Early knee status affects self-reported knee function 1 year after non-surgically treated anterior cruciate ligament injury

Objectives: The primary aim was to assess impact of early knee status on self-reported knee function at 3 and 12 months and on quadriceps strength at 12 months after non-surgically treated ACL injury. The secondary aim was to describe the recovery of muscle strength during the ﬁ rst year after the injury. Design: Prospective cohort study. Participants: 70 patients (42 males; mean age 27 ± 7 years) with acute ACL injury. Main outcome: Knee symptoms, knee function and sporting activities were assessed with the International Knee Documentation Committee Subjective Knee Form (IKDC-SKF). Muscle strength was assessed with an isokinetic dynamometer. Clinical assessment performed at baseline was used to evaluate early knee status. Results: Global knee function, knee joint stability during ADL, gait pattern and one-legged squat assessed in mean 2 weeks after injury hadimpact on self-reported knee function at 3 and 12 months ( r 2 0.105 e 0.267). Mean limb symmetry index (LSI) of muscle strength and jump performance were 91 e 98% at 12 months. Conclusion: Early knee symptoms affect self-reported knee function at 3 and 12 months, while other factors are important for gaining muscle strength. Muscle strength recovered during the ﬁ rst year after ACL injury and reached mean LSI above 90%. Level of evidence: Prospective cohort study, level II. © 2021 The Authors. Published by Elsevier


Introduction
An anterior cruciate ligament injury (ACL) is a serious knee injury that affects individuals in the short and long term.Early after the injury, knee function is reduced due to knee joint instability, swelling, pain, reduced range of motion and muscle function, and an altered movement pattern.Surgical treatment with ACL reconstruction aims to restore knee stability, but non-surgical treatment is also a viable option for some patients (Grindem, Wellsandt et al., 2018) and in many cases it is recommended as first-line treatment (Diermeier, Rothrauff et al., 2020).Both treatment strategies result in similar short term self-reported knee function (Frobell, Roos et al., 2010).Though, despite treatment targets to restore the knee function, many patients have persistent deficits (Abrams, Harris et al., 2014) and do not return to sport (Ardern, Taylor et al., 2014).
There is limited evidence on recovery of knee function and muscle strength after non-surgical treatment and what factors impact the recovery.Understanding the early factors that may impact medium to long term recovery, can help to inform decision making for treatment early after ACL injury.A comprehensive rehabilitation programme using structured progression can improve knee function (Tagesson, Oberg et al., 2008;Eitzen, Moksnes et al., 2010), and quadriceps strength (Tagesson, Oberg et al., 2008), but early symptoms related to the knee trauma might hinder progress in rehabilitation.In accordance, an asymptomatic knee has been an inclusion criterion for participation in rehabilitation interventions (Eitzen, Moksnes et al., 2010;Logerstedt, Lynch et al., 2013;Grindem, Wellsandt et al., 2018).Early after the injury, the rehabilitation targets are to eliminate joint swelling and pain, normalise range of motion and movement pattern during basic movements, such as walking, and to increase load tolerance (Eitzen, Moksnes et al., 2010;Wilk & Arrigo, 2017;Grindem, Wellsandt et al., 2018).The next phase in rehabilitation focuses on muscle strengthening.Quadriceps strength is an important outcome to identify treatment success (Lynch, Logerstedt et al., 2015) and is used as a criterion to determine return to sport (Grindem, Snyder-Mackler et al., 2016).Resolution of the early impairments to achieve an asymptomatic knee is important to advance to later rehabilitation phases with the goal to return to full activity and sport (Eitzen, Moksnes et al., 2010;Wilk & Arrigo, 2017).
It is unclear how various impairments after the initial injury affect future knee function.A symptomatic knee might only tolerate a restricted load, as higher loads may result in joint effusion and pain.That means limited possibilities to perform an effective exercise programme with increased progression to regain muscle function and dynamic knee stability.Therefore, presence of early knee symptoms might affect both knee function in the short term and knee function and muscle strength 1 year after the injury.Increased knowledge of the importance of different early knee symptoms and deficits forms the basis for improved rehabilitation strategies targeting functional impairments and provides conditions for tailored rehabilitation.The primary aim of the present study was to assess the impact of early knee status on self-reported knee function at 3 months and self-reported knee function and quadriceps strength at 12 months after non-surgically treated ACL injury.The secondary aim was to describe the recovery of muscle strength during the first year after the ACL injury.The hypothesis was that the degree of knee joint impairment early after an ACL injury affects functional recovery.

Methods
This prospective study is part of NACOX cohort study, designed to describe the natural corollaries and recovery after acute ACL injury (Kvist et al., 2018).The present study included patients with more extensive follow-up from one centre who formed a subgroup.

Patients and recruitment
Between May 2016 and October 2018 patients seeking medical care for an acute knee injury in our region were consecutively recruited.Inclusion criteria were ACL injury sustained no more than 6 weeks prior to inclusion, verified with MRI, and age between 15 and 40 years at the time of the ACL injury.Patients were excluded if they had a previous ACL injury to the same knee, serious concomitant injury (e.g. total posterior cruciate ligament rupture or fracture that required specific treatment), if they were unable to understand written and spoken swedish language, had cognitive impairments or other illnesses or injuries that impaired function (e.g.fibromyalgia, rheumatic diseases or other diagnoses associated with chronic pain).In the NACOX study, questionnaires are sent to the participants via short messages on smartphones or by e-mail at various time points after injury and had extended follow-up with clinical and functional examination at baseline and 3, 6, 12, and 24 months after injury (Kvist, Gauffin et al., 2018).In the present study, we analysed data from patients with non-surgically treated ACL injury who had completed the IKDC assessment at 12 months after the ACL injury.Participation in this study did not alter the usual course of treatment for patients with ACL injury at the recruitment centre: 1. Patient receives a clinical diagnosis from an orthopaedic surgeon, verified by MRI. 2. Initial treatment with rehabilitation of approximately 3 months' duration.
3. Scheduled follow-up after approximately 3 months, where orthopaedic surgeon and patient decide on need for further treatment.
Patients received information about the study at their initial contact with the healthcare provider.The recruitment procedure is described in the study protocol (Kvist et al., 2018).Patients who accepted participation signed a written informed consent form.Ethical approval was granted by the Swedish Ethical Review Authority, Dnr: 2016/44/31, 2018/123e32.2.2.Assessment of impact of early knee status on functional recovery

Dependent variables
The International Knee Documentation Committee Subjective Knee Form (IKDC-SKF) at 3 and 12 months after the injury and limb symmetry index (LSI) of quadriceps strength at 12 months after ACL injury were analysed as dependent variables.

Independent variables
Clinical assessment of knee status was performed by a physiotherapist and the baseline assessment was used for analysis.Knee joint effusion was assessed by knee joint girth measured with a measurement tape over the mid of the patella and with the Stroke test (Sturgill, Snyder-Mackler et al., 2009).Passive range of motion during knee extension and flexion was measured with a plastic goniometer.Knee extension was measured while the patient was supine with the heel on a 15 cm box, relaxing the knee into full extension.Flexion was measured with the patient seated with their legs on the examination table performing a passive maximum knee flexion using their arms.Functional performance was assessed by visual inspection of their movement pattern during walking and during one-leg squat.Gait pattern was graded during 10 m walk at self-selected speed for symmetrical load distribution and terminal knee extension (normal gait, slight deviation, and significant deviation) and one-legged squat was graded (able to perform >45 , able to perform up to 45 , unable to perform).Knee laxity was assessed using the KT-1000 arthrometer at 133N (mean value of three repetitions) at 3 months after the ACL injury.

Questionnaires
Patients answered the IKDC-SKF at 4 weeks, 3 months, and 12 months after the injury.The IKDC-SKF is an 18-item knee-specific questionnaire, covering symptoms, function and activity level (Irrgang, Anderson et al., 2001).Single assessment numeric evaluation (SANE) rating was used to assess global knee function ("On a scale from 0 to 100, where 100 is perceived as best, how would you rate your knee today?") (Shelbourne, Barnes et al., 2012).Patients graded knee joint stability during activities of daily living and during physical activities/sport (1 ¼ completely unstable to 10 ¼ completely stable) and knee pain (0 ¼ no pain to 100 ¼ worst imaginable pain) on numeric rating scales.Patient-reported activity level before the injury and at each follow-up was classified according to the Tegner Activity Scale (Tegner & Lysholm, 1985), and the modified IKDC activity level (Hefti, Muller et al., 1993;Grindem, Eitzen et al., 2014).In the Tegner Activity Scale, sporting activities and work are graded on an 11-point scale, according to functional demands on the knee.Level 0 is sick leave due to knee problems and level 10 is participation in elite football.The modified IKDC activity level classifies activity into four levels: level I is pivoting and contact sport, level II is pivoting non-contact sport, and level III is neither pivoting nor contact sport; patients who were not participating in physical activity, or who were completing rehabilitation were classified as level IV.

Assessment of muscle strength and jump performance
Assessment of muscle strength and jump performance at 3-, 6-, and 12-months' follow-up after the ACL injury was analysed.Patients performed 5 repetitions at 60 /second and 15 repetitions at 180 /second in an isokinetic dynamometer (Biodex Medical Systems Inc, Shirley, NY).Following strength testing patients performed 4 single-legged hop tests in the following order: single leg hop for distance, crossover hop for distance, triple hop for distance, and 6-m timed hop (Kvist, Gauffin et al., 2018;Logerstedt, Di Stasi et al., 2014;Noyes, Barber et al., 1991;Reid, Birmingham et al., 2007).Patients performed hop tests if they had performed hop exercises during rehabilitation without subsequent effusion or pain.For muscle strength and hop test, the non-injured knee was tested before the injured knee.For hop tests, patients performed 1 to 2 practice trials followed by 2 recorded valid trials for each leg and the best trial was used for analysis.Hop tests were performed with indoor shoes, with hands kept on their backs.For the single leg hop, the crossover hop, and the triple hop for distance, a trial was considered valid if the landing was stable.For the distance hop tests, participants stood at the starting position on one leg, then hopped as far as possible.For the 6-m timed hop, the participants performed a series of one-legged hops over a distance of 6 m.

Statistical analysis
Descriptive statistics of continuous data are presented as mean ± standard deviation and median with interquartile range on all self-reported knee function variables.Nominal data and Likert scales are displayed as number of patients and percentages.
Simple linear regression models were used to assess the impact of early knee status on IKDC-SKF score at 3 and 12 months after ACL injury and on LSI of quadriceps strength at 12 months after ACL injury.Results from the regression models are presented as beta with 95% confidence interval, standardized beta, p-value, and r 2 (proportion of variance in the dependent variable explained by the independent variable).
Linear mixed models, using maximum likelihood estimation with the assumption of unstructured covariance, were used to estimate muscle strength at 3, 6, and 12 months after ACL injury and jump performance in single-legged hop tests at 6 and 12 months after ACL injury.Recovery in muscle strength and jump performance were assessed as mean change between time-points.The main reason for using linear mixed models was to handle the unbalanced design between time-points and use as many observations and patients as possible to estimate mean scores at each time- ADL, activities of daily living; CI, confidence interval; IKDC-SKF, international knee documentation committee subjective knee form; PROM, passive range of motion; SANE, single assessment numeric evaluation; Std, standardized; 1 On a scale from 0 to 100, where 100 is perceived as best, how would you rate your knee today? 2 On a numeric rating scale 1e10, where 1 ¼ totally unstable, 10 ¼ totally stable; 3 Dichotomized: ˂ 0 /! 0 (reference); 4 Dichotomized: ! 1 cm/˂ 1 cm (reference); 5 Symmetrical load distribution and terminal knee extension: significant deviation/slight deviation/normal (reference); 6 Dichotomized: unable to perform/able to perform (reference).

Results
From the subgroup of patients with extensive follow-up (total 139 patients) in the XX cohort, 69 were excluded from the current analyses due to ACLR within 12 months of the ACL injury (n ¼ 58) or were excluded due to not responding to the IKDC-SKF at 12 months (n ¼ 11).Finally, 70 patients (42 males; mean age 27 ± 7 years) were included.Twenty-six patients had an isolated ACL injury, 33 patients had one concomitant injury (ligament/meniscal/cartilage), 10 patients had two concomitant injuries and one patient had three concomitant injuries.Three patients had an earlier ACL injury in the contralateral knee, treated with reconstruction, and had returned to sport after their first ACL injury.Patient characteristics, knee laxity and mean time from the ACL injury to the questionnaires are displayed in Table 1.There was a mean of 2 ± 4 days (maximum 13 days) between baseline questionnaire and clinical assessment.Sixteen patients (23%) had undergone knee arthroscopy during the first year after injury (9 meniscectomies, 2 diagnostic, 1 meniscal suture and 4 other), but none of them had knee arthroscopy within 15 days before 6-or 12-months' follow-up.The meniscal suture was performed 2 months after the injury and this patient did not perform muscle strength testing at 3 months.Self-reported knee function assessed with IKDC-SKF score and activity level according to Tegner score are reported in Table 2. Knee status at baseline revealed functional limitations in most of the patients (Table 3).

Impact of early knee status on functional recovery
Worse single assessment numeric evaluation of global knee function, knee joint stability during ADL, gait pattern and performance on one-legged squat assessed at baseline were associated with worse self-reported knee function at 3 months (r 2 0.124e0.267)and 12 months (r 2 0.105e0.200).In contrast, passive range of motion and knee joint girth were not associated with selfreported knee function at 3 and 12 months.Neither of the early knee status variables assessed at baseline was associated with LSI of quadriceps strength at 12 months (Table 4).In a secondary analysis, the effect of confounding factors in the regression models were analysed.The presence of associated injuries and age did not significantly change the results in the models.There were small effects of sex on the association between early knee status and selfreported knee function at 3 months.Separate regression models by sex are presented in appendix 1.

Recovery of muscle function
Isokinetic strength test at 60 /second displayed quadriceps weakness at both 3 months (mean LSI 78%, 95% CI 74e82%) and 6 months (mean LSI 86%, 95% CI 82e89%), reaching mean LSI of 91% (95% CI 88e94%) at 12 months.For quadriceps strength 60 /second an LSI above 90% was achieved for 28% of the patients at 3 months, 45% at 6 months, and 59% at 12 months.For hamstring strength 60 /second an LSI above 90% was achieved for 45% of the patients at 3 months, 77% at 6 months, and 67% at 12 months (Fig. 1).Quadriceps peak torque in the injured knee and LSI increased from 3 to 6 months and from 6 to 12 months.Hamstring peak torque in the injured knee increased from 3 to 6 months and from 6 to 12 months but increase in LSI was only seen up to 6 months.Quadriceps peak torque at 60 /second in the non-injured leg was unchanged during the first year after injury.Quadriceps and hamstring peak torque at 180 /second increased in the non-injured knee from 3 to 6 months, and hamstring peak torque at 60 /second and 180 /second from 6 to 12 months (Table 5).
Only 3 patients at 3 months and 34 patients at 6 months met criteria to perform the hop tests.At 12 months mean LSIs were 91e97% for muscle strength tests and 95e98% for hop tests.However, 8 patients did not meet the criteria to perform hop tests, i.e., not included in analysis of hop performance (Table 5).

Discussion
Global rating of knee function (SANE), knee joint stability during ADL, gait pattern and one-legged squat at baseline had an impact on self-reported knee function at both 3 and 12 months, but not on quadriceps strength.Quadriceps and hamstring strength on the injured leg recovered continuously up to 12 months after the ACL injury, whereas quadriceps strength in the non-injured leg remained unchanged.Mean LSI of muscle strength tests were 91e97% and mean LSI of jump performance were 95e98% at 12 months, though 90% LSI for quadriceps strength at 60 /second was only reached by 59% of the patients.At 6 months only half of the patients had sufficient knee function to perform hop tests.
The hypothesis that the degree of knee joint impairment early after an ACL injury may affect functional recovery was confirmed.There was an association between the early knee symptoms global rating of knee function, knee joint stability during ADL, gait pattern and one-legged squat and self-reported knee function at both 3 and 12 months.These are all measures of body functions and displays to what extent the injury affects performance in daily activities, giving an overall picture of impact on daily life.In contrast, assessment at body structure level, i.e. passive range of motion and knee joint girth early after injury did not affect the outcome at 3 or 12 months.This means that single measures at body structure level early after injury are of less importance for mid-and long-term outcome.This finding suggests that single impairments at body structure level may be resolved, allowing for progress in rehabilitation.The association between global rating of knee function and knee joint stability during ADL with self-reported knee function at 3 months was stronger for females compared to males.The explanation for these differences between sexes needs to be elaborated on further in future studies.This finding is in with earlier research concluding that individual factors such as age and sex could affect the outcome (Grindem, Wellsandt et al., 2018).It should be kept in mind that associations between early knee status and self-reported knee function were weak and there were no associations between early knee symptoms and muscle strength, indicating that other variables are also of importance for the recovery after ACL injury.Despite early knee symptoms the muscle function recovers, and resolution of the early knee impairments is probably a prerequisite for tolerance of high loads in strength training.Structured progressive strength training including adequate load is almost certainly crucial for recovery of muscle function.Satisfactory muscle strength may be achieved through a comprehensive structured rehabilitation programme containing sufficient strength training (Tagesson, Oberg et al., 2008;Welling, Benjaminse et al., 2019).
To our knowledge, no previous study has reported the impact of early knee symptoms on short or long-term knee function, and therefore our results can guide future rehabilitation strategies.Our results also show that the extent of physical impairments varies between patients, as demonstrated by the standard deviations of the variables.Concordant with the current findings, previous studies evaluating rehabilitation interventions after ACL injury with the inclusion criterion of an asymptomatic knee, i.e., impairment resolution, show a variation of time-points when these impairments resolve (Logerstedt, Lynch et al., 2013;Wellsandt, Failla et al., 2017).Effective early rehabilitation strategies that address knee joint impairments, such as joint effusion or pain, should be a high priority since resolution of these impairments is a prerequisite for progressing to more advanced rehabilitation phases including higher exercise load and more demanding exercises (Logerstedt, Scalzitti et al., 2017;Wilk & Arrigo, 2017).
Muscle strength and hop performance increased over time.It was expected that functional performance would increase with time from the ACL injury, due to increased time for rehabilitation and return to demanding activities and sport (Abrams, Harris et al., 2014), although, only few studies have reported functional performance recovery after non-surgically treated ACL-injury (Ageberg, Thomee et al., 2008;Tagesson, Oberg et al., 2008;Eitzen, Moksnes et al., 2010;Sonesson, Osterberg et al., 2021).At 12months' follow-up, the mean LSI in muscle strength and hop tests were above 90%, indicating symmetrical performance.However, for quadriceps strength at 60 /second LSI above 90% was achieved only for 28% of the patients at 3 months, 45% at 6 months, and 59% at 12 months.Moreover, 8 patients did not complete the hop test battery at 12 months due to insufficient knee function.At 6 months after the injury, about half of the patients had not reached sufficient knee function or had not performed specific hop exercises that were required to perform the hop tests.At 3 months' follow-up only 2 patients met the criteria to perform the hop test.Limb symmetry index displays muscle strength in the injured leg relative to the non-injured leg and knee function might therefore be overestimated.It may be expected that muscle strength decreases also in the non-injured leg during the period after injury (Wellsandt, Failla et al., 2017) due to decreased activity participation (Fomin, Gauffin et al., 2020).Isokinetic testing at 60 /second measures strength, and we found that quadriceps peak torque 60 / second in the non-injured knee did not change over time, indicating that muscle strength loss in the non-injured leg due to the injury may be negligible when rehabilitation starts immediately after the injury.In contrast, the peak torque values in the noninjured knee increased in quadriceps and hamstrings at 180 /second from 3 to 6 months, and in hamstrings at 60 /second and 180 / second from 6 to 12 months, which affected the LSI.
The level of recovery of muscle function in the current study at one year (59%) is comparable to an earlier study reporting recovery in muscle function in about two-thirds of the patients (Ageberg, Thomee et al., 2008).Deficits in muscle function do not differ between patients treated with rehabilitation and ACL reconstruction or rehabilitation only (Ageberg, Thomee et al., 2008).Persistent deficits in muscle strength is reported also after ACL reconstruction (Gokeler, Welling et al., 2017;Herrington, Ghulam et al., 2018;Welling, Benjaminse et al., 2018), and may be associated with the fairly low return-to-sport rates (Ardern, Taylor et al., 2014).Strength deficits are associated with worse present and future outcomes on patient-reported outcome measures and performance-based functional tests (Zwolski, Schmitt et al., 2015;Flosadottir, Roos et al., 2016;Ithurburn, Altenburger et al., 2018;Joreitz, Lynch et al., 2016).These deficits may also increase the risk for future knee injuries and development of knee osteoarthritis (Arhos, Thoma et al., 2020).
The present study followed patients during recovery after ACL injury and the rehabilitation process according to the clinical routine.Therefore, the extent and content of interventions may vary (Carter, Webster et al., 2021) and might be affected by the clinical experience of the physiotherapists (Grindem, Granan et al., 2015).Most patients received a tailored rehabilitation programme supervised by a physiotherapist, but some patients may not have achieved sufficient exercise load during rehabilitation.Structured rehabilitation (Buckthorpe, 2019), high patient motivation (Sonesson, Kvist et al., 2017) and compliance to the rehabilitation programme (Della Villa, Andriolo et al., 2019) are probably key factors for a successful outcome., and.
Strengths of the current study include a consecutive cohort of patients with ACL deficiency who were followed with clinical examinations and questionnaires regularly during the first year after the index injury.All injuries were verified with MRI and there was thorough recording of additional injuries and adverse events during the study period.None of the patients had arthroscopy within 15 days before 6-or 12-months' follow-up.The 12-month outcome is of significance as most individuals with ACL injury have completed rehabilitation and returned to their sporting activity at this time point.Since bilateral deficits are common in patients with ACL injury (Gokeler, Welling et al., 2017;Wellsandt, Failla et al., 2017), use of the LSI may underestimate performance deficits and should therefore be analysed with caution, therefore peak torque values are also reported.This prospective study is a "real life" study following the natural course of ACL injury for patients receiving standard care according to usual routine in the actual clinical context.Since the rehabilitation was not altered within the study context, the external validity of the results is high.Evaluation of effects of a specific rehabilitation programme was not within the scope of this paper.A limitation this approach is the lack of a united standardized rehabilitation protocol which means that there is limited information about the details of the rehabilitation received by each patient.Moreover, there were no established criteria to end the rehabilitation or to clear the patient for return to sport.Instead, physiotherapists formed the rehabilitation and cleared each patient for return to sport according to conventional criteria.Therefore, the implications of the importance of the rehabilitation programme in functional outcome needs to be interpreted with caution.
Clinical implications of the present study are that despite early knee impairment muscle function can recover during the first year after injury.It is probably important to resolve deficits in the early phases of rehabilitation to be able to progress into the next rehabilitation phase, as an asymptomatic knee without swelling and pain is a prerequisite to tolerate increased load in rehabilitation.In this study following the natural course of ACL injury for patients receiving standard care according to usual clinical routine, 41% of patients still had deficits in quadriceps strength and 33% had deficits in hamstrings strength one year after the injury.Clinicians need to be aware of that over one-third of patients might have persistent strength deficits one year after ACL injury.Future research should evaluate which factors may be related to persistent strength deficits, and if interventions such as structured progressive strength training can address these deficits.

Conclusion
Early knee symptoms affect self-reported knee function at 3 and 12 months, while other factors are important for gaining muscle strength.Muscle strength recovered during the first year after the ACL injury and reached mean LSI above 90%.
S. Sonesson, H. Gauffin and J. Kvist Physical Therapy in Sport 50 (2021) 173e183 point and recovery between time-points.All on muscle strength included 69 of 70 patients and 174 of 210 observations.Models on the single hop test included 58 of 70 patients and 86 of 140 observations, models on the crossover hop test included 52 of 70 patients and 75 of 140 observations, models on the triple hop test included 53 of 70 patients and 78 of 140 observations, and models on the 6-m timed hop test included 53 of 70 patients and 77 of 140 observations.Results from the linear mixed models are presented as mean with 95% confidence interval and p-value.All statistical analyses were performed with IBM SPSS Statistics for Windows (Version 25.0.Armonk, NY: IBM Corp).The significance level was set at 5% for all analyses.

Table 1
Patient characteristics.
a Medial collateral ligament/lateral collateral ligament/posterior cruciate ligament.bMedial/lateral meniscus.Table 2 Self-reported knee function and activity level.ADL, activities of daily living; IKDC-SKF, international knee documentation committee subjective knee form; IQR, interquartile range; SANE, single assessment numeric evaluation; SD, standard deviation.a On a scale from 0 to 100 where 100 is perceived as best, how would you rate your knee today?. b 1 ¼ completely unstable, 10 ¼ completely stable.c 0 ¼ no pain, 100 ¼ worst imaginable pain.d Level 0 ¼ no participation in physical activity due to knee problems, level 10 ¼ participation in elite football.e Answered at the 4 weeks questionnaire.S. Sonesson, H. Gauffin and J. Kvist Physical Therapy in Sport 50 (2021) 173e183

Table 3
Descriptive statistics of clinical assessment at baseline, 0e7 weeks after the ACL injury.Simple linear regression models on IKDC-SKF and muscle strength as dependent variables, with knee status at baseline as independent variables.