Gustavo Leporace, Jomilto Praxedes, Glauber Ribeiro Pereira, Sérgio Medeiros Pinto, Daniel Chagas, Leonardo Metsavaht, Flávio Chame, Luiz Alberto Batista
Currently, decision making is an important task in managing a program of sports training. Among other things coaches should decide what physical exercise should be used to improve athletic performance most efficiently. From this point of view, selecting exercises that simultaneously act positively on more than one performance variable, seems an interesting choice. There is evidence that training programs, including plyometric, balance and lumbar-pelvic stability exercises, can contribute to reduce the incidence of ACL injuries in female athletes (Heidt, Sweeterman, Carlonas, Traub, & Tekulve, 2000). Biomechanically, it is believed that these exercises propitiate changes in the kinematic and kinetic lower-limb behaviours related to the mechanism of this type of injury, such as dynamic knee valgus displacement, maximum knee flexion and peak ground reaction forces (Hewett, Stroupe, Nance, & Noyes, 1996; Irmischer et al., 2004; Myer, Ford, Brent, & Hewett, 2006; Myer, Ford, McLean, & Hewett, 2006; Myer, Ford, Palumbo, & Hewett, 2005). Moreover, there is evidence that such training programs can also result in direct improvement of athletic performance variables, mainly those related to vertical jump height (VJH), power and agility. Therefore, it is reasonable to assume that such programs can be used for the purpose of improving the overall performance of athletes (Di Stefano et al., 2010; Luebbers et al., 2003; Myer et al., 2005; Myklebust, Maehlum, Holm, & Bahr, 1998; Newton, Kraemer, & Hakkinen, 1999; Villareal, Gonzalez-Badillo, & Izquierdo, 2008). Nevertheless, although such information might be important for the development of training strategies, its practical use and great widespread requires the overcoming of limitations not addressed in previous studies, particularly with regard to the preventive practice. One of the limitations previously mentioned is the epidemiological indicators associated with the gender of the subjects investigated. In this case, females tend to be more studied, probably because they are more susceptible to ACL injuries, although the prevalence is higher in the male population, likely because more men participate in sports (Renstrom et al., 2008). It seems that ACL injuries result from multi-planar movements and it has been argued that both the risk factors and the injuryinducing mechanisms differ between genders (Hewett, Myer, & Ford, 2005; Krosshaug, Slauterbeck, Engebretsen, & Bahr, 2007; Zazulak, Hewett, Reeves, Goldberg, & Cholewicki, 2007). In women, the injury is primarily associated with dislocations and mechanical load in the sagittal, frontal and transverse planes (Hewett et al., 2005), whereas injuries in men seem to be primarily related to movements and loads in the sagittal plane (Quatman & Hewett, 2009). Specifically, kinematic risk factors in females are supposed to be related to knee valgus and flexion, associated with tibial rotation and hip adduction (Hewett et al., 2005). Otherwise, the risk factors in male athletes seem to be related mainly to decreased knee flexion (Quatman & Hewett, 2009). Thus, although there are studies demonstrating that males can reduce their injury rates with an injury prevention program (Caraffa, Cerulli, Projetti, Aisa, & Rizzo, 1996; Junge, Rosch, Peterson, Graf-Baumann, & Dvorak, 2002), the differences between sex-related mechanisms of injury lead us to suppose even that certain injury prevention exercises are effective in female populations, there is no guarantee that they will offer the same benefit to similar male populations. Another limitation of previous studies is related to the motor tasks used in the experiments. Researchers tend to use double-leg landings (Irmischer et al., 2004; Myer et al., 2005) to evaluate injury risk, which may also limit the generalisation of results because in some sports, such as volleyball, single-leg landings are common after a jump (Tillman, Hass, Brunt, & Bennett, 2004). In addition, approximately 25% of ACL injuries occur after single-leg landings (Krosshaug et al., 2007). According to DiStefano et al. (2010) the specificity of training should be considered carefully to improve both biomechanics and performance. However, the effect of a preventive training program on the kinematic of different motor tasks is not known. Finally, despite evidence of effectiveness of plyometric training to improve sports performance by increasing VJH, power and agility (Meylan & Malatesta, 2009; Newton et al., 1999; Villareal et al., 2008), as well as the potential effectiveness of PTPs in reducing the risk factors for and incidence of ACL injuries (Irmischer et al., 2004; Myer, Ford, Brent et al., 2006, Myer, Ford, McLean et al., 2006, 2005; Grindstaff, Hamiill, Tuzson, & Hertel, 2006), the degree of influence of PTPs on specific performance variables, such as those listed above, remains unproven. To our knowledge, only one study, by Myer et al. (2005), has evaluated the influence of PTPs in improving sports performance as well as reducing risk factors for ACL injury. The authors demonstrated the efficacy of neuromuscular training in increasing athletic performance and reducing ACL injury risk in female athletes. Kilding, Tunstall, and Kuzmic (2008) and DiStefano et al. (2010) also demonstrated the influence of preventive programs on athletic performance, but they did not examine biomechanical aspects. DiStefano et al. (2010) suggested that future studies should examine performance and lower-limb biomechanical behaviours to identify the effects of PTPs. Therefore, the aim of this study was to examine the effects of a neuromuscular training program on lower-limb kinematics during single-leg and double-leg landings and vertical jump height. The experimental hypotheses were that: (i) the training-related changes would be specific to each landing; (ii) the training program would improve vertical jump height; and (iii) the training program would induce kinematic changes, such as increased knee and hip range of motion.
Fifteen male volleyball athletes from a regional team (age: 13 0.7 years, height: 1.70 0.12 m and body mass 60 12 kg) with no history of lower-limb joint injuries participated in this study. At the time of the intervention, all the athletes had at least three years’ experience in the sport and all of them were used to playing in regional and national competitions. A parent/guardian signed informed consent for each athlete and authorised his participation in the study, which was approved by the State University of Rio de Janeiro Ethics Committee. In the week before starting and in the week after finishing training, the subjects underwent tests to examine lower-limb kinematics during landing after vertical jumps and the maximum height reached in these tasks. The training was performed in the beginning of the season. Two vertical jumps were used to induce the targeted kinematic behaviour. For each sequence, the athletes performed the propulsive phase of jumping twice with both legs, but in one they landed with the dominant leg (SL) and in the other they landed with both legs (DL) (Fig. 1A and B). The dominant leg was defined as the leg the participant would use to kick a ball as far as possible (Myer, Ford, Brent et al., 2006, 2005). Initially, the athletes familiarised themselves with the tasks to reduce the influence of a learning effect on the biomechanical variables. At least 5 min after the familiarisation session, three SL and three DL were filmed in the sagittal plane and stored in a computer for later analysis. A trial was considered successful if the athlete could land without losing balance. No athlete had more than two unsuccessful trials. The landing tasks were performed in a random sequence to minimise the possible effects of fatigue, beginning with SL or DL. A one-minute rest interval was allowed between attempts. The subjects performed all tests and training with the shoes they used to play regularly. The only
instruction provided to the athletes was to jump as high as possible and land in the specific condition (double leg or single leg). The execution of landing tasks was filmed with a camera (Sony DCR HC 46) positioned in the sagittal plane 2 m away from the place of execution, with the optical axis projected onto the centre of the capture area perpendicular to the vertical and horizontal orientations and a sample frequency of 30 Hz, which allowed the interlaced processing of 60 frames per second. Six spherical, 20-mm reflective markers were positioned at each athlete’s iliac crest, greater trocanter, lateral condyle of the femur, lateral malleolus, lateral calcaneus and fifth metatarsal to allow us to examine the angular behaviour of the hip and knee in the sagittal plane (Fig. 2). The metatarsal point was used only to generate the corporal model. To assure marker placement reliability, the same researcher, with great experience in palpatory anatomy, positioned all markers in all subjects. A static trial was collected A static trial was collected by asking the subject to stand still while he was aligned with the laboratory (global) coordinate system. This measurement was used to define each subject’s neutral (zero) alignment, with subsequent dynamic kinematic measures quantified relative to this position. To calibrate the images, four non-collinear points were positioned at the vertices of a 50-cm-sided cube positioned parallel to the capture plane and located in the area where the landing tasks were executed (Robertson & Caldwell, 2004). The data for all subjects were captured in the same environment. Therefore, the devices were not relocated between data collection events. After capture, the images were transferred to a personal computer. The raw coordinates of the markers were transformed into global coordinates (Abdel-Aziz & Karara, 1971, p.1) and processed through the Skill Spector software (Geeware, Version 1.2.4, USA), according to the protocol validated by McLean et al. (2005). They found a high correlation between a single camera 2D and a 3D movement analysis system for inter-subject difference for coronal plane knee kinematics (r: 0.76e0.80). The between day intra-tester reliability of this systemwas measured by Button, van Deursen, and Price (2008). They found an ICC ranging from 0.75 to 0.87 for hip, knee and ankle sagittal plane kinematics. We used a low-pass, fourth order Butterworth filter in the forward and reverse directions to prevent phase distortions, with a cut-off frequency of 6 Hz to smooth the kinematic signal.
To determine the instant of ground contact, we used a custommade footswitch transducer (FootPress LaBiCoM) set in the sole of the subjects’ shoes in the region of the first metatarsal. When the electrical circuit of the footswitch was triggered, a light-emitting diode (LED) positioned in the camera’s capture field was activated, indicating the presence or absence of ground contact. The signal produced by the footswitch was also used to calculate the time of the flight phase. This variable was used to estimate the maximum vertical height according to the formula 0.5*g*(t/2)2, where g is the acceleration of gravity (9.81 m/s2), and t is the flight phase duration, measured between the instants of loss and resumption of ground contact, according to the strategy validated by Leard et al. (2007). The largest vertical displacement achieved by athletes in three trials was considered the maximum height, as Moir, Shastri, and Connaboy (2008) have demonstrated that this strategy has an excellent reliability level for males (ICC > 0.9) and should be used instead of the arithmetic mean. MATLAB version 6.5 (The Mathworks, USA) was used for signal processing.
2.1. Preventive training program
The frequency of training was three times/week for six weeks, 45e60 min per session. Participants were monitored during training by a team of four or five physical education coaches and students trained to correct execution errors and to facilitate performance improvements, which provided an average of one coach for every three athletes, allowing effective and individualised control of the training process. The training program was implemented in the final half of the competitive season. The PTP used in this study consisted of a compilation of strategies used in previous studies (DiStefano et al., 2010; Grindstaff et al., 2006; Heidt et al., 2000; Hewett, Lindenfeld, Riccobene, & Noyes, 1999; Hewett et al., 1996; Irmischer et al., 2004; Meylan et al., 2009; Myer, Ford, Brent et al., 2006, 2005; Myer, Ford, McLean et al., 2006; Paterno, Myer, Ford, & Hewett, 2004; Pollard, Sigward, Ota, Langford, & Powers, 2006). The exercises aimed to allowa higher degree of specificity between training and volley all performance. Plyometric, balance and core stability exercises were used (Appendix 1 and 2). After each PTP session, the athletes performed their technical training routine. The PTP was divided into three phases according to Hewett et al. (1996): 1) The technique phase, focused mainly on basic aspects such as correct posture, body alignment throughout the jump, soft landings and instant recoil preparation for the subsequent jump; 2) The fundamentals phase, focused on proper technique to increase power and ability and 3) The performance phase, focused on achieving maximal vertical jump height through improved technique. The training programwas modified every two weeks to increase the difficulty of the exercises because the Principle of Overload indicates that a two-week period is long enough to allowathletes to assimilate the difficulties of previous exercises (Hewett et al.,1996). The degree of difficulty was increased through the use of single leg exercises, increased repetitions and intensity and the use of unstable surfaces with eyes open and closed during specific volleyball techniques (dual task). In each session, approximately eight or nine plyometric exercises, four or five core stability exercises and four or five balance exercises were conducted. A 2-min rest interval was allowed between exercises.
2.2. Statistical analysis
The vertical jump height and angular and temporal lower-limb kinematics related to two landing tasks were compared during bilateral vertical jumps (single-leg landings (SL) and double-leg landings (DL)) (Fig. 1A and B) between before and after the application of a preventive training program (PTP). The influence of the PTP was measured in the following variables: vertical jump height (VJH); angular position of the hip and knee upon ground contact after the flight phase (IAPH and IAPK, respectively); angular position of the hip and knee measured when the body’s centre of gravity reached its lowest position of vertical displacement (MAPH and MAPK, respectively); hip and knee range of motion (HRoM and KRoM) from the initial contact to the maximum angle of each joint; and the time of landing (TL), designated as the time, in seconds, from the initial contact to the lowest position of vertical displacement. These biomechanical variables were selected on the basis of studies that proposed a relationship between the behaviour of these variables and ACL injury risk
factors in men (Quatman & Hewett, 2009; Renstrom et al., 2008) and on the relationship between VJH and performance in volleyball players (Barnes et al., 2007). The reliability of all dependent measures among three attempts for each landing task before and after training was determined using the intraclass correlation coefficient (ICC2,1) and standard error of measurement (SEM) (Weir, 2005). To examine the influence of the training programme on the tested variables, the pre-test and post-test values were compared using the Wilcoxon ranked test, with a significance level of 5%. We used a non-parametric test due to the limited sample size and in consideration of the KolmogoroveSmirnov test results, which suggested a non- Gaussian distribution. The statistical analysis was performed using GraphPad Prism, Version 5.00 for Windows (GraphPad Software, San Diego, California, USA).
All individuals completed at least 80% of the training sessions and performed the tests before and after training, with a mean of 89% (16 from 18 total sessions) of compliance. No athlete was injured or had musculoskeletal pain during or after the training period.
3.1. Reliability analysis
The ICC2,1 of the dependent variables achieved in three attempts at each landing before and after training are shown in Table 1. Excellent reliability values (>0.8) were found for all the situations in both landings.
3.2. Vertical jump height
No statistically significant differences were found between the pre- and post-training in relation to the VJH values with SL or DL (Tables 2 and 3). The means, standard errors, confidence intervals and p values of SL and DL, the dependent variables of this study, are presented in Tables 2 and 3, respectively, and described below.
3.3. Single-leg landing
There were no significant differences between training periods for IAPH, IAPK, MAPH, MAPK, HRoM and KRoM. However, after training, athletes showed significantly longer TLs compared to the pre-training period.
3.4. Double-leg landing
There were no significant differences between training periods for IAPH, IAPK, MAPH, MAPK, HRoM and TL. After training, the athletes showed a statistically significant increase in KRoM.
Improving the overall athletic performance involves, among others things, acting on several variables specific to each situation, seeking that athletes achieve better performance with a low injury risk. In this context, an important goal of preventive exercise is to develop the ability to perform movements with less aggressive mechanical characteristics. In this study, male volleyball athletes participated in a PTP, and their subsequent performance was examined to determine the PTP’s influence on athletic performance and lower-limb kinematics in two landing tasks with different constraints. To our knowledge, this is the first study to examine the influence of preventive training on both the athletic ability and lower-limb kinematics of young male athletes during different types of landings. Other investigations that have addressed this type of training focused on its influence on women (Myer et al., 2005) or on athletic performance alone in young people of both sexes or just males (DiStefano et al., 2010; Irmischer et al., 2004). As has been already described, the risk factors and mechanisms of injury, although not proved by a prospective study, seem to be different between genders, therefore, the inference of the results obtained from female to male athletes should be done carefully. The results obtained in a previous investigation with the same population (Leporace et al., 2010) showed that the valgus angle of the examined group during landings tasks are regarded as being at low risk for ACL injuries (Hewett, Myer, & Ford, 2004; Schmitz, Kulas, Perrin, Riemann, & Shultz, 2007; Swartz, Decoster, Russell, & Croce, 2005; Yu et al., 2005). Although the knee flexion could be considered low, the adequate alignment in frontal plane may be related to low risk for ACL injury. Myer, Ford, Brent, and Hewett (2007) showed that youths with low susceptibility to this injury are less sensitive to preventive training and have a tendency to make minor biomechanical adaptations to exerciseinduced stimuli. Thus, it was generally expected the subjects in the present study to have a low response to training, which does not imply that this training program does not generate an increase in sports performance or that the results would be similar for both types of landings, which in fact could be confirmed in this study. It was verified for SL that, despite the increase in the angular range of motion of the hip and knee, the training effect on the kinematicswas minimal. This low expression for the angular lowerlimb kinematics during unilateral landing may be partially explained by the characteristics of the chosen exercises of the PTP (Tables 1 and 2). The programs have higher number of bilateral plyometric exercises than unilateral, and the stability exercises required no deep knee flexion and no pronounced anterior trunk displacement. Paterno et al. (2004) used plyometric and stability exercises with backward-forward displacement for 6 weeks, which resulted in improved control of body centre of gravity (CoG) displacement in the anterioreposterior direction, although no improvement in medialelateral control occurred. Louw, Grimmer, and Vaughan (2006), who used a training protocol consisting of unilateral plyometric exercises in a population similar to that of the present study, found significant changes in knee-joint displacement after training. This set of findings highlights the importance of respecting the Principal of Specificity when selecting an exercise training program. It is possible that the athletes presented a greater landing duration after training to compensate for their inability to stabilise the knee-joint in deep flexions during the SL. In general, this strategy may have been adopted to reduce the mechanical loads around a joint, given that with more time, the mechanical impulse generated is more likely to be distributed with lower peak forces, leading to greater absorption of the energy generated by a task.We suggest that exercise programs with longer durations and more specific exercises be conducted to evaluate the kinematic and kinetic adaptations in this population. Regarding the maximum flexion of the hip in DL, although the p value did not reach significance, there was a strong trend (p ¼ 0.0645) and a large effect size (Cohen’s d ¼ 0.83). The mean difference was approximately 10, which resulted in a joint ROM increase of approximately 12 in the post-training (Cohen’s d ¼ 0.78), mainly due to anterior trunk inclination. This kinematic behaviour suggests the CoG moving to coordinates in which it remains horizontally aligned with the axis of knee motion. This indicates decreasing extensor torque in this joint, which also reduces the ACL strain (Blackburn & Padua, 2008). This result may be related to the individual’s increased ability to control the CoG projection over the support basis, which is typically obtained as an effect of neuromuscular training programs (Paterno et al., 2004). The knee movement, regarded as a hip angular behaviour-associated movement, showed a significant difference (p ¼ 0.0371, Cohen’s d ¼ 0.80) of approximately 6 in the flexion displacement, which, according to Blackburn and Padua (2008), corroborates the ACL tension reduction. However, this alteration was probably due to the decrease in the knee flexion at initial contact, which may actually increase injury risk (Hewett et al., 2005; Quatman & Hewett, 2009). Future studies are needed to discuss the effect of the paradigm related to the increase of knee displacement as a result of decreased knee flexion at initial contact. Although other studies have suggested that neuromuscular training induces changes in athletic performance (Bobbert, 1990; Kilding et al., 2008; Myer et al., 2005; Villareal et al., 2008), we found no statistically significant difference when comparing pre- and post-training, either for SL or DL. It is possible that 6 weeks of neuromuscular training is insufficient to produce changes in these sports performance aspects in young male athletes. However, previous studies have reported that increases of approximately 10% in jump height with countermovement are associated with improvements in sports performance (Bobert, 1990; Markovic, 2007; Villareal et al., 2008). In this study, based on the evaluation of differences in preand post-training absolute values and on the medium effect size (Cohen’s d ¼ 0.52), we observed that improvements, although not statistically significant, in both landings types were compatible and, in some cases, even higher than the values that were associated in the literature with a positive impact on athletic performance. Villareal et al. (2008) demonstrated in a meta-analysis that plyometric training programs generally generate increases of approximately 7%, corresponding to 3.9 cm in jump height after about 10 weeks. Regarding the preventive programs, Myer et al. (2005) showed improvements of about 3.3 cm (8.3%) after six weeks of training; Kilding et al. (2008) reported improvements of 2 cm (6%) and Di Stefano et al. (2010) reported 1.7-cm improvements (6.9%) after nine weeks of training. The present study found differences of 2.7 cm (10%) and 3.5 cm (11.3%) for the vertical jumps with SL and DL landings, respectively. Despite the absence of statistical differences in the pre- and post-training values, the training program employed in this study increased the athletic performance of young male volleyball athletes in a way similar to that reported in the literature after a training period of only 6 weeks. Moreover, as described earlier the PTP was implemented in the final half of the season. The focus of the training at this stage was on tactical aspects of volleyball while technical aspects related to performance variables were already finished. This emphasizes the importance of implementation of PTP despite the cycle of periodization of the training seeking also performance improvement. The duration of neuromuscular training seems to be a decisive factor, given the significant changes in the young athletes’ performance. Thus, further studies using longer training periods are encouraged to determine the course of the changes in athletic performance over time and establish the minimum period necessary to achieve changes in biomechanical behaviours associated with performance improvements at different stages of training. Unlike other explanations presented in the literature, it can be suggested that the lack of statistical differences for performance in the vertical jump height could be associated with the research design. The majority of previous studies used pre-post-test differences to compare the training effect between intervention and control groups (DiStefano et al., 2010; Kilding et al., 2008; Myer et al., 2005). In contrast, this study adopted the strategy of matching samples. It is suggested the implementation of randomized controlled trials to test the hypothesis of an increase of vertical jump height after the performance of PTPs. Another interesting finding of this study is related to the reliability of the jump height in the three attempts with both landing conditions. Although the findings for the DL jump support the literature (Moir et al., 2008), no studies have examined the reliability of DL vertical jumps with successive SL landing, which is common in volleyball practice (Tillman et al., 2004). In this sense, our results suggest that both landing forms started from jumps of bilateral propulsion can be employed in training programs to evaluate maximal jump height, due to the high reliability value found. Although much of our findings obtained in this investigation are consistent with the literature, three facts must be re-examined carefully. One, which is a limitation of this study, concerns the training time used. Because of technical staff planning and the sports calendar, it was only possible to apply the preventative training for a sixeweek period. Although several studies intending to change movement patterns related to ACL injury risk have used a same time interval, the literature recommends that training programs with PTP features should be conducted throughout the sports season with adequate periodization (Grindstaff et al., 2006). The second aspect is related to the absence of a control group. Thus, it is unclear whether the observed improvements are due to the results of the applied training or other, uncontrolled factors. Additional, controlled studies aimed at measuring the effect of neuromuscular training on lower-limb kinematics and kinetics after different periods of practice are recommended to determine the time required to obtain increased training efficiency without causing excessive stress on joints. The third aspect is related to the possible clinical significance of the results. Despite the present study’s values for the angular variables and height of jumps changes, alterations presented by athletes suggest that the exercise program employed has a mechano-inductive capacity, even in limited magnitude. However, it was also possible to verify that the exercises tended to induce more important developments in a type of landing that corroborates with the proposition that the biomechanical changes in different motor behaviours after a neuromuscular training program are specific to a given stimuli (DiStefano et al., 2010; Myer, Ford, McLean et al., 2006; Myer et al., 2005). Thus, a training program can be effective in inducing changes in motor behaviour related to preventing the incidence of ACL injuries and athletic performance in one motor conduct and not for another.
The PTP seems to induce changes to the kinematic behaviour of the lower limbs. In the single leg condition, the time of landing was increased, while in the double leg condition the knee range of motion was improved. This highlights the importance of the selection of the tasks used to compare the kinematics of lower limbs. Also, although the vertical jump height was not statistically different, the improvement of 10% in both jumps tasks agrees with the values described in the literature and is an important performance improvement in volleyball. Therefore, the results allow us to conclude that PTP may induce specific changes on lower limbs kinematics and improve variables related to sport performance. This supports the idea that training programs similar to the one used in this study should be used in sports; however, they must be applied for longer periods, with adequate periodization during the season.
Conflict of interest
This studywas approved by the State University of Rio de Janeiro Ethics Committee.
This study was partially supported by the Brazilian Research Council (CNPq), Carlos Chagas Filho Foundation for Research Support of Rio de Janeiro (FAPERJ) and Coordination for the Improvement of Higher Level Education (CAPES).
Leia o artigo completo: https://www.ncbi.nlm.nih.gov/pubmed/23312730