## Introduction

Anterior cruciate ligament (ACL) reconstruction surgery is a standard procedure for treating ACL injury. The goals of ACL reconstruction are to restore normal knee biomechanics and laxity, to help patients recover to pre-injury activity levels, and to improve the quality of daily life1,2. However, many patients still experience muscle weakness and fail to return to pre-injury activity level, even after successful surgery and rehabilitation3.

Adequate quadriceps strength, functional outcomes, and psychological preparedness are standard criteria for returning to sports4. Quadriceps weakness is critically detrimental to the return to sports because it is significantly associated with poor functions and deteriorated quality of life after ACL reconstruction5,6,7. Furthermore, patients with an asymmetry of quadriceps strength are also at higher risk of re-injury after ACL reconstruction than the normal population8,9. Although quadriceps muscle strength deficit is expected10,11,12 and essential for ACL reconstruction outcomes, it is not convenient to regularly assess the isokinetic muscular strength during follow-up due to financial and time costs. Instead, self-reported questionnaires evaluate the knee functional status by identifying pain, functional problems, and physical activity levels. The Lysholm score, International Knee Documentation Committee form (IKDC), Hospital for Special Surgery score, and Knee Outcome Survey score are popular for qualitatively assessing subjective knee functions after ACL reconstruction13,14.

Preoperative quadriceps strength has been proposed as a predictor for ACL reconstruction functional outcomes measured15,16. Severe preoperative quadriceps strength deficit is a factor for persistent quadriceps weakness after reconstruction15,16,17,18. Furthermore, previous studies have found that greater preoperative muscular strength can improve outcomes12,15. Therefore, several studies have suggested adding preoperative exercise to increase quadriceps strength before ACL reconstruction surgery to achieve better functions after reconstruction18,19. However, we doubt whether this relationship is applicable in the setting of different grafts and different functional scores at different time periods for evaluation. Eitzen et al.15 evaluated knee functions in ACL reconstructed patients with bone-patellar -tendon-bone graft (BPTB) graft 2 years after surgery and Logerstedt et al.16 reported isometric muscle strength and functional score 6 months after ACL reconstruction. In addition, studies reported the mean time between surgery and resumption of sports of 7.3 (range 2–24) months20. Hence, 1 year may be appropriate to evaluate knee joint with the adequate recovery of muscle strength and functional ability for return to play.

Therefore, we evaluated isokinetic knee strength and subjective knee functions (Lysholm and IKDC scores) in participants with ACL reconstruction using hamstring graft 1 year after surgery. The primary aim of this study is to investigate the relationship between preoperative quadriceps strength and postoperative subjective knee functions after ACL reconstruction. The secondary aim is to see in which population the effect of quadriceps strength deficit on functional outcomes was most predominant.

## Methods

### Participants

This retrospective study enrolled all consecutive patients who underwent primary ACL reconstruction surgery between 2014 and 2017 in Samsung Medical Center. We included male patients with primary ACL reconstruction surgery who were preoperative evaluated and followed for more than 1 year. We excluded patients with a revision ACL reconstruction surgery, multi-ligamentous knee injury, prior ipsilateral lower limb surgeries, inflammatory joint disease, or other conditions affecting the lower limbs. All participants provided informed consent before participation in this study. The study was approved by the Sungkyunkwan University School of Medicine Clinical Research Ethics Board (SMC 2016-09-074-001). All methods were carried out in accordance with relevant guidelines and regulations.

### Surgical technique: ACL reconstruction

An anatomical, single-bundle ACL reconstruction surgery with quadruple semitendinosus and gracilis tendon autografts was performed by the same senior surgeon for all patients included in this study.

### Rehabilitation

All patients received an appropriate, graduated exercise rehabilitation program, consisting of home-based rehabilitation and supervised checkups while visiting the hospital at 15 days, 1, 3, and 9 months after surgery. In the first postoperative week, patients received light-intensity isometric and range of motion (ROM) exercises. The degree of knee flexion was increased by 15° each week in the ROM exercise program. In the 4th week, the knee joint ROM reached 90° and was increased to 120° in the 6th week. The patients were assigned for partial weight-bearing at 4–6 weeks postoperatively, progressing to full weight-bearing. At between 6 and 12 weeks, subjects performed combined strength, endurance, and balance exercises without pain. Light running was permitted at 6 months. Directional running and sports-related training were allowed at 9 months after surgery, in the absence of any swelling, pain, or knee instability.

### Muscle strength evaluation

A bioimpedance analyzer (InBody 170, BIOSPACE, KOREA) was used to measure the height, weight, body composition. It provides separate body mass readings for different segments of the body and uses an algorithm incorporating impedance, age and height in order to estimate total and regional body fat and fat-free mass. We evaluated quadriceps and hamstring muscle strength by the isokinetic dynamometer (CSMI Medical Solutions, Stoughton, MA, USA). After warm-up using a stationary bike for 10 min, the patients sit on the dynamometer and practiced the test protocol to become familiarized. The knee joint was moved 0°–90° at a speed of 60°/s (Nm) for isokinetic strength measurement. Finally, the patients were verbally encouraged to exert maximum effort on the machine. The patients performed four maximum repetitions of knee flexion and extension at 60°/s (Nm), and a 1 min resting period was allowed between consecutive tests. We recorded the hamstring quadriceps peak torque strength, quadriceps peak torque strength, and other indicators. The peak torque refers to the maximum acting torque released by the muscle during contraction. It is the torque value at the highest point of the torque curve and represents the maximum muscle force produced by muscle contraction. The mean peak torque of four repetitions for each velocity was calculated and compared with that on the uninjured side and described as absolute values and the percent of strength deficit calculated with the following formula:

$$Strength~\,deficit = \left( {{\raise0.7ex\hbox{{noninjured\, limb - injured\,limb}} \!\mathord{\left/ {\vphantom {{noninjured\,limb - injured\,limb} {noninjured\,limb}}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{{noninjured\,limb}}}} \right) \times 100\%$$

All isokinetic strength measurements were performed by the same experienced clinical physiologist, who was blinded to the testing result.

### Patient-reported outcome measures

The IKDC21,22 and Lysholm scores23 were used to measure subjective knee function. The IKDC score is a joint-specific outcome measure for assessing symptoms, function, and sports activity pertinent to various knee conditions21. The IKDC contains items regarding symptoms and disabilities important to patients with ACL deficiency22. The Lysholm score, a validated questionnaire, is used to evaluate and quantify patient satisfaction using 100 points scale that measures instability, limping, orthosis utilization, stair climbing, squatting, walking, running, jumping in daily life, pain, swelling, and thigh muscle atrophy23. The IKDC and Lysholm score forms are two widely used questionnaires among researchers and clinicians capable of measuring patient-reported knee function outcomes with good reliability and validity14,24.

### Follow-up

We assessed each participant’s knee function, including isokinetic muscle strength and functional score at baseline (preoperative) and at 12 months after index surgery.

### Statistical analysis

A priori calculated power > 0.80 at an alpha level equal to 0.05 was used to determine that a sample size of 45 for each group was necessary to determine R2. In this study, descriptive statistics were used to describe the basic features of the data. All the values are presented as median with interquartile range (IQR). The statistical analysis to determine associations between the two variables (preoperative quadriceps muscle strength deficit and the knee functional scores at 1 year after surgery) was conducted by the Spearman rank correlation analysis. We divided the patients into three groups according to the quadriceps muscle strength deficit. Although previous study used a 20% cut-off level for assessing the muscle strength deficit15, we set the muscle strength deficit 25% rather than 20% considering patients with unreconstructed ACL insufficiency knee may have a poor strength than those with ACL reconstructed knee. Meanwhile, ACL reconstructed patients could have a quadriceps strength deficit more than 45% during follow-up12 Therefore, the three identified groups (Q1–Q3) were classified according to the preoperative quadriceps muscle strength deficit differences. Group Q1 had the smallest difference with < 25% deficit in quadriceps muscle strength, Q2 showed a 25–45% deficit, and Q3 included patients who had > 45% deficit. Kruskal–Wallis test was used to compare parameters among three groups and Dunn test was used for post hoc analysis. Wilcoxon signed-rank test was used to compare baseline and postoperative functions within each group. A partial correlation analysis was then performed to determine the relationship between the knee functions at 1 year after surgery and the quadriceps muscle strength deficit in all patients and by level of strength deficit. Spearman’s partial correlation analysis, a nonparametric method, was used since some variables did not satisfy normality assumptions. Age, height, weight, and body fat percentage may affect the quadriceps muscle strength and were adjusted. The significance level was set at 0.05. All statistical analyses were performed using SPSS version 13.0 (SPSS Inc., Chicago, IL, USA) and MedCalc for Windows, version 18.6 (MedCalc Software, Ostend, Belgium).

## Results

Seventy-five male patients who underwent primary ACL reconstruction surgery were included in this study. The mean age was 32 (IQR, 26.3–39.8). Forty-seven (62.7%) had concomitant meniscus injury. Thirty-four patients received meniscus repair, and the remaining thirteen patients had partial meniscectomy. The patients were further divided into three groups according to the differences in preoperative quadriceps muscle strength deficit. Group Q1 (n = 26) had the smallest difference with < 25% deficit in quadriceps muscle strength, Q2 (n = 29) showed a 25–45% deficit, and Q3 included patients with a > 45% deficit (n = 20). There were no statistically significant differences in base demographics and the type of meniscus surgery between the three groups.

The baseline knee functional scores differed between the three groups, including IKDC score (p < 0.001) and Lysholm score (p < 0.001). There were also significant differences between the groups in knee extensor (p < 0.001) and flexor strength (p < 0.001). The baseline characteristics of the patients are shown in Table 1. At 1 year after reconstruction, the isokinetic strength and knee functional scores had improved (Supplemental Table 1). Q1 had significantly better isokinetic strength and knee function scores, including IKDC and Lysholm scores than Q3 (Table 2). Compared with Q2, Q1 had a trend of higher knee functions and isokinetic muscle strength, though not reaching the statistical significance.

### Relationship between preoperative quadriceps muscle strength and postoperative knee functional score

The preoperative quadriceps muscle strength deficit had a negative association with the knee functional scores at 1 year follow-up including the IKDC score (rs =−0.397, p = 0.005, Fig. 1a) and the Lysholm score (rs = − 0.454, p < 0.001, Fig. 1b).

### Postoperative knee functional score by levels of quadriceps muscle strength deficit

When comparing knee function at 1 year after surgery, the three groups were significantly different in knee functional score, including IKDC (p = 0.010) and Lysholm score (p < 0.001) (Fig. 2a,b).

Partial correlation analysis was performed to examine the correlation between knee functional scores by different preoperative quadriceps muscle strength deficit levels (Table 3). Only the Q1 group (< 25% difference in preoperative quadriceps muscle strength) showed a statistically significant correlation in postoperative knee functions, measured by the IKDC score and Lysholm score. The Q2 and Q3 groups, which had more than 25% difference in preoperative quadriceps muscle strength, showed no significant association with knee functional scores 1 year after surgery, neither by IKDC nor Lysholm scores. In addition, preoperative quadriceps muscle strength deficit level had a statistically significant negative correlation with subjective knee functions including IKDC score and Lysholm score (Supplemental Table 2).

## Discussion

This study’s principal findings showed that preoperative quadriceps strength deficit negatively predicted knee functional scores 1 year after ACL reconstruction surgery. Patients with more significant preoperative quadriceps strength deficit would have poor knee functions even after rehabilitation. Therefore, preoperative rehabilitation, including quadriceps strengthening, may be crucial to functional recovery after ACL reconstruction surgery.

Muscle weakness in the lower limbs appears to be due mainly to neuromuscular dysfunction and decreased activity level after ACL injury25,26 and the restricted ROM during the recovery phase after ACL reconstruction. In particular, weakening of the quadriceps muscles is known to negatively affect knee functions after reconstruction11,27,28 The degree of strength deficit after ACL reconstruction is one of the important indicators when evaluating the state of the knee joint, and clinically no more than a 10% difference should be permitted for patients returning to sport3,29. However, studies conducted on the extent of strength deficit after ACL reconstruction have found quadriceps muscle strength deficits of 10–27% after 1 year, and the deficiency remained at 6–10% even after more than 5 years following surgery30,31,32.