| | Shoulder muscle recruitment patterns during a kayak stroke performed on a paddling ergometerReceived 7 April 2005; received in revised form 16 September 2005; accepted 18 November 2005. Abstract Precise muscle co-ordination is required to maintain normal shoulder function and alterations in synchrony between shoulder muscles can result in loss of full range of movement and pain. Although shoulder pain in kayakers is high with 53% of elite international paddlers reporting shoulder injuries, little information is available regarding the pattern of shoulder muscle recruitment during paddling. The aim of this study was to investigate the normal recruitment pattern of shoulder muscles during the kayak stroke. Nine recreational paddlers without shoulder pain were examined. EMG data from eight shoulder muscles of the dominant arm were collected simultaneously with video data during simulated paddling on an ergometer. EMG data was normalized to time and peak amplitude. Intersubject consistency was evaluated using Pearson correlation analysis. The results of this study indicated a fair to high correlation in at least one phase of the kayak stroke in five of the muscles examined: upper trapezius, supraspinatus, latissimus dorsi, serratus anterior and rhomboid major. This normative data will enable comparisons with the shoulder muscle recruitment patterns in kayakers with shoulder pain in order to determine the role of altered motor control in the painful kayaking shoulder. 1. Introduction  The number of shoulder injuries among paddlers seen by primary care physicians, sport medicine specialists and orthopaedists is high [9]. In the only study investigating the prevalence of shoulder injuries in kayakers, 53% of the elite international athletes surveyed reported shoulder injuries [5]. Mechanical dysfunction at the shoulder is considered to be the major contributing factor to the development of shoulder pain in paddlers [18]. Normal shoulder function predominantly relies on precise muscle co-ordination [16]. Synchronous contraction of the rotator cuff muscles functions to stabilize the glenohumeral joint during movement by maintaining optimal contact between the humeral head and the glenoid fossa [20]. Co-ordinated movement of the scapula with the humerus is necessary to achieve full range of movement at the shoulder while maintaining adequate dynamic stability [1]. Abnormal shoulder muscle function, including asynchrony between shoulder muscles, may contribute to loss of full range of movement and shoulder joint pain [12]. Shoulder muscle activity during kayaking has only been investigated in one small study [22]. Using surface EMG, this study compared the shoulder muscle activity levels between four skilled kayakers and seven subjects who had very limited or no paddling experience. Data was collected during paddling in a kayak chained in a paddling tank. By visually comparing the unprocessed EMG recordings the authors concluded that there were differences between the shoulder muscle activity levels in skilled and unskilled kayakers. To date no data describing the recruitment pattern of shoulder muscles during the kayak stroke is available. The aim of this study was to use EMG to investigate the recruitment pattern of normal shoulder muscles in skilled, recreational paddlers during the kayak stroke. Such normative data would provide a baseline from which to investigate the role of changes in shoulder muscle activation patterns on paddling performance and in the development of shoulder pain in paddlers. 2. Methods 2.1. Subjects Nine recreational kayakers (three females and six males) volunteered to participate in this study. Their average age was 45 years (range 28–62 years). All subjects were right-dominant as determined by the hand with which they wrote. No subjects had pain in their shoulders while paddling and all had been participating in regular kayaking activities for at least three years. Subjects were fully informed of the nature of the experimental work and signed a consent statement. The University of Sydney Human Ethics Committee approved the study. 2.2. Equipment and procedures EMG recordings were made from eight muscles of the dominant shoulder. Muscle activity was recorded from subscapularis, supraspinatus, infraspinatus, serratus anterior, rhomboid major and latissimus dorsi using fine-wire, stainless steel intramuscular electrodes and from middle deltoid and upper trapezius using standard silver/silver chloride surface electrodes (Paediatric Red Dot, 3M). Fine-wire electrode placement for the majority of muscles was based on the position and procedures described by Geiringer [6]. Insertion of fine-wire electrodes into subscapularis was based on the procedure described by Kadaba et al. [8]. The fine-wire electrodes were inserted using hypodermic needles as cannulas. The electrodes consisted of two Teflon insulated 0.14 mm diameter wires with 3mm of the insulation coating stripped from either end. Prior to needle insertion, the skin was cleaned with an alcohol wipe and sprayed with local anesthetic. The two surface electrodes were placed 1cm apart in the middle of the belly of the upper trapezius and middle deltoid. A ground electrode was placed over the contralateral acromion process. The skin under the sites of the surface electrode placement was shaved, sanded with light sand paper and the skin cleaned with alcohol wipes to reduce impedance to a level below 20kΩ as measured using an ohm meter. Following connection of all electrodes subjects were asked to perform sub-maximal contractions of each test muscle to confirm correct electrode placement and to check EMG signal quality. Because of the significant contribution of muscles to shoulder region stability, muscle activity can be expected in many shoulder muscles with any shoulder movement. Therefore, in this study, an acceptable signal to confirm that the wire electrodes were correctly inserted into the target muscle, was defined as high activity during the movement considered to be the action of the target muscle and low activity during the opposite movement. After confirmation of correct electrode placement by the above tests, subjects performed a simulated paddling task on a K1 paddling ergometer (Roger Cahill, ACT, Australia). Following a 5min warm-up period, synchronized video and EMG data were collected for 15s. EMG signals were amplified at 1000×, bandpass filtered between 20Hz and 1000Hz (Isodam 8, World Precision Instruments) and sampled at a rate of 2kHz with an analogue to digital converter (National Instruments, model BNC-2081, Boston) using custom written data acquisition software (BioAcq-Tim Turner, University of Sydney). At the end of each paddling session the EMG electrodes were removed. On removal, each wire electrode was checked to confirm that they were intact and the skin at the insertion site was cleaned and dressed. A video recording of the paddling cycle was made with a tripod-mounted video camera (Pansasonic) in order to identify the beginning and end of each phase of the paddling cycle. The paddle ergometer was fitted with reflective markers on the paddle and a trigger light was attached to the rear of the ergometer. The video recording and EMG recording were synchronized through a pressure switch that activated the trigger light on the ergometer and triggered the computer signal capture program. The trigger was activated by one of the investigators when the paddle on the tested side was in a horizontal position. The trigger signal recorded represented approximately one stroke cycle. Video data was entered into the computer via a video capture card (Matrox TV450, Canada) at 25 frames per second. 2.3. Data analysis Acknowledge (SDR Scientific, Sydney) was used for data analysis. The raw EMG signals were band pass filtered between 20 and 1000Hz for the wire electrodes and between 20 and 500Hz for the surface electrodes. All signals were then rectified and smoothed using a smoothing factor of 300 samples and normalised to peak amplitude per cycle. To investigate the consistency in muscle activity between paddling cycles for each subject Pearson correlation analysis was performed on three typical paddle-stroke cycles for each subject for each muscle. To perform this analysis the cycle time for each of the paddling stroke cycles needed to be equal. Therefore, to account for variations in cycle time, each paddling cycle was normalised to total time per cycle by dividing each cycle by 100 points and averaging a 50ms window at each point. Normalised data were then analysed by correlating each cycle to the average of all three cycles for each muscle for each subject. The Pearson correlation analysis revealed fair to high correlation between each cycle and the average of the three cycles for each muscle for each subject [3]. As a result only the trial with the highest correlation with the average for each subject was used to examine the consistency of the muscle activation patterns between subjects. The kayak stroke was divided into three phases. Phase 1 or pull-through phase was defined as the paddle position of the tested arm while moving from the most forward to the most backward position: phase 2 or exit phase from the most backward paddle position until it reached a horizontal position: and phase 3 or recovery phase from the horizontal position to the most forward excursion of the paddle. In order to determine an average muscle activity pattern across the nine subjects it was necessary to time normalise the EMG data per phase within a stroke cycle, due to variations between subjects in the relative length of each phase within the paddling stroke cycle. To reflect the differences in duration time of each of the phases of the paddling stroke each phase was normalised to a relative number of time increments: phase 1 was divided into 60 (1–60) time increments, phase 2 into 20 (61–80) time increments and phase 3 was into 80 (81–160) time increments. A 50ms window was averaged at each time increment. Following time normalisation the mean EMG trace for all subjects for each phase was calculated for each muscle. In order to determine the consistency of this mean trace the muscle activity for each phase in each subject was compared to the mean trace of all subjects for the same phase using Pearson correlation analysis. Average correlation coefficients across all subjects were then calculated. 3. Results Average correlation coefficients across all subjects between the average EMG activity patterns and individual subjects, for each muscle examined for each phase of the kayak stroke are shown in Table 1. Fair to high correlation was demonstrated in at least one phase of the kayak stroke in five of the muscles examined: upper trapezius, supraspinatus, latissimus dorsi, serratus anterior and rhomboid major (Fig. 1). | | |  | Stroke phase | Upper trapezius | Middle deltoid | Supraspinatus | Serratus anterior | Rhomboid major | Latissimus dorsi | Infraspinatus | Subscapularis | |
|---|
| Pull through | 0.60 | 0.38 | 0.81 | 0.46 | 0.33 | 0.54 | 0.23 | 0.33 | | | Exit | 0.28 | 0.24 | 0.25 | 0.57 | 0.53 | 0.58 | 0.23 | 0.00 | | | Recovery | 0.70 | 0.48 | 0.67 | 0.48 | 0.26 | 0.12 | 0.46 | 0.14 | | | | |
During the pull through phase a consistent pattern of activity was demonstrated in supraspinatus, upper trapezius and latissimus dorsi muscles (Fig. 1). Mean EMG activity in supraspinatus increased in a linear fashion from 20% of the average maximum activity recorded during the paddling cycle to nearly 80% average maximum activity. In contrast, the increase in mean activity in latissimus dorsi from 20% to ≈65% average maximum activity, occurred in the first half of the pull through phase, while a mean increase in upper trapezius of 40% average maximum paddling activity occurred predominantly in the second half of pull through phase. During the exit phase a consistent pattern of activity was demonstrated in latissimus dorsi, rhomboid major and serratus anterior muscles (Fig. 1). All these muscles demonstrated a linear decrease in activity in this short phase of the paddling cycle of between ≈15% and 30% of the average maximum activity recorded during the paddling cycle for each muscle. During the recovery phase a consistent pattern of activity was demonstrated in supraspinatus and upper trapezius muscles (Fig. 1). Both muscles demonstrated an initial, rapid linear decrease in activity of between ≈40% to 50% of average maximum activity, followed by a small increase in activity in mid recovery phase of short duration, before returning to low activity levels of ≈20% average maximum activity to complete this phase of the paddling cycle. 4. Discussion This study identified fair to high consistency in the activity pattern in five of the eight shoulder muscles investigated during some phases of the paddling cycle in recreational kayakers. This information can begin to form a normative database to be used to investigate the influence of alteration in shoulder muscle recruitment on paddling performance and the development of shoulder pain in kayakers. The EMG data in the current study was analysed using Pearson correlation analysis in order to assess the repeatability [3] of the muscle activation patterns across subjects. This method of analysis was seen as preferable to the investigation of relative muscle activity levels for a number of reasons. Firstly, EMG in six of the eight muscles examined was recorded using fine wire, indwelling electrodes to ensure accurate recording from the target muscles, because these muscles are deep to the surface or move considerably with respect to the skin surface during the range of shoulder movement required to perform the kayak stroke. However, because indwelling electrodes only record from a very small area of the target muscle, it cannot be assumed that the activity pattern recorded using this method reflects the activity pattern in the muscle as a whole. Secondly, investigation of shoulder muscle activation patterns reflects the vital role of muscle co-ordination in normal shoulder function [4], [10], [17] Consideration of these consistent recruitment patterns reflects the roles of these muscles as both torque generators and dynamic stabilizers of the shoulder region. During pull through phase of the kayak stroke latissimus dorsi, supraspinatus and upper trapezius demonstrated a consistent recruitment pattern. During this forceful phase of the kayak stroke, the abducted shoulder joint extends and internally rotates against the resistance of the paddle [11]. The increase in latissimus dorsi activity during pull through phase to peak in mid phase, reflects the role of latissimus dorsi as a prime mover for both shoulder extension and internal rotation. This increase in latissimus dorsi activity during the pull through phase of paddling has been reported previously [22]. In addition, an EMG study during normal freestyle swimming demonstrated that latissimus dorsi was highly active during the propulsive phase of the swim stroke [13]. This study concluded that latissimus dorsi was the primary muscle of propulsion in these swimmers. As the pull through phase of the kayak stroke has been likened to the propulsive phase of the freestyle swim stroke, the results of the current study also support these findings. The highly consistent pattern of increasing activity in supraspinatus activity during pull through probably reflects the dynamic stabilising function of supraspinatus during this propulsive phase of the paddling cycle. Supraspinatus contributes to shoulder joint stability by providing a medial force to the humeral head to keep it accurately positioned in the glenoid fossa during all active shoulder movements [14], [15]. As the upper limb extends and medially rotates against significant resistance during pull through, shearing forces generated by the shoulder muscles producing these movements will produce unwanted translation of the humeral head on the glenoid fossa. Supraspinatus must provide a medial force to counteract these shearing forces in order to create a stable fulcrum about which shoulder extension and internal rotation can occur [7], [17], [21]. In addition to extension and internal rotation at the glenohumeral joint, retraction and medially rotation of the scapula occurs during pull through phase, once again against the resistance of the paddle [11]. The increased activity in upper trapezius, an elevator and lateral rotator of the scapula, during this phase would therefore, seem contradictory. However, this consistent increase may be explained when the dynamic stabilizing function of the scapular muscles, to provide a stable base from which other muscles can work, is considered. For example, supraspinatus attaches to the scapula and, as described above, contributes to creating a stable fulcrum at the glenohumeral joint to enable other shoulder muscles to generate forceful movements to propel the boat through the water. The consistent increase in activity in upper trapezius demonstrated in this study may reflect its function to contribute to scapular stability to enable adequate supraspinatus function as a dynamic stabilizer of the glenohumeral joint, during the forceful shoulder movements that occur during the pull through phase of the kayak stroke. During exit phase serratus anterior, rhomboid major and latissimus dorsi demonstrated a consistent recruitment pattern in this study. During this phase, the paddle is taken out of the water as the scapula laterally rotates simultaneously with glenohumeral abduction [11]. The consistent decrease in latissimus dorsi activity, an adductor of the glenohumeral joint, and in rhomboid major activity, a medial rotator of the scapula, is an efficient motor strategy to facilitate shoulder abduction and scapular lateral rotation torque generation. As serratus anterior is a prime mover for lateral rotation of the scapula an increase in serratus anterior activity during exit phase might be expected [2]. In this study, however, serratus anterior activity consistently decreased by ≈20% average maximum activity recorded during the paddling cycle. One explanation of this unexpected finding might reflect the major role this muscle plays in stabilizing the scapula against the chest wall during shoulder region movements [19]. Therefore, during the forceful pull through phase of the kayak stroke, the serratus anterior would be expected to be providing a significant proportion of the force required to provide scapular stability to enable adequate rotator cuff function. As the paddle force decreases and the paddle is lifted from the water at the beginning of exit phase, the force required to stabilize the scapula against the chest wall, and thus serratus anterior activity, greatly decreases. If this decrease in serratus anterior activity is greater than the activity required to contribute to scapula lateral rotation during the exit phase of the paddling stroke, a resultant decrease in serratus anterior activity, as demonstrated in this study, would occur. During recovery phase the upper trapezius and supraspinatus demonstrated a consistent recruitment pattern. During this phase the abducted shoulder moves into flexion to reach the catch position, which is the beginning of pull through phase for the next kayak stroke. At the same time, the opposite shoulder is involved in the forceful pull through phase of the kayak stroke. In this study, where paddling was simulated on an ergometer, the pulley attaching the paddle to the ergometer recoiled simultaneously with the shoulder moving into flexion at the beginning of recovery phase. This paddle recoil, therefore, assisted the movement of the shoulder at the beginning of recovery phase and may have contributed to the initial decrease in activity observed in upper trapezius and supraspinatus muscles in this study. At mid recovery phase, the opposite paddle is in the catch position about to begin pull through phase. In order to effectively perform pull through the paddle is pushed into the water by the recovery arm until it is fully submerged. This forceful activity coincided with a consistent increase in activity in supraspinatus and upper trapezius. Increased activity in suprapinatus might be expected as it performs its dynamic stabilzer role to protect the glenohumeral joint from this potentially destabilizing force [10]. Increased activity in upper trapezius, as well as other scapular muscles, might therefore, be required to provide a stable base to enable effective supraspinatus stabilizer function. In the final stage of recovery phase activity decreased in supraspinatus and upper trapezius. This coincides with the opposite arm beginning pull through phase, and thus providing the majority of the force moving the paddle. In addition, as described previously, recovery phase movements were assisted by ergometer recoil in this study which could contribute to less activity in shoulder muscles than would be expected in the “on water” situation. Several design aspects of this study need to be considered when drawing conclusions from the results presented. As already discussed, this study was conducted on a paddling ergometer that may have influenced muscle recruitment patterns. Further studies will need to be conducted to determine if the consistent muscle recruitment patterns demonstrated in this study truly reflect the “on water” situation. In addition, several factors may have contributed to lack of consistency in recruitment pattern demonstrated in some muscles in this study. Firstly, in order to ensure that activity was being recorded from the muscle of interest and not neighbouring muscles, activity in six of the eight muscles examined was recorded using fine wire, indwelling electrodes. Because indwelling electrodes record activity from a very small area the activity levels recorded with this method may not be representative of the activity in the whole muscle. Secondly, although the subjects in this study were skilled paddlers with at least three years of regular paddling experience, they were not competing at an elite level. Differences in the range of their paddling skills may be reflected in variation in muscle recruitment patterns. Additional testing of a larger number of paddlers of similar skill levels may produce more consistent results. Despite these limitations consistent patterns of shoulder muscle activation during parts of the paddling stroke have been identified in five of the shoulder muscles examined. In addition, the correlation coefficients in two other muscles were approaching the “fair” level in some stages of the kayak stroke (Table 1). These data form the initial stages in the development of a database of normal muscle activation patterns during paddling that could be used to determine the contribution of abnormal muscle activity in the development of shoulder pain in recreational paddlers. Acknowledgements The assistance of Dr. Hoang Tran Dinh and Dr. Bulent Turman in the collection of the EMG data is gratefully acknowledged. References [1]. [1]Bagg S, Forrest W. Electromyographic study of the scapular rotators during arm abduction in the scapular plane. Am J Phys Med. 1986;65(3):111–123. MEDLINE [2]. [2]Bagg S, Forrest W. A biomechanical analysis of shoulder rotation during arm abduction in the scapular plane. Am J Phys Med Rehabil. 1988;67(6):238–245. MEDLINE [3]. [3]Berg KE, Latin RW. Essentials of research methods in health, physical education, and recreation. second ed.. Baltimore: Lippincott Williams & Wilkins; 2003;. [4]. [4]Blasier R, Guldberg R, Rothman E. Anterior shoulder stability: contributions of the rotator cuff forces and the capsular ligaments in a cadaver model. J Shoulder Elbow Surg. 1992;1(3):140–150. Abstract |
Full-Text PDF (1966 KB)
|
CrossRef
[5]. [5]Edwards A. Injuries in kayaking. Sport Health. 1993;11:8–12. [6]. [6]Geiringer S. Anatomic localization for needle electromyography. Philadelphia: Mosby; 1994;. [7]. [7]Inman V, Saunders J, Abbott L. Observations on the function of the shoulder joint. J Bone Jt Surg. 1944;26(1):1–31. [8]. [8]Kadaba M, Cole A, Wooten M, McCann P, Reid M, Mulford G, et al. Intramuscular wire electromyography of the subscapularis. J Orthop Res. 1992;10:394–397. MEDLINE |
CrossRef
[9]. [9]Kizer K. Medical aspects of white water kayaking. Phys Sports Med. 1987;15:128–137. [10]. [10]Kronberg M, Nemeth G, Brostrom L-A. Muscle activity and co-ordination in the normal shoulder. Clin Orthop. 1990;257:76–85. [11]. [11]Logan S, Holt L. The flatwater kayak stroke. NSCA J. 1985;7:4–11. [12]. [12]Nuber GW, Jobe FW, Perry J, Moynes DR, Antonelli D. Fine wire electromyographic analysis of muscles of the shoulder during swimming. Am J Sports Med. 1986;14(1):7–11. MEDLINE |
CrossRef
[13]. [13]Pink M, Perry J, Browne A, Scovazzo ML, Kerrigan J. The normal shoulder during freestyle swimming. An electromyographic and cinematographic analysis of twelve muscles. Am J Sports Med. 1991;19(6):569–576. MEDLINE |
CrossRef
[14]. [14]Poppen NK, Walker PS. Normal and abnormal motion of the shoulder. J Bone Jt Surg. 1976;58(2):195–201. [15]. [15]Saha AK. Dynamic stability of the glenohumeral joint. Acta Orthop Scand. 1971;42:491–505. MEDLINE [16]. [16]Schenkman M, De Cartaya VR. KInesiology of the shoulder complex. J Orthop Sports Phys Ther. 1987;8(9):438–450. [17]. [17]Sharkey N, Marder R, Hansen P. The entire rotator cuff contributes to elevation of the arm. J Orthop Res. 1994;12:699–708. MEDLINE |
CrossRef
[18]. [18]Walsh M. Sports medicine for paddlers. The cause care and treatment of paddlers’ injuries. Canoe. 1989;17:36–38. [19]. [19]Wiater J, Flatow E. Long thoracic nerve injury. Clin Orthop. 1999;368:17–27. [20]. [20]Wuelker N, Korell M, Thren K. Dynamic glenohumeral joint stability. J Shoulder Elbow Surg. 1998;7(1):43–52. MEDLINE |
CrossRef
[21]. [21]Wuelker N, Wirth CJ, Plitz W, Roetman B. A dynamic shoulder model: reliability testing and muscle force study. J Biomech. 1995;28(5):489–499. Abstract |
Full-Text PDF (2150 KB)
|
CrossRef
[22]. [22]Yoshio H, Takagi K, Kumamoto M, Ito M, Ito K, Yamashita N, et al. Electromyographic study of kayak paddling in the paddling tank. Res J Phys Ed. 1974;18(4):191–198.  Bev Trevithick is a physiotherapist who received her Masters in Biomedical Sciences from the University of Sydney in 2004. Bev has worked as a clinical tutor for both undergraduates and postgraduate physiotherapy students for the School of Physiotherapy, Faculty of Health Sciences, The University of Sydney. She is currently working in a private sports physiotherapy practice in Mooloolaba, Queensland, and has a special interest in hand and upper limb biomechanics in athletes.  Dr. Karen Ginn is a Senior Lecturer in Anatomy in the Faculty of Health Sciences at The University of Sydney and a musculoskeletal physiotherapist in part time private practice. She is involved in research related to the treatment of shoulder dysfunction including electromyographic studies investigating shoulder muscle activity patterns, clinical trials to investigate the effectiveness of treatment for shoulder dysfunction and validation of shoulder examination techniques.  Mark Halaki was awarded his PhD from The University of Sydney in the area of human motor control in 2004. He is currently a Post-doctoral fellow at The University of Sydney, Australia. His research interests include motor control and biomechanics of human movement.  Dr. Ron Balnave is an Honorary Associate of the University of Sydney, in the School of Biomedical Sciences, Faculty of Health Sciences. He has researched and published in areas of motor control and muscle strength. He gained his Bachelors Degree and Ph.D. in physiology and pharmacology from the University of New South Wales, Sydney, Australia. Faculty of Health Sciences, School of Biomedical Sciences, University of Sydney, P.O. Box 170, Lidcombe, NSW 1825, Australia  Corresponding author. Tel.: +61 2 9351 9352; fax: +61 2 9351 9715.
PII: S1050-6411(06)00006-X doi:10.1016/j.jelekin.2005.11.012 © 2006 Elsevier Ltd. All rights reserved. | |
|
FULL TEXT00006-X/journalimage?img=PIIS105064110600006X.gr1.lrg.gif&fig=fig1&kwhquery=&issn=1050-6411&ishighres=false&allhighres=false&free=yes','journalimage');)
