Knee strength, hop performance and self-efficacy at 4 months are associated with symmetrical knee muscle function in young athletes 1 year after an anterior cruciate ligament reconstruction.

OBJECTIVES: We investigated whether patient demographics, 4-month patient-reported outcomes (PRO) and muscle function predicted young athletes regaining symmetrical muscle function in five tests of muscle function 1 year after ACL reconstruction.
METHODS: We extracted data on patient demographics, PROs and the results of five tests of muscle function from a rehabilitation-specific register. Athletes were 15-30 years of age, involved in knee-strenuous sport and had undergone a primary ACL reconstruction. The primary outcome was achieving a Limb Symmetry Index of ≥90% for the battery of tests 1 year after ACL reconstruction. Patient demographics, muscle-function data and results for PROs at the 4-month follow-up were analysed.
RESULTS: In all, 237 athletes (59% female; mean age 22±4 years) were included in the study. One year after ACL reconstruction, 26% (62/237) of the included athletes had achieved symmetrical muscle function. Univariable analysis showed that symmetrical muscle function was associated with present self-efficacy, OR 1.28 (95% CI 1.04 to 1.58, p=0.011), knee-extension strength, OR 1.73 (95% CI 1.28 to 2.34), knee-flexion strength, OR 1.39 (95% CI 1.07 to 1.81), vertical hop, OR 1.77 (95% CI 1.27 to 2.45), single-leg hop for distance, OR 1.98 (95% CI 1.24 to 3.17) and side hop, OR 1.64 (95% CI 1.15 to 2.33).
CONCLUSION: Symmetrical knee-extension and knee-flexion strength, a more symmetrical hop performance and higher present self-efficacy at an early stage all increased the odds of achieving symmetrical muscle function in young athletes 1 year after ACL reconstruction.

Young athletes with more symmetrical knee-extension and knee-flexion strength, a more symmetrical hop performance and higher present self-efficacy, early after ACL reconstruction, have increased odds of achieving symmetrical muscle function 1 year after ACL reconstruction.

Only one in four young athletes achieves symmetrical muscle function 1 year after ACL reconstruction.

Clinicians should inform athletes not to return to sport before symmetrical muscle function is achieved.

To set realistic expectations, athletes with asymmetrical knee muscle function and lower present self-efficacy 4 months after ACL reconstruction should be informed that achieving symmetrical muscle function may take longer than 1 year.

Introduction

After an anterior cruciate ligament (ACL) reconstruction, athletes who have regained symmetrical muscle function have lower risk of a subsequent ACL injury than those who continue to have asymmetry.1 2 In young athletes, 15–30 years of age, a minority regain symmetrical muscle function before returning to sport after a primary ACL reconstruction.3 4 This is a challenge for clinicians, as young individuals have a high reinjury risk after ACL reconstruction, where one-fourth of patients younger than 25 years of age sustain a second ACL injury once they return to sport (RTS).5–10

Symmetrical muscle function is a cornerstone of rehabilitation after an ACL injury as it is associated with a higher rate of RTS.11 Moreover, Toole et al 3 reported that a greater proportion of athletes, who have symmetrical knee strength at time of RTS, maintained the same level of sports participation over the year following the RTS, compared with those who do not meet the criteria.3 We see clinical value in finding modifiable factors early in the rehabilitation that predict the recovery of symmetrical muscle function prior to RTS. A better understanding of patient characteristics as well as of modifiable factors is important to set realistic expectations and goals for patients and optimise rehabilitation. However, knowledge relating to these factors is limited in young athletes.

Many factors probably contribute to the achievement of symmetrical muscle function. Patients who early have lower levels of pain and symptoms, less associated problems and a more symmetrical muscle function may be better off when it comes to recovering symmetrical muscle function at 1 year after the ACL reconstruction.12–16 Moreover, psychological outcomes, such as self-efficacy, that is, how certain an individual is about his/her own ability to perform an activity despite pain or discomfort, might be important, as an association between self-efficacy and symmetrical hop performance has been reported in patients after an ACL reconstruction.17 However, our knowledge of psychological outcomes as predictors is limited, specifically in young ACL patients.

The objective of this study was to investigate whether patient demographics, short-term patient-reported outcomes (PRO) and muscle function are able to predict the achievement of symmetrical muscle function in five tests of muscle function in young athletes 1 year after ACL reconstruction. The hypothesis was that adolescent athletes, who perceived higher levels of self-efficacy and self-reported knee function, and athletes with more symmetrical muscle function at 4 months after ACL reconstruction would have higher odds of achieving symmetrical muscle function 1 year after reconstruction.

Materials and methods

This study was performed as a prospective cohort study based on data from a rehabilitation outcome register, Project ACL.4 18 All the athletes (patients) were given written information about the study and informed consent was obtained. The data have been coded and none of the included athletes could be identified during analyses.

Project ACL includes patients with an ACL injury irrespective of their age, sex, time from injury or choice of treatment. The project uses a web-based platform for regular assessments with PROs and muscle function, according to a predefined schedule of follow-ups after the latest ACL injury or reconstruction.

Patients

Patients who had sustained a unilateral ACL injury and undergone ACL reconstruction between 1 April 2013 and 1 April 2017 were eligible for inclusion. Patients were included in the study if they were between 15 and 30 years of age and had data from the 4-month follow-up, including PROs and/or tests of muscle function, and had performed all tests of muscle function 1 year after the ACL reconstruction. A further inclusion criterion was that the patients were regarded as athletes, defined as having participated in knee-strenuous sport before sustaining their ACL injury, that is, a preinjury Tegner Activity Scale (Tegner) of ≥6.19

Predictors

The included athletes were evaluated with regard to PROs and with tests of muscle function at the 4-month follow-up and at 1 year after the ACL reconstruction. All the patient demographics were self-reported.

Patient-reported outcomes

To assess the preinjury and present level of physical activity, the Tegner was used.14 The Tegner has acceptable test–retest reliability, with an intraclass correlation coefficient (ICC) of 0.8 for patients with an ACL injury or reconstruction.19 20

To assess perceived knee-related self-efficacy, that is, how certain a patient is about the performance of knee activities in the present and in the future, a shorter version of the Knee Self-Efficacy Scale (K-SES) was used.21 The shorter version contains 4 items less than the original K-SES and some items are somewhat rephrased. Reliability, structure and validity is the same or better (unpublished data).21

The Knee injury and Osteoarthritis Outcome Score (KOOS)22 was used to assess patients’ opinions of their knee and associated problems after ACL reconstruction. The KOOS has been reported to have acceptable test–retest reliability, with an ICC ranging from 0.85 to 0.93 for each of the five subscales in patients with an ACL injury or reconstruction.22 In the present study, the KOOS4, an average score of the four subscales of pain, knee-related symptoms, function in sport and recreation and knee-related quality of life, was used. In the KOOS4, the dimension of function throughout daily living is excluded to avoid ceiling effects due to the fact that relatively young and active patients rarely have difficulty with activities of daily living.23

Tests of muscle function

Table 1 shows the detailed information on the test procedure.4 24 The test procedure, including the warm-up procedure, familiarisation and maximum repetitions in both strength and hop tests, has been described in previous studies.4 24

Table 1

Tests of muscle function

	Knee angles (°)	Practice trials, n(% of 1RM)	Maximum repetitions (n)	Rest between repetitions (s)
Knee extension
Isometric*	60	3 (70, 80, 90% of 1RM)	3–5	40
Isokinetic†	0–90	1–2 (90% of 1RM)
Knee flexion
Isometric‡	60	3 (70, 80, 90% of 1RM)	3–5	40
Isokinetic†	0–90	1–2 (90% of 1RM)
Unilateral vertical hop§	–	2	3	20–30
Single-leg hop for distance	–	2	3	20–30
Side hop¶	–	10	1	360

*Measured with David F200 DMS-EVE (David Health Solutions, 2013, Finland).

†Measured with Biodex System 4 (Biodex Medical Systems, Shirley, New York, USA) at 90°/s.

‡Measured with David F300 DMS-EVE (David Health Solutions, 2013, Finland).

§Measured with MuscleLab, Ergotest Technology, Oslo, Norway.

¶As many hops as possible in 30 s over two lines, 40 cm apart.

RM, repetition maximum.

At the start of Project ACL, isometric tests of knee extension and knee flexion were performed using David F200 DMS-EVE (David Health Solutions, 2013, Finland). These results contribute about 20% of the total muscle strength data in the present study. The isometric tests were then replaced by isokinetic concentric strength tests of knee extension and knee flexion using Biodex System 4 (Biodex Medical Systems, Shirley, New York, USA). Both devices have been reported to have high reliability.25–28

After strength testing, three hop tests were performed in the following order: unilateral vertical hop, unilateral hop for distance and unilateral side hop. High test–retest reliability for the three different tests in the battery of hop tests has been reported (ICC 0.93–0.97).29

Outcome

In the present study, achieving a Limb Symmetry Index (LSI) of ≥90% in all five tests of muscle function was defined as the achievement of symmetrical muscle function30 and was used as the primary outcome. The LSI is defined as the ratio between the results for the injured side and the non-injured side expressed as a percentage.

Statistical analysis

Statistical analysis was performed using the Statistical Analysis System (SAS/STAT, V.14.2, 2016; SAS Institute).

Descriptive statistics were reported as count and proportion, as the median with minimum and maximum or as the mean with SD. Comparisons between groups were performed using Fisher’s exact test (lowest one-sided p value multiplied by 2) for dichotomous variables, while the Mantel-Haenszel χ2 test was used for ordered categorical variables. The χ2 test was used for non-ordered categorical variables and the Mann-Whitney U test was used for continuous variables.

Binary logistic regression was performed to analyse the association between the independent variables and the primary outcome, that is, the achievement of symmetrical knee muscle function 1 year after ACL reconstruction.

Finally, a forward stepwise multivariable logistic model including all independent variables was used to predict the achievement of symmetrical muscle function at 1 year. A secondary model, excluding the three hop tests, was also performed. In this second analysis, knee extension symmetry was entered first, based on previous publications.12 16–21 The variables which had a p value of <0.20 in the univariable analyses were entered in the two stepwise regression models, where the variable with the highest area under the receiver operating characteristics (ROC) curve was entered first.

The results from the logistic regression models were presented with an OR, that is, the ratio between the odds of an increase in the predictor, 95% CIs and p values. The area under the ROC curve was given as a measurement of goodness of fit, using limits of 0.90–1=excellent, 0.80–0.90=good, 0.70–0.80=fair, 0.60–0.70=poor and 0.50–0.60=fail.31 All significance tests were conducted at the 5% significance level.

Results

A total of 237 athletes (59% female; mean age 22±4 years) were included in the study (figure 1). Moreover, 122 athletes were excluded because they were not able to perform any of the hop tests and 31 athletes were excluded as they only performed one or two hop tests. There were no differences in patient demographics at the 4-month follow-up between included and excluded athletes (table 2).

Figure 1

Flow chart of inclusion and exclusion.

Table 2

Baseline demographics and dropout analysis for included and excluded patients

	Eligible(n=390)	Included(n=237)	Excluded(n=153)	P value*
Patient sex
Female	236 (60.5%)	139 (58.6%)	97 (63.4%)	0.41
Height (cm)	174.0 (150.0; 200.0)	174.0 (150.0; 200.0)	175.0 (155.0; 197.0)	0.21
Weight (kg)	70.0 (45.0; 118.0)	70.0 (45.0; 102.0)	72.0 (49.0; 118.0)	0.091
BMI (kg/m2)	23.3 (17.9; 38.7)	23.1 (18.5; 35.4)	23.9 (17.9; 38.7)	0.099
Age at index surgery	21.7 (4.3)	21.5 (4.2)	22.1 (4.4)	0.18
Preinjury physical activity level†
6 (%)	22 (6)	13 (6)	9 (6)
7 (%)	58 (15)	33 (14)	25 (16)
8 (%)	116 (30)	71 (30)	45 (29)
9 (%)	136 (35)	82 (35)	54 (35)
10 (%)	58 (15)	38 (16)	20 (13)	0.45
Graft choice
Hamstring (%)	285 (85.1)	190 (86.0)	95 (83.3)
Patella (%)	42 (12.5)	27 (12.2)	15 (13.2)
Allograft (%)	5 (1.5)	3 (1.4)	2 (1.8)
Other graft (%)	2 (0.6)	1 (0.5)	1 (0.9)	0.67

For categorical variables, n (%) is presented.

Patient’s height, weight and BMI are presented as the median (min; max).

Patient’s age at index surgery is presented as the mean (SD).

For comparisons between groups, Fisher’s exact test (lowest one-sided p value multiplied by 2) was used for dichotomous variables, the χ2 test was used for non-ordered categorical variables and the Mann-Whitney U test was used for continuous variables.

*Comparison between included and excluded patients.

†As measured with the Tegner Activity Scale.

BMI, body mass index.

In all, 26% (62/237) had achieved symmetrical muscle function 1 year after ACL reconstruction. At the 4-month follow-up, male athletes had significantly higher K-SESpresent (difference in medians: 1 unit, p<0.0002) and higher KOOS4 (difference in means: 5.5 units, p=0.011) compared with female athletes, table 3. Moreover, male athletes had a higher LSI in the single-leg hop for distance (difference in means: 8.2 cm, p=0.0054) compared with female athletes.

Table 3

Four-month outcome (independent variable) and between-group analysis between female and male athletes

	Included(n=237)	Female(n=139)	Male(n=98)	P value*
K-SESpresent	6.3 (1.6; 9.7) n=176	6.1 (1.6; 9.1) n=110	7.1 (2.1; 9.7) n=66	0.0002
K-SESfuture	7.8 (0.0; 10.0) n=176	7.8 (3.0; 10.0) n=110	8.0 (0.0; 10.0) n=66	0.084
KOOS4	63.4 (13.2) n=176	61.3 (13.2) n=110	66.8 (12.4) n=66	0.011
LSI knee extensor (%)	82.5 (15.5) n=166	80.3 (16.1) n=102	86.1 (13.8) n=64	0.066
LSI knee flexor (%)	92.6 (14.3) n=166	92.2 (14.2) n=102	93.3 (14.4) n=64	0.83
LSI vertical hop (%)	82.6 (74.8) n=96	86.3 (102.5) n=50	78.7 (18.5) n=46	0.16
LSI hop for distance (%)	82.7 (15.7) n=90	78.5 (16.2) n=44	86.7 (14.3) n=46	0.0054
LSI side hop (%)	80.2 (23.3) n=61	77.2 (23.6) n=27	82.5 (23.2) n=34	0.21

K-SESpresent and K-SESfuture are presented as the median (min; max).

The KOOS4 and the results of muscle function tests are presented as the mean (SD).

For comparisons between groups the Mann-Whitney U test was used for continuous variables.

*Comparison between females and males.

KOOS, Knee injury and Osteoarthritis Outcome Score; K-SES, Knee Self-Efficacy Scale;LSI, Limb Symmetry Index.

Univariable analyses

ORs, CIs and p values for the tested variables are summarised in table 4 and figure 2. None of the patient demographics were found to be significant when attempting to predict the achievement of symmetrical muscle function 1 year after ACL reconstruction. Athletes with higher self-efficacy at 4 months had increased odds for symmetrical muscle function at 1 year after ACL reconstruction; K-SESpresent, OR 1.28 (95% CI 1.04 to 1.58, p=0.020). Moreover, increased odds for achieving symmetrical muscle function were found in athletes with more symmetrical knee-extension strength, OR 1.73 (95% CI 1.28 to 2.34, p=0.0004) and knee-flexor strength, OR 1.39 (95% CI 1.07 to 1.81, p=0.015). In terms of hop tests, increased odds were found in athletes with more symmetrical hop performance in the vertical hop, OR 1.77 (95% CI 1.27 to 2.45, p=0.0006), the single-leg hop for distance, OR 1.98 (95% CI 1.24 to 3.17, p=0.0043) and the side hop, OR 1.64 (95% CI 1.15 to 2.33, p=0.0059).

Table 4

Univariable logistic regression model with muscle function recovery as the dependent outcome*

Variable	Missing (n)	Value	Muscle function recovery (1 year)*	Univariable
				OR (95% CI)Muscle function recovery (1 year)	P value	Area under ROC curve (95% CI)
Patient sex	0	Female	32 (23.0%)
		Male	30 (30.6%)	1.48 (0.82 to 2.64)	0.19	0.55 (0.48 to 0.62)
Height (cm) (OR per 10 units)	0	150 to <170	17 (22.4%)
		170 to <179	22 (25.3%)
		179–200	23 (31.1%)	1.30 (0.96 to 1.78)	0.094	0.56 (0.48 to 0.65)
Weight (kg) (OR per 10 units)	0	45 to <66	20 (24.4%)
		66 to <76	19 (23.5%)
		76–102	23 (31.1%)	1.19 (0.92 to 1.55)	0.18	0.55 (0.47 to 0.64)
Age (dichotomised)	0	15–20	26 (22.2%)
		21–30	36 (30.0%)	1.50 (0.84 to 2.69)	0.17	0.55 (0.48 to 0.62)
Tegner preinjury	0	6	5 (38.5%)
		7	14 (42.4%)
		8	8 (11.3%)
		9	24 (29.3%)
		10	11 (28.9%)	0.93 (0.71 to 1.21)	0.59	0.51 (0.42 to 0.59)
KOOS4 (OR per 10 units)	61	24 to <58	17 (29.3%)
		58 to <70	11 (18.6%)
		70–93	20 (33.9%)	1.17 (0.90 to 1.52)	0.23	0.54 (0.45 to 0.64)
K-SESpresent	61	1.6 to <5.5	12 (21.4%)
		5.5 to <7	15 (25.4%)
		7–9.7	21 (34.4%)	1.28 (1.04 to 1.58)	0.020	0.60 (0.51 to 0.69)
K-SESfuture	61	0 to <5.5	6 (27.3%)
		5.5 to <7	9 (28.1%)
		7–10	33 (27.0%)	1.03 (0.85 to 1.24)	0.79	0.52 (0.42 to 0.62)
Graft choice	20	Hamstring	52 (27.4%)
		Patella	6 (22.2%)	0.76 (0.29 to 1.98)	0.57	0.51 (0.47 to 0.56)
LSI knee extension (OR per 10 units)	71	38 to <79	6 (9.8%)
		79 to <90	15 (31.3%)
		90–153	26 (45.6%)	1.73 (1.28 to 2.34)	0.0004	0.69 (0.61 to 0.77)
LSI knee flexion (OR per 10 units)	71	42 to <90	15 (21.7%)
		90–141	32 (33.0%)	1.39 (1.07 to 1.81)	0.015	0.62 (0.52 to 0.71)
LSI vertical hop (OR per 10 units)	141	35 to <65	1 (3.7%)
		65 to <90	18 (36.0%)
		90–149	9 (47.4%)	1.77 (1.27-2.45)	0.0006	0.75 (0.64 to 0.85)
LSI hop for distance (OR per 10 units)	147	37 to <81	4 (13.3%)
		81 to <90	8 (28.6%)
		90–106	13 (40.6%)	1.98 (1.24 to 3.17)	0.0043	0.72 (0.60 to 0.84)
LSI side hop (OR per 10 units)	176	25 to <69	2 (11.1%)
		69 to <90	4 (21.1%)
		90–125	13 (54.2%)	1.64 (1.15 to 2.33)	0.0059	0.74 (0.61 to 0.87)

OR is the ratio for the odds of an increase in the predictor of 1 unit.

Muscle function recovery achieving an LSI of ≥90% in all five tests included in the battery of tests.

All tests are performed with univariable logistic regression.

P values, OR and area under the receiver operating characteristics (ROC) curve are based on original values and not on stratified groups.

*Achieving an LSI of ≥90% in five tests of muscle function.

KOOS, Knee injury and Osteoarthritis Outcome Score; K-SES, Knee Self-Efficacy Scale; LSI, Limb Symmetry Index; ROC, receiver operating characteristics.

Figure 2

Univariable logistic regression models, ORs and 95% CIs for muscle function recovery, defined as achieving a Limb Symmetry Index of ≥90% in five tests of muscle function 1 year after an ACL reconstruction. For patient sex, an OR of >1 indicates a result favouring males. For age at primary ACL reconstruction, an OR of >1 favours the older age group (21–30 years old). For graft choice, an OR of >1 indicates a result favouring patellar tendon autografts. KOOS4, Knee Osteoarthritis Outcome Score (four dimensions); K-SESfuture, future Knee Self-Efficacy Scale; K-SESpresent, present Knee Self-Efficacy Scale; LSI, Limb Symmetry Index; Tegner preinjury, Tegner Activity Scale prior to the injury.

Multivariable analyses

A total of 61 athletes were able to perform all five tests of muscle function at the 4-month follow-up and were included in the multivariable analysis to predict the achievement of symmetrical knee muscle function 1 year after ACL reconstruction. No predictive multivariable model could be created because of data missing from hop tests and the small number of patients who had achieved the dependent outcome of symmetrical muscle function.

When excluding athletes with missing data (hop tests, n=105) from the three hop tests at 4 months, a total of 166 athletes were included in the second multivariable model. No combinations of variables were significant when performing the multivariable model.

Discussion

The principal findings in this study were that more symmetrical knee-extension and knee-flexion strength, a more symmetrical hop performance and higher present self-efficacy, 4 months after ACL reconstruction, were associated with symmetrical muscle function in young athletes at 1 year. In addition, just over one in four of the athletes had recovered their muscle function, defined as an LSI of ≥90% in all five muscle function tests 1 year after the ACL reconstruction.

A multivariable logistic model, including all the independent variables from the 4-month follow-up after ACL reconstruction, was not created, as only 19 of the total of 61 athletes had complete data from all independent variables and had achieved symmetrical muscle function at 1 year. Creating a multivariable model based on these 19 events would have resulted in an overfitted model and thereby an unreliable result.32 When we excluded the hop tests as independent variables, it became impossible to create a predictive multivariable model for symmetrical muscle function 1 year after ACL reconstruction as the inclusion of any other variable resulted in a non-significant model.

Demographics

Younger age and being male are both associated with better outcome, in terms of RTS and perceived knee function.33–35 There are conflicting results related to age as a predictor of achieving symmetrical muscle function in a battery of tests in previous studies.24 36 However, none of the patient demographics were found to be associated with the achievement of symmetrical muscle function in the present study.

Patient-reported outcomes

In the present study, the psychological outcome of self-efficacy was associated with regaining symmetrical muscle function; this extends findings from a previous study17 where higher levels of preoperative self-efficacy were associated with achieving symmetrical muscle function in the single-leg hop for distance. According to the theory of self-efficacy,37 a person with high self-efficacy who confronts a setback will increase his/her level of effort and persistence. We assumed that athletes with high self-efficacy would be more motivated to participate in, and be compliant with, rehabilitation and therefore have greater potential to achieve more symmetrical muscle function compared with athletes with low self-efficacy. This is the first study investigating the association between self-efficacy and symmetrical muscle function across a battery of tests. However, this result must be considered with caution, as the area under the ROC curve was 0.60, meaning that the predictive value of this model is poor.31 This is one of the first studies to use psychological outcomes as predictors of symmetrical muscle function. Future studies are needed to further evaluate the association between early psychological outcome and symmetrical muscle function.

Muscle function

In the present study, all three hop tests were associated with symmetrical muscle function 1 year after ACL reconstruction. The areas under the ROC varied between 0.72 and 0.75 for all three models, indicating that these models had a fair goodness of fit.31

The finding that single-leg hop tests are associated with enhanced outcome has previously been reported in the literature. Logerstedt et al 38 reported that single-leg hop tests, conducted at 6 months after ACL reconstruction, were associated with self-reported knee function 1 year after surgery. Interestingly, more than 50% of the athletes in the present study who were already able to perform a symmetrical side hop at the 4-month follow-up achieved symmetrical muscle function 1 year after the ACL reconstruction. However, about 40% of the included patients who had performed the knee strength tests on both follow-up occasions (n=166) were able to perform all three hop tests as early as 4 months after the reconstruction. This finding could, however, be expected as many rehabilitation protocols recommend beginning with hop exercises at about 4 months after reconstruction.39–41 In addition, this finding provides valuable information for all clinicians and athletes in order to set realistic expectations during rehabilitation and highlights the fact that not all patients will be able to perform hop tests early in their rehabilitation. In comparison, 61% (237/390) of the athletes who were evaluated with tests of muscle function 1 year after the ACL reconstruction were able to perform all three hop tests (figure 1).

The finding that knee-extension strength was associated with symmetrical muscle function in all five tests of muscle function confirms the findings reported by Hartigan et al,36 who reported that patients with more symmetrical isometric knee-extension strength preoperatively predicted the fulfilment of their RTS criteria, including strength and hop tests, 6 months after ACL reconstruction. In addition, Nawasreh et al 11 reported that patients who pass their RTS criteria, including tests of muscle function and PROs, 6 months after ACL reconstruction, continued to maintain higher knee function, compared with patients who did not fulfil the criteria early in the rehabilitation process. However, in the present study, the area under the ROC was 0.69 for the effect of knee-extension strength on achieving symmetrical knee function, which corresponds to a poor goodness of fit.31

In the present study, almost 40% of the athletes who had an LSI of 80% or higher in knee extension at 4 months following the reconstruction achieved symmetrical muscle function at 1 year compared with fewer than 10% of the athletes who had an LSI lower than 80%. This finding suggests that patients who have a deficit in knee-extension strength of more than 20% early after ACL reconstruction may need more than 1 year of rehabilitation to achieve symmetrical muscle function. Considering the positive effects of better short-term outcomes in terms of enhanced self-reported function12 13 15 16 and enhanced muscle function,42–45 clinicians are recommended to focus on optimising knee-extension strength during the entire rehabilitation process.

Limitations and strengths

This prospective cohort study has some limitations that were taken into account before conclusions were drawn. First, a minority of the athletes who performed the tests of muscle function were able to perform all three hop tests at the 4-month follow-up. This resulted in an overly limited population and a multivariable model could therefore not be created.

Second, as two different tests to assess the athletes’ muscle strength were used, we chose not to analyse the absolute value for each test in the statistical analysis. The outcomes from the two methods are regarded as comparable, as they are presented as an LSI. However, the use of the LSI has some limitations in itself.46 47 Since the LSI is calculated as a ratio between the values for the injured and uninjured legs and deficits in strength are common in both legs after ACL reconstruction, the LSI might underestimate muscle function requirements.46 48

This is the first study to investigate predictors of achieving symmetrical muscle function in young athletes, 15–30 years old, after ACL reconstruction. A large population of 237 young athletes was included with results generalisable to a large extent to the general ACL population, as the included athletes did not differ from the excluded 153 athletes. Moreover, the methods used for assessing strength and hop performance, as well as the PROs that were used, have all been reported to be reliable and valid for evaluating patients after an ACL reconstruction.

Conclusions

Symmetrical knee-extension and knee-flexion strength, a more symmetrical hop performance and higher present self-efficacy at an early stage all increased the odds of achieving symmetrical muscle function in young athletes 1 year after ACL reconstruction.