Impact Differences among the Landing Phases of a Drop Vertical Jump in Soccer Players

The aim of this study was to examine the diff erences of landing phase biomechanics between the players who had anterior cruciate ligament (ACL) reconstruction and healthy participants during single leg drop vertical jump. In this study, 11 soccer players who had anterior cruciate ligament reconstruction (aged 23.0±3.6 years, height 177±5.0 cm, weight 83.8±11.7 kg) and 9 healthy soccer players( aged 22.2±2.4 years, height 178±3.0 cm, weight 74.3±6.1 kg) participated voluntarily. During the data collection phase three high speed cameras synchronized to each other and force plate were used. Visual analysis programme and MATLAB were used to calculate kinetic and kinematic variables. Landing techniques of the subjects’ were examined by fl exion angle of knee, ground reaction force and moment parameters. The statistical analyses of the measured results were performed by t-test and Pearson Correlation analysis. According to the results, it was determined that peak vertical ground reaction force exhibited significant phase diff erences (p=0.00, and p=0.00, respectively) between the groups. Obtained results can be explained with “quadriceps avoidance” motion pattern which is characterized by decreased quadriceps activity and lower external knee fl exion moment in an eff ort to control anterior translation of the tibia in subjects with ACL reconstruction. A better understanding of the diff erent phases during single-leg landings can shed a light on mechanism of non-contact anterior crucaite ligament injuries therefore future researches should assess how phase diff erences aff ect drop vertical jump performance.

A number of reports have shown that sports-related ACL injuries generally occur during non-contact situations that are characterized by landing, rapid deceleration, and sudden changes of direction and most of them occur during single-leg landings (Boden et al., 2010) which are common tasks performed from varying vertical heights and horizontal distances during sporting events such as volleyball, basketball and soccer ( Pappas, Zampeli, Xergia, & Georgoulis, 2013). Soccer players sustain the greatest number of ACL injuries (53% of the total) with skiers and gymnasts also at high risk (Hewett, Myer, & Ford, 2005 related to ACL injury risk factors, including vertical ground reaction force (VGRF), joint angles and moment contribute to knee instability and are a primary loading mechanism of the knee joint and ACL (Hewett et. al., 2005;Schroeder, Krishnan, & Dhaher, 2015;Siegmund, Huxel, & Swanik, 2009). Greater GRF upon landing increases the probability of ACL injury, prior to injury, participants who sustain ruptures exhibit 20% larger peak vGRFs during landing than participants who remain healthy. Moreover, the knee angle was signifi cantly more extended in the injured athletes when the foot was completely fl at at the initial foot contact kl So far many studies have been focused on initial contact phase of landing tasks (Čoh, Berić, & Bratić, 2013;Zahradnik, Uchytil, Farana, & Jandacka, 2014). But the other phases such as moment of jump and last contact with the ground can have an impact on biomechanical factors that present a risk for the occurrance of ACL injuries.
Th e purpose of this study was to determine how ground reaction forces, moments and knee fl exion angles diff er between healthy controls and reconstructed subjects during single leg landing phases. We suggested two hypotheses respectively: (1) the knee fl exion angle correlated with the kinematics at the landing phases in both groups; and (2) force, moment and angle values will diff er between each phase and also between the groups.

Participants
Th e participant population consisted of two groups-ACL reconstructed group (n=11 patellar tendon autograft ) and an uninjured control group (n=9). All participants were soccer players performing at amateur soccer clubs and matched for age, height, weight, sports age as shown in Table 1. ACL-reconstructed palyers were included who had an isolated ACL rupture and a subsequent surgical reconstruction using either a hamstring tendon (HT) or patellar tendon (PT) autograft at least 6 months and up to 15 months prior to the study sessions and also only the subjects whose dominant leg were right to the study. Exclusion criterias were; history of signifi cant knee pain prior to the injury and/or at time of testing, contralateral knee injury/ surgery, or prior injury/surgery to the reconstructed knee. Th e dominant leg was determined as the leg used by the participant to kick a ball. All experimental procedures were approved by Th e Ethic Comittee and complied with the principles of the Declaration of Helsinki. Informed consent was obtained from all subjects prior to participation in the research study.

Landing task
Players were instructed to warm up for 5min. and instructed to perform drop jump from a custom made takeoff platform from 20cm vertical height that were placed next to the edge of a force plate (Ali, Robertson, & Rouhi, 2014). Th e command of 'ready' was given to the participants before the start of each landing task. For each landing task all participants began with a standard take-off position by standing on a take-off platform with hands placed on the hips, legs shoulder width apart, and the toes of both feet aligned with the edge of the take-off platform. Participants were then instructed to stand on their dominant leg, drop off , and land as naturally as possible with their dominant foot only centered on the force plate and jump vertically as soon as possible. Th e participants were asked to keep their hands on their hips when landing to reduce any variability from swinging arms. Each subject was asked to perform three successful trials, and the best result was used for further analysis.
All participants wore their own sports shoes throughout data collection. Motion analysis was performed on all subjects using a-camera motion capture (SIMI Reality Motion Systems GmbH, GER) system with three cameras (Basler A602f-HDR GmbH, GER) which were set at 100 frames per second as shown in Figure 1. For digitization, 7 retrorefl ective markers attached to right side of the body; trochanter major, spina iliaca anterior superior, patella, condylus lateralis, tuberositas tibia, condylus lateralis tibialis, malleollus lateralis. A force plate (FP4060-10, BERTEC, USA) measured ground reaction forces (GRFs ) at a sampling rate of 1000 Hz. Videographic and force plate data were time synchronized. Th e vertical ground reaction force (VGRF) was defi ned as the reaction to the force the body exerts on the ground in the vertical direction.

Data reduction and Analysis:
One trial was selected from the best of three trials for data analysis. Th e best trial was determined as the one in which the participant did not remove their hands from the hip during landing, did not allow their non-dominant leg to impact the force plate during landing, or did not lose a marker during impact with the ground. Joint kinematics and kinetics were determined for the dominant leg. Joint kinematic data were calculated using a SIMI Motion Analysis System and analog data was imported into MATLAB (Version 5.3 , Th e Mathworks Inc., Natick, MA). Maximum vertical ground reaction force was calculated aft er initial contact with the force plate during the task which was divided into three phases. Initial contact (IC) phase was defi ned as the instant where the force plate reported values greater than 20 N VGRF, Moment of Jump phase (MoJ) phase defi ned as peak VGRF and last contact (LC) phase was defi ned as the greatest force value aft er moment of jump. According to Ford, Myer and Hewett (2014) study, marker trajectories were fi ltered using a low-pass 2nd-order Butterworth fi lter with a cut-off frequency of 12Hz, chosen aft er conducting a residual analysis. Ground reaction forces were normalized to each subject's body weight and moments normalized by the product of body mass and body height. Th e knee fl exion angle was defi ned as the angle between the thigh and leg segment. Kinetic raw data was collected at 1000Hz and kinematic raw data was collected at 100Hz. Th erefore sampling frequency of both data equated at 250Hz. Kinematic data were low-pass fi ltered using a second-order Butterworth fi lter at 100Hz and analog data were fi ltered at 25Hz.

Statistical analysis
Groups were tested for normal distribution by means of the Kolmogorov-Smirnov test. Homogeneity of the variances was ascertained by Levene's F test. Th e Independent Samples t-test and Pearson correlation analysis were used for variables depending on the normality of distribution. Th e level of significance was set at p<0.05.

Results
Th e overall means and standard deviations of the vertical ground reaction force and knee fl exion angle for the vertical height test among all subjects and t-test coeffi cients of peak VGRF with knee fl exion angle and moments are provided in Table 2. Th e fi ndings from the t-test conducted, revealed for the single landing test that VGRF was signifi cantly higher in uninjured group at the initial contact (p=0.00) and at the moment of jump (p=0.00) but there was no signifi cant diff erence at the moment of last contact between the groups.
As Shown in Table 3, peak VGRF was signifi cantly and negatively correlated knee fl exion angle (r=−0,569 p=0.009 ) in reconstructed group at the moment of initial contact. It is also worth noting from Table 3 that peak VGRF was unsignifi cantly and negatively correlated to knee fl exion at the last contact in both groups, too. But VGRF was unsignifi cantly and positively correlated to knee fl exion angle at the moment of jump. Th ere was no signifi cant correlation amongst VGRF, knee fl exion angle and y component of moment.

Discussion
Th e purpose of this study was to investigate how VGRF, moments and knee fl exion angles diff er between healthy controls and reconstructed subjects during single leg landing phases. It was found that though knee fl exion angles and moment values are equivalent between the groups, diff erences in VGRF indicate that each phase has its own biomechanical mechanisms.
Previous research has suggested that a relationship exist between demographics which supported by Robinson et al. (2014) stated that females to exhibit greater hip internal rotation and hip adduction moment than males (Abdel-Aziz, & Karara, 1971;Pollard et. al., 2007;Ford et al., 2014).
Additionally, stronger support for the ''quadriceps dominance'' theory as a potential mechanism for the sex disparity in ACL epidemiology is provided by studies that found females to demonstrate preferential quadriceps activation compared to males (Ford et al., 2014). Th erefore, only male subjects were included in our study (Table 1).
Our results showed that there is a signifi cant diff erence in VGRFs at IC and MoJ phases but there is no signifi cant diff erence in VGFR at LC between the groups (Table 2). Th is can be explained with "quadriceps avoidance" motion pattern.
Early biomechanical researches that investigated kinetic and kinematic diff erences between healty subjects and reconstructed subjects indicated that many of the subjects perform the tasks with a "quadriceps avoidance" which is characterized by decreased quadriceps activity and lower external knee fl exion moment in an eff ort to control anterior translation of the tibia (Ali et al., 2014). We found no signifi cant diff erence neither in knee fl exion angle nor in moment values between the groups (Table 2). Podraza and White (2010) found similar results; given that ground reaction forces are more likely to be greatest and knee extensor moments smallest when landing in an extended knee position; it is possible that ACL strain from noncontact deceleration may be related to rapid trans- lational joint forces that propagate up the kinetic chain rather than resulting from quadriceps overload induced anterior tibial translation. Boden et al. (2010) also proposed that a lack of absorption of ground reaction forces who were injured, landed with a mean knee fl exion angle of 17.6° compared to uninjured controls that landed with a more plantar fl exed ankle and had a knee fl exion angle of 39.3°.Previous studies indicated that the impact on the lower extremities increases as the peak vertical ground reaction force increases (Ali et al., 2013;Pappas et al., 2013;Podraza et. al., 2010;Wang, 2011). Pappas et al. (2013) compared the ground reaction force between single-leg drop landings and double-leg drop landings. Pappas et al. (2013) found that single-leg drop landings from a height of 0.4 m produced a higher peak vertical ground reaction force than stop jump. Th e results from the work of Boden et al. (2010) suggested that the propagation of reaction forces when landing with the knee near full extension could be an important component of non-contact ACL injuries. Support moment is the net summation of ankle plantar fl exion, knee extension and hip extension moment. Hewett et al. (2005) measured landing biomechanics at baseline for female athletes participating in high school basketball and soccer and followed them for one to two seasons. Th ey found that high knee valgus angle and moment and high side-to-side diff erences in knee valgus angle and moment during landing from a jump were strong predictors of future ACL injury. Since landing from a rebound is the task most commonly associated with ACL rupture in basketball (Sugimoto et al., 2015),it is possible that the fi rst drop landing task does not suffi ciently simulate all the biomechanical mechanisms enacted when landing from a maximal jump. Greater fall heights prior to landing incrementally increase perturbations and, consequently, vGRFs on the lower extremity (Peng, 2011;Abdel-Aziz, & Karara, 1971). Within the fi ndings and limitations of this study, we observed that VGRF s and knee angles diff er among the phases. Additionally, other potential limitation to the current study includes that all participiants performed single leg drop vertical jump, future researches may include double leg drop vertical jump task and add diff erent heights to their studies. A better understanding of the diff erent phases during single-leg landings can shed a light on mechanism of non-contact ACL injuries.
As a conclusion, there are diff erences between the landing phase kinetics and kinematics of single leg drop veritical jumps. We suggest that a higher risk of ACL injury could result from the fact that the single-leg drop jumps exhibites greater peak forces and moments during the landing than the moment of jump and initial contact phases. Th is indicates that non-contact injuries occur during landing phase of jump tasks.
Future research is necessary to evaluate the injury-specifi c infl uences of landing phases. Researchers should attempt to extrapolate these fi ndings to more dynamic and challenging tasks that are more representative of scenarios during which ACL injury occurs and to the populations at heightened risk of ACL injury and also hip, trunk, core, and upper body mechanics are associated with lower extremity biomechanical and neuromuscular factors of each landing phase should be better to be examined for futher information.