Evaluation of eccentric and concentric force during vertical jump after anterior cruciate ligament reconstruction: a comparative study

Abstract

Background

Anterior cruciate ligament reconstruction (ACL-R) significantly affects lower limb biomechanics, particularly during dynamic movements like jumping. The countermovement jump (CMJ) is a commonly used test to evaluate force production and asymmetry in post-ACL-R individuals. This study aimed to compare eccentric and concentric forces during the CMJ between ACL-R patients and healthy controls.

Methods

This retrospective study included 56 ACL-R patients and 47 recreational athletes. All participants performed CMJs on a dual force plate to measure eccentric mean force (ECC), Concentric mean force (CON), and limb symmetry index (LSI). A mixed ANOVA was performed to compare ECC LSI and CON LSI between the ACL-R and control groups.

Results

CMJ height was significantly lower in the ACL-R group compared to the control group (-25.4%, p < 0.001), with a lower ECC LSI (-7.7%, p < 0.001) and CON LSI (-10.0%, p < 0.001).

Conclusion

Six months after ACL-R, both eccentric and concentric force production remain compromised, with significant inter-limb asymmetries. These findings highlight the need for targeted rehabilitation strategies to restore force symmetry and optimize return-to-sport readiness.

What is Known

• ACL reconstruction (ACL-R) affects lower limb biomechanics, particularly force production and asymmetry during dynamic movements like the countermovement jump (CMJ).

• Previous studies have shown that concentric force asymmetry is prevalent after ACL-R, but findings on eccentric force asymmetry remain inconsistent.

What is New

• This study demonstrates that six months after ACL-R, both eccentric and concentric force production remain impaired, with significant inter-limb asymmetries.

• Eccentric force asymmetry is as pronounced as concentric asymmetry, emphasizing the need for targeted rehabilitation strategies.

• These findings suggest that conventional return-to-sport criteria may not adequately address residual force deficits, necessitating more specific interventions to restore symmetry.

Introduction

Anterior cruciate ligament (ACL) injuries are among the most common and devastating sports-related knee injuries, particularly prevalent in activities involving pivoting, jumping, and sudden changes in direction [1, 2]. The ACL plays a crucial role in ensuring knee joint stability, especially during dynamic movements [3]. Thus, ACL injuries can significantly impact an individual’s athletic performance and overall knee function [4]. ACL reconstruction (ACL-R) is often pursued to restore knee stability and function, particularly in individuals who wish to return to high-demand physical activities such as landing and pivoting sports [2]. While ACL-R surgery has become increasingly common, the process of rehabilitation and return to sport remains complex and multi-faceted [5–7]. One important aspect of this process is the assessment of the biomechanical changes that occur post-surgery, particularly during functional tasks such as jumping [7, 8].

The countermovement jump (CMJ) is a widely used functional test to assess lower limb biomechanics, power generation, and symmetry [9]. Successful CMJ performance depends on several physical and functional determinants, including neuromuscular coordination, balance, joint mobility, and lower-limb strength, particularly of the quadriceps and plantar flexors [9, 10]. The eccentric phase involves the rapid deceleration of the body as it descends towards the ground, which requires eccentric muscle contractions to brake the descent and store kinetic energy. The concentric phase of the jump refers to the upward movement, where muscles restore the kinetic energy and contract to generate vertical force and propel the body off the ground [10–13]. Understanding the characteristics of the CMJ following ACL-R is essential for optimizing rehabilitation protocols and guiding return-to-sport decisions [11].

Several studies have investigated performance and kinetic characteristics during the CMJ in individuals with ACL-R [11–14]. Jump height is likely to be lower after ACL-R even when patients were cleared for returning to sport. In terms of kinetic analysis, conflicting results were found for the asymmetry during the eccentric phase of the CMJ while greater asymmetry was consistently found for the concentric phase after ACL-R. Variability in surgical techniques, rehabilitation protocols, and testing methodologies complicates result’s comparisons and generalizability. While Kotsifaki et al. provides crucial insights into jump performance asymmetries, its limitations—including a homogenous male-only sample and a small cohort of non-professional athletes—highlight the need for more inclusive and diverse research [11]. The most studied kinetic parameter during the performance of a CMJ is the impulse during both eccentric and concentric phases. The impulse is the product of the force (vertical ground reaction force) by contact time. Several studies have emphasized the relevance of impulse analysis in post-ACL-R populations, particularly to assess neuromuscular recovery and inter-limb asymmetries during vertical jump tasks [11, 14, 15, 16]. However, as the performance is likely to be lower after ACL-R, the contact time should increase, potentially hiding or underestimating the differences of impulses between participants after ACL-R and controls. It refers to the quantity of force developed during the eccentric phase or concentric phase of the CMJ. While the impulse is a crucial parameter for monitoring the recovery of a specific athlete performance, it can be a limit when comparing groups, especially when differences in contact time are observed. As the performance is likely to be lower after ACL-R, the contact time should increase, potentially hiding or underestimating the differences of impulses between participants after ACL-R and controls. An analysis of the mean force is therefore recommended when comparing different groups to avoid this bias.

The main objective of the present study was therefore to add data to the current body of knowledge for the performance and kinetic characteristics of the CMJ after ACL-R. The first aim was to assess the performance and the force value during the eccentric and concentric phases of the CMJ between participants after ACL-R and healthy controls. The second aim was to compare the asymmetries during the eccentric and concentric phases of the CMJ in participants after ACL-R and healthy controls. We hypothesize that the jump height and the absolute force values (both eccentric and concentric) will be lower in participants after ACL-R and that the asymmetries will be higher in participants after ACL-R than in healthy controls.

Methods

Study design

This retrospective comparative study received approval from the Clinic of Domont ethics committee (IRB PCE309N-003). All patients were treated in agreement with the Declaration of Helsinki and provided written informed consent. The STROBE guidelines (Strengthening the Reporting of Observational studies in Epidemiology) were followed in this study.

Participants

Data were retrieved from the records of the Clinic of Domont Orthopedic Surgery department between January 2023 and March 2024.

From the original 140 ACL-R patients, participants were screened based on inclusion/exclusion criteria (Fig. 1), a total of 56 recreational athletes (32 females and 24 males) after an ACL-R surgery were recruited (ACL-R group). The ACL-R was performed by 3 orthopedic surgeons following the same surgical procedure. All ACL reconstructions were performed arthroscopically using a standardized technique. A quadrupled hamstring tendon autograft (semitendinosus and gracilis) was used in all patients. Femoral fixation was achieved using an adjustable-loop cortical suspensory device, and tibial fixation was performed with a bioabsorbable interference screw. No additional ligament reconstructions were performed. Participants were retrieved from the 3 rehabilitation centers following the same ACL-R rehabilitation program. Although the rehabilitation protocol was not strictly standardized across centers, all patients followed programs grounded in similar clinical principles aimed at restoring function after ACL-R. However, specific details regarding training content (e.g., eccentric overload, single-limb tasks), intensity, and adherence were not documented, which limits our ability to evaluate their potential impact on the study outcomes. Inclusion criteria were: (1) participants aged between 18 and 40 years old, (2) ACL-R using a hamstring graft, (3) testing performed between 5 and 7 months after the surgery, and (4) body mass index (BMI) lower than 30 kg.m². Exclusion criteria were: (1) history of contralateral or ipsilateral ACL injury, (2) history of lower-limb injury in the 6 months before testing, (3) pregnancy, and (4) missing information or data from the medical record and associated injuries other than meniscal injuries (treated either by meniscal suture or meniscectomy) such as osteochondral tears and/or other complex ligament injuries.

Fig. 1.

Flowchart of participant inclusion and exclusion in the ACL-R group

A total of 47 healthy recreational athletes (36 female and 31 male) were recruited for the control group (Control group). Inclusion criteria were: (1) participants aged between 18 and 40 years old, and (2) BMI lower than 30 kg.m². Exclusion criteria were: (1) history of ACL injury, (2) history of lower-limb injury in the 6 months before testing, (3) pregnancy, and (4) missing information or data from the medical record.

All participants in the control group had a Tegner Activity Scale score greater than 4 and a Marx Activity Scale score greater than 6.

Assessment protocol and data analysis

The participant’s height (cm), mass (kg), age and sport activity level (pre-injury for the ACL-R participants) (Tegner score and Marx activity scale) and dominant leg for the Control group were recorded before the start of the tests. Before the jump testing, participants performed a standardized warm-up that consisted of 5 min on a bicycle, 5 two-legged squats and 2 submaximal CMJ. Participants were instructed to stand on two legs on the force platforms. With their hands on their hip, they were asked to perform a CMJ, jumping as fast and high as possible without upper limb movement [17]. Three trials were recorded, and the best was used for analysis [18].

The evaluation of CMJ was performed by dual force plate (Delta Force Plate; Kinvent; V2; 2000 Hz; Montpellier; FRANCE). The eccentric mean force (ECC) and concentric mean force (CON) were calculated for both legs during the eccentric and concentric phases of the CMJ. The data were collected with Kinvent Physio App (v2.7.1). The mean forces were normalized by the mass (N/kg) to allow between group comparisons. The Limb Symmetry Index (LSI) was also calculated (injured leg [IL]/non-injured leg [NIL]*100 for the ACL-R group and non-dominant leg [NDL]/dominant leg [DL]*100 for the Control group).Jump height was calculated using the flight time method, derived from force plate data, which has been shown to provide reliable estimates of vertical jump performance.

Statistical analysis

Descriptive statistics for both groups were generated. Specifically, the means and standard deviations of each variable in our population, namely: age, height, weight, jump height, Tegner and Marx scores, ECC LSI and CON LSI.

To perform the statistical analyses, we set a confidence interval of 95% and a significance level of α = 0.05. These analyses were performed using the JASP^® software. To determine homogeneity between the two groups, a T-test was performed for the quantitative variables that follow a normal distribution (size, weight, age and jump height).

After verification of the normality of the distribution and homogeneity, we conducted mixed ANOVA with 3 factors (intra-subject: LEG [injured/non-dominant vs. contralateral]; PHASE [ECC and CON]; inter-subject: GROUP [ACL-R vs. Control]). Post-hoc analyses were performed with a Holm correction for multiple comparisons.

Results

Participants

From the original 140 ACL-R patients considered for inclusion (Fig. 1), 84 were excluded for the following reasons: 31 patients did not meet the inclusion criteria (e.g., age outside the range, different graft type, insufficient post-op delay, or high BMI),50patients had incomplete or missing data, 2 patients declined to participate,1 patient was excluded due to language comprehension difficulties (English and/or French).

Demographic data

Participants in the ACL-R group were significantly older than those in the control group (28.3 ± 5.1 vs. 24.5 ± 1.9 years; p < 0.001). No significant differences were observed between groups for sex distribution (p = 0.352), height (p = 0.99), or weight (p = 0.26).

BMI was slightly higher in the ACL-R group, but the difference did not reach statistical significance (24.5 ± 2.4 vs. 21.6 ± 2.0 kg/m²; p = 0.13). The distribution of operated sides (right/left) in the ACL-R group and dominant sides in the control group was not significantly different (p = 0.21). No significant differences were found in Tegner Activity Score (6.5 ± 2.0 vs. 7.0 ± 1.5; p = 0.67) or Marx Activity Score (11.9 ± 3.1 vs. 10.2 ± 3.3; p = 0.46) between the groups (Table 1).

Table 1.

Demographic data

Variable	ACL-R (n = 56) Mean ± SD	Control (n = 47) Mean ± SD	p-value
Age (years)	28.3 ± 5.1	24.5 ± 1.9	< 0.001
Sex (M/F)	32 / 24	31 / 16	0.352
Height (m)	1.71 ± 0.09	1.71 ± 0.08	0.99
Weight (kg)	73.7 ± 10.5	66.9 ± 9.0	0.26
BMI (kg/m²)	24.5 ± 2.4	21.6 ± 2.0	0.13
Operated Side (R/L)	32 / 35	28 / 19	0.21
Tegner Activity Score	6.5 ± 2.0	7.0 ± 1.5	0.67
Marx Activity Score	11.9 ± 3.1	10.2 ± 3.3	0.46

Note: All values are expressed as mean ± standard deviation. P-values were calculated using independent t-tests for continuous variables and chi-square tests for categorical variables. Abbreviations: M/F, male/female; R/L, right/left; BMI, body mass index; LSI, limb symmetry index; ECC, eccentric; CON, concentric

Analysis of jump phases

Jump height was significantly lower in the ACL-R group compared to the control group (21.4 ± 7.0 cm vs. 28.7 ± 10.0 cm; p < 0.001), as shown in Table 2. No significant three-way interaction was found between leg, phase, and group (p = 0.169), nor between phase and group (p = 0.401). However, a significant interaction was found between leg and group (p < 0.001), indicating that the difference between limbs varies depending on the group (ACL-R vs. control).

Table 2.

Biomechanical outcomes: jump height and limb symmetry index

Variable	ACL-R (n = 56) Mean ± SD	Control (n = 47) Mean ± SD	p-value
Jump Height (cm)	21.4 ± 7.0 (–25.4%)	28.7 ± 10.0	< 0.001
ECC Mean Force LSI (%)	84.5 ± 15.4 (–7.7%)	91.5 ± 13.8	< 0.001
CON Mean Force LSI (%)	86.7 ± 12.7 (–10.1%)	96.4 ± 9.6	< 0.001

Note: ACL-R; Anterior Cruciate Ligament Reconstruction, ECC; Eccentric, CON: Concentric, LSI; Limb Symmetry Index, SD ; Standard Deviation

Eccentric mean force

The ECC LSI was 84.5 ± 15.4% in the ACL-R group and 91.5 ± 13.8% in the control group (p < 0.001).

Post-hoc analyses showed that ECC was lower on the IL versus the NIL in the ACL-R group (5.79 ± 1.18 N/kg vs. 7.34 ± 1.55 N/kg; p < 0.001). In comparison with the control group, ECC on the IL was significantly lower than both the NDL and DL (NDL: 6.89 ± 1.59 N/kg, p = 0.008; DL: 7.07 ± 1.49 N/kg, p < 0.001).

No significant differences were observed between the NIL of the ACL-R group and the NDL or DL of the control group (p = 0.281 and p = 0.408, respectively).

Results are presented in Tables 2 and 3.

Table 3.

Descriptives eccentric and concentric mean forces in ACL-R ad control groups

Phase	Leg	ACL-R (n = 56) Mean ± SD	Control (n = 47) Mean ± SD	% Difference ACL-R vs. Control	p-value
ECC	IL / NDL	5.79 ± 1.18	6.89 ± 1.59	–16.0%	IL vs. NIL (p < 0.001), IL vs. NDL (p = 0.008), IL vs. DL (p < 0.001)
ECC	NIL / DL	7.34 ± 1.55	7.07 ± 1.49	+ 3.8%	NIL vs. NDL (p = 0.281), NIL vs. DL (p = 0.408)
CON	IL / NDL	7.77 ± 1.44	8.45 ± 1.50	–8.0%	IL vs. NIL (p < 0.001), IL vs. DL (p < 0.001), IL vs. NDL (p = 0.093)
CON	NIL / DL	9.23 ± 1.70	9.07 ± 1.93	+ 1.8%	NIL vs. NDL (p = 0.093), NIL vs. DL (p = 0.081)

Note : IL – Injured Leg; NIL – Non-Injured Leg; DL – Dominant Leg; NDL – Non-Dominant Leg; ECC – Eccentric phase; CON – Concentric phase; Statistical comparisons were derived from post-hoc analyses with Holm correction following mixed ANOVA. Bolded p-values indicate statistically significant differences

Concentric mean force

The CON LSI was 86.7 ± 12.7% in the ACL-R group and 96.4 ± 9.6% in the control group (p < 0.001).

Post-hoc analyses showed that CON was lower on the IL compared to the NIL in the ACL-R group (7.77 ± 1.44 N/kg vs. 9.23 ± 1.70 N/kg; p < 0.001).

CON on the IL was also significantly lower than on the DL in the control group (8.45 ± 1.50 N/kg; p < 0.001).A trend toward lower CON on the NDL in the control group compared to both the NIL in ACL-R (p = 0.093) and the DL in controls (p = 0.081) was observed but was not statistically significant.

Results are presented in Tables 2 and 3.

Discussion

The primary objective of this study was to assess the performance and the force value during the eccentric and concentric phases of the CMJ between participants after ACL-R and healthy controls. The second aim was to compare the asymmetries during the eccentric and concentric phases of the CMJ in participants after ACL-R and healthy controls. Our results confirmed our hypothesis with lower jump height, ECC and CON mean forces on the IL of ACL-R participants than in healthy control participants. Moreover, asymmetries during both eccentric and concentric phases of the CMJ were more important in ACL-R participants 6 months after surgery than in healthy controls.

The CMJ performance at 6 months after ACL-R was 25% lower when compared to healthy participants. This result is in accordance with the results of Read et al. [13]. However, no difference was found in the studies by Miles et al. between ACL-R and healthy participants. The difference in timing after surgery, 6 months in our study versus 9 months in the one by Miles et al. could explain this difference [11].

Both ECC and CON mean forces were lower on the IL than to the NIL, consistent with Read et al., who also reported lower values on the IL during both CMJ phases 6 months after ACL-R. Additionally, in our study, both ECC and CON on the IL were lower than those on the dominant leg of healthy individuals, while no difference was observed between the NIL and both legs of healthy controls [13]. At time of return to sport after ACL-R, both storage and production of force on the IL is altered during CMJ.

In terms of asymmetries, our results align with the findings of Read et al., who also found greater asymmetries for both eccentric and concentric phases in participants after ACL-R when compared to healthy controls. In addition to surgical or neuromuscular factors, it is also important to consider sport-specific biomechanics. Certain sports such as tennis, fencing, or handball involve frequent asymmetric movements that can lead to functional inter-limb asymmetries. These may not reflect deficits but rather sport-related adaptations. Since our participants engaged in various sports, this variability may have contributed to the observed asymmetries.

Both ECC and CON LSIs are lower in the ACL-R group than the conventional threshold of 90% [11, 16]. Although LSI values approached 90%, they did not meet the clinical threshold, reinforcing the idea that neuromuscular deficits may persist 6 months after ACL-R. It is important to note that the variability of LSIs in both groups is high, specifically in the eccentric phase, with coefficient of variation of about 25% for the eccentric phase and 15% in the concentric phase. Moreover, in the studies by Kotsifaki et al. and Miles et al., the authors reported greater asymmetries after ACL-R only during the concentric phase of the CMJ. This discrepancy could be partly explained by the timing of assessment—6 months post-op in our study versus 9 months in Miles et al.—as asymmetries tend to decrease over time after ACL-R. However, other methodological differences may also contribute [11, 14]. For instance, surgical techniques and rehabilitation approaches vary across studies and can influence recovery trajectories [5]. More importantly, the kinetic parameter used may impact findings: Kotsifaki et al. assessed asymmetries using impulse, while we used mean force. Impulse is influenced by contact time, which tends to be longer in ACL-R patients and may obscure real deficits [17, 18]. In contrast, mean force offers a more stable metric for group comparisons and may enhance the detection of eccentric asymmetries that impulse-based analyses could underestimate [19, 20]. Indeed, Read et al. showed that asymmetry between limbs decreased over time after ACL-R, which may explain the higher asymmetry values in our study compared to others [13]. The diversity of rehabilitation provided to the patients may also explain this discrepancy. Without information on the patients’ rehabilitation protocols, it is possible that they were rehabilitated using different methods, which could be a reason for this difference.

Psychological state could help explain the asymmetries observed in our study. Fear of reinjury may lead to unintentional IL protection and therefore to underload the IL when performing a CMJ [21, 22]. Another explanation would be the persistence of quadriceps strength deficit 6 months after surgery [23–26]. In the study by Miles et al., concentric isokinetic knee extensor strength asymmetry was significantly associated with concentric impulse asymmetry but not eccentric impulse asymmetry [14]. It is possible that eccentric strength would be associated with eccentric force asymmetry.

Interpretation of asymmetries in rehabilitation context

The observation of persistent eccentric and concentric inter-limb asymmetries six months after ACL-R suggests that certain neuromuscular deficits or compensatory movement strategies may remain unresolved within this time frame. Inter-limb asymmetry should be viewed as a multifactorial phenomenon. While such asymmetry may reflect incomplete neuromuscular recovery, it can also result from compensatory strategies, psychological factors such as fear of reinjury, or sport-specific adaptations—rather than insufficient rehabilitation alone [13, 21, 22].

For example, Paterno et al. and Hsu et al. have demonstrated that fear of reinjury may alter motor behavior and reduce loading on the injured limb, even in the absence of measurable strength deficits [21, 22]. Jordan et al. also reported that elite alpine ski racers—including those without ACL injury—displayed significant inter-limb asymmetries, suggesting that habitual sport-specific loading does not necessarily indicate dysfunction [12].

Although restoring symmetry is often a goal in ACL-R rehabilitation, these findings support the need for a broader interpretative framework. Asymmetries should be assessed alongside psychological readiness, sport-specific biomechanics, and strength evaluations in order to distinguish between maladaptive deficits and functional adaptations [27–30].

It is equally important to interpret between-group differences in the context of heterogeneity within both groups. In the ACL-R group, variability in rehabilitation exposure, progression, or adherence—though not documented—may have influenced outcomes [31]. Similarly, differences in sport type, training frequency, and movement demands in the control group may have contributed to the observed asymmetries.

The use of objective tools—such as force plates or limb symmetry indices—has been widely explored in the literature to quantify functional performance [32–34]. Although our study does not assess rehabilitation protocols or monitoring tools, the presence of asymmetries in both groups highlights the potential utility of such instruments in future research aimed at characterizing neuromuscular recovery trajectories and individual variability [28, 29, 32].

Strengths and limitations

This study has strengths. First, we recruited a homogenous sample of ACL-R participants with hamstring graft. This increases the internal validity of our results. Second, we analyzed the mean forces and not the impulse as in previous studies to prevent the influence of propulsion time that may reduce the difference between ACL-R participants and healthy participants.

We also must acknowledge some limitations of our work. There was a significant difference in terms of age and weight between groups. The ACL-R participants were older and heavier. These differences may have influenced the performance of the CMJ and therefore overestimate the differences we found for jump height and mean forces. However, we normalized the force data by body mass to allow for group comparison, and we think that these differences had no influence on the asymmetry analyses. Another limitation of this study is the absence of detailed data regarding the sport level and training frequency of participants in the control group. Although all controls were recreational athletes without recent injury history, the lack of information on their specific athletic background or training load may have introduced variability in countermovement jump performance, potentially affecting between-group comparisons.

To deepen the interpretation of the results on CMJ, psychological state evaluation, such as with the ACL-RSI, and strength evaluation would have been meaningful. Future research should consider investigating these factors. Additionally, one important limitation is the lack of quadriceps strength evaluation (isokinetic or isometric). This absence limits our ability to discern whether the reduced force output and asymmetries observed in CMJ performance were due to actual neuromuscular deficits or compensatory motor behaviors. Future studies should integrate objective strength measurements to better clarify these mechanisms.

Another limitation is the absence of psychological readiness assessment, such as the ACL-RSI or TSK-11. Fear of reinjury and psychological factors have been shown to influence limb loading strategies and return-to-sport outcomes after ACL reconstruction. Including validated psychological measures would have provided a more comprehensive understanding of the observed asymmetries and should be considered in future studies.

Finally, the lack of detailed information regarding rehabilitation content and progression. Although all participants followed rehabilitation programs developed under consistent guiding principles, we could not verify the exact exposure to key training modalities such as eccentric overload exercises, single-limb strength training, or jump-landing drills. Furthermore, no data on adherence or asymmetrical monitoring during rehabilitation was available. These factors may have influenced the magnitude of inter-limb asymmetries observed at testing and should be addressed in future prospective studies with controlled rehabilitation tracking.

Conclusion

The results of this study indicate a lower CMJ performance six months after ACL-R compared to healthy controls. Force values during both the eccentric and concentric phases of the CMJ were lower in the IL compared to the NIL in ACL-R participants and the dominant leg of the control group. Additionally, greater asymmetries were found in ACL-R participants compared to healthy participants in both eccentric and concentric phases.

6 months after ACL-R, participants in the ACL-R group did not achieve the 90% LSI threshold in either movement phase, with average values of 84.5% (eccentric) and 86.7% (concentric), highlighting persistent inter-limb asymmetries. Although our findings suggest persistent asymmetries in vertical jump performance six months after ACL-R, the study did not evaluate rehabilitation content or progression. These results should be interpreted descriptively and may help inform future research exploring how specific interventions impact neuromuscular recovery and return-to-sport readiness.

Abbreviations

ACL

Anterior cruciate ligament

ACL-R

Anterior cruciate ligament reconstruction

CMJ

Countermovement jump

STROBE

Strengthening the Reporting of Observational studies in Epidemiology

CON

Concentric mean force

ECC

Eccentric mean force

LSI

Limb symmetry index

BMI

Body mass index

RTS

Return to Sport

IL

Injured leg

NIL

Non-injured leg

DL

Dominant leg

NDL

Non-dominant leg

Author contributions

Conceptualization, F.F. and A.M.-S.; methodology, F.F., A.M.S. and B.P.F.; validation, F.F., J.M. and B.P.F.; investigation, F.F., A.M.S. and T.H.; writing—original draft preparation, F.F., A.M.-S. and W.L.; writing—review and editing, F.F., T.H., M.B., M.D. and B.P.F.; visualization, J.M. and M.D.; and supervision, F.F., A.M.S. and B.P.F. All authors have read and agreed to the published version of the manuscript.

Funding

This study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study received approval from the Clinic of Domont ethics committee (IRB PCE309N-003). All patients were treated in agreement with the Declaration of Helsinki and provided written informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.