IN, SP
REO (%MVIC)
18.4 ± 17.1
15.9 ± 11.4
100.8 ± 55.8 > HR, IN, SL, SP, VS
21.6 ± 19.1
32.9 ± 41.1
54.4 ± 39.9 p < 0.0001 > HR, IN, SL
LRA (%MVIC)
7.4 ± 4.1
5.6 ± 6.4
65.7 ± 35.6 > HR, IN, SL, SP, VS
23.4 ± 34.4 > IN
24.8 ± 44.1
23.8 ± 14.4 > HR, IN
p < 0.0001
RRA (%MVIC)
5.3 ± 3.7
5.0 ± 6.0
53.8 ± 25.5 > HR, IN, SL, SP, VS
7.3 ± 4.8 > HR
10.2 ± 9.5
21.6 ± 15.0 > HR, IN, SL, SP
p < 0.0001
LLD (%MVIC)
21.7 ± 17.9 > SL, VS
37.4 ± 21.9 > HR, PL, SL, SP, VS
14.3 ± 9.9
10.5 ± 4.2
12.2 ± 7.0
9.2 ± 8.2
p < 0.0001
RLD (%MVIC)
22.4 ± 12.9 > SL, VS
44.9 ± 29.3 > HR, PL, SL, SP, VS
17.5 ± 27.7
14.2 ± 9.6
15.0 ± 12.8
10.8 ± 11.3
p < 0.0001
LES (%MVIC)
48.5 ± 25.8 43.9 ± 19.7 6.5 ± 3.2 > PL, SL, SP, VS > PL, SL, SP, VS
14.7 ± 19.5
5.7 ± 3.6
6.5 ± 3.1
p < 0.0001
RES (%MVIC)
43.6 ± 20.3 40.4 ± 18.2 6.5 ± 2.9 > PL, SL, SP, VS > PL, SL, SP, VS
7.3 ± 10.2
5.1 ± 4.3
7.0 ± 4.6) > SP
p < 0.0001
Compressive 1519.4 ± 570.0 1142.4 ± 512.5 force (N) > IN, PL, SL, SP, > SL, SP, VS VS
965.9 ± 309.3 > SL, SP, VS
P-values
541.4 ± 284.4 447.6 ± 224.6 762.0 ± 242.0 p < 0.0001 > SL, SP
HR: High Release; IN: Inversion; PL: Plank; SL: Side-lying; SP: Spider; VS: V-sit; LEO: left external oblique; REO: right external oblique; LRA: left rectus abdominus; RRA: right rectus abdominus; LLD: left latissimus dorsi; RLD: right latissimus dorsi; LES: left erector spinae; RES: right erector spinae; MVIC: maximum voluntary isometric contraction; Significant pair wise comparisons were indicated using “>” at a p-value level of 0.05.
Plank position showed the next highest level of estimated compressive force. High activation of the rectus abdominus and external oblique would be expected in this position because the upper and lower body segments were relatively far away from the L5/S1 (as there was no flexion or extension of the upper or lower body toward the spine) and created large external torques. This increase in external torque required the abdominals to generate internal trunk and hip flexion torques. However, compared to low back muscles, the cross-sectional areas of the abdominals are smaller.21 This is confirmed by the greater coefficients for cross-sectional areas of low back muscles compared to abdominal muscles (EO, RA, LD, and ES were 0.0148, 0.006, 0.0351, and 0.0389, respectively). With the same level of activation, low back muscles would generate more force compared
to abdominal muscles. Therefore, the estimated compressive force in Plank was lower than that in High Release and Inversion. V-sit was another position that generated high demands for the abdominal muscles. Compared to the Plank, the upper and lower body segments in V-sit were closer to the L5/S1 in the vertical axis and therefore imposed smaller external torques on L5/S1.22 The decreased EMG in V-sit compared to Plank supports this notion. In all positions tested, the amount of muscular work had two purposes: to maintain the position and to stabilize the body in the position. In all of the movements tested, co-contraction of the agonists and antagonists would have contributed to the muscular work and compression. The lowest combined means were seen with Side-lying and full hyperex-
tension with the limbs hanging below the pelvis (Spider). The Side-lying is similar to Plank, but the external loading is mainly in the coronal plane compared to sagittal plane because of the rotation of the body in the harness. Also, dancers may have utilized passive tissues (e.g., bone, fascia, and ligaments) to support themselves during Side-lying. In this instance, the lack of muscle activation would correspond to increased stress on the spine. In addition, the dancers were likely utilizing lateral trunk muscles (quadratus lumborum, for example), but these were not measured in this study. For the Spider position, without engaging the abdominal muscles, the dancers were using passive tissues. In this hyperextended position, dancers may have reached the limit of their extension, and passive tissues were required to support the external
170
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
torques, so muscle activations were the lowest. The post-test survey revealed that the dancers felt most comfortable holding the positions Spider and Sidelying attitude. These positions can be held with a minimum of muscular contraction, thus the dancer’s perception of the action was that it could be maintained for longer. However, when cued to use the abdominals in these positions, one of the dancers commented: “When I went into Spider the first time, I let gravity completely take over, but this caused a lot of strain on my lower back where the harness was holding me, and when I tried to get out of the position, it was much more difficult. When [my abdominals were] engaged in the position, I felt much more active in my entire body, and the position felt more dynamic. I was able to support myself much more effectively and could easily move from this position to any other.” The findings of this study provide useful information for performance training and injury prevention in vertical dance. First, vertical dance is a demanding movement form, specifically taxing for muscles as they often are contracting in a different relationship to gravity. Individuals without good low back and abdominal strength and endurance may not be ready to participate in vertical dance. Muscular strength and endurance are needed to maintain postures and spine stability; a lack of strength is likely to cause injuries. Once again, the dancer is working in a novel relationship to gravity and must use the lower trunk to stabilize, control, and help initiate movement.23,24 Specific to vertical dance, control of the trunk and limbs must take place without the feet on the ground, so “normal” compressive forces in the skeleton as seen in standing cannot contribute to posture. Second, individuals are recommended to progress the tasks according to the difficulty of the movement. The results from this study suggest that individuals may start with Side-lying and V-sit and practice high-release and Inversion last. Muscular strength and endurance are needed to support the
dancer in various movements during vertical dance sequences. The dancer is often in a harness for up to 20 minutes in performance but as much as 30 minutes at a time during rehearsals. Third, as previously mentioned, a tradeoff between stability and loading should be considered in vertical dance. A certain amount of muscle co-activation may be needed to maintain stability; however, too much coactivation may generate great loading to the spine and increase the risk of injury. Therefore, while the dancer must have strength in all trunk muscles for balance, excessive co-contraction of the muscles will increase compression loading and limit the dancer’s range of motion.25 For tasks that induce high loading of the spine, instructors should ensure that the dancers have had adequate conditioning, training, and movement knowledge to avoid high loading in extreme positions. Dancers should also pay attention not to overly co-activate muscles and impose unnecessary loading of the spine. For tasks that cause low loading to the spine, attention should be paid to good posture and form as well as muscle co-activation to maintain spine stability. Finally, dancers should endeavor to use skeletal alignment wherever possible in movements to support the loading and avoid taxing the muscles and tendons.
Limitations In this research, we chose to study static postures because of the difficulties involved in collecting EMG data during dynamic dance movement. While considered a valid and reliable research tool, EMG has inherent limitations, such as cross-talk from other muscles and a nonlinear relationship between EMG and muscle forces.15 Each of the muscles has complex origins and attachments, and the placement of the electrodes could substantially affect the EMG signals. To minimize these limitations in the current study, the electrodes were placed by one researcher, according to established literature. In addition, the participant was only tested during one session. There is a margin of error in
EMG lead placement, but as the current study was a within-subject design, potential errors in placing electrodes should have had a mininal effect on the comparisons within each subject. Additionally, as the participant began to sweat, the adhesive lost contact with the skin, and while the movements tested were static, transitioning to another movement might affect adhesion of the EMG sensors. In this case, the EMG sensors were held in place by one of the researchers while the participant changed position to ensure that there was no movement between the electrode and skin. A modified or instrumented harness may be necessary for further research in vertical dance. The model used in the current study to estimate the spinal compressive force had several limitations. We assumed a linear relationship between muscle activation and muscle forces and a constant muscle stress without considering the muscle force-length relationship. We only included eight surface muscles of the trunk, excluding other surface muscles and deep muscles, so we likely underestimated the spinal loading. The purpose of measuring estimated force was to compare the change in spinal loading across different postures, but the exact value of the estimated compressive force should be interpreted with an awareness of these limitations. The vertical dance ability of the participants was varied, as the dancers were not currently training for vertical dance during the data collection time period. There was no comparison to performing the movements while in a weightbearing position, as this would not have been possible for the side lying, High Release, or Inversion positions. However, this information could be valuable in assessing the effects of a non-weightbearing environment on muscle activation and used for comparison. Thus, future studies in this area are needed. Finally, did the testing conditions replicate performance conditions? There was no music provided for the testing, and the dancers were required
Journal of Dance Medicine & Science • Volume 19, Number 4, 2015
to hold the positions for 5 seconds, not something usually measured in the rehearsal process. However, gradually increasing the length of time for each position in training is important for developing strength and endurance in the movements. One participant’s comments in the post-test survey also revealed: “In terms of physical exertion, I do think it [the testing] is similar to class or rehearsal. The positions used in this research are commonly used in rehearsal or class. Staying on the rope and constantly doing something in one of the positions is a lot like being in class. Although we aren’t focusing on the aesthetic of the position or transition in the research, we spend a lot of time holding positions in class.”
Conclusion In this study, muscle activation (RA, EO, ES, and LD) was measured with EMG in typical vertical dance movements. The EMG data were expressed as a percentage of MVIC and were used to estimate the compressive force on the spine. The greatest muscle activation for low back muscles and estimated compressive forces were seen in movements in the vertical axis, with the trunk in a hyper-extended position. The greatest muscle activation for abdominal muscles and compressive forces were seen in a supine position with the arms and legs extended away from the COM. Although this study has helped analyze the stresses on the spine in a range of static vertical dance positions, it has also pointed to the need for future research on dynamic motions. In particular, positions such as Spider, when not performed with muscular activation have the potential to place stress on the spine when entering and exiting the position. These transitional movements are important both for the aesthetics of the dance form and should be monitored for potential stress on the dancer’s body. Vertical dance represents a novel environment for dance training, and the findings of the current study suggest that it is a demanding movement form. Individuals are recommended
to progress the tasks according to the difficulty and physical demands of the movement. Systematic training leads to the development of co-contraction of the trunk muscles; therefore, loading the positions in testing should be done incrementally so that the dancer can develop the strength to sustain the position for 5 seconds or longer. Measuring the relative contribution of muscles, expressed as muscle activation and estimated spinal compression, provides a measure of how much the muscle groups were working to support the spine and the rest of the dancer’s body in the different movements tested. The four muscle groups tested are only a portion of the trunk stabilizers used in vertical dance, but identifying typical muscle recruitment patterns in these groups will help identify key exercises for training. Exercises that target awareness and utilization of all trunk muscles to support the spine and limbs are being developed, and an intervention protocol is being developed that is specific to vertical dance. As vertical dance is increasing in popularity, identification of the stresses involved and development of conditioning programs to ensure the safety of the dancers is warranted.
References 1. Wilson M, Kwon YH. The role of biomechanics in understanding dance movement: a review. J Dance Med Sci. 2008;12(3):109-16. 2. Krasnow D, Wilmerding MV, Stecyk S, et al. Biomechanical research in dance: a literature review. Med Probl Perform Art. 2011 Mar;26(1):3-23. 3. Simpson KJ, Kanter L. Jump distance of dance landings influencing internal joint forces: I. Axial forces. Med Sci Sports Exerc.1997 Jul;29(7):91627. 4. Kulig K, Fietzer AL, Popovich JM Jr. Ground reaction forces and knee mechanics in the weight acceptance phase of a dance leap take-off and landing. J Sports Sci. 2011 Jan;29(2):125-31. 5. Shippen J, May B. Calculation of muscle loading and joint contact forces during the rock step in Irish dance. J Dance Med Sci. 2010;14(1):11-8.
171
6. Mayers L, Bronner S, Agraharasamakulam S, Ojofeitimi. Lower extremity kinetics in tap dance. J Dance Med Sci. 2010;14(1):3-10. 7. Personal communication, Wanda Moretti, Il Posto Dance Company, 4/19/2013. 8. Long R. Yoga Mat Companion 3: Anatomy for Backbends and Twists. Plattsburgh, NY: Bandha Yoga Publications LLC, 2010. 9. Clippinger K. Dance Anatomy and Kinesiology. Champaign, IL: Human Kinetics, 2007. 10. Marras WS, Sommerich CM. A three-dimensional motion model of loads on the lumbar spine: II. Model validation. Hum Factors. 1991 Apr;33(2):139-49. 11. Marras WS, Granta KP. The development of an EMG-assisted model to assess spine loading during wholebody free-dynamic lifting. J Electromyogr Kinesiol. 1997 Dec;7(4):25968. 12. Escamilla RA, Babb E, DeWitt R, et al. Nontraditional abdominal exercises: implications for rehabilitation and training. Phys Ther. 2006 May;86(5):656-71. 13. Beim GM, Giraldo JL, Pincivero DM, et al. Abdominal strengthening exercises: a comparative EMG Study. J Sports Rehab. 1997;6:11-20. 14. Floyd WF, Silver PHS. Electromyographic study of patterns of activity in the anterior abdominal wall muscles in man. J Anat. 1950 Apr;84(2):132-46. 15. De Luca CJ. The use of surface electromyography in biomechanics. J Appl Biomech. 1997;13:135-63. 16. Dai B, Sorensen CJ, Derrick TR, Gillette JC. The effects of postseason break on knee biomechanics and lower extremity EMG in a stop-jump task: implications for ACL injury. J Appl Biomech. 2012 Dec;28(6):70817. 17. de Leva P. Adjustments to Zatsiorsky-Seluyanov’s segment inertia parameters. J Biomech. 1996 Sep;29(9):1223-30. 18. McGill SM, Grenier S, Kavcic N. Cholewicki J. Coordination of muscle activity to assure stability of the lumbar spine. J Electromyogr Kinesiol. 2003 Aug;13(4):353-9. 19. Vera-Garcia FJ, Brown SHM, Gray JR, McGill SM. Effects of different levels of torso coactivation on trunk muscular and kinematic responses
172
Volume 19, Number 4, 2015 • Journal of Dance Medicine & Science
to posteriorly applied sudden loads. Clin Biomech (Bristol, Avon). 2006 Jun;21(5):443-55. 20. Stokes IAF, Gardner-Morse MG, Henry SM. Abdominal muscle activation increases lumbar stability: analysis of contributions of different muscle groups. Clin Biomech (Bristol, Avon). 2011 Oct;26(8):797803. 21. Marras WS, Jorgensen MJ. Granata KP, Wiand B. Female and male trunk geometry: size and prediction of the spine loading trunk muscles derived
from MRI. Clin Biomech (Bristol, Avon). 2001 Jan;16(1):38-46. 22. Quint U, Wikle HJ, Shirazi-Adl A, et al. Importance of the intersegmental trunk muscles for the stability of the lumbar spine. A biomechanical study in vitro. Spine (Phila Pa 1976). 1998 Sep;23(18):1937-45. 23. Dias JM, Menacho Mde O, Mazuquin BF, et al. Comparison of the electromyographic activity of the anterior trunk during the execution of two Pilates exercises—teaser
and longspine—for healthy people. J Electromyogr Kinesiol. 2014 Oct;24(5):689-97. 24. Marques NR, Morcelli MH, Hallal CZ, Goncalves M. EMG activity of trunk stabilizer muscles during Centering Principle of Pilates Method. J Bodyw Mov Ther. 2013 Apr:17(2):185-91. 25. Liebenson C, Karpowiz AM, Brown SH, et al. The active straight leg raise test and lumbar spine stability. PM R. 2009 Jun;1(6):530-5.
