Morphological and isokinetic strength differences: bilateral and ipsilateral variation by different sport activity

2.1 Study design

Cross-sectional study design was used in this investigation. The design was explained to all participants. Informed consent was signed and collected from parents before assessments were performed. The study was approved by the ethical committee of the Faculty of Physical Education and Sport, Charles University, Prague, Czech Republic. Measurements were recorded in accordance with the ethical standards of the Declaration of Helsinki and ethical standards in sport and exercise science research [19].

2.2 Subject sample

The research participants were 96 volunteered young male athletes from the highest youth division of their sport specializations and 43 non-active subjects. Participants were grouped by their sport activity, namely, soccer (S, n = 76, age = 14.7 ± 0.3 years), floorball (F, n = 22, age = 14.4 ± 0.5 years), and non-active (N, n = 43, age = 14.6 ± 0.4 years). The basic anthropometric data are presented in Table 1. The average training period for soccer was 7.2 ± 2.9 years, during which, they had 5–6 training sessions per week, each lasting 1.5 hours, and 1 match. The average period of floorball practice was 5.9 ± 3.3 years, comprising 4-5 training session per week of duration 1.5 hours and 1 match. Non-active participants had no organized physical activity, except for physical education (2 hours per week, with each session lasting 45 minutes) in schools.

2.3 Data collection

2.3.1 Anthropometric data, body composition, morphological asymmetry

Body height was measured using a digital stadiometer (SECA 242, Hamburg, Germany) and body mass was measured using a digital scale (SECA 769, Hamburg, Germany). Data identifying body composition were recorded under the same conditions, during morning hours. The participants used no medication, and had not reduced their body weight prior to the measurement, either radically or in the long-term before competitions. To assess body composition, we used a bioelectrical impedance analysis (Tanita MC-980MA, Tanita Corporation, Japan). Standardized conditions for bio-impedance measurement were maintained [20]. The following indicators were observed: body height (BH), body mass (BM), relative fat free mass (FFM%), percentage of fat mass (FM%), and proportion of muscle mass (non-dominant and dominant arm (BAD_Δ) and non-dominant and dominant leg - bilateral differences (BLD_Δ)).

2.3.2 Isokinetic dynamometry

The muscular strength of the lower limbs was assessed using an isokinetic dynamometer (Cybex NORM®, Humac, CA, USA). The maximum peak muscle torque (PT) of the knee extensors (PT_Q) and flexors (PT_H) of the dominant (DL) and non-dominant leg (NL) during concentric contraction were measured at three angular velocities (60, 180, and 300°s^-1). Due to better comparison of isokinetic strength, we expressed strength in relative values (normalized to body mass). Limb dominance was resolved by determining which foot the participant preferred to use for kicking a ball. The test subject sat on an ergonomically set dynamometer seat, with the arm of the dynamometer adjusted according to the instructions and individual somatic characteristics of the participant. The axis of the dynamometer arm rotation was visually adjusted to the axis of knee rotation with a laser point. PT was controlled and modified by gravitational influence at each velocity. The motion range was 90° (maximum extension was marked and set as anatomic zero "0°"). The participant’s trunk and thigh of tested leg were fixed by straps because of the isolation of the examined movement. The participant held the side handles of the device during the measurement. The testing protocol consisted of three attempts at knee flexion and extension at the monitored velocities from the lowest to the highest velocity). Before testing at each velocity, the participants completed four training trials at submaximal intensity. Visual feedback and verbal stimulation were given during the testing.

2.4 Statistical analyses

Data are expressed as the mean ± standard deviation (SD). Parametric procedures were chosen after verification of normality in data distribution using the Shapiro-Wilk test. Multivariate analysis of variances (MANOVA) with Bonferroni post-hoc tests were used for between group comparisons (independent variable) of all observed parameters (depended variables). When the criterion of sphericity as an assumption for data processing, assessed by the Mauchly’s test (χ²), was not met, degrees of freedom were adjusted by means of Greenhouse-Geisser’s (GG) sphericity correction, and the statistical significance was assessed based on the degrees of freedom. To evaluate the equality of error variances, Levene’s test was used. Single data points were identified as an outlier value if found to be out of the interval of mean ± 2*standard deviation. The “Partial eta square” coefficient ( η p 2 ), which explains the proportion of variance of the factor, was used to assess effect size as follows: η p 2 = 0.02 was considered a small effect, η p 2 = 0.13 was considered a medium effect, and η p 2 = 0.26 was considered a large effect [21]. The p-value was set at 0.05 in all statistical analyses. The probability of type II error (beta) was controlled by post-hoc (retrospective) analysis and was set at 0.2 (conventional value). Power analysis was performed using G*Power (version 3.1.9 Kiel University, GER). Statistical analyses were carried out using IBM^® SPSS^® v24 (Statistical Package for Social Sciences, Inc., Chicago, IL, USA, 2012).

3.1 Anthropometric and body composition parameters

The independent variable (Sports activity) had significant effects on BM (F_2,138 = 5.77, p = 0.004, η p 2 = 0.08), BF (F_2,138 = 15.48, p = 0.000, η p 2 = 0.18), and FFM (F_2,138 = 15.48, p = 0.000, η p 2 = 0.18). Bonferroni post hoc analysis revealed significant differences between groups (Table 1).

3.2 Fluid segmental distribution differences between limbs

A significant effect of sports activity on fluid segmental distribution in all the observed parameters was observed (Table 2). Significant morphological differences were detected in both upper (BAD_Δ) and lower extremities (BLD_Δ). Bonferroni post hoc analysis revealed significantly higher values of BAD_Δ in non-active individuals (0.19 ± 0.11 kg) than in soccer players (0.10 ± 0.11 kg). Moreover, we found significantly lower value of BLD_Δ in floorball players (0.32 ± 0.11 kg) than in soccer players (0.58 ± 0.27 kg) and non-active boys (0.63 ± 0.28 kg).

3.3 Isokinetic strength differences of knee extensors and flexors

We found the independent variable (Sports activity) to have significant effects on isokinetic strength of knee extensors in all observed contraction velocities (Table 3). Results revealed significantly higher strength of knee extensors (dominant and non-dominant leg) in soccer players compared with non-active boys and floorball players (except the performance at slowest contraction velocity in non-preferred leg). In contrast, no significant effect (p > 0.05) of sport activity (between groups effect) was found in strength of knee flexors during slowest angular velocity (Table 3). Soccer players reached the highest performance of knee flexors in higher angular velocities (180 and 300°s^-1).

3.4 Bilateral and ipsilateral strength ratio differences

In terms of bilateral strength ratios, we found a significantly higher quadriceps to quadriceps (Q:Q) ratio in soccer players (10.99 ± 7.75%) than in non-active boys (7.47 ± 5.92%). No other significant differences in bilateral strength comparison were found (Table 4). On the other hand, we found a significant effect of sport activity factor on ipsilateral ratio of both legs (dominant, non-dominant) at each velocity (60,180,300 °s^-1). Bonferroni post hoc tests revealed the highest ratio for soccer players, and lowest ratio in non-active boys.

4.1 Anthropometric and body composition parameters

This study revealed the lowest level of body proportion in floorball players (lowest body height and mass) and higher relative fat free mass in soccer and floorball players compared with non-active boys (Table 1). Results indicate greater quality of body composition in favour of sporting population. The amount of fat free mass and its components is determined by genetics (constitutional type), age (on-going ontogenesis), and type of training (energy coverage) that the athletes undergo. Sportsmen with high-quality strength training will thus achieve a higher value of muscle mass than the general adolescent population of similar age [22]. The groups of soccer and floorball players registered the higher quality of body composition compare to non-active boys (Table 1). The effort to eliminate some percentage of fat mass and the preference of a higher proportion of active components is associated with a closer relationship between active mass and success in the sport. Franchini et al. [23] reported that there is a significant relationship between the percentage of fat mass and the level of maximum oxygen consumption by athletes (r = –0.83; p ˂ 0.01). The reported percentage of body fat in non-active boys ranges from 18 – 24.2% [3, 24, 25] and 9.9 –16.0% within active ones [3, 26, 27]. We found a significant difference in the fat mass distribution of active vs. non-active adolescents. The groups of floorball and soccer players had significantly lower fat mass and higher FFM than the non-athletes (p ˂ 0.01). The reason for this difference, as suggested by the higher quality of BC for athletes, could be on-going ontogenesis and a high frequency of training sessions and large volume of loading in team sports such as soccer and floorball. Reduction of excessive fat mass in adolescents through controlled physical activity creates appropriate conditions for a high quality of life among the general population. However, it is important to consider a very important triangle comprising “load – nutrition – regeneration”. Optimal growth and muscle development are only possible when training intensity, nutritional intake, and recuperation are well balanced [28].

4.2 Fluid segmental distribution differences between limbs

By monitoring maladaptive changes based on sport practice, it was assumed that in adolescent athletes, when frequented muscle groups of particular sport activity (upper and lower limbs) are being selected, motor performance is executed unilaterally. When one limb becomes more preferred by neural-motor pattern (leg technique within soccer or arm technique of floorball) there will be bilateral morphological asymmetry in the proportion of muscle mass in upper or lower limbs. This hypothesis was confirmed when the bilateral comparison of the muscle mass proportion between the preferred and non-preferred lower limbs showed a significantly higher proportion of muscle mass (soccer players = 0.58 ± 0.27 kg compared with floorball players = 0.32 ± 0.11 kg). Surprisingly, we found the highest morphological asymmetry (BAD_Δ) in upper extremities in non-active boys in contrast with athlete groups. This could be caused by every day preference of dominant limb during daily activities. We consider the biggest evaluated asymmetry we monitored for non-active adolescents to be the consequence of the low level of physical activity and reason is that sedentary lifestyle has a significant impact on the postural parameters in many young boys and girls [29]. Excessive fat mass and low representation of active muscle body mass in turn may contribute to the development of many health disorders, including disorders of the musculoskeletal system. Moreover, it could be the reason that players from athlete groups perform strengthening exercises, including exercises focused on both upper extremities within the training process which could prevent from these asymmetries.

Floorball players showed higher levels of bilateral morphological asymmetry (BAD_Δ = 0.15 ± 0.08 kg) compared with soccer players (BAD_Δ =0.10 ± 0.11 kg), although the difference was not statistically significant. Evaluated asymmetries may be caused by functional adaptations. In other words, the body successfully adapts to the asymmetrical loading demands of the sport in order to decrease the excessive strain on some tissues and some level of asymmetry may be a natural effect of lateral dominance due to the unilateral nature of the sports discipline.

It is obvious that morphological side–to–side diversification depends on sport specificity [7]. Asymmetry of floorball is caused by non-linear bending, torque movements, and overloading of lumbar spine with shortening of pectoral muscle groups. The absence or insufficient use of compensational exercises leads to disruption of motor patterns and formation of muscle asymmetries, incorrect body posture, shoulder protraction, hyperlordosis, hyperkyphosis, and scoliosis. Studies have shown that muscle asymmetries and overloading of muscle tissue of floorball players can lead to injuries [30, 31].

In soccer, lateral dominance can be defined as footedness: the preferred foot for kicking the ball, and the limb that mobilizes through an open kinetic chain. Asymmetry is a product of unilateral kicking and greater movement repetition of the dominant limb. Participation in asymmetric sport disciplines is connected with asymmetric changes in soft tissues [32]. Asymmetries and shortened muscles may have severe similar physical consequences, increasing the risk of injury and negative affected sport performance [33].

4.3 Isokinetic strength differences of knee extensors and flexors

Soccer players showed the highest values of isokinetic muscle force performance for knee extensors and flexors regardless of monitored contraction speed (Table 3). Floorball and soccer players reached greater muscle force of knee extensors (Preferred lower limb: ~ 2.50 Nmkg^-1, Non-preferred lower limb: ~ 2.37 Nmkg^-1) compared with the non-active group (Preferred lower limb: 2.25 Nmkg^-1, Non-preferred lower limb: ~ 2.20 Nmkg^-1). Holm et al. [34] indicated a greater muscle force in 14 year old soccer players (2.69 ± 0.29 N.m.kg^-1). De Ste Croix et al. [35] reported the muscle force of non-active boys as 2.44 N.m.kg^-1. The greatest muscle force of knee flexors was reported for soccer players (Preferred lower limb: 1.48 ± 0.27 N.m.kg^-1, Non-preferred lower limb: 1.40 ± 0.25 N.m.kg^-1). Holm et al. [34] reported larger values (1.72 ± 0.28 N.m.kg^-1) within the same age group of soccer players. Differences between monitored groups were not statistically significant (Table 3).

The isokinetic strength of extensors and flexors significantly declined with increasing contraction velocity (Table 3). This is in line with other studies [33, 36]. Generally, with increasing velocity of movement in the concentric contraction, the force that muscle is able to exert decreases. This relationship between muscular strength and speed of contraction is known as Hill’s curve. Based on this curve, we expected a greater production of muscle force. Expectations were mostly at high velocity results of active groups (300°s^-1), mainly because of potentially larger neuro-muscular adaptation aimed for substantial production of force in a relatively short time period. (Rate of force development, RFD). In sport, activities requiring a high level of RFD (rebounds, impacts, speed of direction change, acceleration, deceleration, shooting, etc.) occur frequently and can cause an adaptation effect in favour of athletes. The results of our study showed significantly higher muscle strength of soccer player’s knee extensors and flexors in comparison with floorball players and non-active boys.

4.4 Bilateral and ipsilateral strength ratios differences

Bilateral difference in soccer players` knee extensor muscle force (10.99 ± 7.75%) was significantly higher compared with the non-active group (7.47 ± 5.92%). Bilateral difference of floorball players` knee extensors was lower (9.18 ± 8.01%), although not significant in comparison with other groups (Table 4). Greater bilateral strength asymmetries were found in knee flexion (H:H) than in knee extension (Q:Q) in the group of soccer players and non-active boys (Table 4). For adult soccer players, Magalhaes et al. [37] also reported greater level of bilateral muscle asymmetry of knee flexors (HH = 10.6 ± 8.0%) in comparison with extensors (QQ = 7.3 ± 6.5%). The bilateral deficit of knee flexors was greater at higher angular velocities and could have been caused by low neuromuscular adaptation to this type of muscle contraction [12]. Kellis et al. [38] reported significant differences in muscle strength in favour of preferred leg in 13-year old soccer players. This difference was caused by the fact that the preferred leg is “more preferred for passing, shooting”, compared with the non-preferred leg “which is used for support during (standing leg) the kicking phase”. Davies, et al. [39] reported that values differing by 10% to 15% are usually considered to be indicative of significant asymmetry. Higher values of deficit indicate a muscle imbalance, which may predispose the subject to possible muscle-related injuries such as strains and tendinitis. Knapik et al. [40] stated that the athletes with muscle strength imbalances higher than 15% at bilateral comparison of extremities had 2.6-times higher frequency of injuries, compared with athletes where this difference was lower than 15%. With increasing angular velocity as well, bilateral difference of knee flexors increased. This is in line with the other studies [37].

Houweling et al. [41] considered monitoring of the bilateral ratio of the knee flexors strength (H:Q) at velocity 60°s^-1 as the most valid indicator reflecting previous injury of this risk muscle group in soccer. Moreover, the H:Q ratio has been used to examine the similarity between hamstring and quadriceps moment-velocity patterns and to assess the knee functional ability and muscle balance [42]. We found the highest values of hamstring-to-quadriceps ratio for all angular velocities in the soccer players group (Table 4). Soccer players had significantly higher HQ ratio compared with non-active boys (all measurements) and floorball players (in angular velocity 180°s^-1). This contrast can be explained by the advanced knee flexors specific load adaptation that soccer requires. The lowest ratio was observed for non-active boys. This result is presumably a result of insufficient neuromuscular specific adaptation for particular movement pattern. Baratta et al. [43] reported that participants who do not regularly exercise their hamstrings had a significant decrease in hamstring activation compared with normal healthy subjects and athletes with regularly activity. Hoffman et al. [44] considered 6:10 as the accepted value of the H:Q ratio. Kong & Burns [45] reported the range as 42–80% for healthy men and women. In a recent study [46], the authors reported a significant relationship between the ipsilateral isokinetic strength ratio and shooting accuracy of the kicking leg.