Evaluating the effectiveness of the association between different biofeedback methodologies and therapeutic exercise protocols concerning pain, it seems that just two studies ^[1][2][32,37] with a low bias risk underlined a significantly better improvement in favor of the experimental groups that followed multimodal interventions (mirror and verbal feedback or electromyography (EMG) biofeedback and taping); this does not let us discriminate which feedback type significantly impacts the result. The other three randomized controlled trials (RCTs)RCTs ^[3][4][5][31,35,38] found actual effectiveness in this intervention. However, the type of control groups used does not let us draw any conclusion because Baldon et al. (2014) ^[3][31] used different therapeutic protocols between interventions and control groups and found a significant improvement only after the follow-up (12 weeks). Emamvirdi (2019) ^[4][35] and Ebrahimi (2021) ^[5][38] accounted only for the controls with an educational treatment. Instead, Aghakeshizadeh et al. (2021) ^[6][36] found a significant improvement in the case group compared to the control group treated only with therapeutic exercise and did not find differences if compared to a third group with a different protocol (exercise and internal focus) if compared to the experimental protocol (exercise and external focus).

2. Biofeedback and Therapeutic Exercise Association on Functionality

The efficacy of functionality was measured through the Kujala Anterior Knee Pain Scale (AKPS)KPS, and results were significant in only two of the works ^[2][7][33,37]. These conclude that the improvement in the score of the Kujala Scale is significantly in favor of the experimental group only if the variable evaluated in the follow-up in the short term (6 or 12 weeks) is considered. In comparison, there are no differences between the groups if done intra-intervention (2 weeks), post-intervention (4 weeks), or middle-term follow-up. In addition, in this case, though, being the proposal multimodal (mirror, verbal and proprioceptive feedback or EMG biofeedback and taping), it is impossible to discriminate which protocol type is most relevant to the outcome. As highlighted for pain, Ebrahimi et al. (2021) ^[5][38] noticed a significant improvement for the experimental group also on functionality (AKPS), but with the same “problems” mentioned above in the control group. Aghakeshizadeh et al. (2021) ^[6][36], similar to the outcome pain, registered a significant improvement in the AKPS for the intervention group to the control one performing exclusively exercise, while compared with the third group, assigned to a different protocol, did not find differences.

3. Biofeedback and Therapeutic Exercise Association on Strength

3.1. Hip Abductors

Only one study ^[8][34] showed a significant improvement in efficacy in favor of the intervention group on the hip abductors muscle strength, underlining that an association between visual and auditory feedback with therapeutic exercise protocol can favor strength improvement in these areas if compared to exercise only. In addition, Aghakeshizadeh et al. (2021) ^[6][36] found some improvements but, similar to pain and functionality outcomes, these were found only in the comparison between the exercise and external focus group compared to the control group treated with therapeutic exercise, which was non-significant when compared to the group with exercise and internal focus. The RCT of Baldon et al. (2014) ^[3][31] presented critical findings because there was a significant improvement in hip abductors’ strength related to the experimental group, but it must be noted that this group was doing concrete work on these muscle districts. On the other side, only Rabelo and colleagues (2017) ^[7][33] did not find differences between the two groups analyzed, considering the bias risk. Strangely, Emamviridi et al. (2019) ^[4][35] did not find any differences between the groups analyzed, even if the control one did not exercise; this can be explained by the fact that the duration of the protocol (6 weeks) was not sufficient to register adaptations of this type, and also by the nature of the intervention itself.

3.2. Knee Extensors

Regarding knee extensors, three RCTs ^[2][8][9][28,34,37] found a significant improvement in favor of the experimental group regarding this variable, even with a high bias risk. Two works ^[2][9][28,37] found that the strength improvement was verified only after the intervention (4 or 8 weeks) and at the following eventual follow-up (6 weeks). In comparison, there were no differences between the groups at the evaluation intra-intervention (2 or 4 weeks), underlining the necessity of a longer protocol duration to obtain a positive adaptation. Bennell et al. (2010) ^[10][30] found a significant improvement in favor of the intervention group related to the eccentric strength evaluated just after the intervention (6 weeks), but was canceled at the following follow-up (14 weeks); instead, in terms of the concentric and isometric strength, there were no found differences between the two groups at post-intervention nor follow-up. Moreover, all the studies that reported knee extensor strength improvement used visual feedback, especially the EMG biofeedback, in the protocols proposed for the interventions. Finally, in the studies run by Baldon (2014) ^[3][31] and Rabelo (2017) ^[7][33], even if this former had some bias risk, they did not find differences between the groups studied.

4. Biofeedback and Therapeutic Exercise Association on Lower Limb Kinematic

Aghakeshizadeh et al. (2021) ^[6][36] reached concrete conclusions regarding the lower limb kinematic of the knee valgus, underlining significant improvements in favor of the experimental group (exercise and external focus) compared to both in the control group, with just exercise, and the group doing a protocol with the association of exercise and internal focus. Roper’s study (2016) ^[1][32], with a low bias risk, also underlines a reduction of the knee valgus related to IC while running, which was significative only in the experimental group and exclusively after the follow-up (4 weeks); this underlines that the intervention length (2 weeks) is not sufficient to obtain this adaptation. Significant improvement related only to the intervention group was encountered by Emamvirdi et al. (2019) ^[4][35], even when a patient education protocol was administered to the control group, and the RCT may have had some biases. Both studies use an intervention protocol composed of therapeutic exercise, mirror and verbal feedback. Rabelo et al. (2017) ^[7][33] did not find differences in the examined groups, but the short duration of the intervention (4 weeks) is a limiting factor in the development of this adaptation; moreover, theis study presents some biases.

5. Biofeedback and Therapeutic Exercise Association on VM and VL EMG Activity

In the included studies, different aspects relating to the influence of the intervention on the VM (vastus medialis) and VL (vastus lateralis) EMG activity of the quadriceps muscle were discussed. Dursun et al. (2001) ^[11][27] registered significant improvements in the VM medium EMG activity in favor of the experimental group during all the recordings. Regarding VL, they found that the medium EMG activity increased in favor of the intervention group only with the evaluation intra-intervention (4 weeks); instead, no differences were registered in the remaining groups (8 and 12 weeks). Any differences between the analyzed groups regarding the maximum VM and VL EMG activity were described. However, it should be highlighted that theis study presents a high bias risk. Ng et al.’s (2008) ^[12][29] RCT, presenting a high bias risk, found a significative optimization of the VMO/VL EMG ratio in favor of the experimental group. Finally, the study done by Bennell et al. (2010) ^[10][30] showed a significant improvement in the intervention group if compared to the control one, exclusively associated with the VMO-VL EMG onset timing in going down the stairs during the evaluation post-intervention (6 weeks) that disappeared at the following follow up (12 weeks), no differences were described in this outcome related to ascending the stairs in both the evaluations. The analysis of theis study’s results had to consider the possible presence of bias. It is important to underline how the cited studies associated therapeutic exercise with EMG-biofeedback with eventual auditory feedback in the experimental group. This combination can represent a valuable aid to optimize VM and VL EMG activity.