AN ASSESSMENT OF THE CONTRACTILE PROPERTIES OF THE SHOULDER MUSCULATURE IN ELITE VOLLEYBALL PLAYERS USING TENSIOMYOGRAPHY

Abstract

Background:

In volleyball, offensive (Hitters) and defensive players (Non-Hitters) perform differing actions that vary both kinematically and in terms of intensity. This may impose contrasting demands on the musculature involved in performing these actions. Previous research has identified differences in the muscle activation and contractile properties of the lower-body musculature between positions. Additionally, asymmetries between dominant and non-dominant limbs of the upper-body musculature has been observed in athletes performing overhead movements.

Purpose:

The aim of this study was to use Tensiomyography (TMG) to examine the contractile properties of the shoulder musculature in elite volleyball players.

Study Design:

Cross-sectional study

Methods:

Thirty-one elite volleyball players participated in this study (Age: 23 ± 2 yrs, Body Mass: 76.5 ± 9.8 kg, Stature: 181 ± 9.3 cm), 26 of which displayed right-limb dominance and five displayed left-limb dominance. Contractile properties of the shoulder musculature including the anterior deltoid (AD), biceps brachii (BB), posterior deltoid (PD), and the upper trapezius (UT) were assessed bilaterally using TMG measures on one occasion prior to any training or exercise. The contractile measures provided by TMG included the maximal displacement (Dm), contraction time (Tc), delay time (Td), sustain time (Ts), and the relaxation time (Tr).

Results:

No statistically significant differences were observed between positions or limbs, except that Hitters displayed a significantly lower Ts of the left AD compared to Non-hitters (p = 0.01, ES = 1.02), and significant differences between dominant and non-dominant sides in the Td of the UT in Non-hitters were present (p = 0.05, ES = 0.8).

Conclusion:

These data suggest that irrespective of playing position and limb dominance, contractile properties of the shoulder musculature in elite volleyball players, as measured using TMG, display few significant differences.

Levels of Evidence:

3b

INTRODUCTION

In volleyball, the objective of offensive players is to attack to score points, typically achieved by hitting the ball at high speeds in an action called “spiking.” The action of spiking is complex and is the compilation of technical skill and muscular qualities. The front row players (i.e outside hitters, opposites and middle blockers) will take a majority of these swings, for elite players practicing 16-20 hours per week, spike counts can reach 40,000 in a single season.¹ On the other hand setters are responsible for handling the second contact, their job is to position the ball for the hitters to perform an overhead strike or “spike” manoeuvre and to put the hitters in the best possible position to score. Finally, liberos are defensive specialists whose main roles are to receive serves and play defence. Setters and liberos still perform overhead skills such as serving and setting however, the intensity of these actions is lower than that seen during spiking performed by hitters.² The highly-specific actions required of each position leads to a difference in demands placed on the body and requires different muscular actions. Furthermore, positional specialization may result in repetitive forceful overhead actions in some volleyball athletes that could create differences in musculature between positions and in dominant (D) versus non-dominant (ND) arms.³

It is unclear whether asymmetries between D and ND limbs are a necessary adaptation or a cause of injury within volleyball. Shoulder overuse injuries are common in volleyball and account for about 8-20% of all volleyball injuries.⁴ Despite the prevalence of shoulder injuries in volleyball, most research examining overhead athletes has studied athletes participating in other biomechanically similar sports such as baseball or tennis.^5,6 The risk of injury in addition to the importance of spiking in volleyball indicate further analysis into the specific musculature involved in this action is needed. This current lack of evidence specific to volleyball means muscular adaptations associated with performing positional specific actions regularly in technical practices is unclear.

Analysis of muscular properties and muscular activity during various movements have been used to gain insight into understanding various types of athletic performance. Reeser, et al.⁷ used electromyography (EMG) to analyze the volleyball spike and determined that muscle activity differed in the various phases of the swing. In preparation for spiking the hitter must cock their arm back by abducting and externally rotating at the shoulder. During the “wind-up” phase of the spike, peak activity was found in the anterior deltoid, infraspinatus and supraspinatus. Then as the arm is cocked back the infraspinatus and teres minor work to externally rotate the shoulder. Finally, as the arm accelerates toward the ball the internal rotators (teres major, subscapularis, pectoralis major and latissimus dorsi) were found to be at their highest activity.⁸ The glenohumeral joint is inherently quite unstable, thus, the muscles of the shoulder, primarily the rotator cuff, are critical to stabilizing the joint while spiking.⁹ Furthermore, studies have found that ball velocity is correlated to strength measures of the internal rotators of the shoulder which, supports the findings seen from the EMG study.¹⁰

Tensiomyography (TMG) is a novel and non-invasive method of quantifying the contractile properties of the muscle including; maximal displacement (Dm), contraction time (Tc), delay time (Td), sustain time (Ts), and relaxation time (Ts). The method involves applying extramuscular electrical stimulation to the muscle in a relaxed state and quantifying the radial displacement of the muscle belly, subsequent calculations can provide information on the magnitude and speed of contraction, speed of relaxation, responsiveness of the skeletal muscle assessed and an estimation the ratio of type I to type II fibers.^11-15 This process can provide novel information compared to EMG as it describes an athletes’ muscular profile, rather than their specific activation patterns.

While previous work has examined the spike action using EMG⁷, what is not yet known is whether the differing positional demands of volleyball result in a different contractile profile of the shoulder musculature at rest. This assessment of contractile properties may provide physiotherapists and support staff valuable information relating to positional specific adaptations of the shoulder musculature in response to practices and game play. This information may assist practitioners in the decision-making process when designing training and rehabilitation interventions, as it relates to accounting for positional differences and potential asymmetries.

TMG has not yet been used to assess differences in the contractile properties of the shoulder musculature between hitters and non-hitters along with the presence of asymmetries between dominant and non-dominant arms. Therefore, the aim of this study was to use TMG to examine the muscular properties of the shoulder in elite volleyball players. It was hypothesized that differences in the contractile properties of the shoulder musculature would be present with hitters displaying faster contraction times, shorter delay times and a greater maximum displacement than non-hitters. Furthermore, it was hypothesized that differences between D and ND arms would be present, with the D arm displaying faster contraction times, shorter delay times and a greater maximum displacement than the ND arm.

METHOD

Participants

A total of 31 elite volleyball players volunteered to participate in the study and were allocated into two groups based on whether they were hitters (H) or non-hitters (NH). All players were competing in the Volleyball England Men's or Women's Super League (Tier 1 of English Volleyball). Participants allocated to the hitters group were characterised by their primary position being considered a front row attacker (i.e. outside hitters, middle blockers and opposites). Participants not characterised by this definition (i.e. setters, defensive specialists and liberos) were allocated to the non-hitter group. Right limb dominance was displayed in 14 participants in the hitters group and all 12 participants in the non-hitters group. Left limb dominance was displayed in five participants in the hitters group. Both males and females were included with participant characteristics presented in Table 1. All none injured squad members participated in the testing procedures, which determined the n. Participants were informed of the procedures and risks associated with the study and provided written informed consent before participating. Ethical approval for the study was granted by the Institutional Ethical Committee.

Table 1.

Descriptive characteristics of Hitters and Non-Hitters (Mean ± SD)

Group	N	Male	Female	Age (years)	Body Mass(kg)	Stature (cm)
Non-Hitters	12	5	7	23 ± 1	72.4 ± 7.9	176.1 ± 8.5
Hitters	19	8	11	24 ± 3	78.9 ± 10.2	185 ± 8.3
Total	31	13	18	23 ± 2	76.5 ± 9.8	181.7 ± 9.3

Protocol

A cross-sectional, comparative study was conducted to assess differences of the muscular properties of the shoulder between H and NH, as well as D vs ND muscle property differences within both groups. Measurements of the muscular properties of the shoulder were assessed in all participants using TMG. The methodology for TMG assessment was identical for all participants with values taken by the same investigator, who had experience with TMG ( > 1 year). All measurements were taken prior to the start of the athlete group's training session before any exercise had been undertaken.

TMG measurements were taken of the anterior deltoid (AD), biceps brachii (BB), posterior deltoid (PD), and the upper trapezius (UT) on both left and right sides. These muscle groups were selected based on previous studies investigating the muscle firing patterns of the shoulder during the volleyball serve and spike along with similar overhead throwing motions.^2,8 In these studies, the AD, BB and PD were identified, among other deep muscles, as key muscle groups acting on the glenohumeral joint and the UT was identified as a key muscle group, among other deep muscles, acting on the scapular joint. These selected muscles were chosen for measurement as they are superficial and therefore able to be assessed using TMG. All measurements were taken when participants were in a seated upright position, as recommended by the TMG user guidelines. The values recorded from these measurements were then used to assess differences between H and NH, along with differences between left and right limbs within groups. The reliability of TMG measurements has previously been established, with intra-class correlation coefficient (ICC) ranges of 0.77-0.97 and 0.86-0.98 reported for all muscle contractile properties of the vastus medialis and biceps brachii, respectively. ^16,17 Specific information pertaining to calculation of metrics derived from contractility the authors refer the reader the cited document. ¹⁸

Procedures

TMG creates a radial displacement using a portable device via an electrical stimulus (up to 100mA, for 1 ms) that is applied percutaneously, eliciting a muscular contraction that is detected by a digital transducer applied above the muscle belly (Figure 1).¹¹ This digital transducer records the displacement from the muscle belly using a spring loaded displacement sensor at the surface of the skin (TMG-BMC Ltd, Ljubljana, Slovenia). The sensor was consistently retracted to 50% of its length to ensure a consistent initial pressure for all muscles measured.¹³ The sensor was positioned perpendicular to the thickest part of each muscle group, identified through visual inspection and palpation of the muscle during a voluntary contraction.¹² The electrical stimulus was delivered through self-adhesive electrodes that were placed approximately 5cm on either side of the sensor for all muscle groups.

Figure 1.

Visual representation of the equipment and set up for the measurement of the contractile properties of the biceps brachii using tensiomyography.

A series of contractions of increasing amplitude (approximately 10mA) was used to obtain a maximal response. This maximal response was determined by a plateau of muscle displacement in the twitch response curves.¹³ Only the maximal output data were used for subsequent analyses. The variables measured using TMG for all muscle groups were maximal displacement, contraction time, delay time, sustain time, and relaxation time (Figure 2). Maximal displacement (Dm): The maximal radial displacement of the muscle belly. Contraction time (Tc): the contraction time between 10 and 90% Dm. Delay time (Td): the time taken from the onset of the electrical stimulus to 10% of the maximal radial displacement. Sustain time (Ts): the time between the instant when the Dm reached 50% of its value until, during relaxation, the Dm returned to 50% of its maximal value. Relaxation time (Tr): the time taken for Dm to fall from 90% to 50% (relaxation time, Tr).¹⁷

Figure 2.

Example of a radial twitch responses to an electrical stimulation with the measured variables annotated. Dm = Maximal displacement, Tc = Contraction time, Td = Delay time, Ts = Sustain time, Tr = Relaxation time.

Statistical Analysis

Data are presented as mean ± SD. Statistically significant differences were noted when p < 0.05. All statistical analyses were conducted using the Microsoft Excel 2013 statistical package. Differences between groups for all variables in the muscle groups measured were assessed using a paired samples independent t-test. Differences between D and ND limbs within groups for all variables in the muscle groups measured were assessed using a within-subjects dependent t-test. Percentage differences were calculated for differences between groups as well as for differences within groups for all variables in the muscle groups measured. Within group differences were also assessed by calculating 90% confidence intervals for all variables in the muscle groups measured.

In addition, effect sizes were calculated along with associated qualitative inferences for both differences between groups and within groups. Between group effect sizes were calculated using the formula from Hedges (Equation 1)¹⁸ as the Hitters and Non-Hitters contained different group sizes. Within group effect sizes were calculated using the formula from Cohen (Equation 2).¹⁹ The qualitative inferences associated with the calculated effect sizes were defined as trivial ( < 0.2), small (0.2 - 0.6), moderate (0.6 - 1.2), large (1.2 - 2.0), very large (2.0 - 4.0), and nearly perfect ( > 4.0).²⁰

$$Hedges\text{'}g=\frac{M_1-M_2}{S{D}_{pooled}^{*}}$$

$$Cohen\text{'}sd=\frac{M_1-M_2}{S{D}_{pooled}^{*}}$$

RESULTS

Descriptive characteristics of participants are presented in Table 1.

Between groups

Hitters (76.77 ± 50.91 m·s⁻¹) showed significantly different (404% lower, p = 0.01) sustain time of the left AD compared to NH (186.34 ± 159.44 m·s⁻¹) with a moderate effect size (ES = 1.02).

Moderate effect sizes were observed in the right AD of H showing a 143.27% lower sustain time (83.36 ± 93.62 vs 211.31 ± 270.86 m·s⁻¹, p = 0.07, ES = 0.69); a 364.26% lower relaxation time (26.12 ± 31.00 vs 58.5 ± 62.99 m·s⁻¹, p = 0.07, ES = 0.70); and a 47.11% lower delay time (16.96 ± 6.85 vs 22.38 ± 10.38 m·s⁻¹, p = 0.09, ES = 0.65) of the right AD compared to NH, respectively, although none of these differences were significantly different. No significant differences were observed between H and NH (p > 0.05) with trivial and small effect sizes calculated in all other muscle groups for all other variables measured (Table 2).

Table 2.

Differences in the contractile properties of the muscle groups measured using tensiomyography between Hitters and Non-Hitters (Mean ± SD)

Muscle	Variable	Hitters	Non-Hitters	P value	% Difference	Effect Size (g)	Qualitative Inference
Left Bicep	Tc (m·s−1)	24.98 ± 6.98	22.06 ± 8.37	0.30	11.67	0.39	small
	Ts (m·s−1)	105.77 ± 74.27	118.86 ± 97.04	0.67	-12.37	0.16	trivial
	Tr (m·s−1)	47.79 ± 32.03	52.01 ± 68.63	0.82	-8.83	0.09	trivial
	Dm (mm)	6.47 ± 11.53	3.12 ± 2.13	0.33	51.78	0.37	small
	Td (m·s−1)	28.01 ± 8.29	28.31 ± 8.99	0.93	-1.07	0.03	trivial
Right Bicep	Tc (m·s−1)	23.87 ± 5.18	25.38 ± 7.03	0.50	-6.32	0.25	small
	Ts (m·s−1)	74.18 ± 38.16	130.67 ± 153.50	0.13	-76.15	0.57	small
	Tr (m·s−1)	32.36 ± 21.18	36.97 ± 28.36	0.61	-14.26	0.19	trivial
	Dm (mm)	3.45 ± 2.54	3.36 ± 1.77	0.92	2.53	0.04	trivial
	Td (m·s−1)	28.09 ± 5.94	28.01 ± 4.18	0.97	0.31	0.02	trivial
Left Anterior Deltoid	Tc (m·s−1)	16.21 ± 5.03	16.49 ± 4.59	0.88	-2.06	0.06	trivial
	Ts (m·s−1)	76.77 ± 50.91	186.34 ± 159.44	0.01*	-404.16	1.02	moderate
	Tr (m·s−1)	27.38 ± 21.60	41.68 ± 31.06	0.15	-140.05	0.56	small
	Dm (mm)	1.35 ± 0.98	1.23 ± 0.94	0.73	11.96	0.13	trivial
	Td (m·s−1)	20.97 ± 6.14	19.54 ± 3.97	0.48	8.05	0.27	small
Right Anterior Deltoid	Tc (m·s−1)	14.21 ± 5.70	17.08 ± 3.99	0.14	-34.70	0.56	small
	Ts (m·s−1)	83.36 ± 93.62	211.31 ± 270.86	0.07	-143.27	0.69	moderate
	Tr (m·s−1)	26.12 ± 31.00	58.5 ± 62.99	0.07	-364.26	0.70	moderate
	Dm (mm)	1.15 ± 1.07	1.03 ± 0.54	0.73	32.95	0.13	trivial
	Td (m·s−1)	16.96 ± 6.85	22.38 ± 10.38	0.09	-47.11	0.65	moderate
Left Posterior Deltoid	Tc (m·s−1)	14.22 ± 5.69	13.77 ± 4.01	0.82	2.83	0.09	trivial
	Ts (m·s−1)	70.57 ± 80.78	46.73 ± 51.26	0.37	84.08	0.34	small
	Tr (m·s−1)	10.19 ± 9.16	7.59 ± 4.53	0.37	26.61	0.34	small
	Dm (mm)	0.48 ± 0.39	0.64 ± 0.62	0.36	-32.85	0.34	small
	Td (m·s−1)	20.70 ± 13.51	17.19 ± 4.15	0.39	16.24	0.32	small
Right Posterior Deltoid	Tc (m·s−1)	13.81 ± 3.81	13.82 ± 5.56	1.00	-0.07	0.00	trivial
	Ts (m·s−1)	80.45 ± 87.11	45.84 ± 66.69	0.25	111.07	0.43	small
	Tr (m·s−1)	12.07 ± 9.19	11.23 ± 8.86	0.80	7.28	0.09	trivial
	Dm (mm)	0.42 ± 0.35	0.56 ± 0.29	0.25	-34.03	0.43	small
	Td (m·s−1)	22.94 ± 14.78	18.26 ± 6.77	0.31	21.80	0.38	small
Left Trapezius	Tc (m·s−1)	55.22 ± 24.27	58.30 ± 28.88	0.75	-5.84	0.12	trivial
	Ts (m·s−1)	316.77 ± 174.11	359.95 ± 214.58	0.54	-10.92	0.23	small
	Tr (m·s−1)	111.82 ± 125.89	127.14 ± 128.85	0.75	-16.63	0.12	trivial
	Dm (mm)	2.57 ± 1.46	2.57 ± 1.17	0.99	-0.40	0.00	trivial
	Td (m·s−1)	30.86 ± 12.24	41.67 ± 26.24	0.13	-28.74	0.57	small
Right Trapezius	Tc (m·s−1)	52.36 ± 26.44	54.05 ± 29.47	0.87	-3.07	0.06	trivial
	Ts (m·s−1)	326.75 ± 167.04	372.54 ± 202.25	0.50	-23.68	0.25	small
	Tr (m·s−1)	116.17 ± 82.70	115.14 ± 74.06	0.97	2.27	0.01	trivial
	Dm (mm)	2.52 ± 1.39	2.57 ± 1.29	0.93	-3.46	0.03	trivial
	Td (m·s−1)	27.43 ± 8.53	26.66 ± 4.67	0.78	2.78	0.10	trivial

Tc = contraction time, Ts = sustain Time, Tr = relaxation time, Dm = maximum displacement, Td = delay time.

*denotes statistically significant difference between Hitters and Non-Hitters at p < 0.05.Effect size inferences are defined as; < 0.2 = Trivial, 0.2-0.6 = Small, 0.6-1.2 = Moderate, 1.2-2.0 = Large, 2.0-4.0 (Very Large), > 4.0 (Nearly Perfect).

Within groups

Significant differences and moderate effect sizes were observed in the delay time between the D and ND UT in NH (36.01%, 90% CI = -26.08 to -3.93 m·s⁻¹, ES = 0.8, p = 0.05) and the contraction time of the D and ND BB in H (15.72%, 90% CI = -4.04 to 1.83 m·s⁻¹, ES = 0.72, p = 0.01). A moderate effect size was observed between D and ND sides for the relaxation time of the PD in NH (48.02%, 90% CI = 0.23 to 7.06 m·s⁻¹, ES = -0.52, p = 0.11), however this was non-significant (Table 3).

Table 3.

Differences between dominant and non-dominant sides of Hitters and Non-Hitters of the contractile properties of the muscle groups measured using tensiomyography (Mean ± SD). Right limb dominance was displayed in all 12 Non-Hitters and 14 Hitters, left limb dominance was displayed in 5 Hitters.

Muscle	Group	Variable	Non-Dominant	Dominant	p-value	% Difference	90% CI	Effect Size (d)	Qualitative Inference
Bicep	Hitters	Tc (m·s−1)	26.51 ± 64.73	22.34 ± 6.68	0.01*	15.72	-4.04 to 1.83	0.72	moderate
		Ts (m·s−1)	106.39 ± 73.42	73.56 ± 39.24	0.04*	30.85	-57.05 to -6.14	0.56	small
		Tr (m·s−1)	47.32 ± 26.71	32.83 ± 27.85	0.12	30.62	-30.04 to -0.83	0.53	small
		Dm (mm)	4.23 ± 2.53	3.29 ± 2.61	0.03*	22.25	-6.70 to 0.67	0.37	small
		Td (m·s−1)	29.16 ± 6.49	26.96 ± 7.71	0.46	7.51	-4.79 to 4.98	0.31	small
	Non-Hitters	Tc (m·s−1)	22.06 ± 8.37	25.38 ± 7.03	0.18	-15.03	-0.53 to 7.17	-0.43	small
		Ts (m·s−1)	118.86 ± 97.04	130.67 ± 153.50	0.79	-9.93	-60.79 to 84.40	-0.09	trivial
		Tr (m·s−1)	52.01 ± 68.63	36.97 ± 28.36	0.39	28.92	-42.40 to 12.31	0.29	small
		Dm (mm)	3.12 ± 2.13	3.36 ± 1.77	0.52	-7.72	-0.35 to 0.83	-0.12	trivial
		Td (m·s−1)	28.31 ± 8.99	28.01 ± 4.18	0.89	1.04	-3.87 to 3.28	0.04	trivial
Anterior Deltoid	Hitters	Tc (m·s−1)	15.89 ± 4.86	14.54 ± 5.94	0.24	8.50	-3.61 to -0.17	0.25	small
		Ts (m·s−1)	78.38 ± 52.12	81.74 ± 93.05	0.89	-4.29	-34.60 to 47.08	-0.04	trivial
		Tr (m·s−1)	29.68 ± 23.88	23.82 ± 28.98	0.41	19.74	-12.59 to 10.20	0.22	small
		Dm (mm)	1.39 ± 1.14	1.11 ± 0.89	0.30	19.66	-0.61 to 0.23	0.27	small
		Td (m·s−1)	19.91 ± 1.14	18.02 ± 7.67	0.45	9.51	-7.53 to -0.07	0.28	small
	Non-Hitters	Tc (m·s−1)	16.49 ± 4.59	17.08 ± 3.99	0.69	-3.60	-1.78 to 2.97	-0.14	trivial
		Ts (m·s−1)	186.34 ± 159.44	211.31 ± 270.86	0.63	-13.40	-56.86 to 106.81	-0.11	trivial
		Tr (m·s−1)	41.68 ± 31.06	58.5 ± 62.99	0.28	-40.37	-7.74 to 41.39	-0.34	small
		Dm (mm)	1.23 ± 0.94	1.03 ± 0.54	0.50	15.71	-0.64 to 0.26	0.25	small
		Td (m·s−1)	19.54 ± 3.97	22.38 ± 10.38	0.36	-14.52	-2.06 to 7.73	-0.36	small
Posterior Deltoid	Hitters	Tc (m·s−1)	13.84 ± 4.14	14.19 ± 5.46	0.75	-2.58	-2.22 to 1.41	-0.07	trivial
		Ts (m·s−1)	73.00 ± 78.08	78.02 ± 89.75	0.82	-6.88	-26.75 to 46.51	-0.06	trivial
		Tr (m·s−1)	11.44 ± 9.38	10.82 ± 9.06	0.78	5.39	-1.64 to 5.39	0.07	trivial
		Dm (mm)	0.46 ± 0.43	0.43 ± 0.31	0.69	6.61	-0.18 to 0.07	0.08	trivial
		Td (m·s−1)	22.45 ± 14.98	21.19 ± 13.35	0.75	5.56	-4.15 to 8.62	0.09	trivial
	Non-Hitters	Tc (m·s−1)	13.77 ± 4.01	13.82 ± 5.56	0.98	-0.38	-3.19 to 3.30	-0.01	trivial
		Ts (m·s−1)	46.73 ± 51.26	45.84 ± 66.69	0.97	1.90	-41.65 to 39.87	0.01	trivial
		Tr (m·s−1)	7.59 ± 4.53	11.23 ± 8.86	0.11	-48.02	0.23 to 7.06	-0.52	moderate
		Dm (mm)	0.64 ± 0.62	0.56 ± 0.29	0.65	12.82	-0.37 to 0.21	0.17	trivial
		Td (m·s−1)	17.19 ± 4.15	18.26 ± 6.77	0.68	-6.19	-3.02 to 5.15	-0.19	trivial
Trapezius	Hitters	Tc (m·s−1)	52.32 ± 25.87	55.26 ± 24.87	0.70	-5.63	-15.27 to 9.56	-0.12	trivial
		Ts (m·s−1)	321.45 ± 171.78	322.07 ± 169.59	0.99	-0.19	-58.68 to 78.66	0.01	trivial
		Tr (m·s−1)	111.53 ± 124.91	116.46 ± 84.15	0.88	-4.42	-47.27 to 55.98	-0.05	trivial
		Dm (mm)	2.55 ± 1.43	2.54 ± 1.42	0.98	0.27	-0.57 to 0.49	0.01	trivial
		Td (m·s−1)	29.71 ± 12.41	28.57 ± 8.60	0.57	3.83	-6.44 to 0.42	0.11	small
	Non-Hitters	Tc (m·s−1)	58.30 ± 28.88	54.05 ± 29.47	0.68	7.29	-20.95 to 12.45	0.15	trivial
		Ts (m·s−1)	359.95 ± 214.58	372.54 ± 202.25	0.88	-3.50	-117.433 to 142.60	-0.06	trivial
		Tr (m·s−1)	127.14 ± 128.85	115.14 ± 74.06	0.73	9.44	-66.99 to 42.99	0.11	trivial
		Dm (mm)	2.57 ± 1.17	2.57 ± 1.29	1.00	0.06	-0.78 to 0.78	0.00	trivial
		Td (m·s−1)	41.67 ± 26.24	26.66 ± 4.67	0.05*	36.01	-26.08 to -3.93	0.80	moderate

Tc = Contraction Time, Ts = Sustain Time, Tr = Relaxation Time, Dm = Maximum Displacement, Td = Delay Time. 90% CI = 90% Confidence Interval

*denotes statistically significant difference between dominant and non-dominant sides within the muscle group measured p < 0.05. Effect size inferences are defined as; < 0.2 = Trivial, 0.2-0.6 = Small, 0.6-1.2 = Moderate, 1.2-2.0 = Large, 2.0-4.0 (Very Large), > 4.0 (Nearly Perfect).

Significant differences were observed between D and ND sides in the sustain time (30.85%, 90% CI of -57.05 to -6.14 m·s⁻¹, P = 0.04) and maximum displacement (22.25%, 90% CI of -6.70 to 0.67 mm, p = 0.03) of the BB in H, however only small effect sizes were calculated (ES = 0.56 and ES = 0.37, respectively). No significant differences were observed between D and ND muscle groups in any of the variables measured for H and NH (p > 0.05), with trivial and small effect sizes measured and 90% CI's spanning 0.

As can be seen from Figure 3, no significant differences were observed in the overall contractile symmetry of the muscle groups measured in H or NH.

Figure 3.

Panel A: Mean Contraction Time and standard deviation (m·s-1) of Left and Right muscle groups of Hitter and Non Hitter groups. Panel B: Mean Sustain Time and standard deviation (m·s-1) of Left and Right muscle groups of Hitter and Non Hitter groups. *statistically significant difference (P < 0.05) between hitters and non-hitters of the Left Anterior Deltoid muscle. Panel C: Mean Relaxation Time and standard deviation (m·s-1) of Left and Right muscle groups of Hitter and Non Hitter groups. Panel D: Mean Displacement and standard deviation (mm) of Left and Right muscle groups of Hitter and Non Hitter groups. Panel E: Mean Delay Time and standard deviation (m·s-1) of Left and Right muscle groups of Hitter and Non Hitter groups. *statistically significant difference (P < 0.05) within non-hitters between left and right sides of the Trapezius muscle. Panel F: Mean Lateral Symmetry and standard deviation (%) of muscle groups of Hitter and Non Hitter groups.

DISCUSSION

The aim of the current study was to assess if differences exist in the contractile properties of the shoulder musculature in elite volleyball players between playing positions, and if asymmetries in the contractile properties exist between dominant and non-dominant arms as measured by TMG. Key findings indicate that H and NH display similar contractile properties and no asymmetries between dominant and non-dominant arms in the shoulder muscle.

Positional differences between Hitters and Non-Hitters

The findings of the current study are unexpected due to the differences in actions performed between H and NH.^1,2 Specifically, H perform spiking actions requiring forceful muscular contractions performed at high velocities and intensities to strike the ball with the greatest velocity.^1,7 In contrast, NH perform setting actions requiring lower intensity, less forceful muscular contractions to place the ball in specific positions at lower velocities in preparation for H to strike.^2,10 The differences expected are supported by performance evaluations of playing position conducted by Mielgo-Ayuso, et al.²³ who observed significantly greater overhead medicine ball throw scores in hitters compared to non-hitters. Similarly, Marques, et al.²⁴ observed greater upper-body strength performances in hitters compared to setters and liberos in the bench press and overhead medicine ball throw. Marques, et al.²⁴ suggested the diminished upper-body strength qualities exhibited by liberos and setters may be a reflection of the limited upper-body movement during play compared to hitters. On the contrary, when examining overhead medicine ball throw performance, Milic´, et al.²⁵ observed no differences between playing position, suggesting contractile properties of the upper body musculature may not differ. These contrasting findings however, may be due to the populations studied. Mielgo-Ayuso, et al.²³ and Marques, et al.²⁴ studied professional female and male volleyball players respectively, whereas Milic´, et al.²⁵ studied young female volleyball players. This difference in playing level and experience may contribute to the contrasting findings, due to professional players having had longer to develop specialist characteristics for their playing position;²⁶ thus exhibiting greater differences than their less experienced counterparts. Of course, these variances in findings could also be due to the different outcome measures selected in the respective studies.

While the lack of differences in TMG outcomes observed in the current study are in contrast with previous findings,^23,25 this may be a reflection of the assessment of the individual musculature rather than the assessment of synergistic muscles which co-contract to perform an action. Specifically, the bench press and overhead medicine ball throw utilize a synergistic contraction of multiple upper body muscle groups to produce force,^27,28 thus, the differences in performance observed in these studies between playing position may not be a result of differing contractile properties of the shoulder musculature. Rather, the differences exhibited may be a result of the overall contribution of the musculature involved in performing the overhead medicine ball throw and bench press. Moreover, these movements require active recruitment of the muscle by the participant, whereas the use of TMG (as in the current study) requires no physical effort from the participant as an electrical stimulus is used to evoke a muscular contraction.¹¹ This may suggest that contractile differences may not be apparent, as observed in the current study; however, the contribution of other muscle groups within the upper body may contribute to the differences observed between playing positions in previous studies.

The current study is the first to use TMG to assess the contractile properties of the shoulder musculature in volleyball players. However, previous research has used TMG to assess the contractile properties of lower body musculature in volleyball players. These studies have observed lower maximal displacement values in the biceps femoris in defensive specialists compared to blocking players²⁹ and faster normalized response speeds in liberos and setters (NH) compared to opposites and middles (H) in the biceps femoris, rectus femoris, vastus medialis and vastus lateralis.³⁰ These differences may be attributable to NH performing knee flexion movements during controlled jumps and lateral movements requiring greater stability and isometric contractions; whereas hitters perform movements requiring greater velocities of knee flexion to extension with less stability and isometric contractions.

The current data results indicate that contractile properties of the shoulder musculature are similar between positions and dominant and non-dominant arms. This in contrast to previous work reporting that playing position in volleyball can influence the muscular contractile properties of the lower body^29,30 These novel findings may be a result of the muscle group assessed. As suggested by Rodríguez-Ruiz, et al.²⁹ and Rodriguez-Ruiz, et al.³⁰ non-hitters perform receiving and setting actions with involvement from the lower body musculature, which may contribute to the limited upper body involvement in non-hitters exhibited during play compared to hitters.²⁴ Consistent with this, while hitters perform forceful explosive actions; the lower body and hip-trunk musculature significantly contribute to performing these actions.^7,31 The contributions of the lower body musculature in performing these respective actions may lead to less involvement of the upper body musculature than expected.

Additionally, the strength-training practices of the participants in the current study may have further contributed to the lack of differences between playing positions observed. All participants had an extensive strength-training background and completed a regular, comprehensive strength-training programme alongside their technical practices. Of note, the strength programs completed by the participants did not differ between playing position. This similarity in the strength-training practices between positions may off-set any positional adaptations in the shoulder musculature caused during technical practices and game play. The combination of similarity in strength-training programmes and the lesser contribution of the upper body musculature than expected may lead to a lack of positional specific adaptations in the shoulder; resulting in the observation of no positional differences in the contractile properties of the shoulder.

Symmetrical differences between Dominant and Non-Dominant limbs

When considering inter-limb differences in the contractile properties of the shoulder, the current study observed no differences in any variables assessed in both H and NH, contradicting previous research suggesting inter-limb differences would be prevalent in volleyball athletes, irrespective of position.^3,32,33 Hadzic, et al.³³ showed that the strength of the internal rotators (IR) of the shoulder muscle (assessed isometrically) were greater in the dominant shoulder than in the non-dominant shoulder. Interestingly, Hadzic, et al.³³ did not find any differences in strength ratios or asymmetry between playing positions in the internal or external rotators. Supporting Hadzic, et al.³³, several studies have reported greater IR strength in the dominant shoulder compared to the non-dominant shoulder assessed isokinetically and isometrically.^34-36 The asymmetries reported in these studies are in contrast to the findings of the present study. The lack of differences observed however, may be a result of methodological differences in assessing muscle asymmetry along with the musculature assessed. The present study used TMG analysis to assess the contractility of the shoulder muscle, which is limited to the measurement of superficial muscles due to the non-invasive nature of the measurement.^11,37,38 Furthermore, TMG assessments are passive in nature, thus comparisons with active strength measures are problematic. Previous studies reporting asymmetries in muscle strength have used isokinetic and isometric strength measures able to assess combinations of deep musculature such as the teres major and subscapularis for IR, and the teres minor and infraspinatus for ER of the shoulder, as opposed to singular muscle groups. The limitation of TMG in assessing only superficial musculature may mean that although asymmetries were not able to be identified in these muscle groups, they may exist in deep muscles.

The lack of asymmetry observed in the current study may also be a result of the training practices of the participants. Volleyball practices typically result in 16-20 hours per week of training on-court with spike counts in excess of 40,000 in a single season.¹ It may be expected this high volume of technical skill performance would result in asymmetrical adaptations of the upper-body musculature; however, this is without consideration of strength training practices. In the current study, the population of players measured were well-trained and had an extensive strength-training background, typically training bilaterally. Previous research has shown that stronger individuals who strength train bi-laterally display greater symmetry than relatively weaker individuals;³⁹ with bi-lateral strength training performed in weaker individuals reducing the presence of strength asymmetries in the lower and upper body.^40,41

CONCLUSION

The results of the current study show that irrespective of playing position, the contractile properties of the AD, PD, BB, and UT muscles of the shoulder in elite volleyball players, as measured using TMG, display no significant differences. Furthermore, TMG measures displayed no differences between dominant and non-dominant arms. The assessment and monitoring of contractile properties using TMG methods may be a suitable, non-invasive option for assessing the musculature of the shoulders of volleyball players. Future research should continue to assess upper body musculature in elite volleyball players using a combination of TMG and electromyography (EMG) measures as TMG is limited to superficial muscles only. This would help to further understand if differences in contractile properties relate to neuromuscular measures.