According to FIFA, football is the world's most popular sport with more than 265 million players (1), and female professional, semi-professional, and community football is one of the fastest growing sport worldwide. Along with its popularity, football accounts for a higher rate of lower-limb injuries accompanied by a higher incidence of anterior cruciate ligament (ACL) rupture and associated burden than do other sports (2–5). Females have a higher risk of concussion and ankle injuries, and the risk of serious knee injuries is at least double when compared with men; meanwhile, men present greater risk of hamstring and groin injuries (6–12). In particular, ACL injury rate is higher during match play compared with training, and in cutting and pivoting sports (13) compared to other sports. In this scenario, non-contact ACL injuries, that typically occur during high-impact tasks such as decelerations, landing, and cutting represent one of the most devastating injuries, accounting for 70% of ACL annual injuries with an incidence rate of 0.42 per 1,000 player-hours (14). The long return to play period (6–24 months) and an increased risk in future development of osteoarthritis represent the associated burdens. It should be mentioned that injury, fear of injury, and lack of physical skills or strength represent the current barriers to sport and physical activity participation in females (15). All these reasons support the introduction of new methodologies to prevent or reduce the risk of injury. Nowadays, a large variety of functional screening tools are adopted during pre-season tests to assess athletes’ readiness to compete and to establish a baseline measurement at the beginning of a rehabilitation program. These tests allow tracking athletes’ progress and assisting in return to play decisions, along with identifying specific movement faults, muscle weakness, imbalances. In terms of functional evaluation, the most commonly used methods are the Functional Movement Screen, the Star Excursion Balance Test, the Y Balance Test, the Drop Jump Screening Test, the Landing Error Scoring System, and the Tuck Jump Analysis (16): they imply visual scoring of different movements as well as pain, and the best score is associated with the ability to correctly complete the movement without compensation. In considering the possible causes of ACL rupture, non-contact ACL injuries have been reported as the most frequent that commonly occur when landing from a jump, during cutting maneuvers, or sudden decelerations (17). There is a plethora of studies reporting about the association between ACL rupture and a poor landing mechanism (18): with the knee in a valgus position and the femur in adduction and internal rotation; with the knee in an extended position and with excessive quadriceps activation relative to the hamstrings; with deficits in trunk control; and with large leg-to-leg asymmetries (18). These considerations are at the basis of the standard functional tasks currently adopted in the pre-season tests (19). However, the large number of ACL injuries still reported annually (80,000 per year in the United States alone) (20) poses the need for screening athletes through other tasks that can better portray the multifactorial mechanism of ACL rupture. Among the possible ones, cutting maneuvers has recently attracted a large attention as they represent the primary cause of non-contact injuries in cutting dominant sports (i.e., rugby, American football, handball, basketball, football) (14, 21). In this context, the injury mechanisms that have been most commonly described include frontal plane trunk misalignment with respect to the cutting direction, internally rotated hip, valgus knee alignment at loading response, an excessive extended knee, hip abduction, increased foot progression angle, rearfoot landing, and excessive lateral foot-plant distances. All these conditions evoke the greatest multiplanar knee joint loading (14, 22). In particular, when females perform cutting maneuvers, landing with an extended hip and knee, with a valgus knee alignment, an internal rotation of the tibia, and a pronated foot (i.e., “position of no return”) have been indicated as the “high-risk postures” (23). We should further consider that, when ACL is placed under stress, hamstrings are activated by both a fast-to-respond reflex arc from ACL mechanoreceptors to the hamstrings, and a secondary reflex arc from mechanoreceptors in the muscle or joint capsule. The first one allows the hamstrings to act as a torque regulator “on demand” during ligament overloading, thus supporting in maintaining joint stability. The second one, thanks to a longer response time than the first one, and an inhibitory input to the quadriceps, provides activation of the hamstrings to assist knee instability conditions (24). However, elite athletes are subjected to nonphysiologically large forces coupled with fast rates of load applied to the knee, which do not allow these two reflex arcs to respond fast enough (18). This is one of the main reasons why thigh muscle exercises and strength improvements may not be sufficient to provide the necessary support to reduce incidences of instability during high-impact tasks (i.e., decelerations, landing, and changes of direction). It is in this general framework that the inclusion of proprioceptive training finds its applicability in prevention and rehabilitation programs (25). Given the importance of motor control in restoring athletes motor function after injury, therapies have been recently proposed targeting at training the proprioceptive system to achieve coordinated sensory–motor patterns in human movements. On the one hand, we can find neuromuscular and proprioceptive training exercise (26) and, on the other, vibrations applied to muscles or tendons that can induce different effects depending on the vibrated muscle, the sensory context, and the task. In the study by Melnyk et al. (27) whole-body vibrations were applied on stretch reflexes involved in knee joint control as ACL injuries prevention treatment. Results of the study showed improved knee stability and reduced anterior tibial displacement upon shock provocation (28). Another non-invasive therapeutic modality commonly used is transcutaneous electrical nerve stimulation; this technique is based on the stimulation of the large diameter mechanosensitive afferent nerve fibers in the skin but it is mostly used in pain control (29). Recently, a neurorehabilitation device, Equistasi®, based on focal mechanical vibrations (30) has been applied to modify cyclists’ pedaling posture (31), and this was reflected in a better distribution of the pressure on the saddle of a group of amateur cyclists. The device was originally adopted for therapeutic purposes in Parkinson's disease targeting at improving trunk sway (32) and gait (33). The Equistasi® device, once applied on specific sites of the body (i.e., muscles, trigger points), self-generates focal mechanical vibrations that interacts with the mechanoreceptors, the Golgi tendon organs, and the neuromuscular spindles. To the best of the authors’ knowledge, vibrations have been applied for ACL injury prevention purposes as a whole-body solution (27, 28); therefore, the aim of the present study is to evaluate the impact of providing a proprioceptive stimulus through the Equistasi® device on the biomechanical risk factors associated with ACL injury. For this purpose, a biomechanical evaluation was carried out directly on the field and involved a group of elite female football players performing a series of sidestep cutting maneuvers. This task was selected as it is similar to the sport gestures involved in competitions, and accounts for the majority of non-contact injuries. From a biomechanical point of view, externally applied knee abduction, internal rotation, flexion-extension torques, knee flexion angle, and anterior tibial shear forces have all been identified as the currently agreed biomarkers of non-contact ACL injury risk (14). Furthermore females have been reported to use both an increased knee valgus kinematics and decreased hamstrings-to-quadriceps peak torque ratio compared to males (17). Of these biomarkers, in the present study, lower-limb joint angles and torques were considered, and a female elite team was selected, due to the higher rate of incidence of ACL rupture in females than males. The main hypothesis of the present study is that by improving proprioception through the use of the Equistasi® device, the multiplanar loading condition associated with increased ACL strain could be reduced, and consequently this could be reflected in the performance of a high-risk task such as sidestep cutting maneuver. In order to verify the effective elicitation of the somatosensory system, and its link with motor control, through the application of the vibratory stimulus, the Romberg test was performed similar to previous studies aiming at restoring balance and motor control in individuals with Parkinson's disease (32, 33).

The main findings of the current study should be considered: the statistically significant reduction observed on the knee abduction torque following the proprioceptive stimulation both after the immediate application of the stimuli (T1) and after 2 and 4 weeks of treatment (T2 and T4, respectively). From a biomechanical point of view, according to Bates et al. (45), this parameter has the potential to generate the greatest change in ACL strain in isolated conditions than other parameters such as anterior tibial shear force or tibial internal rotation. However, in order to assess the overall efficacy of an injury prevention treatment, not only this parameter should be considered (18). In our study, this reduction in knee abduction torque was detected in association with a progressive reduction on hip abduction torque and ground reaction forces from T1 up to T4. A further reduction was noted in the knee flexion-extension angle at T4 in combination with a reduction of the ankle plantarflexion angle. Between T0 and T1 also, lower knee and hip flexion-extension and adduction–abduction torques were detected. Compressively, our results showed the adoption of a landing strategy that complies with the requirement of reducing the biomechanical factors contributing to multiplanar loading (14). In terms of the task chosen for assessing the efficacy of the injury prevention approach, the present study investigated athletes based on the biomechanical analysis of the sidestep cutting maneuver, given the higher rate of incidence of ACL rupture in cutting dominant sports. As the rate is further increased when females are considered rather than males, the study focused on elite female athletes. Given the paucity of research reporting about the efficacy of training programs in terms of biomechanical analysis, an important novelty of the current study should be considered that a complete biomechanical analysis of the lower-limb joints was performed before and after the treatment at different time points. In particular, as no consensus was reached about the most crucial variables for describing cutting maneuvers (33), the ones available for identifying a poor landing strategy were chosen. These are indeed associated with the ligament, quadriceps, trunk, and leg dominance theories developed for landing tasks, namely, hip, knee, and ankle flexion angles (i.e., quadriceps dominance theory); hip, knee, and ankle flexion and adduction–abduction torques (i.e., ligament dominance theory); and vertical ground reaction forces (i.e., softer landings associated with better trunk control) (18). According to the findings of the present study, the proprioceptive stimulation showed to have the potential to improve cutting biomechanics related to the ligament theory, as a decreased knee abduction torque was registered immediately after the device application (T1) and at the two following evaluations (T2 and T4). Noticeably, the observed reduction on hip flexion and abduction torques as well as on hip and knee flexion angles might indicate the achievement of a movement strategy able to overcome the quadriceps dominance theory. Further reduction detected in the vertical ground reaction forces at T1 and T2 suggests an improvement in the ability to perform “softer” contact with the ground. This could be interpreted as an indirect measure of a better trunk control (18). The immediate effects of proprioceptive stimulation can be found in the posturographic analysis results, which showed the promotion of an activity in both the vestibular and somatosensory apparatus and in COP path x (anterior–posterior direction) following the application of the device in T1 respect to T0 (Figure 3 and Supplementary Data A).

These results together with the ones from the Friedman Test are in agreement with recent findings that detected improvements in motor controls of healthy and pathological individuals including athletes (31, 43) after the application of a vibratory stimulus. Even though localized muscle vibration basic principles are known, their effects and the scenario of all possible applications both in sport and clinics are still poorly investigated (46–50). With respect to this, it is important to highlight that the present study adopted localized muscle vibration, which, by targeting directly the muscle–tendon unit of interest, has the advantage of strongly stimulating the muscle spindles and the excitatory drive from the afferent neurons to the homonymous motor unit pool (46–51). This is further supported by a recent work that investigated the effect of Equistasi® device on the somatosensory pathway through the analysis of high-frequency oscillations, and demonstrated that vibrotactile afference, delivered by the device, were able to interfere with the somatosensory processing (35). A number of limitations should be acknowledged: first of all the small sample size does not allow generalization of the current findings and, in particular, this could be the reason why the statistically significant differences highlighted by the Friedman test in the posturographic parameters were not confirmed in the post-hoc analysis. The trunk motion was not analyzed; therefore, the potential improvement on the trunk control during high-risk postures was not accounted for. The placebo effect of the application of the device cannot be neglected as the current study lacks a comparison with a placebo device, differently from what was previously performed by the authors in an assessment of a group of individuals with Parkinson's disease (37). The study lacks a comparison with a control group performing a different prevention training, especially the most common ones such as muscle strengthening and stretching. No follow-up was performed in order to define the treatment duration effects; joint kinetics was assessed through plantar pressure data combined with 3D video analysis, and thus possibly affected by similar errors as the one reported in the study by Guiotto et al. (16) when compared with stereophotogrammetry and force plates. However, it should be noted that there is agreement toward encouraging the assessment of the biomechanical risk factors (52) directly on the field, especially when complex gestures are analyzed such as sidestep cutting maneuvers.

In conclusion, the overall study demonstrated the possibility to improve the biomechanics of cutting on a group of female elite soccer players through the application of a vibratory based proprioceptive stimulation with the Equistasi® device. Noticeably, this study assessed the biomechanical risk factors associated with ACL injury risk directly on the field, while performing a task more similar to the sport gestures involved in competition, such as sidestep cutting, and which accounts for the majority of non-contact injuries. Results were encouraging when compared with similar studies reporting about the application of injury prevention programs to improve the biomechanics of landing. Future research is needed to support the current findings, especially by including a larger number of athletes, by adding a placebo device for removing any possible placebo effect, by comparing the treatment with currently used strengthening and stretching programs, and by adding further time point evaluations to detect the duration of the treatment effects.