Limited positive effects on jump-landing technique in girls but not in boys after 8 weeks of injury prevention exercise training in youth football

For inclusion, the players had to be physically healthy and able to participate in testing with maximum effort. Background information about the players was collected from questionnaires at baseline.

Two different IPEP versions were introduced by randomisation in four teams each. Four teams were instructed to use the original Knee Control IPEP (Knäkontroll, SISU Idrottsböcker, 2005) with the added running warm-up that was introduced as a mobile application/webpage in 2012, and four teams used the Knee Control + . Both interventions were preceded by 5 min of standard warm-up running exercises with focus on knee control, followed by the same six principal neuromuscular preparation exercises or training aims for approximately 15 min: one-legged knee squats, pelvic lifts (hamstring strengthening), two-legged knee squats, the bench (core stability), lunges and jump/landing technique. The differences between the original Knee Control IPEP and Knee Control + were mainly the amount of possible progressions, with four different levels and an additional partner exercise in the original Knee Control IPEP, while Knee Control + had more options for progression with 6–10 different levels of increasing difficulty of exercises. Both IPEPs were led by the coach and, for practical reasons, all players in the team used the same level of the exercises, meaning that the exercises were progressed on a team-basis when the coach assessed that the players’ technique and neuromuscular control had improved.

The IPEP was implemented during an ordinary training session where all coaches received written and oral information from physiotherapists about the programme, and also practised the exercises. When possible, the players took part as well. The teams were recommended to use the programme at every training session (2–3 times per week) during the 8-week study period. Teams were recommended to use the 5-minute warm-up running exercises before matches as well. Teams were instructed to use all six IPEP exercises and start with the basic level, and then progress after 2 weeks of training. At this time the first author phoned the coaches to follow-up on the training and to discuss whether any exercises needed to be modified or replaced, and also encouraged the coaches to start progressing the training. Coaches were asked to report any adverse events throughout the intervention period.

Jump-landing technique was assessed by studying neuromuscular control in the DVJ and TJA. Both tests were filmed with two GoPro Hero5 cameras, one from the frontal and one from the lateral view. The cameras were synchronised with a GoPro smart remote control to start and stop filming at the same time. The films were scrutinised by the first author, a physiotherapist with 18 years’ experience as a group training instructor, using the Windows media player as many times as necessary to be certain about the judgment and in both real-time and slow-motion.

The study was approved by the regional ethical review board in Linköping, Sweden: Dnr 2017/294-31. Players and their legal guardians received written information about the study and gave written informed consent before study commencement. Players depicted in the additional files specifically consented to their pictures being shown in research presentations.

For the within-group comparisons effect sizes, Cohen’s d was calculated for parametric data using mean values and standard deviations, and r was calculated for non-parametric data based on Z/\(\sqrt{N}\), where N equals the number of observations. Effect sizes were interpreted as: small d = 0.2, medium d = 0.5, and large d = 0.8 or small r = 0.1, medium r = 0.3 and large r = 0.5 [6].

The small improvements seen in TJA among the girls are positive, as girls have a higher risk of acute knee injury than boys [39]. The potential for improvement may also have been higher in girls as they performed worse at baseline. According to Myer et al. [25] individuals presenting ≥ 6 flaws in the TJA should be specifically targeted for preventive actions. The players in the current study showed a median of three to four flaws, i.e., they performed rather well already at baseline and had less room for improvement, but a significant improvement in the girls was found nonetheless. Another study also found improvement in the TJA after a neuromuscular training intervention, but similar improvement was observed also in the control group [16]. In future studies it is possible that using the modified TJA scoring with a three-graded scale [8], instead of the dichotomous scale used in the current study, might capture smaller changes in neuromuscular control as a result of preventive training.

A critique of the TJA is that it may not measure fatigue in a structured way, as the speed of jumping is not controlled for and one way for athletes to improve the total score may be to jump at a slower rate [36]. In the current study, the improvement seen in the TJA total score in the girls occurred in combination with an increased number of jumps. This strengthens the results on the TJA total score, as an increase in jump rate is more demanding for the player and may challenge the neuromuscular control more. This increase in jump rate may also be interpreted as a training effect of the Knee Control IPEPs suggesting increased strength, power and plyometric performance.

In the DVJ no effect was shown on landing biomechanics, except for an increase in knee flexion angle at initial contact in girls. The results are in line with two other studies in girls’ football that also reported no effect from 2D or 3D motion analyses of DVJ from preventive training using the WIPP (Warm-up for Injury Prevention and Performance) programme [11] and the 11 + [2]. The results do, however, diverge from Lim et al., [20] who showed improvements in both knee flexion angles and knee separation distance during rebound jumps after using a modified version of the PEP (Prevent injury and Enhance Performance) programme in youth female basketball for 8 weeks. In future studies, inclusion of electromyographic activity measures may be valuable to understand better the effect mechanisms of the IPEPs. As neither boys nor girls in the current study presented with distinct valgus alignment (normalised knee separation distances < 60%) according to the criteria by Noyes et al., [29] the room for improvement in the DVJ may have been limited, explaining the absent effect.

Knee valgus seemed to be unaffected by the interventions as assessed in the TJA criteria of lower extremity valgus as well as in the subjective and objective assessments of the DVJ. Movement patterns that are coupled to high knee valgus moment and valgus angles have been shown to be associated with an increased risk of ACL injury [14], although other subsequent studies have not confirmed these results [18, 19]. While some argue that knee valgus is a natural movement, that needs to be controlled rather than avoided [7], decreasing lower extremity valgus and increasing knee flexion angles in jumping/landing and cutting are often seen as important targets for IPEPs [17, 19] and in the current study the girls’ knee flexion angles improved. It should be noted that even though the effect size was large, the minimal clinically relevant change in knee flexion angle is not known. Additionally, an overall better jump-landing technique at follow-up was seen among the girls according to the TJA total score. How this affects the overall risk of injury is not known, as few of the individual TJA criteria showed significant improvements. Considering that the risk of acute knee injury is higher in girls than in boys, the study shows positive results by indicating that movement patterns may be changed by the interventions.

Two-dimensional tests cannot assess rotation, such as internal and external rotation of the lower extremity but have been shown to have high reliability and adequate validity compared with 3D motion analysis [32]. However, due to the imprecision of the subjective assessments we saw considerable ceiling and floor effects, resulting in less room for improvement or decline in performance. One strength was that the results of the subjective DVJ assessment and the objective 2D motion analysis using normalised knee separation distances corroborated each other as no change in either test was seen. Another strength was the use of two different tests of jump-landing technique assessed in the same population making it possible to study the consistency of the results across assessments. Utilisation fidelity (if the programme was used as intended regarding exercise selection, dosage and progression) or exercise fidelity (whether the exercises were done with the correct technique) was not monitored as part of the study, which is a possible limitation, especially as the exercise fidelity with the Knee Control IPEP has been shown to be sub-optimal [23]. The potential for an IPEP to improve jump-landing technique may be influenced by, for instance, coaches’ instructions on exercise technique. Therefore, the coach may play a vital part in any observed effect or lack of effect on jump-landing technique.

Some other limitations include a shorter study duration than intended, due to the short time period between school starting after the summer break and the end of the football season. However, regarding neuromuscular control, a single exercise session with adequate feedback may be enough to affect jump-landing technique [5], even though this of course may not have long-lasting effects. Other limitations were the lack of blinding, as the analyst for practical reasons could not be blinded. Furthermore, other limitations were the lack of a pure control group, limited time for progression of the training and large drop out, especially among the girls, which limits the statistical power. When interpreting these results, one must bear in mind that the population was young, with rather large intra-individual differences within the same session. Additionally, the tests were unfamiliar to most and they had no established movement pattern when doing DVJs and TJAs at baseline, and hence, there is a possibility of a learning effect for the follow-up measurements. Additionally, results may be obscured by a maturing study population potentially changing their movement technique spontaneously. As in similar studies, we cannot be certain that changes seen in the tests will be transferred to real-life potentially harmful situations.

The results were not analysed based on the players’ performance at baseline even though individuals at high risk of ACL injury are believed to respond more favourably to IPEPs from a neuromuscular and biomechanical point of view [9, 24]. To make all players benefit more from IPEPs it has been suggested that training should be tailored towards the observed movement deficits [9, 15], which may be valuable in future studies. The feasibility of individualised tailoring of IPEPs in the real-world may, however, be questioned as this requires more experience and time from the coach.

This study extends understanding of the effect mechanisms of IPEPs, which is important from, for example, an implementation point of view. Earlier studies have shown that coaches modify the IPEPs, with unknown effects on prevention efficacy. As understanding of the effect mechanisms grows it may be possible to tailor programmes to fit the teams and, as a result, support successful adoption and maintenance of the IPEPs.