Journal Logo

APPLIED SCIENCES

Match Play–induced Changes in Landing Biomechanics with Special Focus on Fatigability

SMEETS, ANNEMIE; VANRENTERGHEM, JOS; STAES, FILIP; VERSCHUEREN, SABINE

Author Information
Medicine & Science in Sports & Exercise: September 2019 - Volume 51 - Issue 9 - p 1884-1894
doi: 10.1249/MSS.0000000000001998

Abstract

During match play, athletes experience fatigue as a result of playing time. Growing evidence suggests that this incremental fatigue has a negative effect on lower extremity neuromuscular control and might increase, for example, anterior cruciate ligament (ACL) injury risk (1,2). Fatigue studies showed that proprioception (3), postural control (4,5), and movement coordination (6,7) alter under fatigued circumstances. In addition, more and more evidence exists that simulation of playing time-related fatigue might also influence muscle activation patterns (8,9) and kinematics and kinetics (10–12). For example, Zebis et al. (8) and Lessi et al. (9) found altered hamstrings activation in healthy athletes who completed a fatigue protocol that, respectively, simulated a handball match (8) or team sport–related tasks (FAST protocol) (9). Other studies found that in fatigued circumstances, athletes have a more extended landing pattern, characterized by smaller hip and/or knee flexion angles, compared with unfatigued circumstances (11–17). Such extended landing pattern is in turn associated with increased injury risk (18). However, the existing evidence around how fatigue alters neuromuscular control during landing tasks is not consistent, as shown in a systematic review by Santamaria and Webster (2). More recently, Barber-Westin and Noyes (1) also concluded that studies assessing the effect of fatigue on lower limb biomechanics during single leg landings have mixed findings (1).

A likely explanation for the inconsistencies between fatigue studies is the diversity in fatigue protocols being used. For example, some studies use a localized muscle fatigue protocol that induces fatigue of specific muscle groups (4,19), whereas other studies use general fatigue protocols that simulate playing time-related fatigue by inducing a combination of cardiovascular fatigue and muscle fatigue (8,12,20,21). It is evident that the latter ones are most suitable to assess whether match-related fatigue modifies neuromuscular control and, thereby, increases injury risk (2). However, less evident is the level of fatigue that should be induced by such a match play simulation protocol. One can standardize the protocol for the level of fatigue in each individual subject, for example, by continuing a protocol until exhaustion or until a certain drop in performance can be recorded (e.g., 20% decrease in jump height). However, an important disadvantage is that it might then not be representative of what happens in match play, with fitter players typically experience less fatigue than players who have a poor fitness level. To overcome this problem, fixed match play demand protocols (12,22,23) have been developed. In these protocols, a fixed distance or duration is used to standardize the demands of the protocol to real match play demands. The important caveat of standardized demands is that the observed effects of fatigue are then confounded by the individual’s fitness level or, in other words, their fatigability. Therefore, it is feasible that overall pre- and postfatigue effects may reveal a more consistent pattern when taking into account between-subject differences in fatigability.

Another methodological limitation of previous studies is that the symmetry between limbs is often assumed to simplify data collection and/or data analysis (e.g., unilateral data collection or pooling of the data of both limbs). This is in contrast with growing evidence suggesting that uninjured subjects do show between-limb differences in biomechanical parameters during gait and running (24,25) but also during more dynamic tasks (26,27). Hence, it is possible that also between-limb differences in response to fatigue exists (28,29). So far, studies that assessed the effect of limb dominance on fatigue show inconsistent findings (25,30,31), and therefore we assessed fatigue effects in both the dominant and the nondominant limbs separately in this study.

The first aim of the current study was to assess whether match-related fatigue induced by a fixed match play demand protocol causes overall alterations in landing kinematics, kinetics, and muscle activation. In the second part of the study, we established the relationship between fatigability and alterations in movement patterns to assess the influence of between-subject differences in fatigability. We hypothesized that athletes who experience more fatigue after a fixed match play demand protocol would also show movement patterns that involve greater risk for lower limb injuries.

METHODS

Participants

Eighteen uninjured subjects participated in this study (8 females and 10 males, age = 21.33 ± 1.53 yr, height = 178.14 ± 9.51 cm, weight = 71.31 ± 9.56 kg). All participants were involved in competitive sports on a regular basis (at least 3 times a week) and were free of lower extremity injuries for at least 6 months. None of them had an ACL injury in the past. The participants wore standardized indoor footwear (Indoor Copa, Kelme, Elche, Spain), and where necessary, long hair was tied up to avoid marker occlusion. All participants signed a written informed consent, and the study was approved by the local ethics committee (Commissie Medische Ethiek of the University Hospitals Leuven), with reference number S60182.

Protocol

The test session started with a standardized warm-up consisting of 5 min cycling on a stationary bike, 10 squats, and 10 squat jumps. Subsequently, surface EMG electrodes were placed on seven lower limb muscles according to the SENIAM guidelines (32). We recorded the activation of the following muscles (bilateral) with a wireless EMG system (Zerowire, Aurion, Milan, Italy) sampling at 1000 Hz: vastus lateralis (VL), vastus medialis (VM), biceps femoris (referred to as hamstring lateralis [HL]), semitendinosus (referred to as hamstring medialis [HM]), gluteus medius (GlutMed), gastrocnemius medialis (GM), and gastrocnemius lateralis (GL). All electrode locations were shaved and gently cleaned with 70% isopropyl alcohol to reduce skin impedance. Silver–silver chloride, pregelled bipolar surface EMG electrodes (Ambu Blue Sensor, Ballerup, Danmark) were placed over the muscle belly and aligned with the expected muscle fiber orientations, with 2 cm interelectrode distance.

Three static maximal voluntary contractions (MVC) of 5 s were performed for all muscle groups to normalize EMG data (33). For the MVC of the quadriceps muscles, participants were seated with 90° of hip and knee flexion and asked to extend their leg against a strap fixated around the distal tibia and the physiotherapy bed. For the hamstring muscles, participants were positioned in prone lying with 20° of hip flexion and 45° of knee flexion and asked to flex their knee against a strap fixated around the distal tibia and the back of the test leader. For the gastrocnemius muscles, participants were seated with 20° of knee flexion and had to plantar flex their feet against a strap that was fixated around the ball of their feet and the back rest of the physiotherapy bed. For the GlutMed muscles, participants were positioned in side lying with 30° of hip and knee flexion, and they were asked to abduct their leg against a strap that was fixated around the distal part of their thighs and the physiotherapy bed.

After the performance of the MVC, we attached 44 spherical reflective markers to the participants according to the eight segment “Liverpool John Moores University” model, including feet, upper and lower legs, pelvis, and trunk. This model was previously described in detail and shown to be reliable for measuring kinematics and kinetics during drop vertical jumps (34,35).

Subsequently, the participants performed the actual protocol, which consisted of a set of landing tasks. During the performance of these tasks, three-dimensional kinematic data were collected using 10 MX-T20 optoelectronic cameras (VICON, Oxford, UK) sampling at 100 Hz, synchronized with data recorded from two force plates (AMTI, Watertown, MA) sampling at 1000 Hz. Furthermore, a synchronized wireless EMG system (Zerowire, Aurion) was used to record muscle activity at 1000 Hz of the VL, VM, HL, HM, GM, GL, and GlutMed using surface electrodes.

All landing tasks were performed before and after a 5-min match play simulation protocol (SAFT5 protocol) (23). This match play simulation protocol includes functional movements (e.g., sprinting, jogging, agility ladder, slalom, counter movement jumps, and scissor jumps) and was shown to be valid to simulate peripheral and central match-related fatigue for at least 30 min after finishing the protocol (all subjects were able to perform the set of landing tasks in this time window) (23). The protocol of SAFT5 is described in detail elsewhere (23) and is a reduced version on the SAFT90 protocol, which has been shown to have similar intensity as to match play (22).

The landing tasks included three different sets of unilateral tasks. The order of the sets was randomized using an online tool (www.randomizer.org), and within each set, the task was repeated until three valid trials were performed on both the dominant and the nondominant legs. The dominant leg was defined as the preferred leg to kick a ball (36). For all three tasks, both take off and landing were on the same leg, and a trial was considered valid if the landing was central on the force plate and the subject could maintain their balance on one leg for 5 s after landing. Shuffling on the stance leg was not allowed. Participants were allowed to familiarize themselves with the tasks by performing practice repetitions at the start of the session. The subjects could practice as much as they want but did not receive any feedback or instructions on landing technique as we did not want to influence their natural landing patterns. On the basis of the practice trials, we determined the starting position for the single leg hop for distance to ensure that the subjects landed on the force plate.

  1. Single leg hop for distance. Participants were instructed to jump as far as possible on one leg.
  2. Medial hop. Participants were instructed to perform a jump sideways over a 0.24-m high hurdle (1.5 cm wide) on one leg. The mediolateral distance that had to be covered was calculated based on their leg length (1/2 of the leg length), and leg length was defined as the distance between ASIS and medial malleolus.
  3. Vertical hop with 90° of medial rotation. Participants were instructed to jump as high as possible on one leg, while performing an inward rotation of 90°.

All subjects had to score their perception of fatigue by giving an RPE of the legs (RPE-L) and breathlessness (RPE-B) using a 16-item BORG scale (37), both before and after the SAFT5 protocol. The prefatigue RPE scores were administered after finishing all landing tasks prefatigue, meaning that some subjects already experienced some fatigue at that time point. We decided to use differential RPE to differentiate between central factors (oxygen consumption, central nervous system) represented by RPE-B and peripheral factors (neuromuscular fatigue) represented by RPE-L (37). Besides the fact that RPE is a reliable, valid, and easy-to-use proxy for exercise intensity, RPE also reflects training status with fitter athletes showing lower RPE values for a standardized task (38–40).

Data analysis

All modeling and data processing were undertaken in Visual 3D (v.6.01.07; C-Motion, Germantown, MD). Marker trajectories and forces were filtered using a fourth-order low-pass Butterworth filter with a cutoff frequency of 18 Hz. Subsequently, hip, knee, and ankle kinematics were calculated using a Cardan sequence of rotations (41), and kinetics were calculated using inverse dynamics (42). External joint moments are described in this study (e.g., an external knee flexion moment will flex the knee). Besides joint angles and joint moments, also the vertical ground reaction force was analyzed. Initial contact events were created when the vertical ground reaction force crossed a 10-N threshold. Only the data of the landing were analyzed, from initial contact until 500 ms after initial contact.

All raw EMG signals of the motion trials and of MVC trials were band-pass filtered (6–240 Hz), rectified and low-pass filtered with a fourth-order zero-lag Butterworth filter at a cutoff frequency of 15 Hz. Subsequently, the filtered EMG signals of the jumping tasks were normalized to the peak value obtained during the three isometric MVC.

We decided to not pool the data of the dominant and nondominant legs but treat them as two different groups as there is no strong evidence yet that fatigue effects are similar for the dominant and nondominant leg (28,29).

Statistical analysis

To compare pre- and postfatigue kinematics, kinetics, and muscle activation patterns, repeated-measures ANOVA tests were conducted. A two-way ANOVA was used with main effects for fatigue and leg dominance. As we did not want to reduce the landing kinematics, kinetics, and muscle activation data to discrete values such as mean or peak values, we used statistical parametric mapping (SPM) for our statistical analysis. SPM makes it possible to analyze curve data as it calculates test statistics at each time node, avoiding the problem of multiple comparisons by modeling the behavior of random time-varying signals (43,44). Subsequently, post hoc tests (SPM t-tests) were conducted for all parameters that had a significant main or interaction effect.

Furthermore, correlation analyses were conducted to assess whether fatigability was related with landing pattern alterations (research question 2). More specifically, we used SPM regression analyses to assess the relationship between fatigue indicators (postfatigue RPE-L and postfatigue RPE-B) and landing kinematics, kinetics, and muscle activation. Movement patterns of athletes who are less resistant to fatigue may well be worse not only in a fatigued but also in an unfatigued state, so the regression analyses were performed twice, first for the prefatigue landing patterns and subsequently for the postfatigue landing patterns. The fatigue indicators used were postfatigue RPE-L and postfatigue RPE-B. Both values represent the subjective feeling of fatigue immediately after the performance of the SAFT protocol. We decided to only report the correlations that were significant for at least 10 ms and with r > 0.5 or r < −0.5. All correlation analyses were performed for the dominant legs and for the nondominant legs separately as research question 1 showed differences in landing kinematics and kinetics between the dominant and the nondominant legs. If a significant correlation was found between a fatigue indicator (RPE-score) and a biomechanical parameter, an extra figure was made to further illustrate the relationship. Therefore, we divided the subjects into four quartiles based on their fatigability: quartile 1 represents the subjects who reported the lowest RPE scores, and quartile 4 represents those who reported the highest RPE scores (the detailed range of RPE scores for every quartile can be found in the figure captions). Subsequently, the mean for every quartile was plotted to illustrate the relationship between fatigability and biomechanical variables. This approach is similar to making scatter plots of zero-dimensional data (for example, data of peak angles).

RESULTS

Fatigue Outcomes

Figure 1 shows that after the performance of the SAFT5 protocol, the subjects had significantly RPE-B (prefatigue, 8.22 ± 2.07 [extremely/very light]; postfatigue, 16.11 ± 1.60 [hard/very hard]) and higher RPE-L scores (prefatigue, 9.56 ± 2.01 [very light]; postfatigue, 15.17 ± 2.33 [hard]) (P < 0.001). Furthermore, the average performance was decreased: they covered significant less distance (prefatigue, 180.83 ± 28.94 cm; postfatigue, 172.32 ± 29.7 cm) during the single leg hop for distance (P < 0.001) and jumped less high (prefatigue, 8.57 ± 0.99 cm; postfatigue, 8.14 ± 1.34 cm) during the 90° hop (P = 0.003).

F1
FIGURE 1:
Overview of RPE values, jump distance, and jump height pre- and postfatigue. Every line represents the data of one subject. (Lower row) The full lines represent the individual data of the dominant legs; the dashed lines the individual data of the nondominant legs.

Overall Fatigue Effects

Kinematics

Repeated-measures ANOVA showed no interaction or fatigue effects on joint angles. However, leg dominance effects were found for the knee abduction angles in all tasks (P < 0.001) (Table 1). Post hoc analyses showed increased knee abduction angles in the nondominant legs during the entire landing phase both pre- and postfatigue (P < 0.001), and this across all tasks (Figure, Supplemental Digital Content 1, Post hoc analysis knee abduction angles, https://links.lww.com/MSS/B577).

T1
TABLE 1:
Overview of the two-way repeated-measures ANOVA.

Kinetics

Repeated-measures ANOVA showed no interaction or fatigue effects on joint moments or on the vertical GRF. However, leg dominance effects were found for hip flexion moments (single leg hop and medial hop, P < 0.001; vertical hop with 90° medial rotation, P = 0.024) and hip adduction moments (single leg hop and medial hop, P < 0.001; vertical hop with 90° medial rotation, P = 0.005) (Table 1). Post hoc analyses showed increased hip adduction moments in the beginning of the landing phase in the nondominant legs both pre- and postfatigue, and this across all tasks (P < 0.001). Furthermore, nondominant legs showed decreased hip flexion moments after peak loading both pre- and postfatigue, which was found in the single leg hop for distance (P < 0.001) (Figure, Supplemental Digital Content 2, Post hoc analysis hip flexion moments, https://links.lww.com/MSS/B578) and in the medial hop (P < 0.001) (Figure, Supplemental Digital Content 3, Post hoc analysis hip abduction moments, https://links.lww.com/MSS/B579).

Muscle activation

Repeated-measures ANOVA showed no interaction effects, fatigue effects, or leg dominance effects on muscle activation of all muscles tested (VM and VL, HM and HL, GM and GL, and GlutMed) (Table 1).

Fatigability

RPE-L

Higher postfatigue RPE-L scores were significantly correlated with smaller pre- and postfatigue knee abduction angles in the nondominant legs during the landing phase of the single leg hop for distance (prefatigue: r = −0.65, from 144 to 187 ms; r = −0.63, from 389 to 405 ms; postfatigue: r = −0.65, from 0 to 14 ms; r = −0.63, from 68 to 81 ms of the landing phase) and the vertical hop with 90° of medial rotation (prefatigue: r = −0.65, from 0 to 31 ms of the landing phase; postfatigue: r = −0.65, from 0 to 37 ms of the landing phase) (Fig. 2).

F2
FIGURE 2:
(Upper row) Linear regression analyses between postfatigue RPE-L values and knee abduction angles in the nondominant legs during the landing phase (from IC until 500 ms after IC). If the t-curve (black line) crosses the critical threshold (red dashed line), a significant correlation between RPE-L and knee abduction angles was found for that specific period. The dotted line represents the t-curve for the correlation with prefatigue angles; the full line represents the t-curve for the correlation with postfatigue angles. (Lower row) Visualization of the knee abduction angles (dotted line: prefatigue; full line: postfatigue) for four different quartiles based on RPE-L (Q1 = RPE-L 11–13, Q2 = RPE-L 14–15, Q3 = RPE-L 16, and Q4 = RPE-L 17–19).

No correlations were seen between other joint angles and RPE-L, or between joint moments, GRF, muscle activation, and RPE-L (Table, Supplemental Digital Content 4, Overview of correlation analyses, https://links.lww.com/MSS/B580).

RPE-B

Higher postfatigue RPE-B scores were significantly correlated with smaller postfatigue knee flexion angles in the nondominant legs in the beginning of the landing phase of the single leg hop for distance (r = −0.61, from 65 to 183 ms of the landing phase) and with smaller pre- and postfatigue knee flexion angles in the nondominant legs during the medial hop (prefatigue: r = −0.62, from 9 to 34 ms of the landing phase; postfatigue: r = −0.61, from 16 to 38 ms of the landing phase) (Fig. 3).

F3
FIGURE 3:
(Upper row) Linear regression analyses between postfatigue RPE-B values and knee flexion angles in the nondominant legs during the landing phase (from IC until 500 ms after IC). If the t-curve (black line) crosses the critical threshold (red dashed line), a significant correlation was found for that specific period. The dotted line represents the t-curve for the correlation with prefatigue angles; the full line represents the t-curve for the postfatigue angles. (Lower row) Visualization of the average knee flexion angles (dotted line: prefatigue; full line: postfatigue angles) for four different quartiles based on RPE-B (Q1 = RPE-B 13–14, Q2 = RPE-B 15–16, Q3 = RPE-B 17, and Q4 = RPE-B 18–19).

Furthermore, higher postfatigue RPE-B scores were significantly correlated with higher postfatigue vertical GRF (r = 0.68) and knee flexion moments (r = 0.70) in the dominant legs around peak loading during the vertical hop with 90° of medial rotation (208–286 and 193–265 ms of the landing phase, respectively) (Fig. 4).

F4
FIGURE 4:
(Upper row) Linear regression analyses between postfatigue RPE-B values and vertical GRF (left) legs and between RPE-B and knee flexion moments (right) in the dominant legs during the landing phase (from IC until 500 ms after IC). If the t-curve (black line) crosses the critical threshold (red dashed line), a significant correlation was found for that specific period. The dotted line represents the t-curve for the correlation with prefatigue angles; the full line represents the t-curve for the postfatigue angles. (Lower row) Visualization of the average GRF and average knee flexion moments (dotted line: prefatigue; full line: postfatigue angles) for four different quartiles based on RPE-B (Q1 = RPE-B 13–14, Q2 = RPE-B 15–16, Q3 = RPE-B lowest RPE-B 17, Q4 = RPE-B 18–19).

Finally, higher postfatigue RPE-B scores were significantly correlated with decreased postfatigue knee abduction moments (r = −0.69) and decreased postfatigue hip abduction moments (r = 0.69) in the nondominant legs during the performance of the single leg hop for distance (199–256 and 166–216 ms of the landing phase, respectively) (Fig. 5).

F5
FIGURE 5:
(Upper row) Linear regression analyses between postfatigue RPE-B values and hip abduction moments (left) and between RPE-B and knee abduction moments (right) in the nondominant legs during the landing phase (from IC until 500 ms after IC). If the t-curve (black line) crosses the critical threshold (red dashed line), a significant correlation was found for that specific period. The dotted line represents the t-curve for the correlation with prefatigue angles; the full line represents the t-curve for the correlation with postfatigue angles. (Lower row) Visualization of the average hip abduction and knee abduction moments (dotted line: prefatigue; full line: postfatigue angles) for four different quartiles based on RPE-B (Q1 = RPE-B 13–14, Q2 = RPE-B 15–16, Q3 = RPE-B lowest RPE-B 17, Q4 = RPE-B 18–19).

No correlations were seen between other joint angles and RPE-B, or between joint moments, GRF, muscle activation, and RPE-B (Table, Supplemental Digital Content 4, Overview of correlation analyses, https://links.lww.com/MSS/B580).

DISCUSSION

This study showed no alterations in landing kinematics, kinetics, or muscle activation patterns when a group of athletes was exposed to a match play simulation protocol (SAFT5) (23). This lack of movement alterations with match-related fatigue leads to rejection of the hypothesis that a fixed match play simulation protocol influences overall movement patterns. Nonetheless, we found an important role of fatigability in explaining alterations in movement patterns.

The SAFT5 protocol induced clear overall effects on fatigue outcomes: subjects reported an average RPE-B value of 16.11 (which means very hard to hard) and an average RPE-L value of 15 (hard); furthermore, their general performance decreased (4.7% on the single leg hop for distance, 5.2% on the vertical hop with 90° medial rotation). However, this did not reflect in any overall postfatigue alterations in landing kinematics, kinetics, or muscle activation patterns. Individual responses to the SAFT5 protocol were very diverse: postfatigue RPE-L ranged from 11 to 19 and RPE-B from 13 to 19. This implies that some subjects described the protocol as “fairly light” (RPE score of 11) or “somewhat hard” (RPE score of 13), in contrast to others who described it as “extremely hard” (RPE score of 19) (Fig. 1). This underscores the notion that a fixed match play simulation protocol (e.g., predefined in time/distance) is not suitable for inducing a standardized level of fatigue in all subjects, and that the overall pre- and postcomparison in our first part of the experiment is biased to achieving a false-negative outcome as some subjects were very resistant to match-related fatigue and only reported low levels of fatigue. The large variability in perception of fatigue depends on several factors. In the first place, it is influenced by someone’s fitness level; when an athlete becomes fitter because of effective training sessions, a fixed external load will be perceived less and less demanding (45). However, also other factors such as psychosocial well-being and recovery state determine perceptual abilities of an external load. For example, poor recovery or increased stress levels can cause that athletes feel more fatigued (46). The same external load (such as a fixed fatigue protocol) will thus be perceived very differently between subjects (47). In previous work, little attention was paid to this methodological issue, which likely helps explain the inconsistencies in fatigue effects on landing kinematics and kinetics reported in the systematic reviews (1,2). Therefore, we advise that future fatigue studies always administer RPE values to take into account between subjects’ differences in perception of fatigue.

In the second part of our study, we assessed whether fatigability was related with movement alterations. The results confirmed our hypothesis that high fatigability was associated with unfavorable landing kinematics and kinetics. Two relationships between fatigability outcomes and landing kinematics and kinetics were found across tasks and are therefore of major interest. First, athletes who reported higher RPE-B scores showed smaller postfatigue knee flexion angles in the nondominant legs during the landing phase of the single leg hop and the medial hop. Smaller knee flexion angles may be an indication of a stiff landing pattern, and a stiff pattern is in turn associated with injury risk (48). It should be emphasized that in our study, only a significant correlation between RPE-B scores and knee flexion angles was found, whereas RPE-L scores were not correlated with knee flexion angles. This might suggest that feeling of breathlessness/cardiovascular fatigue and not muscle fatigue modifies knee flexion angles. Therefore, it might be interesting for future studies to further investigate the role of RPE-B on knee joint angles in larger populations. Possibly, the reduction of knee flexion range of motion is not a physical limitation but rather a protective strategy to avoid further detrimental fatigue of the quadriceps muscle by reducing its eccentric work. Besides reduced knee flexion angles, increased GRF and knee flexion moments are also features suggestive for stiff landing patterns (48). In our study, we only found a significant correlation between higher postfatigue RPE-B scores and increased postfatigue GRF and knee flexion moments during the vertical hop with 90° of medial rotation. Possibly, the decrease in performance postfatigue (e.g., athletes jumped less far and high) could explain why the reduced knee joint angles were not accompanied with increased GRF or increased knee flexion moments in all tasks. However, during real match play, one may expect that athletes will push their limits and try to keep performing at the same level, even if they are nearly completely exhausted. In such situations, landing patterns with reduced knee flexion angles might lead to higher ground reaction forces and external knee flexion moments (49). Furthermore, the landing tasks of our protocol were maybe too easy in comparison to match play, and thus the athletes may have had enough reserve to counteract fatigue effects. Future studies should therefore investigate fatigue effects in more challenging situations, for example, by involving decision-making tasks to better simulate all aspects of match play and to improve the external validity of the protocol (47).

Another significant relationship found across tasks was that subjects who reported higher post-RPE-L scores showed smaller pre- and postfatigue knee abduction angles during the performance of the single leg hop for distance and the vertical jump with 90° of medial rotation. A study of Benjaminse et al. (11) found that athletes show smaller knee abduction angles during the landing of a stop jump task after completing a fatigue protocol until full exhaustion. However, in our study, we found that athletes with a high fatigability already showed smaller knee abduction angles prefatigue. Allowing less abduction range of motion might be an indication of a protective mechanism as, for example, less knee abduction might reduce ACL injury risk (50).

Most correlations between fatigability and movement alterations were only found in the nondominant leg, which supports the hypothesis that a similar response to fatigue might not be assumed in both legs. Therefore, one should be careful with pooling of the data of both legs and with generalization of the findings of the dominant legs to the nondominant legs. So far, very few studies have assessed limb dominance effects on fatigue responses, and therefore future studies should further explore this relationship. Probably asymmetries in strength (51) and landing patterns (52–54), but also the fact that a functional asymmetry exists between both limbs, might introduce asymmetry in fatigability. Sadeghi et al. (24) described the different functions of both legs during unilateral tasks: the dominant limb is used to mobilize and manipulate (e.g., kicking a ball), whereas the nondominant leg has a stabilization role as supporting limb. The nondominant leg is thus less selected during unilateral daily life activities that require coordinated force generation (55), which might cause an asymmetry in fatigability as the nondominant leg experiences less training.

This was the first study that assessed the relationship between fatigability and landing kinematics and kinetics, but it comes with some limitations: both the single leg hop for distance and the vertical hop with 90° of medial rotation were maximal performance tests. As already mentioned earlier, subjects jumped less far/high postfatigue, which probably influenced kinematics and kinetics. However, as there is no evidence yet if and how jump distance and jump height are associated with alterations in kinematics and/or kinetics, we were not able to correct for this decrease in performance. Nevertheless, large influences are not expected as the medial hop, which has equal jump height and distance before and after fatigue, showed similar associations between fatigability and altered postfatigue landing kinematics and kinetics as the two tasks that had a decrease in performance. A second limitation is that participants were asked to rate their perception of fatigue to quantify fatigability. RPE remains just a subjective perception of intensity, although global RPE is a reliable and valid proxy of oxygen consumption (V˙O2) and heart rate based measures (39,40,56), so follow-up studies should still focus on such more objective measures. Furthermore, as earlier mentioned, perception of fatigue varies widely between individuals, and thus the same fatigue protocol can be perceived very differently between subjects. Perceptual abilities of fatigue can be influenced by not only differences in physical fitness levels between subjects but also psychosocial factors and recovery/sleep (46,47).

Finally, future studies could use a progressive fatigue design in which a certain task is repeatedly evaluated while becoming more and more fatigued, similar to Borotikar et al. (13) who found that fatigue-induced changes in kinematics during single leg landings were already visible at 50% level of fatigue during a progressive fatigue protocol. We expect that progression of fatigue goes along with progressive worsening of movement patterns, considering that our results showed stronger associations between fatigue outcomes and landing patterns postfatigue than prefatigue.

From a practitioner’s point of view, the finding that being less resistant to match-related fatigue is associated with a reduction of knee range of motion in the sagittal and frontal plane (less knee flexion and abduction) might be an indication to adapt the current approach of prevention programs. If future prospective studies could confirm the role of fatigability in lower limb injury risk and even investigate whether fatigability might be seen as a modifiable injury risk factor as such, prevention programs should also focus on the improvement of resistance to match-related fatigue, for example, by increasing someone’s fitness level to ensure that athletes perform the same activities at a lower percentage of their maximal capacity and, thus, fatigue less rapidly and do not reach the very high RPE-B levels, which were associated with reduced knee flexion angles in the present study.

In conclusion, this study did not find alterations in landing kinematics, kinetics, and muscle activation when a group of athletes was exposed to a fixed match play simulation protocol. The between-subject variance in perception might explain why we found no overall match-related fatigue effects, and why inconsistencies between previous studies have been reported. When fatigability was taken into account, we found that being less resistant to match-related fatigue was related with less optimal landing kinematics and kinetics. Future prospective studies should therefore assess the role of fatigability in lower extremity injury risk.

This research did not receive specific grant from any funding agency in the public, commercial, or nonprofit sectors.

The authors declare that there is no conflict of interest.

The results of the study do not constitute endorsement by the American College of Sports Medicine and are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.

REFERENCES

1. Barber-Westin SD, Noyes FR. Effect of fatigue protocols on lower limb neuromuscular function and implications for anterior cruciate ligament injury prevention training: a systematic review. Am J Sports Med. 2017;45(14):3388–96.
2. Santamaria LJ, Webster KE. The effect of fatigue on lower-limb biomechanics during single-limb landings: a systematic review. J Orthop Sports Phys Ther. 2010;40(8):464–73.
3. Hiemstra LA, Lo IK, Fowler PJ. Effect of fatigue on knee proprioception: implications for dynamic stabilization. J Orthop Sports Phys Ther. 2001;31(10):598–605.
4. Marchetti PH, Orselli MI, Duarte M. The effects of uni- and bilateral fatigue on postural and power tasks. J Appl Biomech. 2013;29(1):44–8.
5. Caron O. Effects of local fatigue of the lower limbs on postural control and postural stability in standing posture. Neurosci Lett. 2003;340:83–6.
6. Cowley JC, Gates DH. Inter-joint coordination changes during and after muscle fatigue. Hum Mov Sci. 2017;56(Pt B):109–18.
7. Samaan MA, Hoch MC, Ringleb SI, Bawab S, Weinhandl JT. Isolated hamstrings fatigue alters hip and knee joint coordination during a cutting maneuver. J Appl Biomech. 2015;31(2):102–10.
8. Zebis MK, Bencke J, Andersen LL, et al. Acute fatigue impairs neuromuscular activity of anterior cruciate ligament-agonist muscles in female team handball players. Scand J Med Sci Sports. 2011;21(6):833–40.
9. Lessi GC, dos Santos AF, Batista LF, de Oliveira GC, Serrão FV. Effects of fatigue on lower limb, pelvis and trunk kinematics and muscle activation: gender differences. J Electromyogr Kinesiol. 2017;32:9–14.
10. Lessi CG, Fábio SV. Effects of fatigue on lower limb, pelvis and trunk kinematics and lower limb muscle activity during single-leg landing after anterior cruciate ligament reconstruction. Knee Surgery, Sport Traumatol Arthrosc. 2017;25(8):2550–8.
11. Benjaminse A, Habu A, Sell TC, et al. Fatigue alters lower extremity kinematics during a single-leg stop-jump task. Knee Surg Sports Traumatol Arthrosc. 2008;16(4):400–7.
12. Cortes N, Greska E, Kollock R, Ambegaonkar J, Onate JA. Changes in lower extremity biomechanics due to a short-term fatigue protocol. J Athl Train. 2013;48(3):306–13.
13. Borotikar BS, Newcomer R, Koppes R, McLean SG. Combined effects of fatigue and decision making on female lower limb landing postures: central and peripheral contributions to ACL injury risk. Clin Biomech (Bristol, Avon). 2008;23(1):81–92.
14. Thomas AC, Lepley LK, Wojtys EM, McLean SG, Palmieri-Smith RM. Effects of neuromuscular fatigue on quadriceps strength and activation and knee biomechanics in individuals post-anterior cruciate ligament reconstruction and healthy adults. J Orthop Sports Phys Ther. 2015;45(12):1042–50.
15. McLean SG, Samorezov JE. Fatigue-induced ACL injury risk stems from a degradation in central control. Med Sci Sports Exerc. 2009;41(8):1661–72.
16. Lucci S, Cortes N, Van Lunen B, Ringleb S, Onate J. Knee and hip sagittal and transverse plane changes after two fatigue protocols. J Sci Med Sport. 2011;14(5):453–9.
17. Quammen D, Cortes N, Van Lunen BL, Lucci S, Ringleb SI, Onate J. Two different fatigue protocols and lower extremity motion patterns during a stop-jump task. J Athl Train. 2012;47(1):32–41.
18. Leppänen M, Pasanen K, Krosshaug T, et al. Sagittal plane hip, knee, and ankle biomechanics and the risk of anterior cruciate ligament injury: a prospective study. Orthop J Sports Med. 2017;5(12):232596711774548.
19. O’Connor KM, Johnson C, Benson LC. The effect of isolated hamstrings fatigue on landing and cutting mechanics. J Appl Biomech. 2015;31(4):211–20.
20. Marshall PW, Lovell R, Jeppesen GK, Andersen K, Siegler JC. Hamstring muscle fatigue and central motor output during a simulated soccer match. PLoS One. 2014;9(7):1–11.
21. Wright M, Chesterton P, Wijnbergen M, O’Rourke A, Macpherson T. The effect of a simulated soccer match on anterior cruciate ligament injury risk factors. Int J Sports Med. 2017;38(8):620–6.
22. Small K, McNaughton LR, Greig M, Lohkamp M, Lovell R. Soccer fatigue, sprinting and hamstring injury risk. Int J Sports Med. 2009;30(8):573–8.
23. Bossuyt FM, García-Pinillos F, Raja Azidin RMF, Vanrenterghem J, Robinson MA. The utility of a high-intensity exercise protocol to prospectively assess ACL injury risk. Int J Sports Med. 2015;37(2):125–33.
24. Sadeghi H, Allard P, Prince F, Labelle H. Symmetry and limb dominance in able-bodied gait: a review. Gait Posture. 2000;12(1):34–45.
25. Brown AM, Zifchock RA, Hillstrom HJ. The effects of limb dominance and fatigue on running biomechanics. Gait Posture. 2014;39(3):915–9.
26. Sinsurin K, Srisangboriboon S, Vachalathiti R. Side-to-side differences in lower extremity biomechanics during multi-directional jump landing in volleyball athletes. Eur J Sport Sci. 2017;17(6):699–709.
27. Whyte EF, Richter C, O’Connor S, Moran KA. The effect of high intensity exercise and anticipation on trunk and lower limb biomechanics during a crossover cutting manoeuvre. J Sports Sci. 2018;36:889–900.
28. Willems ME, Ponte JP. Divergent muscle fatigue during unilateral isometric contractions of dominant and non-dominant quadriceps. J Sci Med Sport. 2013;16(3):240–4.
29. Radzak KN, Putnam AM, Tamura K, Hetzler RK, Stickley CD. Asymmetry between lower limbs during rested and fatigued state running gait in healthy individuals. Gait Posture. 2017;51:268–74.
30. Girard O, Brocherie F, Morin J, Millet GP. Human movement science lower limb mechanical asymmetry during repeated treadmill sprints. Hum Mov Sci. 2017;52:203–14.
31. Whyte EF, Kennelly P, Milton O, Richter C, O’Connor S, Moran KA. The effects of limb dominance and a short term, high intensity exercise protocol on both landings of the vertical drop jump: implications for the vertical drop jump as a screening tool. Sport Biomech. 2017;3141:1–13.
32. Hermens H, Freriks B, Merletti R, et al. SENIAM 8: European Recommendations for Surface Electromyography. Enschede, the Netherlands: Roessingh Research and Development, 1999.
33. Konrad P. The ABC of EMG—A Practical Introduction to Kinesiological Electromyography. Arizona: Noraxon; 2006. p. 61.
34. Malfait B, Sankey S, Firhad Raja Azidin RM, et al. How reliable are lower-limb kinematics and kinetics during a drop vertical jump? Med Sci Sports Exerc. 2014;46(4):678–85.
35. Sankey SP, Raja Azidin RM, Robinson MA, et al. How reliable are knee kinematics and kinetics during side-cutting manoeuvres? Gait Posture. 2015;41(4):905–11.
36. Mcgrath TM, Waddington G, Scarvell JM, et al. The effect of limb dominance on lower limb functional performance—a systematic review. J Sports Sci. 2016;34(4):289–302.
37. Mclaren SJ, Graham M, Spears IR, Weston M. The sensitivity of differential ratings of perceived exertion as measures of internal load. Int J Sports Physiol Perform. 2016;11(3):404–6.
38. Robertson J. Perceived Exertion for Practitioners: Rating Effort with the Omni Picture System. Human Kinetics; 2004. 145 p.
39. Impellizzeri FM, Rampinini E, Coutts AJ, Sassi A, Marcora SM. Use of RPE-based training load in soccer. Med Sci Sports Exerc. 2004;36(6):1042–7.
40. Coutts AJ, Rampinini E, Marcora SM, Castagna C, Impellizzeri FM. Heart rate and blood lactate correlates of perceived exertion during small-sided soccer games. J Sci Med Sport. 2009;12(1):79–84.
41. Cole G, Nigg BM, Ronsky JL, Yeadon MR. Application of the joint coordinate system to three-dimensional joint attitude and movement representation: a standardization proposal. J Biomech Eng. 1993;115(4A):344–9.
42. Winter DA. Biomechanics and Motor Control of Human Movement. 4th ed. Hoboken (NJ): John Wiley & Sons, Inc.; 2009.
43. Robinson MA, Vanrenterghem J, Pataky TC. Statistical parametric mapping (SPM) for alpha-based statistical analyses of multi-muscle EMG time-series. J Electromyogr Kinesiol. 2015;25:14–9.
44. Pataky TC, Robinson MA, Vanrenterghem J. Vector field statistical analysis of kinematic and force trajectories. J Biomech. 2013;46(14):2394–401.
45. Gil-Rey E, Lezaun A, Los Arcos A. Quantification of the perceived training load and its relationship with changes in physical fitness performance in junior soccer players. J Sports Sci. 2015;33(20):2125–32.
46. Windt J, Gabbett TJ. How do training and competition workloads relate to injury ? The workload—injury aetiology model. Br J Sports Med. 2017;51(5):428–35.
47. Benjaminse A, Webster KE, Kimp A, Meijer M, Gokeler A. Revised approach to the role of fatigue in anterior cruciate ligament injury prevention: a systematic review with meta-analyses. Sport Med. 2019;49(4):565–86.
48. Leppänen M, Pasanen K, Kujala UM, et al. Stiff landings are associated with increased ACL injury risk in young female basketball and floorball players. Am J Sports Med. 2016;45(2):386–93.
49. Favre J, Clancy C, Dowling AV, Andriacchi TP. Modification of knee flexion angle has patient-specific effects on anterior cruciate ligament injury risk factors during jump landing. Am J Hear Circ Physiol. 2016;44(6):1540–6.
50. Hewett TE, Myer GD, Ford KR, et al. Biomechanical measures of neuromuscular control and valgus loading of the knee predict anterior cruciate ligament injury risk in female athletes: a prospective study. Am J Sports Med. 2005;33(4):492–501.
51. Rahnama N, Lees A, Bambaecichi E. A comparison of muscle strength and flexibility between the preferred and non-preferred leg in English soccer players. Ergonomics. 2005;48(11–14):1568–75.
52. Brown SR, Wang H, Dickin DC, Weiss KJ. The relationship between leg preference and knee mechanics during sidestepping in collegiate female footballers. Sports Biomech. 2014;13(4):351–61.
53. Brophy R, Silvers HJ, Gonzales T, Mandelbaum BR. Gender influences: the role of leg dominance in ACL injury among soccer players. Br J Sports Med. 2010;44(10):694–7.
54. Pappas E, Carpes FP. Lower extremity kinematic asymmetry in male and female athletes performing jump-landing tasks. J Sci Med Sport. 2012;15(1):87–92.
55. Flanagan EP, Harrison AJ. Muscle dynamics differences between legs in healthy adults. J Strength Cond Res. 2007;21(1):67–72.
56. Wong DP, Carling C, Chaouachi A, et al. Estimation of oxygen uptake from heart rate and ratings of perceived exertion in young soccer players. J Strength Cond Res. 2011;25(7):1983–8.
Keywords:

FATIGUE; NEUROMUSCULAR CONTROL; MATCH SIMULATION; LANDING PATTERNS; INJURY RISK

Supplemental Digital Content

Copyright © 2019 by the American College of Sports Medicine