Orthopedics & Biomechanics
949
Dry-land Bilateral Hand-force Production and Swimming Performance in Paralympic Swimmers
Affiliations
Key words ▶ asymmetry ● ▶ force ● ▶ swim-bench ergometer ●
A. A. Dingley1, D. Pyne2, B. Burkett3 1
Performance Research, Australian Institute of Sport, Canberra, Australia Physiology, Australian Institute of Sport, Belconnen, Australia 3 Faculty of Science, Health and Education, University of the Sunshine Coast, Maroochydore DC, Australia 2
Abstract
▼
The effectiveness of human movement is the culmination of several musculoskeletal factors; asymmetry in movement could reduce optimal performance. The aims of this study were to quantify relationships between bilateral handforce production, swimming performance, and the influence of fatigue. Paralympic swimmers (n = 21, aged 20.9 ± 4.7 yr) were categorised into no, high- and low-range physical disability groups and performed two 100 m time trials to measure swimming performance. Bilateral hand-force was measured over two 60 s maximal tests on a swim-bench ergometer to quantify the degree of asymmetry. Large relationships
Introduction
▼ accepted after revision December 17, 2013 Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1364023 Published online: June 3, 2014 Int J Sports Med 2014; 35: 949–953 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0172-4622 Correspondence Prof. Brendan Burkett Faculty of Science Health and Education University of the Sunshine Coast 4558 Maroochydore DC Australia Tel.: + 61/7/5430 2827 Fax: + 61/7/5459 4682 [email protected]
By the nature of their physical impairment Paralympic swimmers generally exhibit an asymmetrical anthropometric profile. The likely causes of asymmetry include bilateral differences in musculoskeletal profile, limitations in flexibility, and muscular imbalances [15]. This scenario can lead to impaired balance or propulsion [10] and subsequently impaired swimming performance [3, 23]. Asymmetry may compromise a swimmer’s technique, and compensatory action could lead to further muscle imbalances and potential injury [15]. Highlighting the need to investigate the relationships between asymmetry and swim force production. Increases in propulsive force generation are generally associated with improvements in swimming performance [17]. Swimming velocity is strongly related to the net forces produced at the hand [11], upper arm [8] and the muscle mass of the arm [4]. A close relationship is evident between net force, arm co-ordination, and technical proficiency [2]. A swim-bench ergometer targets swimming-spe-
between mean force and swimming velocity were seen for both the high- (r = 0.62, ± 0.45; r-value, ± 90% confidence limits) and low-range (r = 0.62, ± 0.50) groups. Asymmetry was related to level of disability, with the smallest difference of 6.7, ± 2.6 N in the no-musculoskeletal disability group. This difference increased to 13.1, ± 10.0 N and 13.5, ± 16.2 N in the high- and lowrange groups. Between the first and last 15 s of the swim-bench test, reductions in mean force were small for the physical disabilities groups. Similarly, changes in asymmetry were small for both the no-physical and low-range groups. Paralympic swimmers with a more severe physical impairment typically generate substantially lower force and velocity.
cific muscle groups and enables direct bilateral measurement of hand force generation without the confounding drag forces associated with inwater swimming [17, 18]. This data quantifies left and right side muscle strength asymmetry and the dynamical asymmetry produced during propulsion [7]. Assessment of a swimmer’s force profile [18] and his or her ability to generate symmetrical muscular power [14] is fundamental for swimming performance. Fatigue has been found to influence the coordination of propulsive arm movements and generated propulsive forces in able-bodied swimmers [1, 19]. The impact of fatigue on the potentially more asymmetrical Paralympic swimmer is yet to be established. The aim of this study was to quantify the magnitude of the relationship between dry-land bilateral hand-force, swimming performance and the influence of fatigue in Paralympic swimmers. We envisaged that swimmers with greater physical disabilities would generate less force, have greater asymmetry and be more prone to the negative effects of fatigue on performance.
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Authors
950 Orthopedics & Biomechanics
Table 1 Breakdown of categories in class, age, height, mass, skinfolds and propulsive measures (mean ± SD).
no-physical disability high-range physical disability low-range physical disability
Class S13 = 1 S14 = 4 S9 = 4 S10 = 4 S2 = 1 S3 = 1 S6 = 2 S8 = 4
Age (yrs)
Height
Body mass
Skinfolds
Peak force
Mean force
(m)
(kg)
(mm)
(N)
(N)
(m.s − 1)
21.1 ± 2.1
1.7 ± 0.1
69.2 ± 7.7
116.0 ± 43.7
57.4 ± 12.9
31.6 ± 5.4
1.5 ± 0.0
20.0 ± 2.6
1.8 ± 0.1
65.9 ± 8.4
68.0 ± 25.9
65.6 ± 23.4
40.6 ± 12.5
1.6 ± 0.2
21.8 ± 7.1
1.6 ± 0.3
54.7 ± 13.2
66.0 ± 17.6
65.9 ± 12.0
36.1 ± 7.9
1.2 ± 0.2
Methods
▼
Subjects A descriptive cross-sectional study examined 21 elite Paralympic swimmers, from 9 different Paralympic swimming classes ▶ Table 1). The swimmers were across a range of disabilities (● stratified using the International Classification System into nophysical disability, i. e., visually impaired (class S13) and intellectual impairment (S14); high-end physical disability (S 9, 10); and low-end physical disability (S2, 3, 6, 8). All swimmers had competed at international competitions (2010–2012) – their physical and demographic characteristics ▶ Table 1. Written informed consent was obtained are shown in ● prior to voluntarily participation in the study. Human Ethics Committees approval (number 20101213) was obtained [5]. Over a period of 24 h, each athlete undertook 2 different testing sessions in the laboratory and pool. To minimise the effects on test performance of residual or cumulative fatigue from training, swimmers and coaches were asked to perform aerobic-only swimming training on the day before the testing period.
Body composition An anthropometric profile was conducted by a Level 3 accredited anthropometrist according to the International Society for the Advancement of Kinanthropometry guidelines. Body mass, stretch stature and skinfolds at 7 sites were recorded. The 7 sites used to calculate the sum of skinfolds were triceps, subscapular, biceps, supraspinale, abdomen, thigh, and calf.
Strength and force assessment A calibrated bilateral swim-bench ergometer (Weba Sport, Wien, Austria) was used to measure the force produced in the left and right arms during simulated freestyle swimming. The swimmer lay prone on the bench with his or her hips attached to the bench using a strap [22]. Adjustments were made to enable swimmers with an upper body disability to hold the paddle during performances. Each swimmer performed 2 × 60 s maximal freestyle efforts followed by a 5 min rest between trials. Each arm stroke was measured independently and the mean and peak force recorded. Typical error of measurement (intra-swimmer reliability) between trials for the strength and force assessment was 2.0 ± 4.9 N for peak force, and 4.7 ± 4.0 N for mean force. The difference in peak force between the left and right arm was used to quantify the degree of asymmetry of the upper body according to the equation [7]: Asym =
Velocity
where Asym is the coefficient of dynamical asymmetry, N is the number of samples in a cycle, XLi is the force generated by the left upper limb (N), and XPi is the force generated by the right upper limb (N). To determine the effect of fatigue on the change in bilateral differences in force production over 60 s of maximal exercise, the difference between the first and last 15 s was calculated.
Swimming specific testing Swimmers performed 2 maximal dive-start 100 m freestyle time trials on a 12 min cycle in a 50 m pool. Trials used electronic timing (Ares 21, Omega) and were recorded using a digital video camera (50 Hz Sony digital – TRV950) positioned on a 3 m high platform at the 25-m mark. Using the custom built software GreenEye Race Analysis (Version 4.8.550), the following measurements were recorded and averaged per 25 m: total time, average velocity, and drop-off in time between first and second lap. The typical error of measurement (intra-swimmer reliability of performance) between the two 100 m freestyle time trials was 0.07 ± 0.02 m/s for velocity and 1.2 ± 1.0 s for total time. The intra-swimmer variability of time trial velocity and performance expressed as a coefficient of variation was 1.2 % and 1.1 %, respectively.
Statistical analysis Traditional statistical methods and magnitude-based inferences (standardised effects) were employed [20, 21]. All numeric values were log-transformed prior to analysis to normalise the data and reduce the homogeneity of error. Measures of centrality and spread are reported as mean ± standard deviation. Precision of estimation was made with 90 % confidence limits. The independent variable was type of disability (no, low-, and highrange), and the dependent variables were swimming velocity, mean and peak force production, and changes in asymmetry. Mean force production and asymmetrical differences are presented in raw units (N), while differences between disability group, and changes during maximal swim-bench testing are both presented as percent effects. Magnitudes of correlation to characterise the degree of association between measures were classified using the following criteria: r = ± 0–0.1 trivial, ± 0.1– 0.3 small, ± 0.3–0.5 moderate, ± 0.5–0.7 large, ± 0.7–0.9 very large, and > ± 0.9 nearly perfect. An effect was deemed unclear if its confidence interval spanned both a substantially positive (0.1) and negative ( − 0.1) threshold value.
1 N ∑ | X Li − X Pi | N i=1
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Category
Orthopedics & Biomechanics Results
951
High-range physical disabilities
Discussion
▼
The aims of this study were to quantify the magnitude of the relationships between dry-land bilateral hand-force and swimming performance, and the influence of fatigue in Paralympic swimmers. The high- and low-range physical disability groups showed large relationships between mean force and swimming velocity. Asymmetry appeared to be related to level of physical
r = 0.62, ± 0.45
60 Mean Force (N)
45 30 15
Low-range physical disabilities 50
r = 0.62, ± 0.50
40 30 20 0.8
1.0
1.2 1.4 Velocity (m.s–1)
1.6
1.8
Fig. 1 Faster Paralympic swimmers are able to produce higher mean forces (r-value ± 90 % CL). Solid line represents regression line. Dashed line represents 90 % confidence interval.
Table 2 Correlations between propulsive measures and dynamical asymmetry (r-value, ± 90 % confidence limits and qualitative descriptor). Category
Peak force
Mean force
Velocity
no-physical disability high-range physical disability low-range physical disability
− 0.60 ± 0.69 large − 0.27 ± 0.60 unclear 0.49 ± 0.53 moderate
− 0.40 ± 0.77 unclear − 0.60 ± 0.47 large − 0.27 ± 0.60 unclear
− 0.55 ± 0.72 unclear − 0.43 ± 0.55 unclear 0.36 ± 0.58 unclear
Table 3 Bilateral differences in force production and dynamical asymmetry (mean, ± 90 %CL; qualitative descriptor). Category no-physical disability high-range physical disability low-range physical disability
Peak force
Mean force
(N)
(N)
Dynamical asymmetry (N)
2.9 ± 10.7 unclear 11.1 ± 12.0 small 14.5 ± 12.0 moderate
1.5 ± 13.2 unclear 1.0 ± 7.6 unclear 11.1 ± 9.3 moderate
6.7 ± 2.6 small 13.1 ± 10.0 moderate 13.5 ± 16.2 moderate
disability, with more severely disabled swimmers exhibiting greater asymmetry. Fatigue induced reductions in 100 m freestyle time trial performance were evident in all groups, as were small reductions in mean force production over the 60 s period during the maximal effort swim-bench test in swimmers with a physical disability. Small increases in asymmetry were associated with fatigue for both the no-physical and low-range groups. Paralympic swimmers exhibiting higher mean hand-force produced substantially faster swimming velocities, consistent with research on able-bodied swimmers [6, 17]. Large relationships
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Mean and standard deviation for height, mass and sum of 7 skin▶ Table 1. The mean sum of 7 skinfolds for folds are presented in ● swimmers with no-physical disability was substantially higher than those with a physical disability (49, ± 44 mm; difference, ± 90 % confidence limits). Higher propulsive measures on dry-land (swim-bench ergometer) were produced and maintained by swimmers in the high-range physical disability group, with the lowest propulsive measures being observed among swimmers in the no-physical disabilities group. When the swimmers were all grouped together (n = 21) a higher mean force production was associated with faster swimming velocities, with a moderate relationship appearing between mean force and swimming velocity (r = 0.53, ± 0.28; r-value, ± 90 % ▶ Fig. 1 shows the magnitude of the relationconfidence limits). ● ship for high- and low-range groups between velocity and mean force. For the no-physical disability group there was an unclear correlation (r = 0.46, ± 0.76) between mean force production and swimming velocity. Relationships between dynamical asymmetries and generation of propulsive forces on the swim-bench ergometer and swim▶ Table 2. Clearly large ming velocity in the pool are shown in ● relationships were only evident for mean force for the highrange physical disability group ( − 0.60, ± 0.47; r-value, ± 90 % confidence limits) and peak force for the no-physical group ( − 0.60, ± 0.69). In contrast, the low-range physical disability group showed a moderate positive correlation between peak force and dynamical asymmetry – that is the swimmers generating more hand force had a larger asymmetry. Bilateral differences in both peak and mean hand force production on the swim-bench ergometer increased in magnitude as the severity ▶ Table 3). This trend was also of physical disability increased (● seen in dynamical asymmetry with the smallest difference occurring in swimmers with no-physical disability (6.7, ± 2.6 N; difference, ± 90 % confidence limits) and largest in the low-range physical disability group (13.5, ± 16.2 N). The drop-off in mean force production and velocity during the 2 ▶ Fig. 2. While there maximal performance tests are shown in ● were decreases for all categories in both measures, the no-physical disability group showed the greatest drop-off in mean force ( − 26.0, ± 15.7 %; difference, ± 90 % confidence limits) and velocity ( − 8.7, ± 1.9 %). There were changes in bilateral differences for force production and dynamical asymmetry over the 60 s maximal swim-bench ▶ Table 4). Unclear changes were observed for ergometer test (● peak force in all groups and mean force for the no-physical disability group. A small increase in bilateral asymmetry was observed in mean force generation for both the high- and lowrange physical disability groups during the 60 s test. Small increases were also evident in dynamical asymmetry for the nophysical and low-range groups.
Mean Force (N)
▼
1.8
20
1.6
10 1.4
0
1.0 15
eL ast
Low -ra
ng
e1
st 1
15
ng
Low -ra
eL ast
st 1
ang h-r
Hig
Hig
h-r
ang
e1
Las t cal ysi
Ph
s
5s
s 15
15 1 st cal No
ysi Ph No
Mean Force (N)
s
–20 5s
1.2
s
–10
Velocity (m.s-1)
Mean Force (N)
30
Velocity (m.s-1)
Fig. 2 The effect of fatigue on mean force production between the first and last 15 s of a 60 s maximal swim-bench test and velocity between the first and second 50 m in a 100 m time trial (difference ± 90 % CL).
Table 4 Change in bilateral difference in force production and dynamical asymmetry within the 60 s swim-bench ergometer test between the 1st 15 s and final 15 s (mean, ± 90 %CL; qualitative descriptor). Category no-physical disability high-range physical disability low-range physical disability
Peak force
Mean force
Dynamical
(N)
(N)
asymmetry (N)
+ 0.7 ± 15.1 unclear + 1.6 ± 11.5 unclear + 1.1 ± 10.7 unclear
− 2.3 ± 10.6 unclear + 3.0 ± 3.1 small + 8.4 ± 5.9 small
+ 5.2 ± 6.9 small − 0.2 ± 5.1 unclear + 5.4 ± 4.4 small
were evident for both the low- and high-range physical disability groups, suggesting that the net (propulsive less drag) force production influences these swimmers’ performance. A moderate but unclear relationship between force and swimming velocity in the no-physical disability group aligns with previous research on able-bodied swimmers [12]. It appears that the nonlinear force-velocity relationship is directly influenced by the swimmers’ level of physical impairment and the more severely physically disabled swimmers would benefit from increasing their force production. This outcome could be attributed to a variety of imbalances. For example, upper limb disabilities cause imbalance related to differences in arm length and upper body strength and power, while lower limb disabilities cause asymmetry related to crossover balance effects. The importance of muscular strength and power generation in the development of swimming velocity appears to be dependent on severity of physical disability. Higher dynamical asymmetry was observed in the physical disabilities group, which also exhibited higher differences between the arms for peak and mean force. One of the explanations for these bilateral differences could be that increased physical disability in subjects with cerebral palsy diminishes synchronization of motor units to both limbs during bilateral movements [7]. While Paralympic swimmers have developed compensatory strategies to cope with the asymmetries that are caused by their disabilities [13, 16], the asymmetry profiles are constantly
undergoing dynamic change [15] and therefore these changes should be monitored to ensure performance improves. This is evident when investigating fatigue, which reduces an individual’s ability to perform maximally. Over the 60 s period in the maximal effort swim-bench test the largest differences in dynamical asymmetry were for the no- and low-range physical disabilities groups. It appears that increases in asymmetry reduced the mean force production and velocity with the largest drop-off observed in the no-physical disability. This was an unexpected finding given that the physical disability group had numerous disabilities (amputees, cerebral palsy, etc.) that presumably affected manoeuvrability. The negative correlations between propulsive measures and dynamical asymmetry imply that increases in dynamical asymmetry would likely reduce propulsion. This finding, in conjunction with the positive relationships between mean force generated on the swim-bench ergometer and swimming velocity, suggests that the swim-bench ergometer should be a useful evaluation tool for scientists and coaches working with Paralympic swimmers. A focus on improved co-ordination and reduced asymmetries would lead to higher propulsive forces and improved swimming velocity. Simulated swimming likely has advantages in measuring isolated hand force production, generated purely from the arms, since in the water other factors such as technique and body position could be confounders. The mechanical limitation is a restricted body-roll. The customised bench we used in this investigation incorporated graduating foam stiffness and flexible supporting stands to compensate for restricted body-roll [9]. Further research into the extent and mechanisms of fatigue on Paralympic swimmers, especially those with cerebral palsy and intellectual disabilities, would be beneficial. This investigation employed a bivariate analysis relating various physiological and musculo-skeletal characteristics with swimming velocity. Future studies are required with Paralympic swimmers that employ multivariate analyses to better define factors contributing to enhancement of performance. The outcomes of this study should be useful for the Paralympic swimming community. We acknowledge the inherent variability and wide range of physical disabilities of the subject group. We also emphasise caution in interpreting outcomes of correlation analysis as associations between measures rather than strictly indicating a causal relationship. However, the ability to measure the upper body bilateral propulsive force production in a relatively large number of elite Paralympic swimmers can inform both pool-based and dry-land training, and physical therapy support.
Conclusion
▼
It appears that faster Paralympic swimmers have developed strategies to compensate for asymmetries caused by their disability and produce substantial forces. The large positive relationship between hand-force production and swimming time-trial performance suggests that simulated (dry-land) swimming could be useful as a monitoring tool in Paralympic swimmers. Fatigue is associated with increased asymmetry reductions in force-generation and swimming velocity. The relationship between asymmetry and fatigue is influenced by the swimmer’s level of physical impairment. Coaches and swimmers should
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
952 Orthopedics & Biomechanics
focus on training programmes that maintain near symmetrical mean force production over the entire race.
Acknowledgements
▼
The authors acknowledge the Australian Paralympic Committee, Swimming Australia, the Australian Institute of Sport, the Sunshine Coast University, and all the participants for their time and patience during the project.
Conflict of interest: The authors have no conflict of interest to declare. References 1 Alberty M, Potdevin F, Dekerle J, Pelayo P, Gorce P, Sidney M. Changes in swimming technique during time to exhaustion at freely chosen and controlled stroke rates. J Sports Sci 2008; 26: 1191–1200 2 Chollet D, Chalies S, Chatard JC. A new index of coordination for the crawl: description and usefulness. Int J Sports Med 2000; 21: 54–59 3 Daly DJ, Djobova SK, Malone LA, Vanlandewijck Y, Steadward RD. Swimming speed patterns and stroking variables in the Paralympic 100-m freestyle. Adapt Phys Activ Q 2003; 20: 260–278 4 dos Santos MAM, Barbosa Junior ML, Castro Melo WVd, da Costa AV, Cunha Costa Md. Estimate of propulsive force in front crawl swimming in young athletes. J Sports Med 2012; 3: 115–120 5 Harriss DJ, Atkinson G. Ethical standards in sports and exercise science research: 2014 update. Int J Sports Med 2013; 34: 1025–1028 6 Hawley JA, Williams MM. Relationship between upper body anaerobic power and freestyle swimming performance. Int J Sports Med 1991; 12: 1–5 7 Jaszczak M. The dynamical asymmetry of the upper extremities during symmetrical exercises. Hum Mov 2008; 9: 116–120 8 Lecrivain G, Slaouti A, Payton C, Kennedy I. Using reverse engineering and computational fluid dynamics to investigate a lower arm amputee swimmer’s performance. J Biomech 2008; 41: 2855–2859
9 Lee J, Mellifont R, Winstanley J, Burkett B. Body roll in simulated freestyle swimming. Int J Sports Med 2008; 29: 569–573 10 Maglischo EW. Swimming Fastest: The essential reference on technique, training and program design. Champaign, Illinois: Human Kinetics, 2003 11 Marinho DA, Barbosa TM, Reis VM, Kjendlie PL, Alves FB, Vilas-Boas JP, Machado L, Silva AJ, Rouboa AI. Swimming propulsion forces are enhanced by a small finger spread. J Appl Biomech 2010; 26: 87–92 12 Miyashita M, Kanehisa H. Dynamic peak torque related to age, sex, and performance. Res Quart 1979; 50: 249–255 13 Osborough CD, Payton CJ, Daly DJ. Relationships between the front crawl stroke parameters of competitive unilateral arm amputee swimmers, with selected anthropometric characteristics. J Appl Biomech 2009; 25: 304–312 14 Potts AD, Charlton JE, Smith HM. Bilateral arm power imbalance in swim bench exercise to exhaustion. J Sports Sci 2002; 20: 975–979 15 Sanders RH, Thow J, Fairweather MM. Asymmetries in swimming: where do they come from? J Swim Res 2011; 18: 1–11 16 Satkunskiene D, Schega L, Kunze K, Birzinyte K, Daly D. Coordination in arm movements during crawl stroke in elite swimmers with a locomotor disability. Hum Mov Sci 2005; 24: 54–65 17 Sharp RL, Troup JP, Costill DL. Relationship between power and sprint freestyle swimming. Med Sci Sports Exerc 1982; 14: 53–56 18 Smith DJ, Norris SR, Hogg JM. Performance evaluation of swimmers. Sports Med 2002; 32: 539–554 19 Soares S, Silva R, Aleixo I, Machado L, Fernandes RJ, Maia J, Vilas-Boas JP. Evaluation of Force Production and Fatigue using an Anaerobic Test Performed by Differently Matured Swimmers. In: Kjendlie PL, Stallman R, Cabri J (eds.). Biomechanics and Medicine in Swimming XI. Oslo, Norway: Norwegian School of Sport Sciences, 2010; 291–293 20 Stang A, Poole C, Kuss O. The ongoing tyranny of statistical significance testing in biomedical research. Eur J Epidemiol 2010; 25: 225–230 21 Stapleton C, Scott MA, Atkinson G. The ‘so what’ factor: statistical versus clinical [corrected] significance. Int J Sports Med 2009; 30: 773–774 22 Swaine IL. Time course of changes in bilateral arm power of swimmers during recovery from injury using a swim bench. Br J Sports Med 1997; 31: 213–216 23 Tourny-Chollet C, Seifert L, Chollet D. Effect of force symmetry on coordination in crawl. Int J Sports Med 2009; 30: 182–187
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Orthopedics & Biomechanics
Copyright of International Journal of Sports Medicine is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
949
Dry-land Bilateral Hand-force Production and Swimming Performance in Paralympic Swimmers
Affiliations
Key words ▶ asymmetry ● ▶ force ● ▶ swim-bench ergometer ●
A. A. Dingley1, D. Pyne2, B. Burkett3 1
Performance Research, Australian Institute of Sport, Canberra, Australia Physiology, Australian Institute of Sport, Belconnen, Australia 3 Faculty of Science, Health and Education, University of the Sunshine Coast, Maroochydore DC, Australia 2
Abstract
▼
The effectiveness of human movement is the culmination of several musculoskeletal factors; asymmetry in movement could reduce optimal performance. The aims of this study were to quantify relationships between bilateral handforce production, swimming performance, and the influence of fatigue. Paralympic swimmers (n = 21, aged 20.9 ± 4.7 yr) were categorised into no, high- and low-range physical disability groups and performed two 100 m time trials to measure swimming performance. Bilateral hand-force was measured over two 60 s maximal tests on a swim-bench ergometer to quantify the degree of asymmetry. Large relationships
Introduction
▼ accepted after revision December 17, 2013 Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1364023 Published online: June 3, 2014 Int J Sports Med 2014; 35: 949–953 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0172-4622 Correspondence Prof. Brendan Burkett Faculty of Science Health and Education University of the Sunshine Coast 4558 Maroochydore DC Australia Tel.: + 61/7/5430 2827 Fax: + 61/7/5459 4682 [email protected]
By the nature of their physical impairment Paralympic swimmers generally exhibit an asymmetrical anthropometric profile. The likely causes of asymmetry include bilateral differences in musculoskeletal profile, limitations in flexibility, and muscular imbalances [15]. This scenario can lead to impaired balance or propulsion [10] and subsequently impaired swimming performance [3, 23]. Asymmetry may compromise a swimmer’s technique, and compensatory action could lead to further muscle imbalances and potential injury [15]. Highlighting the need to investigate the relationships between asymmetry and swim force production. Increases in propulsive force generation are generally associated with improvements in swimming performance [17]. Swimming velocity is strongly related to the net forces produced at the hand [11], upper arm [8] and the muscle mass of the arm [4]. A close relationship is evident between net force, arm co-ordination, and technical proficiency [2]. A swim-bench ergometer targets swimming-spe-
between mean force and swimming velocity were seen for both the high- (r = 0.62, ± 0.45; r-value, ± 90% confidence limits) and low-range (r = 0.62, ± 0.50) groups. Asymmetry was related to level of disability, with the smallest difference of 6.7, ± 2.6 N in the no-musculoskeletal disability group. This difference increased to 13.1, ± 10.0 N and 13.5, ± 16.2 N in the high- and lowrange groups. Between the first and last 15 s of the swim-bench test, reductions in mean force were small for the physical disabilities groups. Similarly, changes in asymmetry were small for both the no-physical and low-range groups. Paralympic swimmers with a more severe physical impairment typically generate substantially lower force and velocity.
cific muscle groups and enables direct bilateral measurement of hand force generation without the confounding drag forces associated with inwater swimming [17, 18]. This data quantifies left and right side muscle strength asymmetry and the dynamical asymmetry produced during propulsion [7]. Assessment of a swimmer’s force profile [18] and his or her ability to generate symmetrical muscular power [14] is fundamental for swimming performance. Fatigue has been found to influence the coordination of propulsive arm movements and generated propulsive forces in able-bodied swimmers [1, 19]. The impact of fatigue on the potentially more asymmetrical Paralympic swimmer is yet to be established. The aim of this study was to quantify the magnitude of the relationship between dry-land bilateral hand-force, swimming performance and the influence of fatigue in Paralympic swimmers. We envisaged that swimmers with greater physical disabilities would generate less force, have greater asymmetry and be more prone to the negative effects of fatigue on performance.
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Authors
950 Orthopedics & Biomechanics
Table 1 Breakdown of categories in class, age, height, mass, skinfolds and propulsive measures (mean ± SD).
no-physical disability high-range physical disability low-range physical disability
Class S13 = 1 S14 = 4 S9 = 4 S10 = 4 S2 = 1 S3 = 1 S6 = 2 S8 = 4
Age (yrs)
Height
Body mass
Skinfolds
Peak force
Mean force
(m)
(kg)
(mm)
(N)
(N)
(m.s − 1)
21.1 ± 2.1
1.7 ± 0.1
69.2 ± 7.7
116.0 ± 43.7
57.4 ± 12.9
31.6 ± 5.4
1.5 ± 0.0
20.0 ± 2.6
1.8 ± 0.1
65.9 ± 8.4
68.0 ± 25.9
65.6 ± 23.4
40.6 ± 12.5
1.6 ± 0.2
21.8 ± 7.1
1.6 ± 0.3
54.7 ± 13.2
66.0 ± 17.6
65.9 ± 12.0
36.1 ± 7.9
1.2 ± 0.2
Methods
▼
Subjects A descriptive cross-sectional study examined 21 elite Paralympic swimmers, from 9 different Paralympic swimming classes ▶ Table 1). The swimmers were across a range of disabilities (● stratified using the International Classification System into nophysical disability, i. e., visually impaired (class S13) and intellectual impairment (S14); high-end physical disability (S 9, 10); and low-end physical disability (S2, 3, 6, 8). All swimmers had competed at international competitions (2010–2012) – their physical and demographic characteristics ▶ Table 1. Written informed consent was obtained are shown in ● prior to voluntarily participation in the study. Human Ethics Committees approval (number 20101213) was obtained [5]. Over a period of 24 h, each athlete undertook 2 different testing sessions in the laboratory and pool. To minimise the effects on test performance of residual or cumulative fatigue from training, swimmers and coaches were asked to perform aerobic-only swimming training on the day before the testing period.
Body composition An anthropometric profile was conducted by a Level 3 accredited anthropometrist according to the International Society for the Advancement of Kinanthropometry guidelines. Body mass, stretch stature and skinfolds at 7 sites were recorded. The 7 sites used to calculate the sum of skinfolds were triceps, subscapular, biceps, supraspinale, abdomen, thigh, and calf.
Strength and force assessment A calibrated bilateral swim-bench ergometer (Weba Sport, Wien, Austria) was used to measure the force produced in the left and right arms during simulated freestyle swimming. The swimmer lay prone on the bench with his or her hips attached to the bench using a strap [22]. Adjustments were made to enable swimmers with an upper body disability to hold the paddle during performances. Each swimmer performed 2 × 60 s maximal freestyle efforts followed by a 5 min rest between trials. Each arm stroke was measured independently and the mean and peak force recorded. Typical error of measurement (intra-swimmer reliability) between trials for the strength and force assessment was 2.0 ± 4.9 N for peak force, and 4.7 ± 4.0 N for mean force. The difference in peak force between the left and right arm was used to quantify the degree of asymmetry of the upper body according to the equation [7]: Asym =
Velocity
where Asym is the coefficient of dynamical asymmetry, N is the number of samples in a cycle, XLi is the force generated by the left upper limb (N), and XPi is the force generated by the right upper limb (N). To determine the effect of fatigue on the change in bilateral differences in force production over 60 s of maximal exercise, the difference between the first and last 15 s was calculated.
Swimming specific testing Swimmers performed 2 maximal dive-start 100 m freestyle time trials on a 12 min cycle in a 50 m pool. Trials used electronic timing (Ares 21, Omega) and were recorded using a digital video camera (50 Hz Sony digital – TRV950) positioned on a 3 m high platform at the 25-m mark. Using the custom built software GreenEye Race Analysis (Version 4.8.550), the following measurements were recorded and averaged per 25 m: total time, average velocity, and drop-off in time between first and second lap. The typical error of measurement (intra-swimmer reliability of performance) between the two 100 m freestyle time trials was 0.07 ± 0.02 m/s for velocity and 1.2 ± 1.0 s for total time. The intra-swimmer variability of time trial velocity and performance expressed as a coefficient of variation was 1.2 % and 1.1 %, respectively.
Statistical analysis Traditional statistical methods and magnitude-based inferences (standardised effects) were employed [20, 21]. All numeric values were log-transformed prior to analysis to normalise the data and reduce the homogeneity of error. Measures of centrality and spread are reported as mean ± standard deviation. Precision of estimation was made with 90 % confidence limits. The independent variable was type of disability (no, low-, and highrange), and the dependent variables were swimming velocity, mean and peak force production, and changes in asymmetry. Mean force production and asymmetrical differences are presented in raw units (N), while differences between disability group, and changes during maximal swim-bench testing are both presented as percent effects. Magnitudes of correlation to characterise the degree of association between measures were classified using the following criteria: r = ± 0–0.1 trivial, ± 0.1– 0.3 small, ± 0.3–0.5 moderate, ± 0.5–0.7 large, ± 0.7–0.9 very large, and > ± 0.9 nearly perfect. An effect was deemed unclear if its confidence interval spanned both a substantially positive (0.1) and negative ( − 0.1) threshold value.
1 N ∑ | X Li − X Pi | N i=1
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Category
Orthopedics & Biomechanics Results
951
High-range physical disabilities
Discussion
▼
The aims of this study were to quantify the magnitude of the relationships between dry-land bilateral hand-force and swimming performance, and the influence of fatigue in Paralympic swimmers. The high- and low-range physical disability groups showed large relationships between mean force and swimming velocity. Asymmetry appeared to be related to level of physical
r = 0.62, ± 0.45
60 Mean Force (N)
45 30 15
Low-range physical disabilities 50
r = 0.62, ± 0.50
40 30 20 0.8
1.0
1.2 1.4 Velocity (m.s–1)
1.6
1.8
Fig. 1 Faster Paralympic swimmers are able to produce higher mean forces (r-value ± 90 % CL). Solid line represents regression line. Dashed line represents 90 % confidence interval.
Table 2 Correlations between propulsive measures and dynamical asymmetry (r-value, ± 90 % confidence limits and qualitative descriptor). Category
Peak force
Mean force
Velocity
no-physical disability high-range physical disability low-range physical disability
− 0.60 ± 0.69 large − 0.27 ± 0.60 unclear 0.49 ± 0.53 moderate
− 0.40 ± 0.77 unclear − 0.60 ± 0.47 large − 0.27 ± 0.60 unclear
− 0.55 ± 0.72 unclear − 0.43 ± 0.55 unclear 0.36 ± 0.58 unclear
Table 3 Bilateral differences in force production and dynamical asymmetry (mean, ± 90 %CL; qualitative descriptor). Category no-physical disability high-range physical disability low-range physical disability
Peak force
Mean force
(N)
(N)
Dynamical asymmetry (N)
2.9 ± 10.7 unclear 11.1 ± 12.0 small 14.5 ± 12.0 moderate
1.5 ± 13.2 unclear 1.0 ± 7.6 unclear 11.1 ± 9.3 moderate
6.7 ± 2.6 small 13.1 ± 10.0 moderate 13.5 ± 16.2 moderate
disability, with more severely disabled swimmers exhibiting greater asymmetry. Fatigue induced reductions in 100 m freestyle time trial performance were evident in all groups, as were small reductions in mean force production over the 60 s period during the maximal effort swim-bench test in swimmers with a physical disability. Small increases in asymmetry were associated with fatigue for both the no-physical and low-range groups. Paralympic swimmers exhibiting higher mean hand-force produced substantially faster swimming velocities, consistent with research on able-bodied swimmers [6, 17]. Large relationships
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Mean and standard deviation for height, mass and sum of 7 skin▶ Table 1. The mean sum of 7 skinfolds for folds are presented in ● swimmers with no-physical disability was substantially higher than those with a physical disability (49, ± 44 mm; difference, ± 90 % confidence limits). Higher propulsive measures on dry-land (swim-bench ergometer) were produced and maintained by swimmers in the high-range physical disability group, with the lowest propulsive measures being observed among swimmers in the no-physical disabilities group. When the swimmers were all grouped together (n = 21) a higher mean force production was associated with faster swimming velocities, with a moderate relationship appearing between mean force and swimming velocity (r = 0.53, ± 0.28; r-value, ± 90 % ▶ Fig. 1 shows the magnitude of the relationconfidence limits). ● ship for high- and low-range groups between velocity and mean force. For the no-physical disability group there was an unclear correlation (r = 0.46, ± 0.76) between mean force production and swimming velocity. Relationships between dynamical asymmetries and generation of propulsive forces on the swim-bench ergometer and swim▶ Table 2. Clearly large ming velocity in the pool are shown in ● relationships were only evident for mean force for the highrange physical disability group ( − 0.60, ± 0.47; r-value, ± 90 % confidence limits) and peak force for the no-physical group ( − 0.60, ± 0.69). In contrast, the low-range physical disability group showed a moderate positive correlation between peak force and dynamical asymmetry – that is the swimmers generating more hand force had a larger asymmetry. Bilateral differences in both peak and mean hand force production on the swim-bench ergometer increased in magnitude as the severity ▶ Table 3). This trend was also of physical disability increased (● seen in dynamical asymmetry with the smallest difference occurring in swimmers with no-physical disability (6.7, ± 2.6 N; difference, ± 90 % confidence limits) and largest in the low-range physical disability group (13.5, ± 16.2 N). The drop-off in mean force production and velocity during the 2 ▶ Fig. 2. While there maximal performance tests are shown in ● were decreases for all categories in both measures, the no-physical disability group showed the greatest drop-off in mean force ( − 26.0, ± 15.7 %; difference, ± 90 % confidence limits) and velocity ( − 8.7, ± 1.9 %). There were changes in bilateral differences for force production and dynamical asymmetry over the 60 s maximal swim-bench ▶ Table 4). Unclear changes were observed for ergometer test (● peak force in all groups and mean force for the no-physical disability group. A small increase in bilateral asymmetry was observed in mean force generation for both the high- and lowrange physical disability groups during the 60 s test. Small increases were also evident in dynamical asymmetry for the nophysical and low-range groups.
Mean Force (N)
▼
1.8
20
1.6
10 1.4
0
1.0 15
eL ast
Low -ra
ng
e1
st 1
15
ng
Low -ra
eL ast
st 1
ang h-r
Hig
Hig
h-r
ang
e1
Las t cal ysi
Ph
s
5s
s 15
15 1 st cal No
ysi Ph No
Mean Force (N)
s
–20 5s
1.2
s
–10
Velocity (m.s-1)
Mean Force (N)
30
Velocity (m.s-1)
Fig. 2 The effect of fatigue on mean force production between the first and last 15 s of a 60 s maximal swim-bench test and velocity between the first and second 50 m in a 100 m time trial (difference ± 90 % CL).
Table 4 Change in bilateral difference in force production and dynamical asymmetry within the 60 s swim-bench ergometer test between the 1st 15 s and final 15 s (mean, ± 90 %CL; qualitative descriptor). Category no-physical disability high-range physical disability low-range physical disability
Peak force
Mean force
Dynamical
(N)
(N)
asymmetry (N)
+ 0.7 ± 15.1 unclear + 1.6 ± 11.5 unclear + 1.1 ± 10.7 unclear
− 2.3 ± 10.6 unclear + 3.0 ± 3.1 small + 8.4 ± 5.9 small
+ 5.2 ± 6.9 small − 0.2 ± 5.1 unclear + 5.4 ± 4.4 small
were evident for both the low- and high-range physical disability groups, suggesting that the net (propulsive less drag) force production influences these swimmers’ performance. A moderate but unclear relationship between force and swimming velocity in the no-physical disability group aligns with previous research on able-bodied swimmers [12]. It appears that the nonlinear force-velocity relationship is directly influenced by the swimmers’ level of physical impairment and the more severely physically disabled swimmers would benefit from increasing their force production. This outcome could be attributed to a variety of imbalances. For example, upper limb disabilities cause imbalance related to differences in arm length and upper body strength and power, while lower limb disabilities cause asymmetry related to crossover balance effects. The importance of muscular strength and power generation in the development of swimming velocity appears to be dependent on severity of physical disability. Higher dynamical asymmetry was observed in the physical disabilities group, which also exhibited higher differences between the arms for peak and mean force. One of the explanations for these bilateral differences could be that increased physical disability in subjects with cerebral palsy diminishes synchronization of motor units to both limbs during bilateral movements [7]. While Paralympic swimmers have developed compensatory strategies to cope with the asymmetries that are caused by their disabilities [13, 16], the asymmetry profiles are constantly
undergoing dynamic change [15] and therefore these changes should be monitored to ensure performance improves. This is evident when investigating fatigue, which reduces an individual’s ability to perform maximally. Over the 60 s period in the maximal effort swim-bench test the largest differences in dynamical asymmetry were for the no- and low-range physical disabilities groups. It appears that increases in asymmetry reduced the mean force production and velocity with the largest drop-off observed in the no-physical disability. This was an unexpected finding given that the physical disability group had numerous disabilities (amputees, cerebral palsy, etc.) that presumably affected manoeuvrability. The negative correlations between propulsive measures and dynamical asymmetry imply that increases in dynamical asymmetry would likely reduce propulsion. This finding, in conjunction with the positive relationships between mean force generated on the swim-bench ergometer and swimming velocity, suggests that the swim-bench ergometer should be a useful evaluation tool for scientists and coaches working with Paralympic swimmers. A focus on improved co-ordination and reduced asymmetries would lead to higher propulsive forces and improved swimming velocity. Simulated swimming likely has advantages in measuring isolated hand force production, generated purely from the arms, since in the water other factors such as technique and body position could be confounders. The mechanical limitation is a restricted body-roll. The customised bench we used in this investigation incorporated graduating foam stiffness and flexible supporting stands to compensate for restricted body-roll [9]. Further research into the extent and mechanisms of fatigue on Paralympic swimmers, especially those with cerebral palsy and intellectual disabilities, would be beneficial. This investigation employed a bivariate analysis relating various physiological and musculo-skeletal characteristics with swimming velocity. Future studies are required with Paralympic swimmers that employ multivariate analyses to better define factors contributing to enhancement of performance. The outcomes of this study should be useful for the Paralympic swimming community. We acknowledge the inherent variability and wide range of physical disabilities of the subject group. We also emphasise caution in interpreting outcomes of correlation analysis as associations between measures rather than strictly indicating a causal relationship. However, the ability to measure the upper body bilateral propulsive force production in a relatively large number of elite Paralympic swimmers can inform both pool-based and dry-land training, and physical therapy support.
Conclusion
▼
It appears that faster Paralympic swimmers have developed strategies to compensate for asymmetries caused by their disability and produce substantial forces. The large positive relationship between hand-force production and swimming time-trial performance suggests that simulated (dry-land) swimming could be useful as a monitoring tool in Paralympic swimmers. Fatigue is associated with increased asymmetry reductions in force-generation and swimming velocity. The relationship between asymmetry and fatigue is influenced by the swimmer’s level of physical impairment. Coaches and swimmers should
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
952 Orthopedics & Biomechanics
focus on training programmes that maintain near symmetrical mean force production over the entire race.
Acknowledgements
▼
The authors acknowledge the Australian Paralympic Committee, Swimming Australia, the Australian Institute of Sport, the Sunshine Coast University, and all the participants for their time and patience during the project.
Conflict of interest: The authors have no conflict of interest to declare. References 1 Alberty M, Potdevin F, Dekerle J, Pelayo P, Gorce P, Sidney M. Changes in swimming technique during time to exhaustion at freely chosen and controlled stroke rates. J Sports Sci 2008; 26: 1191–1200 2 Chollet D, Chalies S, Chatard JC. A new index of coordination for the crawl: description and usefulness. Int J Sports Med 2000; 21: 54–59 3 Daly DJ, Djobova SK, Malone LA, Vanlandewijck Y, Steadward RD. Swimming speed patterns and stroking variables in the Paralympic 100-m freestyle. Adapt Phys Activ Q 2003; 20: 260–278 4 dos Santos MAM, Barbosa Junior ML, Castro Melo WVd, da Costa AV, Cunha Costa Md. Estimate of propulsive force in front crawl swimming in young athletes. J Sports Med 2012; 3: 115–120 5 Harriss DJ, Atkinson G. Ethical standards in sports and exercise science research: 2014 update. Int J Sports Med 2013; 34: 1025–1028 6 Hawley JA, Williams MM. Relationship between upper body anaerobic power and freestyle swimming performance. Int J Sports Med 1991; 12: 1–5 7 Jaszczak M. The dynamical asymmetry of the upper extremities during symmetrical exercises. Hum Mov 2008; 9: 116–120 8 Lecrivain G, Slaouti A, Payton C, Kennedy I. Using reverse engineering and computational fluid dynamics to investigate a lower arm amputee swimmer’s performance. J Biomech 2008; 41: 2855–2859
9 Lee J, Mellifont R, Winstanley J, Burkett B. Body roll in simulated freestyle swimming. Int J Sports Med 2008; 29: 569–573 10 Maglischo EW. Swimming Fastest: The essential reference on technique, training and program design. Champaign, Illinois: Human Kinetics, 2003 11 Marinho DA, Barbosa TM, Reis VM, Kjendlie PL, Alves FB, Vilas-Boas JP, Machado L, Silva AJ, Rouboa AI. Swimming propulsion forces are enhanced by a small finger spread. J Appl Biomech 2010; 26: 87–92 12 Miyashita M, Kanehisa H. Dynamic peak torque related to age, sex, and performance. Res Quart 1979; 50: 249–255 13 Osborough CD, Payton CJ, Daly DJ. Relationships between the front crawl stroke parameters of competitive unilateral arm amputee swimmers, with selected anthropometric characteristics. J Appl Biomech 2009; 25: 304–312 14 Potts AD, Charlton JE, Smith HM. Bilateral arm power imbalance in swim bench exercise to exhaustion. J Sports Sci 2002; 20: 975–979 15 Sanders RH, Thow J, Fairweather MM. Asymmetries in swimming: where do they come from? J Swim Res 2011; 18: 1–11 16 Satkunskiene D, Schega L, Kunze K, Birzinyte K, Daly D. Coordination in arm movements during crawl stroke in elite swimmers with a locomotor disability. Hum Mov Sci 2005; 24: 54–65 17 Sharp RL, Troup JP, Costill DL. Relationship between power and sprint freestyle swimming. Med Sci Sports Exerc 1982; 14: 53–56 18 Smith DJ, Norris SR, Hogg JM. Performance evaluation of swimmers. Sports Med 2002; 32: 539–554 19 Soares S, Silva R, Aleixo I, Machado L, Fernandes RJ, Maia J, Vilas-Boas JP. Evaluation of Force Production and Fatigue using an Anaerobic Test Performed by Differently Matured Swimmers. In: Kjendlie PL, Stallman R, Cabri J (eds.). Biomechanics and Medicine in Swimming XI. Oslo, Norway: Norwegian School of Sport Sciences, 2010; 291–293 20 Stang A, Poole C, Kuss O. The ongoing tyranny of statistical significance testing in biomedical research. Eur J Epidemiol 2010; 25: 225–230 21 Stapleton C, Scott MA, Atkinson G. The ‘so what’ factor: statistical versus clinical [corrected] significance. Int J Sports Med 2009; 30: 773–774 22 Swaine IL. Time course of changes in bilateral arm power of swimmers during recovery from injury using a swim bench. Br J Sports Med 1997; 31: 213–216 23 Tourny-Chollet C, Seifert L, Chollet D. Effect of force symmetry on coordination in crawl. Int J Sports Med 2009; 30: 182–187
Dingley AA et al. Dry-land Bilateral Hand-force Production … Int J Sports Med 2014; 35: 949–953
953
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Orthopedics & Biomechanics
Copyright of International Journal of Sports Medicine is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
