Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Muscle stretch reflexes are widely considered to beneficially influence joint stability and power generation in the lower limbs. While in the upper limbs and especially in the muscles surrounding the shoulder joint such evidence is lacking.

OBJECTIVE:

To quantify the electromyographical response in the muscles crossing the shoulder of specifically trained overhead athletes to an anterior perturbation force.

METHODS:

Twenty healthy male participants performed six sets of different external shoulder rotation stretches on an isokinetic dynamometer over a range of amplitudes and muscle pre-activation moment levels. All stretches were applied with a dynamometer acceleration of 10,000 ${}^{\circ}$ /s ${}^{2}$ and a velocity of 150 ${}^{\circ}$ /s. Electromyographical response was measured via sEMG.

RESULTS:

Consistent reflexes were not observed in all experimental conditions. The reflex latencies revealed a significant muscle main effect (F (2,228) $=$ 99.31, $p<$ 0.001; $\eta^{2}=$ 0.466; $f=$ 0.934) and a pre-activation main effect (F (1,228) $=$ 142.21, $p<$ 0.001; $\eta^{2}=$ 0.384; $f=$ 1.418). The stretch reflex amplitude yielded a significant pre-activation main effect (F (1,222) $=$ 470.373, $p<$ 0.001; $\eta^{2}=$ 0.679; $f=$ 1.454).

CONCLUSION:

Short latency muscle reflexes showed a tendency to an anterior to posterior muscle recruitment whereby the main internal rotator muscles of the shoulder revealed the most consistent results.

Keywords

Elastic energy storage stretch reflex isokinetic dynamometry throwing shoulder perturbation external shoulder rotation

1. Introduction

High-speed over-head throwing requires a complex sequence of individual joint movements involving the whole body. Several investigations examined which joints and joint-specific angular motions are predominantly involved in the generation of high projectile velocities [1, 2, 3, 4]. Roach et al. [5] and Roach and Lieberman [6] promoted the hypothesis of the ‘kinetic chain’ which transfers momentum, generated in adjacent, connected body segments from joint to joint. The resulting interaction moments are evidence of high angular velocities during throwing in line with velocity-dependent torso rotation and shoulder internal rotation [2]. Studies that analysed muscular activity in throwing and related tasks indicate high muscle activity during internal shoulder rotation throughout the acceleration phase of the movement [7, 8]. However, some findings emphasized that only approximately 50% of the generated shoulder power can be explained by muscle activation patterns [5]. First explanatory approaches suggested that the storage and release of elastic energy also cause significant contribution to power enhancement within the upper body due to a ‘pre-stretching’ of numerous throwing related muscles just prior to their activation [9]. The stretch shortening cycle (SSC) of muscle function has been biomechanically analysed by several investigators over the last decades. Convincing evidence indicates that a SSC results in an enhancement of performance within the final phase of movement (concentric phase) when compared to isolated concentric action [10, 11]. However, so far most investigations have focused on the lower limbs. As described by Roach et al. [5] the posture of the arm at the end of the ‘cocking-phase’ (time of maximum external shoulder rotation) passively stretches elastic elements and therefore enables elastic energy storage and release at the shoulder. The exact magnitude of the rapid stretch contribution to power enhancement still remains largely unknown, especially within the upper extremity.

Former research has outlined an existing reflex arc within the feline model with an extremely short latency (2.7–3.1 ms) between the shoulder capsule and the internal rotator muscles of the shoulder evoked by electrical stimulation of the shoulder capsule [12]. In accordance with these findings the existence of various proprioceptors such as free nerve endings and Golgi mechanoreceptors in the human glenohumeral joint has been reported [13, 14]. Purposeful investigations of the monosynaptic short latency stretch reflex (SLR) indicated that it has considerable importance in stiffness regulation of the muscle, which would consequently enhance joint moments and power during SSCs movements [15, 16]. However only a few studies have used a joint perturbation to simulate specific tasks in the upper extremity which is associated with rapid angular motions of the glenoid joint. An exception was a study by Latimer et al. [17] analysing muscle response to an anterior perturbation force. Observed latencies ranged from 110 to 220 ms whereby the anterior muscles responded first followed by the posterior muscles. Furthermore, Myers et al. [18] demonstrated that muscle reflexive response considerably relied on muscle pre-activation which was expressed in a latency shortening. However, the results of these investigations were in disagreement to the findings of Roach et al. [5]. For an elite thrower up to 70% of the achieved projectile release velocity is developed in the last 100 ms of the movement [19]. The actual arm acceleration of a pitching motion which is initiated at the point of maximal external shoulder rotation and ends at ball release took only 42–58 ms of the total pitch temporal sequence (foot contact to ball release: 145 $\pm$ 15 ms) [20]. Therefore, the observed reflexive responses of the internal rotator muscles of the shoulder appear to be too slow to have any beneficial influence on the generation of projectile velocity.

The purpose of the current research was (i.) to examine the capability of isokinetic dynamometry for the evaluation of muscular stretch reflexes in the muscles surrounding the shoulder and (ii.) to find evidence for SSC mechanisms at the shoulder to be similar to those described for the lower limbs [21]. We hypothesised that (a) muscle pre-activation moment and stretch amplitude would significantly influence reflex latency, reflex amplitude and muscle activation duration in terms of the stretch, (b) that the clavicular part of the pectoralis major muscle and the anterior part of the deltoid muscle in their function as major internal rotators of the shoulder would show the fastest and most consistent response to the anterior perturbation by the dynamometer as formerly reported in the literature [17].

Figure 1.

A subject positioned in the dynamometer during familiarisation. Left: Starting position (90/90/90-position). Right: iMER (position of individual maximal external shoulder rotation).

2. Methods

2.1 Participants

Twenty healthy right-handed men (age: 22.9 $\pm$ 2.38 y, height: 182.7 $\pm$ 5.8 cm, weight: 83.1 $\pm$ 8.5 kg), with no history of shoulder musculoskeletal injury within the last twelve months or any shoulder weakness or discomfort during activity participated in this study. All participants performed an over-head sport (javelin throw, volleyball, handball) for at least 4 years with a minimum of two training sessions per week (2 to 6 sessions/week). Before measurement participants were fully informed of the experimental procedures and provided written informed consent. The study was approved by the Ethic Committee of the German Sport University Cologne (Approval number 121/2017) in compliance with the Declaration of Helsinki. Subjects were assigned to an anonymous ID during the study in order to mask individual identity and warrant data anonymity.

2.2 Experimental protocol

The study consisted of one familiarisation and one test session. Subjects were seated on an isokinetic dynamometer (IsoMed2000, D&R GmbH, Hemau, Germany) in an upright position. A special back pad was used to stabilize the left (non-dominant) shoulder and ensure a sufficient abutment for the moving dominant arm throughout the application of the stretches. The right (also the dominant in all subjects) arm was 90 ${}^{\circ}$ abducted, 90 ${}^{\circ}$ externally rotated and the elbow was flexed 90 ${}^{\circ}$ with the aid of a mechanical goniometer. The resulting position was defined as the start setting (90/90/90-position) for the external shoulder rotation trials. This position mimicked the functional shoulder position during the late cocking phase of a pitching motion [22]. We defined that the shoulder was 0 ${}^{\circ}$ externally rotated in this starting position (Fig. 1).

We determined maximum isometric moment of force (MVC) of the pectoral muscle, the deltoid muscle and the latissimus dorsi muscle from at least 3 trials of maximum isometric contractions of each subject in the 90/90/90-position with 2 minutes break between trials. The highest moment value during a steady plateau was used as MVC.

To minimise the risk of overstrain the individual maximum external stretch amplitude (iMER) for each participant was determined within the familiarisation. We used the same stretch velocity as in the test session (150 ${}^{\circ}$ /s). Therefore, the dynamometer lever arm applied an external perturbation moment to externally rotate the subjects forearm and humerus. The subjects were instructed to keep their muscles in a relaxed state during this procedure. Once they perceived any discomfort within the glenohumeral joint or any surrounding muscle they were encouraged to actively contract their muscle. This released a trigger, which was set at 10% MVC. Once these 10% moment level was reached, the motion direction of the dynamometer lever arm was reversed. The average stretch amplitude of at least 10 trials was defined as individual maximal stretch amplitude.

Table 1
Experimental conditions. The order of stretch conditions was randomized

Condition	Stretch	Stretch	Pre-innervation
	amplitude	velocity	level
	(% of iMER)	( ${}^{\circ}$ /s)	(% of MVC)
Warm-Up	–	15	–
Internal MVC	–	–	–
External stretch	50	150	0
	50	150	25
	50	150	50
	100	150	0
	100	150	25
	100	150	50

iMER: individual maximal external shoulder rotation, MVC: maximum voluntary contraction.

During the test session external shoulder rotation stretches were applied using the formerly determined individual maximal stretch amplitudes (50% of iMER and 100% of iMER) and three different pre-activation moment levels (0% MVC, 25% MVC and 50% MVC) according to the moment values of the MVC-test (Table 1). This set-up results in six different testing conditions. For each condition a minimum of 10 trials was accomplished with a rest period of 10 s between trials to avoid habituation and fatigue of the response [23] and possible influence of post activation depression [24]. Each stretch was applied with a dynamometer acceleration of 10 ${}^{4}$ ${}^{\circ}$ /s ${}^{2}$ and a velocity of 150 ${}^{\circ}$ /s. For trials with 25% or 50% pre-activation moment level, the pre-activation levels were steered via verbal feedback. The participants were requested to contract their internal rotator muscles against a fixed resistance. Once they reached the required moment they were asked to keep their moment at the correspondent level for at least 1 second before the release of the trigger and the onset of the external shoulder rotation movement. The stretch velocity was chosen in accordance to the study of Zanca et al. [25] who determined that significantly higher velocities during external shoulder rotation movements yield unphysiological conditions in terms of force generating capability of the surrounding muscles. Furthermore, Finley et al. [26] verified an acceleration dependency of the short latency stretch reflex components up to stretch accelerations of 5 $\times$ 10 ${}^{4}$ ${}^{\circ}$ /s ${}^{2}$ resulting in an successive increase of stretch response at increasing accelerations. The order of the test conditions was randomized. Moment, lever arm position and sEMG were registered continuously. MVC trials were repeated at the end of the test session to ensure that the test protocol did not induce obvious signs of fatigue.

2.3 Instrumentation

Surface EMG (sEMG) activity of the primary internal rotator muscles of the shoulder was recorded at a sampling frequency of 2000 Hz via an A/D board (National Instruments, Austin, USA) and Templo 7.1 Software (Contemplas GmbH, Kempten, Germany). Data were stored on a computer for further processing. Bipolar, rectangular surface electrode sensors (Bagnoli, Delsys Inc., Boston, USA) with an interelectrode distance of 1 cm were used. In accordance to the SENIAM recommendations [27] the EMG sensors were attached to the subjects skin. One sensor was attached to the clavicular muscle head of the pectoralis muscle (pars clavicularis, PCL), one to the sternal muscle head (pars sternocostalis, PST), one to the anterior deltoid (pars clavicularis, DCL) and one to the latissimus dorsi muscle (LTD) respectively. A 5 $\times$ 5 cm reference electrode was attached to the back of one hand. Before electrode placement the skin was shaved, abraded and cleaned with alcohol to minimise skin resistance. For further sEMG data processing MATLAB R2016a software (MathWorks, Natick, USA) was used. The mechanical delay of the trials was measured by two accelerometers (2-D, 5V signal voltage, 5 g measuring sensitivity), one was attached to the dynamometer and one was placed at the insertion of the pectoral muscle. The delay was determined by comparing the acceleration onset given by the dynamometer to those recorded by the accelerometers. The discrepancy between these data was defined as mechanical delay and therefore subtracted during reflex latency calculation.

2.4 Data analysis

Surface EMG signals were demeaned and bandpass zero-lag filtered (10–450 Hz) with a digital fourth-order Butterworth filter for further analysis. The SLR was manually identified. The onset of the SLR was defined as the first consistent rapid increase in sEMG activity following the trigger stimulus. The second zero crossing of the signal determined the end of the SLR (Fig. 2). The peak-to-peak amplitude was calculated for the selected part of the signal. SLR values were normalized to the peak-to-peak amplitude of the MVC trial, which was selected out of a 1 s plateau. Trials with no obvious stretch reflex response were excluded from further analysis.

Table 2
Descriptive statistics (mean $\pm$ standard deviation) for muscle reflex latencies (in ms), muscle reflex amplitudes (in % of MVC) and muscle activation duration (in ms)

	0% MVC		0% MVC		25% MVC		25% MVC
	50% of iMER		100% of iMER		50% of iMER		100% of iMER
Muscle reflex latencies (ms)
PCL	60.5	$\pm$ 4.4	60.0	$\pm$ 3.8	55.3	$\pm$ 4.5	55.0	$\pm$ 5.5
PST	67.4	$\pm$ 2.7	67.1	$\pm$ 2.3	62.5	$\pm$ 3.9	62.6	$\pm$ 2.9
DCL	62.8	$\pm$ 5.1	62.6	$\pm$ 4.8	57.7	$\pm$ 4.6	57.0	$\pm$ 3.7
Muscle reflex amplitudes (% of T ${}_{\text{MCV}}$ )
PCL	10.7	$\pm$ 5.6	10.6	$\pm$ 5.6	38.3	$\pm$ 10.2	38.7	$\pm$ 8.4
PST	16.5	$\pm$ 6.3	16.0	$\pm$ 6.2	37.6	$\pm$ 10.8	37.2	$\pm$ 10.7
DCL	15.7	$\pm$ 8.2	15.7	$\pm$ 8.3	38.9	$\pm$ 10.2	39.8	$\pm$ 8.6
Muscle activation duration (ms)
PCL	130.88	$\pm$ 32.05	290.34	$\pm$ 79.31	487.74	$\pm$ 79.42	515.44	$\pm$ 65.07
PST	127.34	$\pm$ 27.46	275.58	$\pm$ 77.61	471.53	$\pm$ 82.36	513.79	$\pm$ 55.00
DCL	133.23	$\pm$ 40.38	274.90	$\pm$ 70.71	477.03	$\pm$ 75.79	516.46	$\pm$ 68.77

Figure 2.

Schematic diagram showing the pectoral clavicular head reflexive response (sEMG, upper trace) to an external shoulder perturbation and the dynamometer angle (lower trace) during a 25% MVC 50% of iMER. The mechanical delay was determined via two accelerometers, muscle reflex latency was defined as time from actual perturbation to first muscle response.

Figure 3.

Top: Schematic diagram showing dynamometer angle and stretch velocity in 50% of iMER condition; bottom: Schematic diagram showing dynamometer angle and stretch velocity in 100% of iMER condition.

Figure 4.

Shoulder muscle reflex latencies in all tested conditions and muscles. ${}^{*}$ PST reflex latencies were significantly slower than the PCL and the DCL in all tested conditions. ${}^{**}$ Muscle pre-activation caused significantly faster reflex latencies in all tested muscles.

Figure 5.

Shoulder muscle reflex amplitude in all tested conditions and muscles. ${}^{*}$ Pre-innervation caused significant higher reflex amplitude in all tested muscles.

Figure 6.

Shoulder muscle activation duration following stretch reflex onset. Significant increased activation duration between 50% of iMER and 100% of iMER in 0% MVC condition. Significant longer activation duration in trials with 25% MVC pre-activation compared to 0% MVC trials. No difference between different stretch amplitudes in 25% MVC condition.

2.5 Statistical analysis

All statistical data analyses were carried out on SPSS 25 (IBM Corporation, Armonk, USA). Three within-factor repeated measures ANOVA with Bonferroni post-hoc tests were performed to determine main effects and factor interactions. The within-factors were the level of pre-activation moment (0% MVC, 25% MVC, 50% MVC), the applied stretch amplitude (50% of iMER, 100% of iMER) and the tested muscle (6 muscles). The significance level was set at $p<$ 0.05 for all analyses. If the requirements for the ANOVA were not met by the data a pairwise analysis by Wilcoxon signed-rank test was carried out. To analyse significant discrepancy between pre and post MVC-tests a paired $t$ test was carried out.

3. Results

In all trials the targeted dynamometer acceleration of 10 ${}^{4}$ ${}^{\circ}$ /s ${}^{2}$ and stretch velocity of 150 ${}^{\circ}$ /s was reliably reached with isokinetic phase durations of the movement of 90.3 $\pm$ 0.7% during 50% of iMER trials and 94.8 $\pm$ 0.8% during 100% of iMER trials respectively (Fig. 3). On average the iMER of the participants was 48 $\pm$ 4.38 ${}^{\circ}$ starting from the previously defined 0 ${}^{\circ}$ externally rotation in the 90/90/90-position. Consistent SLRs were not observed in all experimental conditions. For stretches performed with 50% MVC moment pre-activation the SLR could not be determined unequivocally in any muscle. Furthermore, the latissimus dorsi muscle showed no consistent stretch response in all tested conditions. The descriptive statistics for shoulder muscle reflex latencies, reflex amplitudes and the muscle activation duration are displayed in Table 2. Statistical analysis of the reflex latencies revealed a significant muscle main effect (F (2,228) $=$ 99.31, $p<$ 0.001; $\eta^{2}=$ 0.466; $f=$ 0.934) and a significant pre-activation main effect (F (1,228) $=$ 142.21, $p<$ 0.001; $\eta^{2}=$ 0.384; $f=$ 1.418). Comparison among the tested muscles in different state of pre-activation and applied stretch amplitude revealed the sternal pectoral muscle head reflex latency to be significantly slower than the clavicular pectoral muscle head and the anterior deltoid (Fig. 4), furthermore pre-activation led to faster muscle responses in all trials. Analysis of the stretch reflex amplitude yielded a significant pre-activation main effect (F (1,222) $=$ 470.373, $p<$ 0.001; $\eta^{2}=$ 0.679; $f=$ 1.454). Comparison among the tested muscles in different state of pre-activation and applied stretch amplitude presented a significantly stronger muscle response in all muscles to the external rotation stretches in case of muscle pre-activation (Fig. 5). Neither for reflex latencies of the tested muscles nor reflex amplitudes a significant main effect of the stretch amplitude could be determined. The results for the muscle activation duration in response to the stretch showed neither a normal distribution of the residuals nor homogeneity of the variances. Pairwise comparison revealed significantly longer muscle activation in terms of increased stretch amplitudes within trials without muscle pre-activation ( $n=$ 20). Trials with muscle pre-activation showed significant longer muscle activation duration than trials unbiased by stretch amplitudes ( $n=$ 20) (Fig. 6). No differences were observed in MVC moment between the beginning and end of the protocol (pre: 88.68 $\pm$ 15.06 Nm, post: 88.56 $\pm$ 14.83 Nm).

4. Discussion

The results from the current research indicate that muscle pre-activation has a significant effect on the short latency stretch reflex. Contraction levels of 25% of MVC led to a significant quickening of the reflex latencies due to the applied stretches. Significant faster muscle responses during pre-activated test conditions were observed in the clavicular and the sternal head of the pectoral muscle and the anterior head of the deltoid (Fig. 4). In addition pre-activation led to a significant increase of stretch reflex amplitude in all muscles and to an increase of muscle activation duration as an indication of increased excitation of the motoneurons [28]. Variation of the stretch amplitude did not show an effect on reflex latency or amplitude. However, an increase in stretch amplitude influenced muscle activation duration by increasing activation duration significantly. This observation blurred in the presence of muscle pre-activation. Overall the observed reflex latencies of specifically trained overhead athletes tended to be faster than formerly reported latencies [17, 18, 29] for the internal shoulder rotator muscles.

The stretch velocity as an important factor that influences the outcome size of the reflex, since the sEMG amplitude is highly related to the velocity of the stretch [cf. 30], was set to 150 ${}^{\circ}$ /s throughout all test trials. It is reported for the lower limbs that an increase in stretch velocity produces greater reflex response up to a velocity of 300 ${}^{\circ}$ /s in healthy subjects and those with Parkinson’s disease [31]. However, our selected stretch velocity stood in accordance with the results of Zanca et al. [25] who showed that higher velocities seem to be unphysiological in terms of moment generation during external shoulder rotations even in specifically trained overhead athletes. The chosen starting position for the external shoulder rotation stretches mimicked the functional shoulder position during utmost dynamic shoulder movements such as over-head throwing [32]. By the use of a special stabilizing back pad unrestricted mobility of the scapula of the dominant right arm was enabled. Nevertheless, it remains to be noted that this setting yields limitations to subject fixation which potentially allowed evasive trunk movements during the joint perturbation. To control that all stretches were applied with the intended dynamometer acceleration and to adjust the mechanical delay of the measurement two accelerometers were used to allow an accurate determination of the stretch reflex onset.

It is well established that muscle pre-activation decreases the latency of the stretch reflex [cf. 30]. The decrease in reflex latency can be explained by an increase in muscle spindle sensitivity to intramuscular lengthening [15, 33] resulting in a prompt detection of the stretch caused by the anterior perturbation moment and therefore evoking a swifter muscular response. The significantly longer reflex latency of the sternal head of the pectoral muscle seems logical when taking pectoral muscle architecture into account. Despite its relatively small morphological part of total pectoral muscle (approximately 19% of total volume) [34] the clavicular head is the major generator of an internal rotation muscle moment at the shoulder. The pectoral tendon consists of two components one of which is only connected to the clavicular head. The sternal head of the muscle is substantially bigger with a volume of approximately 81% of total muscle volume [34]. Yet it is contracting synergistically during internal shoulder rotation. The normalized reflex amplitude elicited by the stretch allows a rough indication of the proportion of the motoneuronal pool that was excited by the stretch of the muscle. Significant increases in the observed reflex amplitude in 25% MVC pre-activated muscles can be explained by a full activation of the fusimotor system which has been suggested to be reached at the chosen pre-activation level [35]. Furthermore $\alpha$ - $\gamma$ coactivation has been shown to lead to removal of slack from extrafusal and intrafusal muscle fibres which will elevate muscle spindle sensitivity [36]. The increased duration of muscle activation in terms of an enhancement of the applied stretch amplitude may indicate the onset of voluntary muscle contraction since this observation was blurred in pre-activated measurement conditions. In pre-activated muscles the muscle activation duration did not show significant differences between the applied stretch amplitudes indicating that the excitability of the motoneurons was not affected by a continuation of the stretch after reaching the reflex threshold.

In contrast to previous research, in our study a muscle pre-activation level of 50% MVC yielded no consistent stretch reflex in any investigated muscle [18]. We attributed the lack of consistent reflexive muscle responses to the considerably high background activation achieved during the pre-activation phase. Since electromyography measures the degree of muscle activation and the electrophysiological activation of a muscle initiates force generation strong evidence indicates that higher EMG levels correlate with higher muscle force [37]. During isometric contraction the biceps brachii muscle showed a non-linear relationship between measured force and muscle activation. There are also different EMG-force relationships due to differences in muscle architecture reported in the literature [cf. 37]. Furthermore Cronin et al. [38] revealed a negative correlation between higher forces and short-latency stretch reflex behaviour. They showed a decline in fascicle stretch velocity of over 50% between passive conditions and maximal force levels coinciding with progressively smaller muscular stretch response [cf. 38].

The latissimus dorsi muscle showed no response to the applied external shoulder rotation stretches. Given that the subjects were asked to produce an isometric internal rotation force to release the external shoulder perturbation trigger and taking previous findings into account [17, 18] we expected a reflexive response of the muscle. The observed muscle activation patterns during over-head throwing might contribute an explanation for the lack of reflexive muscle response. Jobe et al. [39] and Escamilla and Andrews [8] illustrated that the latissimus muscle is particularly activated during the late cocking phase and the early acceleration phase. In contrast to the starting position in our test setting the shoulder in that case was already located in a state of maximal external rotation which indicates that the reflex threshold for the latissimus dorsi was potentially not reached within our test setting. Furthermore, we cannot exclude a slight stretch evasion of the subjects by arching their backs in the late phase of the external shoulder perturbation which might have had an inhibiting influence on latissimus stretch response.

The relatively limited knowledge about reflex activity at the shoulder results in difficult comparisons among studies. To the best of our knowledge only four publications with comparable test settings to assess reflex latency at the shoulder exist [17, 18, 29, 40]. Wallace et al. [29] tested subjects with and without unstable shoulders. Although they used an accelerometer attached to the dynamometer to determine perturbation onset the results of this investigations are difficult to compare to our findings. The tested subjects showed significantly longer reflex latencies. No pre-activation condition was examined and moreover the authors did not reveal the acceleration which was used to apply the stretches. Latimer et al. [17] assessed muscular reflex response by a rope and pulley system using different weights to simulate an anterior translation force to the humerus. Subjects were placed in a supine position. Further methodological incomparability between this study and our research led to a difficult if not impossible comparability of the results. The results of Myers et al. [18] are similar to those of the current study. The tested muscles showed a significant pre-activation main effect showing the fastest reflexive response in the 20% pre-activation condition. The overall slightly longer latencies can be explained by the non-specifically trained participants examined since Brindle et al. [40] showed that a comparison between healthy subjects and trained intercollegiate pitchers express in faster reflex latencies for the pitchers. However, Brindle et al. [40] assessed reflex with an internal perturbation in contrast to the externally induced acceleration in our test setting.

Although this study suggests some interesting areas for further research, there were some limitations to the study design. One limiting factor was the fixation of the participants. As mentioned before the external shoulder rotations were applied in an upright sitting position by the use of a custom-made back pad to stabilize the non-dominant shoulder of the subjects. This set-up mimics the functional position of over-head throwing but still it cannot be ruled out that evasive trunk movements led to an increase of reflex latency. The increase might be explained by a variation of the limb position and as a consequence an inhibition of the length excitability of the motoneuron pool [cf. 30]. Furthermore, the chosen 90/90/90-starting position resulted in a slight pre-stretch of the internal shoulder rotator muscles of the shoulder joint. The increased muscle length and the pre-activation moment could reduce the impact of muscle slack which might limit the comparison of our outcomes to studies that chose different start settings. The potential influence of various starting positions must be investigated in further research. Finally the steering of the pre-activation moment level was a limiting factor of the study design. An extensive familiarisation was necessary to ensure reproducible pre-activation moment levels throughout the test session. Further research could facilitate this procedure by the use of a visual feedback system.

5. Conclusion

The results of the current research indicate that the chosen measurement setting was capable of eliciting muscular stretch reflexes in the internal shoulder rotator muscles. This observation further enhanced our knowledge of the stretch reflex induction and thus offers a more complex method to analyse neural muscle function for clinical applications such as stroke and spasticity treatment. Although the reflex latencies during pre-activation behaved as hypothesised the question still remains how these observations fit into the model of the ‘kinetic chain’ for high-speed over-head throwing. To get closer to a solution the influence of limb positioning, different stretch velocities and acceleration of the stretches needs to be further investigated to get a better understanding of the role of shoulder muscle stretch and energy storage at the shoulder during specific high-speed tasks such as over-head throwing.

Footnotes

Acknowledgments

The study was not funded and there is no conflict of interest. The authors thank Florian Weiser for assistance in customizing the MATLAB 2016a software for EMG-data processing.

Conflict of interest

None to report.

References

Hirashima

Ohtsuki

. Exploring the mechanism of skilled overarm throwing. Exerc Sport Sci Rev. 2008; 36(4): 205-11.

Hirashima

Yamane

Nakamura

Ohtsuki

. Kinetic chain of overarm throwing in terms of joint rotations revealed by induced acceleration analysis. J Biomech. 2008; 41(13): 2874-83.

Hess

Richardson

Darnell

Friis

Lisle

Myers

. Timing of rotator cuff activation during shoulder external rotation in throwers with and without symptoms of pain. J Orthop Sports Phys Ther. 2005; 35(12): 812-20.

van den Tillaar

Ettema

. A force-velocity relationship and coordination patterns in overarm throwing. J Sports Sci Med. 2004; 3(4): 211-9.

Roach

Venkadesan

Rainbow

Lieberman

. Elastic energy storage in the shoulder and the evolution of high-speed throwing in Homo. Nature. 2013; 498(7455): 483-6.

Roach

Lieberman

. Upper body contributions to power generation during rapid, overhand throwing in humans. J Exp Biol. 2014; 217(Pt 12): 2139-49.

Illyes

Kiss

. Shoulder muscle activity during pushing, pulling, elevation and overhead throw. J Electromyogr Kinesiol. 2005; 15(3): 282-9.

Escamilla

Andrews

. Shoulder muscle recruitment patterns and related biomechanics during upper extremity sports. Sports Med. 2009; 39(7): 569-90.

Wilk

Voight

Keirns

Gambetta

Andrews

Dillman

. Stretch-shortening drills for the upper extremities: theory and clinical application. J Orthop Sports Phys Ther. 1993; 17(5): 225-39.

10.

Komi

. Physiological and biomechanical correlates of muscle function: effects of muscle structure and stretch-shortening cycle on force and speed. Exerc Sport Sci Rev. 1984; 12: 81-121.

11.

Nicol

Avela

Komi

. The stretch-shortening cycle: a model to study naturally occurring neuromuscular fatigue. Sports Med. 2006; 36(11): 977-99.

12.

Diederichsen

Krogsgaard

Voigt

Dyhre-Poulsen

. Shoulder reflexes. J Electromyogr Kinesiol. 2002; 12(3): 183-91.

13.

Guanche

Noble

Solomonow

Wink

. Periarticular neural elements in the shoulder joint. Orthopedics. 1999; 22(6): 615-7.

14.

Gohlke

Janssen

Leidel

Heppelmann

Eulert

. Histopathological findings in the proprioception of the shoulder joint. Orthopade. 1998; 27(8): 510-7.

15.

Shemmell

Krutky

Perreault

. Stretch sensitive reflexes as an adaptive mechanism for maintaining limb stability. Clin Neurophysiol. 2010; 121(10): 1680-9.

16.

Komi

. Stretch-shortening cycle: a powerful model to study normal and fatigued muscle. J Biomech. 2000; 33(10): 1197-206.

17.

Latimer

Tibone

Pink

Mohr

Perry

. Shoulder reaction time and muscle-firing patterns in response to an anterior translation force. J Shoulder Elbow Surg. 1998; 7(6): 610-5.

18.

Myers

Riemann

Y-Y

Hwang

J-H

McMahon

Lephart

. Shoulder Muscle Reflex Latencies Under Various Levels of Muscle Contraction. Clin Orthop Relat Res. 2003; 407: 92-101.

19.

Morriss

Bartlett

. Biomechanical factors critical for performance in the men’s javelin throw. Sports Med. 1996; 21(6): 438-46.

20.

Dillman

Fleisig

Andrews

. Biomechanics of pitching with emphasis upon shoulder kinematics. J Orthop Sports Phys Ther. 1993; 18(2): 402-8.

21.

Komi

. Physiological and Biomechanical Correlates of Muscle Function: Effects of Muscle Structure and Stretch – Shortening Cycle on Force and Speed. Exerc Sport Sci Rev. 1984; 12(1): 81-122.

22.

Atwater

. Biomechanics of overarm throwing movements and of throwing injuries. Exerc Sport Sci Rev. 1979; 7: 43-85.

23.

Rothwell

Day

Berardelli

Marsden

. Habituation and conditioning of the human long latency stretch reflex. Experimental Brain Research. 1986; 63(1): 197-204.

24.

Xenofondos

Patikas

Koceja

Behdad

Bassa

Kellis

, et al. Post-activation potentiation: The neural effects of post-activation depression. Muscle Nerve. 2015; 52(2): 252-9.

25.

Zanca

Oliveira

Saccol

Mattiello-Rosa

. Isokinetic dynamometry applied to shoulder rotators – Velocity limitations in eccentric evaluations. J Sci Med Sport. 2011; 14(6): 541-6.

26.

Finley

Dhaher

Perreault

. Acceleration dependence and task-specific modulation of short- and medium-latency reflexes in the ankle extensors. Physiol Rep. 2013; 1(3): e00051.

27.

Hermens

Freriks

Merletti

Stegeman

Blok

Rau

, et al. European Recommendations for Surface Electromyography. Enschede, the Netherlands: Roessingh Research and Development; 1999. 122.

28.

Levin

Hui-Chan

. Are H and stretch reflexes in hemiparesis reproducible and correlated with spasticity? J Neurol. 1993; 240(2): 63-71.

29.

Wallace

Beard

Gill

Eng

Carr

. Reflex muscle contraction in anterior shoulder instability. J Shoulder Elbow Surg. 1997; 6(2): 150-5.

30.

Voerman

Gregoric

Hermens

. Neurophysiological methods for the assessment of spasticity: the Hoffmann reflex, the tendon reflex, and the stretch reflex. Disabil Rehabil. 2005; 27(1-2): 33-68.

31.

Rothwell

Obeso

Traub

Marsden

. The behaviour of the long-latency stretch reflex in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry. 1983; 46(1): 35-44.

32.

Scoville

Arciero

Taylor

Stoneman

. End range eccentric antagonist/concentric agonist strength ratios: a new perspective in shoulder strength assessment. J Orthop Sports Phys Ther. 1997; 25(3): 203-7.

33.

Myers

Lephart

. The role of the sensorimotor system in the athletic shoulder. J Athl Train. 2000; 35(3): 351-63.

34.

Fung

Wong

Ravichandiran

Agur

Rindlisbacher

Elmaraghy

. Three-dimensional study of pectoralis major muscle and tendon architecture. Clin Anat. 2009; 22(4): 500-8.

35.

Proske

. Kinesthesia: the role of muscle receptors. Muscle Nerve. 2006; 34(5): 545-58.

36.

Ogiso

McBride

Finni

Komi

. Short-latency stretch reflex modulation in response to varying soleus muscle activities. J Electromyogr Kinesiol. 2002; 12(1): 17-26.

37.

Disselhorst-Klug

Schmitz-Rode

Rau

. Surface electromyography and muscle force: limits in sEMG-force relationship and new approaches for applications. Clin Biomech (Bristol, Avon). 2009; 24(3): 225-35.

38.

Cronin

Peltonen

Ishikawa

Komi

Avela

Sinkjaer

, et al. Effects of contraction intensity on muscle fascicle and stretch reflex behavior in the human triceps surae. J Appl Physiol (1985). 2008; 105(1): 226-32.

39.

Jobe

Moynes

Tibone

Perry

. An EMG analysis of the shoulder in pitching. A second report. Am J Sports Med. 1984; 12(3): 218-20.

40.

Brindle

Nyland

Shapiro

Caborn

Stine

. Shoulder proprioception: latent muscle reaction times. Med Sci Sports Exerc. 1999; 31(10): 1394-8.

Evaluation of passively induced shoulder stretch reflex using an isokinetic dynamometer in male overhead athletes

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2.1 Participants

2.2 Experimental protocol

Table 1 Experimental conditions. The order of stretch conditions was randomized

2.4 Data analysis

Table 2 Descriptive statistics (mean ± standard deviation) for muscle reflex latencies (in ms), muscle reflex amplitudes (in % of MVC) and muscle activation duration (in ms)

3. Results

4. Discussion

5. Conclusion

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Experimental conditions. The order of stretch conditions was randomized

Table 2
Descriptive statistics (mean $\pm$ standard deviation) for muscle reflex latencies (in ms), muscle reflex amplitudes (in % of MVC) and muscle activation duration (in ms)