The Effects of a Combined Pre- and Post-Operative Anterior Cruciate Ligament Reconstruction Rehabilitation Program on Lower Extremity Muscle Imbalance

Abstract

This study examined whether the 5-week pre-operative progressive exercise rehabilitation program with weekly monitoring contributed to a significantly lower muscle activity imbalance in the treatment group, both before and immediately after anterior cruciate ligament reconstruction (ACLR), as well as during the next 5 weeks in comparison to the control group. Twelve professional soccer players took part in the study (from among the top three Polish levels of competition) (age: 26 ± 5 years, body mass: 73 ± 7 kg, stature: 180 ± 6 cm, training experience: 15 ± 4 years). The participants were randomly assigned to the treatment group (n = 6) or to the control group (n = 6). Both groups performed the same pre- and post-operative progressive exercise rehabilitation program, while the rehabilitation of the treatment group was extended by supplementary body-weight functional stabilization training. The three-way repeated-measures ANOVA revealed a statistically significant interaction for muscle × group × time (p < 0.0001; F = 24.897; η² = 0.806). The post-hoc analysis for the interaction effect of muscle × group × time indicated a significantly higher muscle activity imbalance for every measured muscle in the control group at any time point than in the treatment group (from p = 0.036 to p < 0.0001). The muscle activity imbalance was significantly higher from the 1st to 4th weeks than in the last week before surgery for quadriceps (p < 0.016 for all) and hamstrings (p < 0.001). However, in the case of gluteal muscles’ activity imbalance, it was significantly higher at every time point of the post-operative phase than in the last week before surgery (p < 0.001). The results of this study showed that the 5-week pre-operative rehabilitation program with weekly monitoring influenced outcomes of the post-operative phase. Nevertheless, performing both pre- and post-ACLR rehabilitation significantly reduced the muscle activity imbalance of lower limbs, but in the case of the quadriceps muscles, not to a sufficient level.

1. Introduction

Rupture of the anterior cruciate ligament (ACL) is common among professional soccer players and a potentially career-threatening injury [1]. A recent study revealed that the ACL incidence rate was 0.4215 per 1000 h of play during matches and 0.0305 per 1000 h of training in Serie A [2]. The time of rehabilitation following a complete rupture of ACL and its reconstruction (ACLR) is highly variable and often requires from 9 to 12 months [3]. Despite the fact that the rate of return to play after ACLR is very high in professional soccer, current evidence shows that game performance is lower up to 2 years after reconstruction [4]. Although surgical techniques are very important, the rehabilitation process plays a crucial role in return to play (RTP) [5].

The ultimate goal of rehabilitation is to restore the function of the reconstructed ACL to meet the specific neuromuscular requirement of a specific sport. Even after a proper ACLR and rehabilitation, the neuromuscular properties are still impaired, leading to increased risk for ACL re-injury [6]. One approach is to conduct combined pre- and post-operative progressive rehabilitation among patients treated for acute injury [7,8]. The rationale for this approach is that optimal pre-operative knee joint function will lead to a faster and safer return of knee joint function in the post-operative phase [7,9]. Eitzen et al. [10] showed that a 5-week progressive exercise rehabilitation program, within 3 months after ACL rupture, was sufficient to significantly improve knee joint function (limb symmetry index and single-leg hop test). However, the authors did not also evaluate knee function during pre-operative rehabilitation. Another study by de Jong et al. [9] indicated that a large pre-operative quadriceps strength deficit results in a significantly lower limb symmetry index at 6 and 9 months after ACLR. Moreover, the study by Failla et al. [11] was the only one to show that pre-operative rehabilitation involving progressive strengthening and neuromuscular training, followed by a post-operative rehabilitation program, resulted in better functional results and RTP rates 2 years after ACLR than the cohort without extended preoperative rehabilitation. Furthermore, Kim et al. [12] found that 4 weeks of pre-operative rehabilitation significantly lowered the deficit in quadriceps strength and improved single-leg hop distance in comparison to patients who did not participate in a pre-operative rehabilitation. Moreover, in addition to the positive effects of preoperative rehabilitation on physical fitness, the psychological factors should also be taken into account as important for return to pre-injury sport activity after ACLR [13,14]. Hence, common psychological barriers, such as the fear associated with the surgery and the course of rehabilitation after it, can be reduced through properly conducted pre-operative rehabilitation [15]. Therefore, it seems necessary to consider both the pre-operative and post rehabilitation phases together as a concept of treatment aimed at performance restoration, which requires careful examination and determination of the optimal approach.

Surface electromyography (sEMG) feedback is used in clinical practice to analyze the magnitude of excitation for a given muscle or muscle group (during activity or at rest) by means of properly positioned surface electrodes on the skin [16]. The data obtained by sEMG analysis provide information on how much effort is required to perform a specific task [17]. The benefits of this technology are useful in the development of sports performance, injury prevention and monitoring the progress of rehabilitation [18,19]. Therefore, EMG analysis in conjunction with physical fitness tests is often used in sports and rehabilitation sciences to provide an insight into how the neuromuscular system behaves under specific exercises, techniques or loads [20,21,22]. The sEMG can be useful during rehabilitation after ACL rupture as the ACL contains mechanoreceptors which affect neuromuscular control of the knee beyond its mechanical function of maintaining knee joint stability [23]. Changes in motor control strategy may reveal differences in proprioception, postural control, movement, and recruitment patterns which can be assessed by sEMG measurements [24].

Since the pre-operative rehabilitation may have a crucial impact on the post-operative rehabilitation protocol, our purpose was to determine whether a progressive 5-week pre-operative exercise rehabilitation program with weekly monitoring would contribute to faster reduction in muscle activity imbalance in the post-operative phase. We hypothesized that the group performing a 5-week rehabilitation program in the pre-operative phase would have significantly lower muscle activity imbalance of the lower extremities both the week before and after the ACLR surgery as well as 5 weeks after surgery.

2. Materials and Methods

2.1. Participants

Twelve professional soccer players took part in the study (from among the top three Polish levels of competition) (age: 26 ± 5 years, body mass: 73 ± 7 kg, stature: 180 ± 6 cm, training experience: 15 ± 4 years). The participants were randomly assigned to the treatment group (n = 6) or to the control group (n = 6). All of them were informed about the experimental procedures and the possible harmful risks and benefits of the conditions and provide written consent before enrolment into the study. All participants suffered injuries during the season and during the game at a time when their motor preparation and sports form were at a high level. None of the participants had previously diagnosed orthopedic contraindications and had no prior injuries. The study was conducted over a period of 24 months due to the need to properly select study participants. The study design was approved by the Bioethics Committee at the Academy of Physical Education in Katowice, Poland. The study sample size was calculated a priori based on a statistical power of 0.8, an effect size of 0.35, and a significance level of <0.05, taking lower-limb muscle activity imbalance as a reference variable. A minimum sample size of individuals was obtained (G*Power, Dusseldorf, Germany).

2.2. Rehabilitation Procedures

Both groups performed the same pre- and post-operative progressive exercise rehabilitation program, while the rehabilitation of the treatment group was extended by supplementary body-weight functional stabilization training (

Table 1. Description of exercises used in supplementary body-weight functional stabilization training.

Exercise	Description
Single Leg Stance with Knee Raise	Start position: Stand tall and wide. Movement: Lift knee up toa 90-degree angle in the hip while simultaneously raising the arms above the head.
Single Leg Stance with Knee Raise and Lateral Quick Step	Same as the single leg stance with knee raise but with a quick lateral step between each repetition.
Single Leg Stance with Knee Raise and Linear Quick Step	Same as the single leg stance with knee raise but with a quick linear step between each repetition.
Single Leg Romanian Deadlift	Start position: Stand tall and wide. Movement: Lift leg and reach back behind towards a wall. Feel the weight going back towards your heel as your body tilts forward as if bowing forward, until parallel to the floor. Simultaneously raise the arms above the head. The opposite knee is slightly bent at the same degree during the whole movement.
Single Leg Romanian Deadlift with Lateral Quick Step	Same as the single leg Romanian deadlift but with a quick lateral step between each repetition.

). Moreover, in each pre- and post-operative rehabilitation session, manual therapy techniques were used to regulate soft tissue tension and drainage procedures to expand the range of motion. The equipment used to improve muscle activity in the injured limb and to compensate for the imbalance in relation to the healthy limb were: elastic bands, gymnastic balls, an element of external loads (dumbbells and kettlebells), adequate to the capability of each participant. Participants were allowed to put full weight on their limbs and walk freely. Due to isolated ACL damage in the examined participants, walking without elbow crutches and without stabilization (brace) was permitted in order to avoid deepening of muscle asymmetry and the creation of intra-knee adhesions.

2.3. Before Surgery

Orthopedic examination, the extent of damage to the knee as seen in magnetic resonance imaging (MRI) and the result of dynamic arthrometry determined the intensity of the pre-operative phase of rehabilitation for which all participants were qualified. The pre-operative rehabilitation sessions began when the participant recovered impairments related to swelling and range of motion deficit. A five-week waiting period before the surgery was necessary due to the damaged areas called bone bruises, visible in the subchondral bone by means of MRI. These spaces should regain their proper structure and absorption features before the tunnels are drilled in preparation for the transplanted tendon. This period of time is recommended to avoid the bone tunnel enlargement syndrome as an undesired side effect of ACLR [25]. Because there is no clear consensus on rehabilitation in the early stages following an ACL injury and preparing for reconstruction, participants of this study performed a supervised rehabilitation program similar to that described by Kim et al. [12]. This program consisted of 20 min stationary cycling and the following exercises performed in two sets of 10 repetitions with a 60 s rest interval between sets: seated leg flexion and extension, supine wall slides, short arc leg extensions, straight leg raises, isometric leg curls, hip thrusts, leg presses, and half squats. Moreover, three sets of 30 s single-leg standing and standing on a balance board were performed.

2.4. Surgery

All patients analyzed were operated on using the same ACLR method, by the same operating team. Transplanted allograft (tibialis anterior tendon) was purchased from the local Tissue Bank with similar diameter and length for all the recipients. Surgeries were performed under epidural anesthesia to allow early movement and walking with crutches even on the same day of surgery. Not harvesting the autograft material from the patient for implantation resulted in relatively diminished pain sensation, encouraging physical activity.

2.5. Post-Surgery

On the first day after the procedure, the participants were instructed by physiotherapists about possible elevation exercises, antithrombotic prophylaxis, non-isolated isometric exercises, joint mobility, static limb loading and walking on crutches to reduce loads on the operated limb according to pain tolerance. Both groups performed the same post-operative progressive exercise rehabilitation program according to the recommendations of the Melbourne ACL Rehabilitation Guide 2.0, while the rehabilitation of the treatment group was extended through supplementary functional body-weight stabilization training (Table 1).

2.6. Supplementary Functional Body-Weight Stabilization Training

Functional stabilization training was used, using the individual’s body weight, taking into account the aspect of physical discomfort associated with the surgery. Protocols of functional stabilization with an emphasis on one–legged standing and muscle tension far from the place of damage, such as the muscles of the trunk, shoulder girdle and upper limbs, were implemented.

The support program was based on the execution of five selected functional stabilization exercises (Table 1) in a specific, constant sequence. In the period preceding the operation, the program started in the third week after the injury and was carried out for a period of 5 weeks before the surgery (until the day of its performance). The same program was carried out after the surgery. It began 4–7 days after surgery, immediately after the early post-surgical symptoms had subsided. The exercises were performed for both limbs.

In both periods (pre and postoperative), the load was gradually increased by adding volume. The training progression included an additional set in each consecutive week of the experiment, from two sets in the first week to six sets in the fifth week of 10 repetitions of each exercise with a 60 s rest between sets. Each participant performed exercises every day, at the same time of the day, with full control of the technique of the movements performed.

2.7. Electromyographic Measurements Procedure

The sEMG data were registered bilaterally from the tight muscles (quadriceps-Q, hamstrings-H, and gluteal-G) using shorts similar to elastic sportswear with properly positioned pairs of bipolar surface electrodes for non-invasive assessment of the electrical muscle activity (Myontec Ltd., Kuopio and Suunto Ltd., Vantaa, Finland). The electrodes were placed on the inner surface of the shorts and recorded the mean rectified sEMG signal on the distal part of the tight muscles, and the reference electrodes were positioned longitudinally along the left and right lateral sides (above the tractus iliotibialis). The shorts provide reliable and repeatable data [17,26,27] with the intra-class correlation for Q, H, G measurements ranging from 0.83 to 0.87 [26]. It has to be highlighted that the shorts are unable to assess particular muscle excitation (e.g., biceps femoris) within a selected muscle group (e.g., hamstrings). For example, the electrodes around the H muscle area collected data from the semimembranosus, semitendinosus and both heads of biceps femoris. The raw sEMG signals were registered with a high sampling frequency of 1000 Hz and a frequency band of 50–200 Hz (−3 dB). Next, they were rectified and averaged for each 100 ms interval without overlapping. These 10 Hz data were stored in ASCII format in the memory module and transferred to the PC via custom software (Muscle Monitor, Myontec Ltd., Kuopio and Suunto Ltd., Vantaa, Finland) for analysis. From the 10 Hz sEMG data, the mean rectified value was calculated from a 1 s window during a constant torque signal. Six channels were recorded (three from each limb). To ensure proper signal conduction, the electrodes were moistened with tap water before donning the shorts. When wet, the membrane covering the electrodes prevents the electrode from drying out at the skin-to-electrode interface. The mean lower-limb sEMG activity imbalance represented a relative (percentage) ratio between lower limbs for each muscle group and was used for further analysis. Maximal voluntary isometric contraction (MVIC) measurement was performed according to the SENIAM procedure [28]. After the warm-up, participants were tested in MVIC of the studied muscles as described in

Table 2. Characteristics of maximal voluntary isometric contraction tests for each muscle group.

Muscle Group	Description
Quadriceps	In a sitting position with a knee and hip joint angle flexed at 90° with hands kept across the chest. Participants placed their feet on the pad and performed a maximum bilateral voluntary isometric contraction. Participant started the contraction via a visual cue. During the maximum isometric effort, a chair with a non-movable back supported the trunk. The contraction was held for 3 s.
Hamstrings	In a prone position with knees flexed to 45°. Manual resistance was provided by the two instructors at the distal portion of the leg towards extension, while the third instructor stabilized the hip and the pelvis. The participants then performed a maximum bilateral voluntary isometric contraction for 3 s.
Gluteal	In a prone position, with the knee flexed to 90°. Manual resistance was provided by the two instructors in the downward direction through the ventral foot and distal posterior thigh while the pelvis was stabilized by the third instructor. The participants then performed a maximum bilateral voluntary isometric contraction by lifting their thighs and leg off the table as hard as possible. The contraction was held for 3 s.

.

2.8. Statistical Analysis

All statistical analyses were performed using SPSS (version 25.0; SPSS, Inc., Chicago, IL, USA). The results were expressed as means with standard deviations (±SD) and 95% confidence intervals. Statistical significance was set at p < 0.05. The normality of data distribution was examined using the Shapiro–Wilk tests, Mauchly’s test of sphericity was conducted to test for the homogeneity of data and if it did not comply (p < 0.05), the Greenhouse–Geisser adjustment value was used. The percentage of differences in muscle activity imbalance between each muscle group (Q; H; G) within treatment group was analyzed by a three-way (muscle × pre_post × time; 3 × 2 × 5) ANOVA with repeated measures. Similarly, the differences between the treatment and control group were analyzed with a three-way (muscle × group × time; 3 × 2 × 6) ANOVA with repeated measures. Effect sizes for main effects and interaction were estimated by calculating the partial eta squared (η²). In the event of a significant main effect, post-hoc comparisons were conducted using the Bonferroni test.

3. Results

3.1. Between Group Differences

The three-way repeated-measures ANOVA indicated a statistically significant interaction for muscle × group × time (p < 0.001; F = 24.897; η² = 0.806) (

Table 3. Difference in muscle activity imbalance in pre- and post-operative phase.

	Pre-Operative Phase					Post-Operative Phase
Week	5	4	3	2	1	1	2	3	4	5
Muscle Imbalance ([%]; Mean ± SD; 95%CI)											p (Muscle × Time
Treatment Group
Q	60 ± 6 (54 to 67)	54 ± 7 (46 to 61)	47 ± 5 (42 to 52)	40 ± 4 (35 to 44)	30 ± 3 * (27 to 33)	71 ± 4 *# (66 to 75)	62 ± 2 *# (60 to 64)	54 ± 3 *# (50 to 57)	45 ± 3 *# (41 to 48)	35 ± 3 * (32 to 38)	<0.001
H	33 ± 4 (29 to 37)	29 ± 3 (26 to 32)	23 ± 3 (20 to 27)	19 ± 2 (16 to 22)	16 ± 2 * (14 to 17)	57 ± 4 *# (53 to 61)	44 ± 3 *# (40 to 47)	32 ± 4 *# (27 to 36)	26 ± 2 *# (24 to 29)	19 ± 3 *# (16 to 22)
G	27 ± 2 (25 to 29)	22 ± 2 (19 to 24)	18 ± 2 (15 to 20)	17 ± 2 (14 to 19)	11 ± 1 * (10 to 12)	58 ± 4 *# (53 to 62)	45 ± 5 *# (40 to 50)	32 ± 4 *# (27 to 36)	19 ± 3 *# (16 to 23)	14 ± 2 *# (11 to 16)
Control Group
Q					44 ± 6	78 ± 3 #	70 ± 2 #	66 ± 2 #	59 ± 2 #	51 ± 2	<0.001
H					33 ± 4	70 ± 3 #	62 ± 3 #	54 ± 3 #	48 ± 2 #	42 ± 2
G					19 ± 3	64 ± 2 #	54 ± 2 #	48 ± 4 #	44 ± 3 #	38 ± 3 #

Q—Quadriceps muscles, H—hamstring muscles, G—gluteal muscles; SD—standard deviation, CI—confidence interval; * significant difference between treatment and control group in corresponding week <0.05; ^# significant difference in comparison to the last week before surgery within group <0.05.

and Figure 1). The post-hoc analysis for the interaction effect of muscle × group × time showed a significantly higher muscle activity imbalance for every measured muscle in the control group at any time point than in the treatment group (from p = 0.036 to p < 0.001).

3.2. Within Control Group Differences

The post-hoc analysis for the interaction effect of muscle × group × time showed a significantly higher muscle activity imbalance for Q in comparison to H and G at every time point (p < 0.018). Moreover, H muscle activity imbalance was significantly higher than G in the last week before surgery and the 1st, 2nd and 4th weeks in the post-operative phase (p < 0.001, p = 0.035, p = 0.004, p = 0.046; respectively) but not in the 3rd and 5th weeks (p = 0.077, p = 0.148; respectively).

The muscle activity imbalance was significantly higher from the 1st to 4th weeks than in the last week before surgery for Q (p < 0.016 for all); H (p < 0.001). However, in the case of G muscle activity imbalance, it was significantly higher at every time point of post-operative phase than in the last week before surgery (p < 0.001).

Moreover, the Q and H muscle activity imbalance significantly and successively decreased from the 1st to last weeks in the post-operative phase (1st vs. 2nd; 2nd vs. 3rd; 3rd vs. 4th; 4th vs. 5th) (p < 0.02 for all; p < 0.04 for all; respectively). As for G, muscle activity imbalance decreased in the 1st vs. 2nd weeks (p = 0.001) and the 4th vs. 5th (p = 0.034) but not in the 2nd vs. 3rd (p = 0.06), and the 3rd vs. 4th (p = 0.172).

3.3. Within Treatment Group Differences

The three-way repeated-measures ANOVA indicated a statistically significant interaction for muscle × pre_post × time (p < 0.001; F = 20.387; η² = 0.803) (Table 3).

The post-hoc analysis for the interaction effect of muscle × pre_post × time showed a significantly higher muscle activity imbalance for Q in comparison to H and G at every time point of the pre- and post-operative phases (p < 0.004). Moreover, H muscle activity imbalance was significantly higher than G from the 5th to 2nd and the last week in pre-operative phase (p = 0.031; p = 0.005; p = 0.015; p = 0.001; respectively) and in the 4th and 5th week in the post-operative phase (p = 0.009 and p = 0.001; respectively).

Furthermore, the Q muscle activity imbalance significantly and successively decreased from the 5th to the last week in the pre-operative phase (5th vs. 4th; 4th vs. 3rd; 3rd vs. 2nd; 2nd vs. last; p < 0.019 for all). It was similar for H and G muscle activity imbalance (p < 0.034; p < 0.015; respectively), except for the 3rd vs. 2nd weeks (p = 0.059; p = 1; respectively).

The Q muscle activity imbalance significantly and successively decreased from the 1st to 5th weeks in the post-operative phase (1st vs. 2nd; 2nd vs. 3rd; 3rd vs. 4th; 4th vs. 5th; p < 0.04 for all). Similarly, for H muscle activity imbalance (p < 0.035 for all) and for G muscle activity imbalance (p < 0.048) except for the 4th vs. 5th weeks in the post-operative phase (p = 0.076).

The Q muscle activity imbalance was significantly greater in the 1st week of the post-operative phase than at all other time points (p < 0.006 for all), while in the 5th week of the post-operative phase, it wasn’t significantly different than in the last week of pre-operative phase (p = 0.057). However, in case of H and G muscle activity imbalance, it was significantly higher in the 5th week of the post-operative phase than in the last week of the pre-operative phase (p = 0.047 and p = 0.01; respectively).

4. Discussion

The main finding of this study was the fact that the 5-week pre-operative progressive exercise rehabilitation program with weekly monitoring contributed to a significantly lower muscle activity imbalance in the treatment group, both before and immediately after ACLR, as well as during the next 5 weeks in comparison to the control group. Moreover, the results showed that the Q muscle activity imbalance was significantly higher in comparison with H and G for both groups. Moreover, in the case of Q, although 5 weeks of pre-operative and post-operative rehabilitation was sufficient to regain the imbalance noted in the last week prior to surgery, it was still not satisfactory. Conversely, in H and G, the muscle activity imbalance was still significantly higher than that prior to ACLR, but reached a satisfactory level at the end of the rehabilitation program.

These results provide valuable information for physiotherapists and physicians concerning the course of the rehabilitation process after ACLR, and indicate the need for rehabilitation immediately after an injury until the planned surgery is carried out and its subsequent continuation. First of all, the 5-week pre-operative rehabilitation program contributed to significantly less muscle imbalance each week after ACLR compared to the control group. Secondly, in the post-operative phase there was a trend of faster imbalance reduction in the treatment group than the control group. Thirdly, the greatest imbalance in muscle activity among the measured muscle was found in Q, so this muscle group may need special attention throughout the rehabilitation process. Lastly, despite the fact that 5-weeks of post-operative rehabilitation successfully restored Q muscle activity imbalance to a similar value as that of the last week before surgery, it still amounted to approximately 35%. An asymmetry above 20% (side to side) is implicated in abnormal movement patterns and asymmetrical limb loading strategies following ACLR which may indicate a high risk of ACL reinjury [29,30,31]. Thus, it is imperative that patients continue the rehabilitation process with addition work aimed at improving Q function. On the other hand, in the case of G and H, despite the fact that it was not possible to obtain the same degree of imbalance as before the surgery, it seems that a satisfactory value was achieved. In summary, it can be concluded, that the use of sEMG in enables the monitoring of rehabilitation and its course, providing valuable feedback that can be used to individualize the entire process and inform decision-making to optimize sports participation following ACLR.

Currently, there is no clear consensus on rehabilitation in the early stages following an ACL injury and preparation for reconstruction. There is no recommendation for the selection of exercises and their exact volume and intensity. For example, in a study by Eitzen et al. [10], the pre-operative rehabilitation was based on the principles outlined in the American College of Sports Medicine position stand for progression models for resistance training for healthy adults. Nevertheless, the authors demonstrated that the 5-week progressive program combined with resistance training, plyometric exercises, and general exercises for balance and stability with perturbation training significantly improved knee function. Moreover, Shaarani et al. [32] showed a beneficial effect of 6-week gym- and home-based pre-operative rehabilitation on the single-legged hop test, which was also maintained 12 weeks after ACLR. Our study confirms previous findings and provides evidence for a positive impact of pre-operative rehabilitation on post-operative outcomes. After the 5-week pre-operative rehabilitation program, the rehabilitation course after the ACLR was faster for each of the measured muscles compared to the control group (e.g., Q by 36% vs. 27%; treatment vs. control group; respectively). It seems that this result can be explained by the “muscle memory” phenomenon. This phenomenon is related to the fact that individuals who have trained previously gain muscle mass and strength faster after retraining [33,34,35,36]. Therefore, it seems this might be achieved through accelerated adaptations to the rehabilitation process, such as a progressive inhibition of overactive thigh muscles due to reduced stiffness and increased range of motion. Furthermore, increased strength of particular muscles and improved intramuscular coordination could decrease muscular imbalance. Hence, performing a rehabilitation program immediately after the injury and its continuation until ACLR seems necessary and may yield meaningful benefits in the post-operative period.

Our study found the greatest muscle activity imbalance in Q among the measured muscles at each time point in the rehabilitation course. While the rehabilitation program used in this study was effective in reducing the muscle activity imbalance in Q, it did not lead to a satisfactory level. It is possible that such results are the result of the program applied, which should be even more focused on Q development. Taking this into account, it seems that the most important goal of both the pre- and post-operative phases should be to optimize the function of this muscle. This seems obvious because Q and H play a vital role in terms of moving and stabilizing the knee joint. Nevertheless, our study shows that the development of Q muscles can be even more important than H in regaining muscle activity balance. Moreover, the pre-operative Q strength is considered to be the single most important predictor for knee function 2 years after ACLR [7]. Therefore, these 5-week phases of pre- and post-operative rehabilitation are insufficient and if possible, should be lengthened.

The highest decrease in muscle activity imbalance from the 1st to 5th weeks of the post-operative phase was reported in G muscles (by 44% in the treatment group). Most likely it is related to the fact that a much lower intensity and volume of exercise can be used on Q and H due to the limited knee range of motion immediately after the injury and ACLR. Nevertheless, our rehabilitation program proved to be sufficient for reducing H muscle activity imbalance, again indicating that Q requires the most attention.

This study has several limitations that are important to mention. Employment of allografts is not common practice for primary ACLR. Allografts are transplants generally designated for revision surgery but are sometimes the preferred option to eliminate post-harvesting symptoms. One of the most relevant of them according to the literature is their influence on the function of the lower extremities, specifically asymmetrical muscle strength noticeable in the objective isokinetic evaluation, which is detrimental to the injured and ipsilateral harvested side. Moreover, the highly specific participants characteristics (highly trained men) may have affected the obtained results and therefore should not be replicated in other groups. Furthermore, no physical fitness tests were performed, therefore it is not known whether the reduction of imbalance in lower limb muscle activity contributed to the improvement of sports performance.

The results of this study showed that the 5-week pre-operative rehabilitation program with weekly monitoring influenced outcomes of the post-operative phase. Nevertheless, performing both pre- and post-ACLR rehabilitation significantly reduced muscle activity imbalance of lower limbs, but in the case of the Q muscles, not to a sufficient level. Therefore, we suggest that the postoperative phase should last longer and focus more on the development of the Q muscles than the one used in our study.