© 2014 John Wiley & Sons A/S.

Scand J Med Sci Sports 2014: ••: ••–•• doi: 10.1111/sms.12215

Published by John Wiley & Sons Ltd

Anterior cruciate ligament injury after more than 20 years. II. Concentric and eccentric knee muscle strength E. Tengman1, L. Brax Olofsson2, A. K. Stensdotter1,3, K. G. Nilsson2, C. K. Häger1 Department of Community Medicine and Rehabilitation, Section for Physiotherapy, Umeå University, Umeå, Sweden, 2Department of Surgical and Perioperative Sciences, Section for Orthopaedics, Umeå University, Umeå, Sweden, 3Faculty of Health Education and Social Work, Section for Physiotherapy, Sør-Trøndelag University College, Trondheim, Norway Corresponding author: Eva Tengman, PhD Student, Section for Physiotherapy, Umeå University, SE-90187 Umeå, Sweden. Tel: +0046 703770680, Fax: +004690 7869267, E-mail: [email protected]

1

Accepted for publication 17 February 2014

The long-term consequences on knee muscle strength some decades after rupture of the anterior cruciate ligament (ACL) are not established. The aims of our study were to examine peak torque more than 20 years after ACL injury and to compare their knee muscle strength to that of healthy controls. We tested 70 individuals with unilateral ACL injury 23 ± 2 years after injury, whereof 33 (21 men) were treated with physiotherapy in combination with ACL reconstruction (ACLR) and 37 (23 men) with physiotherapy alone (ACLPT). These were compared with 33 age- and gender-matched controls (21 men). A Kin-Com® dynamometer (90°/s) was used to measure peak torque in knee flexion and extension in both concen-

tric and eccentric contractions. Knee extension peak torque, concentric and eccentric, was ∼10% lower for the injured leg compared with the non-injured leg for both ACLR (P < 0.001; P < 0.001) and ACLPT (P = 0.007; P = 0.002). The ACLPT group also showed reduced eccentric knee flexion torque of the injured leg (P = 0.008). The strength of the non-injured leg in both ACL groups was equal to that of controls. No difference was seen for those with no-or-low degree of knee osteoarthritis compared to those with moderate-to-high degree of osteoarthritis. ACL injury may lead to a persistent reduction of peak torque in the injured leg, which needs to be considered across the lifespan.

Rupture of the anterior cruciate ligament (ACL) is a common injury to the knee joint, which typically occurs in sports and recreational activities. ACL injury impacts knee function with negative consequences regarding return to sport or leisure activities as well as physical activity level in both the short-term (Ageberg, 2002; Ardern et al., 2011) and the long-term perspectives (Kostogiannis et al., 2007; Streich et al., 2011; Frobell et al., 2013). In the acute stage, ACL injuries are managed either with physiotherapy in combination with ACL surgical reconstruction, or alternatively treated conservatively with physiotherapy alone. There is however insufficient evidence whether any of these treatments is superior in a long-term perspective (Delince & Ghafil, 2012; Frobell et al., 2013). Regardless of other treatments, physiotherapy is recommended and usually includes assessments for joint mobility, muscle strength, functional activities, and neuromuscular control (Ageberg, 2002; Trees et al., 2005). A primary objective of ACL rehabilitation is to restore knee function to the same level as the non-injured leg, commonly evaluated with the “limb symmetry index” (LSI, side-to-side difference), where < 90%, i.e., more than 10% difference between injured and non-injured legs, is generally regarded as unsatisfactory (Augustsson et al., 2004).

Despite rehabilitation, knee muscle weakness is one of the main dysfunctions after ACL injury (Ageberg, 2002; Palmieri-Smith et al., 2008) and may contribute to predisposition to develop knee osteoarthritis (OA) (Keays et al., 2010). Still, short-term strength deficits may diminish with time (Ageberg et al., 2007; PalmieriSmith et al., 2008). Only a couple of studies have addressed the impact of the injury on knee muscle strength after 10–15 years, showing that knee muscle strength of the injured leg may not be prominently reduced compared with that of the non-injured leg (LSI values ranging between 90% and 102%) (Ageberg et al., 2007; Holm et al., 2010; Oiestad et al., 2010a). LSI > 90% does not however necessarily indicate sufficient strength as strength in the contralateral leg may also be reduced. Furthermore, general decline in leg strength during middle age and later has to be considered in long-term follow-up studies (Kim et al., 2010). Therefore, it is highly important to compare with age- and gender-matched controls. Twenty years or more is a long time when several factors may come into play, potentially with negative influence on strength and function, such as increasing age, reduced physical activity level (Mihelic et al., 2011; Streich et al., 2011) (cf. Tengman et al., 2014) and OA, which is common after ACL

1

Tengman et al. injury (von Porat et al., 2004; Mihelic et al., 2011; Streich et al., 2011). Reduced isokinetic knee muscle strength for quadriceps (Moisala et al., 2007) and hamstrings (Tsepis et al., 2004; Moisala et al., 2007) after ACL injury is also shown to be associated with poorer subjectively experienced knee function (Lysholm score). Good knee function includes concentric as well as eccentric strength, and focusing on eccentric quadriceps strength after ACL reconstruction has shown greater improvements in strength and hop performance than traditional exercise (Gerber et al., 2009). Insufficient quadriceps strength may yield inferior performance in activities that require eccentric muscle action such as stair descent (Gao et al., 2012). Concentric and eccentric action in the hamstring muscles is also important in, e.g., jump landing and late swing in running (Thelen et al., 2005). Hamstring co-contraction may also be effective in maintaining knee stability through synergistic action to the ACL, thereby preventing excessive anterior displacement and internal rotation of the tibia (Hirokawa et al., 1991). Moreover, hamstring activation provides reciprocal inhibition on the quadriceps, which contributes to muscular balance across the knee joint (Solomonow et al., 1987). Commonly, the ratio between concentric hamstring strength and quadriceps (H : Q ratio) is examined, whereas the relationship between eccentric hamstring and concentric quadriceps strength would appear more relevant as the stabilizing action of the hamstring is crucial during extension of the knee (Aagaard et al., 1998).

At one hospital, surgery was predominant. This cohort had physiotherapy and reconstructive surgery (ACLR). The other hospital administered a physiotherapy program and strategy that ACL reconstruction should be restricted to those who would gain clear benefit from surgery and not solely on severity of injury or activity level demands. This cohort had physiotherapy only (ACLPT). We did a follow-up on both cohorts. Inclusion criteria for the present study were as follows: unilateral ACL injury with a non-injured contralateral leg, no inflammatory or rheumatic disease or neurological pathology, and willingness to participate. Out of, in total, 113 eligible patients, 70 of the 81 who fulfilled the inclusion criteria agreed to participate; 33 ACLR and 37 ACLPT. Eleven individuals declined (nine in ACLR and two in ACLPT) due to living far away, or considered the testing too time-consuming or unknown reasons; these individuals however did not differ in background variables (e.g., age, body mass index, scores at KOOS, Lysholm and Tegner, and one-leg hop which was tested for in all 113 participants) from those who participated in the present study. In both groups, more than 90% of the patients displayed some form of OA at the time of investigation, varying from Kellgren & Lawrence stage 1 to stage 4 (Kellgren & Lawrence, 1957). The healthy control group consisted of a convenience sample of 33 individuals matched for age and gender recruited through advertisements and via staff and acquaintances. The controls reported no previous knee injuries and were clinically examined for any injury of the menisci or ligaments, which led to exclusion. Background data for the three groups are shown in Table 1. All participants were presented with written and oral information about the study and gave their written informed consent according to the Declaration of Helsinki. The project was approved by the Regional Ethical Review Board.

Description of ACL groups and treatments Both ACL groups and their treatment have been described in detail in the accompanying paper (Tengman et al., 2014).

Aim The main aim of this cross-sectional study was to determine several important aspects of knee muscle strength in the very long perspective, on average, 23 (and up to 28) years after a unilateral ACL injury in two groups of individuals who were either treated with physiotherapy in combination with reconstructive surgery, or with physiotherapy alone. We investigated strength differences between the injured and non-injured legs and compared their knee muscle strength to that of healthy controls. Further, we aimed to determine whether strength was correlated with knee function and kneespecific activity level, and was influenced by the degree of radiological knee OA.

ACLR treated with physiotherapy in combination with reconstructive surgery Participants were injured between the years 1981 and 1993 and were treated with an ACL reconstructive surgery, including preand post-operative physiotherapy in the period of 1987–1993. Nineteen participants had surgery with a patellar tendon quadriceps autograft augmented with a synthetic polypropylene braid (Kennedy LAD). Nine participants received a graft placed through a femoral tunnel using an aiming device (Odensten & Gillquist, 1986). Five participants received a bone-patellar tendon-bone graft according to the prevailing standard procedure (Lipscomb et al., 1981). During the first 14 weeks, a knee brace and crutches were used. The post-operative physiotherapy included functional with gradually increased levels of intensity and difficulty aiming for improved joint mobility, strength, coordination, and balance. The goal was full return to sports activities approximately 22 weeks after surgery.

Materials and methods Participants

ACLPT treated with physiotherapy alone

The patients in the present study are included in a larger unique follow-up study of two cohorts of individuals who injured their ACL, on average, 23 years ago, and who were treated at two separate hospitals in two different geographical regions of Northern Sweden. At both hospitals after the diagnosis of ACL injury had been established, the initial treatment consisted of physiotherapy for at least 3 months. The treatment decision was eventually made by the orthopedic surgeons at the respective hospitals.

Participants were injured between the years 1983 and 1988. Briefly, the treatment consisted of a one-to-one tailored physiotherapy program that focused on activity modification and progressively increased functional stability three times per week. All exercises were performed with both legs, and rehabilitation was regarded complete when the patient could perform the final stage of exercises safely and with good quality (Tegner, 1990). The mean time to reach this level and thus being allowed to return to

2

Knee muscle strength 23 years after ACL injury Table 1. Participant characteristics

Group

n participants Male/female Age at test Height (cm) Weight (kg) BMI (kg/m2) Years since injury Years between injury –surgery Cause of injury Soccer Alpine Other sports Non-sporting Lysholm KOOS Pain Symptoms ADL Sport/Rec QoL Tegner median [range]

ACLR

ACLPT

Knee healthy controls

33 21 / 12 45.6 [4.5] 174.0 [9.1] 83.0 [15.6] 27.2 [3.3] 23.9 [2.8] 3.8 [2.5]

37 23 / 14 48.1 [5.9] 173.5 [8.0] 87.1 [14.9] 28.9 [4.6] 23.1 [1.3] –

33 21 / 12 46.7 [5.0] 176.4 [9.8] 77.4 [14.9] 24.6 [2.5] – –

24 2 6 1 78 [18] 78 [18] 79 [20] 41 [16] 50 [28] 49 [22] 4 [3–7]

25 5 2 5 69 [17] 85 [16] 72 [19] 90 [15] 67 [29] 61 [25] 4 [2–7]

– – – – 100 99 [1] 98 [2] 100 99 [2] 98 [3] 6 [3–7]

Means and SD. ACLR, group treated with physiotherapy in combination with reconstructive surgery; ACLPT, group treated with physiotherapy alone; ADL, activities of daily living; QoL, quality of life.

sports was 22 weeks (range 12–60 weeks). Participation in high risk sports, such as soccer, floorball, and wrestling, was advised against.

Knee function and physical activity level We investigated if there were correlations between the strength and the subjectively reported knee function and knee-specific physical level, respectively. Knee function was assessed with Lysholm score (physician administrated, 100 = optimal knee function) (Tegner & Lysholm, 1985). Knee function and symptoms were also rated on the Knee injury and Osteoarthritis Outcome Score (KOOS) on 5 subscales: pain, symptoms, function in activities of daily living, sports and recreation (Sport/Rec), and kneerelated quality of life (QoL) (Roos et al., 1998). Knee-specific physical activity level was assessed according to the Tegner activity scale (physician administrated) (Tegner & Lysholm, 1985).

Data acquisition Peak torque of concentric and eccentric knee flexion and extension was tested with a Kin-Com® dynamometer at 90°/s (Kinetic communicator 125 Auto Positioning, Chattanooga Group Inc.; Hixon, TN, USA). Isokinetic measurements (concentric and eccentric) are commonly used to assess knee muscle strengths with high to very high reliability (Sole et al., 2007; Pua et al., 2008) and validity (Pua et al., 2008). The test procedure started with a standardized 6-min warm-up on a bicycle ergometer at a moderate intensity. Thereafter, jump tasks were performed in a movement analysis laboratory (reported in Tengman et al., 2014). A 15- to 30-min rest period was allowed prior to the isokinetic assessment, and during which, the principles of strength testing were explained. Participants were seated in the dynamometer with hips in 90° flexion and with arms resting on the lap. Individual adjustments were made for the seat of the chair to accommodate the length of the femur and for the lever arm to align the axis of rotation of the dynamometer with the anatomical axis of rotation of the knee joint (lateral femoral condyle). Stabilization straps were applied across

the chest, pelvis, and thighs. A cuff proximal to the malleolus secured the ankle. Range of motion (ROM) was set from 0–20° (depending on hamstring tightness) to 90° of knee flexion. Trial sessions with submaximal contractions were performed for familiarization with the equipment. All tests were performed on both legs; ACL-injured groups started with the uninjured leg and the control group started with the dominant leg, defined as the preferred leg with which to kick a ball. One set of six maximum repetitions was performed for each exercise starting with knee extension, followed by knee flexion after a period of 8-min rest. For both exercises, each concentric contraction was followed by an eccentric contraction with about 6-s rest in between. Verbal encouragement was provided to facilitate maximal efforts. The same physiotherapist (E. T.) tested all participants.

Torque data Raw data from isokinetic strength tests were exported from the Kin Com to a software program MrD View (custom-made, Department of Biomedical Engineering and Informatics, Norrland’s University Hospital, Sweden). Torque values for the initial and final 4° of ROM were excluded from the analyses to eliminate inertial effects close to the acceleration (start) and deceleration (end) of the movement (Pua et al., 2008). Limb weight was recorded and the torque was gravity-adjusted using the computer software. The repetition with the highest peak torque (normalized to body weight: N·m/kg) was used in the statistical analyses. Peak torque ratios were calculated for concentric knee flexion and knee extension (Hc : Qc) and for eccentric knee flexion and concentric knee extension (He : Qc). All torque values from one subject in the ACLPT and one in the ACLR were missing due to acute dizziness and time constraints, respectively. In addition, three single peak values (one for ACLR, ACLPT, and controls) were lost due to technical problems during data saving.

Statistics All variables were tested for normality using Shapiro-Wilk test, whereas Levene’s test was used for test of homogeneity. For

3

Tengman et al. with-in group comparisons, linear mixed models were used. Outcome variables were peak torque in N·m/kg (knee flexion and extension; concentric and eccentric), Hc : Qc and He : Qc. The fixed factors considered in the models were as follows: leg (injured and non-injured), gender (men and women), and radiographic OA (no-or-low and moderate-to-high degree of OA). Both injured and non-injured legs were included in the analysis for all participants. All two-way and three-way interactions were included in a first model and then all non-significant interactions and factors were successively removed to a final model. “Participant” was included in the model as a random effect. The estimated means, confidence intervals (CIs), F-values, and P-values from the model were presented. Comparisons to the control group were made with independent t-tests. Spearman’s rho correlation coefficient was used to assess associations between peak torque and subjective knee function (Lysholm, KOOS, and Tegner). The significance level was set to P < 0.05. All statistics were performed using the Statistical Package for the Social Sciences, version 20 (IBM SPSS, Armonk, New York, USA).

ACLR The models showed that leg had significant fixed effects for concentric and eccentric knee extension with a lower peak torque (13% and 14%, respectively) for the injured leg compared with the non-injured leg (Table 3). For knee flexion peak torque (concentric and eccentric), there were no differences between the injured and noninjured legs. Women displayed lower concentric knee extension torque (12%) and lower concentric knee flexion torque (20%) compared to men. For eccentric contractions, no statistically significant gender effect on strength was found. There was no difference in peak torque or LSI in any of the test conditions between those who had no-or-low OA compared to those who had a moderate-to-high degree of OA. ACLPT

Results Participants’ characteristics The participants’ characteristics, physical activity level (Tegner activity scale), and knee function (Lysholm score and KOOS) are presented in Table 1. Detailed analysis of the physical activity level, and knee function are reported in the parallel paper (Tengman et al., 2014). Knee muscle strength Table 2 shows the mean values for knee flexion and knee extension peak torques and side-to-side differences. The mean LSI ranged from 86% to 96% for the ACL-injured persons in extension and between 88% and 106% for flexion, whereas the control group showed LSI mean values between 94% and 104%. Table 3 displays the estimated mean values and the results of the statistical analysis, which will be described in more detail below.

The models showed that leg had significant fixed effects for concentric and eccentric knee extension, and eccentric knee flexion with a lower peak torque (ranging from 7% to 12% lower) for the injured leg compared to the non-injured leg (Table 3). For concentric knee flexion peak torque, there was no difference between the injured and non-injured legs. Women produced lower peak torques for knee flexion and extension during both concentric and eccentric contractions compared to men in all test conditions, with differences ranging from 13% to 29%. There was no difference in peak torque between those who had no-or-low OA compared to those who had a moderate-to-high degree of OA. Comparison to controls The injured leg in ACLR and ACLPT had a lower peak torque compared to controls regarding concentric and

Table 2. Knee extension and knee flexion peak torque normalized to body weight (N·m/kg)

Group

Knee extension peak torque ACLR Men Women ACLPT Men Women Healthy Controls Men Women Knee flexion peak torque ACLR Men Women ACLPT Men Women Healthy Controls Men Women Means and SD. *For controls, non-dominant leg. † For controls, dominant leg. LSI, limb symmetry index.

4

Concentric (N·m/kg)

Eccentric (N·m/kg)

Injured leg*

Non-injured leg†

LSI

Injured leg*

Non-injured leg†

1.93 ± 0.4 1.66 ± 0.3 1.90 ± 0.4 1.52 ± 0.4 2.23 ± 0.3 1.82 ± 0.3

2.17 ± 0.3 1.94 ± 0.4 2.06 ± 0.4 1.58 ± 0.3 2.24 ± 0.4 1.84 ± 0.3

89 ± 13% 87 ± 7% 93 ± 16% 96 ± 13% 100 ± 16% 99 ± 16%

2.62 ± 0.7 2.48 ± 0.6 2.49 ± 0.6 2.25 ± 0.7 3.08 ± 0.7 2.60 ± 0.4

3.05 ± 0.7 2.84 ± 0.6 2.89 ± 0.5 2.43 ± 0.5 3.19 ± 0.7 2.73 ± 0.6

86 ± 15% 89 ± 23% 88 ± 22% 93 ± 20% 97 ± 11% 98 ± 17%

1.23 ± 0.3 1.00 ± 0.2 1.23 ± 0.3 0.88 ± 0.2 1.33 ± 0.2 1.00 ± 0.1

1.22 ± 0.2 0.95 ± 0.2 1.22 ± 0.3 0.87 ± 0.1 1.29 ± 0.3 0.98 ± 0.2

101 ± 15% 106 ± 21% 100 ± 18% 101 ± 15% 104 ± 12% 104 ± 20%

1.46 ± 0.5 1.33 ± 0.4 1.29 ± 0.3 1.11 ± 0.7 1.49 ± 0.2 1.22 ± 0.2

1.54 ± 0.4 1.32 ± 0.3 1.47 ± 0.3 1.13 ± 0.3 1.51 ± 0.3 1.32 ± 0.2

94 ± 18% 100 ± 18% 88 ± 20% 98 ± 13% 100 ± 14% 94 ± 10%

LSI

NS NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS NS

NS

NS

NS NS

NS

NS

NS NS

NS

NS

NS NS

NS

Leg dominance

NS

OA*

F = 7.3, P = 0.011 Women 1.12 (0.97–1.27) Men 1.37 (1.26–1.27) NS

F = 7.8, P = 0.008 Injured 1.18 (1.08–1.29) Non-injured 1.31(1.21–1.42) F = 12.7, P = 0.001 Injured 0.62 (0.58–0.69) Non-injured 0.54 (0.50–0.57) F = 5.7, P = 0.022 Injured 0.63 (0.58–0.67) Non-injured 0.58 (0.54–0.63) NS

NS

NS

F = 4.17, P = 0. 045 Women 1.81 (1.61–1.99) Men 2.05 (1.90–2.20) F = 13.6, P < 0.001 Women 1.55 (1.36–1.74) Men 1.98 (1.84–2.12) NS

F = 32.8, P < 0.001 Injured 1.80 (1.58–1.83) Non-injured 2.06 (1.70–1.95) F = 7.8, P = 0.007 Injured 1.70 (1.58–1.83) Non-injured 1.83 (1.70–1.95) F = 17.5, P < 0.001 Injured 2.55 (2.31–2.78) Non-injured 2.95 (2.71–3.19) F = 10.87, P = 0.002 Injured 2.35 (2.16–2.54) Non-injured 2.68 (2.48–2.87) NS F = 4.5, P = 0.036 Women 2.34 (2.07–2.60) Men 2.69 (2.07–2.60) F = 10.9 P = 0.002 Women 1.09 (1.00–1.17) Men 1.11 (1.09–1.19) F = 20.9, P < 0.001 Women 0.87 (0.75–0.99) Men 1.22(1.13–1.31) NS

Gender

Leg

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

Interactions

NS

NS

P = 0.018

P = 0.022

P = 0.015

NS

NS

NS

P = 0.002

P = 0.041

P = 0.002

P = 0.010

Injured leg controls

Comparison to controls

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

Non-injured leg controls

*Bilaterally radiographic OA was graded according Kellgren & Lawrence (KL) and divided into participants with no-to-low (KL 0–1) or moderate-to-high (KL 2–4) degree of OA. The F-values and P-values from the model are presented together with the estimated marginal means (N·m/kg) and confidence interval (CI). “Participant” was included in the model as a random effect and all models had a significant random effect of “participant.” For comparisons between ACL-injured persons and controls, independent t-tests were used. NS, not significant.

He : Qc ACLR ACLPT

ACLPT

Hc : Qc ACLR

Ecc knee flexion ACLR ACLPT

ACLPT

Con knee flexion ACLR

ACLPT

Ecc knee extension ACLR

ACLPT

Con knee extension ACLR

With-in group

Table 3. Linear mix models were used for with-in group comparisons

Knee muscle strength 23 years after ACL injury

5

Tengman et al. Table 4. Hamstrings to quadriceps ratio

Group

ACLR ACLPT Healthy controls

Hc : Qc

Men Women Men Women Men Women

He : Qc

Injured leg*

Non-injured leg†

Injured leg*

Non-injured leg†

0.63 ± 0.12 0.60 ± 0.10 0.65 ± 0.13 0.60 ± 0.14 0.61 ± 0.12 0.56 ± 0.12

0.56 ± 0.08 0.51 ± 0.11 0.60 ± 0.12 0.57 ± 0.14 0.58 ± 0.11 0.53 ± 0.06

0.75 ± 0.16 0.79 ± 0.18 0.68 ± 0.15 0.76 ± 0.17 0.68 ± 0.12 0.68 ± 0.12

0.71 ± 0.14 0.71 ± 0.14 0.72 ± 0.14 0.74 ± 0.16 0.68 ± 0.15 0.72 ± 0.13

Means and SD. “c” stands for concentric and “e” stands for eccentric. *For controls, non-dominant leg. † For controls, dominant leg

eccentric knee extension. ACLPT had, in addition, lower peak torque in eccentric knee flexion. Peak torques in the non-injured leg in both ACLR and ACLPT were similar to healthy controls (Tables 2 and 3). In the control group, peak torques were similar for the dominant and nondominant legs in all conditions. Women produced a lower peak torques than men (P < 0.001–0.034) in all test conditions with differences ranging from 13% to 25%. Hamstrings to quadriceps ratio Mean and SD for Hc : Qc ratio and He : Qc ratios are presented in Table 4. The Hc : Qc ratio was higher in the injured leg compared to the non-injured leg and compared to healthy controls in both ACL-injured groups. In contrast, the ratios for the non-injured leg were similar to controls for both ACL groups (Table 3). No difference was seen between genders (Table 3). For He : Qc ratio, no significant fixed effects were seen neither between legs (injured and non-injured legs) or for gender. For the ACL injured, no difference in Hc : Qc and He : Qc ratios was seen between those who had no-or-low degree of OA compared to those who had moderate-to-high degree of OA. For the control group, there was no difference in any of the H : Q ratio variables between the dominant and non-dominant legs or with regard to gender. Correlations between peak torque and knee function and physical activity level For the participants with ACL injury, peak torque correlated positively with subjective knee function of the injured leg. A low-to-moderate correlation could be seen between all four peak torques obtained and the Lysholm score, where a higher value was related to a higher peak torque (N·m/kg) (concentric knee extension P < 0.001, r = 0.44; eccentric knee extension P = 0.008, r = 0.32; concentric knee flexion P = 0.006, r = 0.34; eccentric knee flexion P < 0.001, r = 0.52). Likewise, a low-tomoderate positive correlation was seen between the peak

6

torques and the subscale KOOS symptoms (concentric knee extension P = 0.003, r = 0.36; eccentric knee extension P = 0.018, r = 0.29; concentric knee flexion P = 0.022, r = 0.28; eccentric knee flexion P < 0.001, r = 0.52). No other correlations were found between peak torque and either of the other subscales of the KOOS or with the Tegner activity scale. Discussion The main aims of the present study were to determine the knee muscle strength in participants who suffered a unilateral ACL injury, on average, 23 years ago and to compare their knee muscle strength to that of healthy controls. As our study was not a randomized controlled trial (RCT), but a cross-sectional study a long time after injury, we have chosen not to directly compare the groups with ACL injury but to present and discuss the results separately. To our knowledge, no study has investigated knee muscle strength in a long-term follow-up after ACL injury regarding all of the following: (a) concentric as well as eccentric contractions in flexion and extension; (b) inclusion of patients who have had physiotherapy with or without surgery as initial treatment; and (c) comparison to healthy controls. The results showed that participants with ACL injury had reduced strength in the injured leg compared with their noninjured leg, but also in relation to the legs of healthy controls. The non-injured leg was however as strong as that of healthy controls irrespective of treatment. Our study showed that there were, regardless of treatment, knee extension strength deficits in the injured leg compared with the non-injured leg as well as controls. An LSI below 90% was seen in concentric and eccentric knee extension torque for ACLR and in the eccentric contractions for ACLPT, indicating that the injured leg was more than 10% weaker than the contralateral leg, which is suggested as an unsatisfactory outcome of rehabilitation (Augustsson et al., 2004). LSI values for knee extension in the present study were lower than in studies that have investigated knee muscle strength in a shorter perspective, i.e., after 10–15 years after injury and that showed LSI values between 90% and 100% (Ageberg

Knee muscle strength 23 years after ACL injury et al., 2007; Holm et al., 2010; Oiestad et al., 2010a). This suggests that the strength deficit of the injured leg may become more evident with time, which is contradictory to earlier suggestions that the strength difference may be reduced over time (Palmieri-Smith et al., 2008). There may, however, be a bilateral quadriceps weakness following a unilateral ACL injury (Urbach & Awiszus, 2002), which may lead to a misleadingly high LSI, which then reflects a bilateral strength deficit rather than normalization of deficits in the injured leg. This is a major weakness of using the LSI measurement only without comparison to norm values. The participants with ACL injury in our study had nevertheless equal peak torque in their non-injured leg compared with the matched controls, which contradicts a general strength reduction in both legs. For ACLR, the largest impact on strength was seen in knee extension torque, with 13%–14% reduced strength compared with the non-injured leg in concentric as well as eccentric contractions. There were no significant sideto-side differences in knee flexion. This is in line with another study that reports approximately 10% strength deficit for concentric knee extension but not for knee flexion measured 4–7 years after ACL reconstruction (Moisala et al., 2007). In the 10-year follow-up of individuals with ACL reconstruction by Holm et al. (2010), there was a smaller knee strength reduction (LSI > 90%) in both flexors and extensors compared with the noninjured leg. None of these studies used a healthy control group and bilateral knee strength reduction could not be excluded. In the short-term, surgery with patellar graft seems to have a greater negative impact on knee extension strength than surgery with a hamstrings graft (Dauty et al., 2005) and vice versa. The ACLPT had a smaller (7%) side-to-side difference for concentric knee extension peak torque, while in addition also a side-to-side difference for eccentric contractions of 12%, almost twice the size of the concentric side difference. Deficits were also found in eccentric knee flexion with a 10% torque reduction in the injured leg, whereas no side differences were found in the concentric phase. The LSI in concentric contractions is in agreement with the results of another long-term study of conservatively treated ACL injuries (15 years after injury), showing a LSI of 96.5% for concentric knee extension and a LSI of 101.7% for concentric knee flexion (Ageberg et al., 2007). No other long-term studies have investigated eccentric knee peak torque, and thus comparisons to the literature cannot be made. In our study, the eccentric knee extension torque was reduced for ACLPT as for the ACLR, but interestingly, the eccentric knee flexion peak torque was impaired only for the ACLPT. This is possibly explained by discomfort because of increased knee joint laxity and perhaps difficulty to control an increased translation in the knee joint due to the ACL deficit. Anterior tibial translation during isokinetic knee extensions in conservatively treated

ACL-deficient knees has showed 38% higher translation in the ACL-injured knee compared with the non-injured knee during eccentric contractions, while there were no differences in translation for concentric contractions (Kvist et al., 2001). Isokinetic eccentric contraction may be influenced of pain, discomfort, and neurophysiological factors so that the contraction may not be fully activated (Westing et al., 1991), which may lead to a less reliable measurement. Numerous different factors may influence strength during a time period as late as more than two decades after injury, such as development of knee OA, decreased physcial activity level, and knee function (von Porat et al., 2004; Meunier et al., 2007; Mihelic et al., 2011; Streich et al., 2011). Therefore, we investigated correlations between strength and subjective knee function and knee-specific activity level and if strength was affected by the degree of radiographic OA. Indeed, although moderate, a higher peak torque correlated with a higher subjective knee function (Lysholm score), a finding consistent with other research (Tsepis et al., 2004; Moisala et al., 2007). For KOOS, only the subscale “symptoms” was associated with peak torque. Other studies have shown that KOOS-Sport/Rec and KOOS-QoL are the subscales most negatively influenced after an ACL injury (von Porat et al., 2004; Meunier et al., 2007), as was also the case in our data. KOOS-Sport/Rec consists of more physical items and it is therefore surprising that no correlations were seen between strength and KOOS-Sport/ Rec. KOOS was, however, primarily developed to assess effects of OA and in our data the degree of OA does not seem to influence knee muscle strength. Another possible explanation to this result is that with increasing age and time after injury, there might be a change in the character of strength and function relationship, possibly also related to the overall expectation of knee function compared to that during young age and shortly after injury. Only a couple of studies have examined knee strength as long as 10–15 years after injury, but neither have taken OA into consideration (Ageberg et al., 2007; Holm et al., 2010). In general, there is conflicting evidence regarding the relationship between strength and OA (Roos et al., 2011). Oiestad et al. (2010b) also did not find any association between quadriceps weakness shortly after ACL reconstruction and knee OA in the long-term. The relationship between knee muscle strength and OA after ACL injury certainly needs further investigation, especially concerning the long-term perspective. Another aim of the present study was to investigate the H : Q ratio in the long-term perspective, as a low H : Q ratio (i.e., a decreased hamstring strength relative to that of the quadriceps) is suggested as a potential risk factor for lower extremity injuries including ACL injury (Knapik et al., 1991). Moreover, low H : Q ratio is strongly correlated with lower functional ability (measured by Cincinnati rating score) in participants with

7

Tengman et al. ACL injury (Li et al., 1996). Therefore, a high H : Q ratio with reasonable strength of both hamstrings and quadriceps is desirable. The ACL-injured participants in the present study displayed comparable H : Q ratios for the non-injured leg when compared to age- and gendermatched controls and similar H : Q ratios that have been reported for both men (0.62) and women (0.57) at the same angular velocity (90°/s) (Yoon et al., 1991). Our results also show that the injured leg generally has a higher H : Q ratio compared to the non-injured leg, but this is explained by reduced quadriceps strength (knee extension torque) rather than sufficient hamstrings torque. It is therefore debatable how useful the H : Q value is alone when there are deficits in quadriceps strength. This pinpoints the importance of reference values obtained from a non-injured control group. The first goal of rehabilitation should be to normalize the strength relative to the non-injured leg and the second goal should be to increase the H : Q ratio by means of improving the strength of the hamstrings. A limitation is obviously that our study was not designed as a RCT many years ago, which makes direct comparisons between ACL groups not valid. Despite that, the present study offered a unique possibility to examine ACL-injured individuals coming from two separate cohorts with different treatments in two separate hospitals. As of today, there are no RCTs with very long-term follow-up studies of both patients treated with or without surgery that have assessed knee muscle strength. The obvious challenge with long-term follow-up studies is the many potential factors that may influence their strength over the years. The present study obviously reflects ACL reconstructive surgical techniques of the 1980s and early 1990s. Although these techniques have progressed, we have to consider that many individuals who have been treated according to these methods are only in their mid-40s today and are expected to live a considerable number of years with the associated outcomes.

Perspectives This cross-sectional study investigated knee muscle strength, on average, 23 years after unilateral ACL injury. Persons with ACL injury displayed strength deficiencies in the knee muscles of the injured leg, while strength in their non-injured leg was comparable to knee healthy controls of the same age and gender. Peak torques showed a low-to-moderate positive correlation with subjective knee function, while the stage of OA did not influence knee strength. This was true independent of the initial treatment. When comparing the injured and non-injured legs, the ACLR had the largest reduction in concentric and eccentric knee extension peak torque, whereas the ACLPT had the largest reductions in eccentric peak torque in both knee extension and flexion. There is a need for more systematic research evaluating long-term treatment effects on knee muscle strength while also addressing eccentric contraction, something that has not been done prior to this study. In rehabilitation, focus should be on restoring normal knee muscle strength of the injured leg in concentric as well as eccentric strength, while also aiming for good function in the long-term perspective. Key words: Isokinetic, peak torque, long-term consequences, cross-sectional design.

Acknowledgements The authors would like to acknowledge funding support from the Swedish Scientific Research Council Project No. K2008-70X20845-01-3, K2011-69X-21876-01-3, and CIF P2012-0008, Umeå University (Young Researcher Awardee C Häger), Västerbotten County Council, foundation for medical research at Umeå University and Ingabritt & Arne Lundbergs Research Foundation. In addition, we would like to thank Dr Håkan Jonsson, Professor Yelverton Tegner, and PT Lars Lundgren for providing background data; and Martin Persson and Monica Edström for assistance in data collection. Sincere appreciation is also expressed to the participants for their commitment.

References Aagaard P, Simonsen EB, Magnusson SP, Larsson B, Dyhre-Poulsen P. A new concept for isokinetic hamstring: quadriceps muscle strength ratio. Am J Sports Med 1998: 26: 231–237. Ageberg E. Consequences of a ligament injury on neuromuscular function and relevance to rehabilitation – using the anterior cruciate ligament-injured knee as model. J Electromyogr Kinesiol 2002: 12: 205–212. Ageberg E, Pettersson A, Friden T. 15-year follow-up of neuromuscular function in patients with unilateral nonreconstructed anterior cruciate ligament injury initially treated with rehabilitation and activity modification:

8

a longitudinal prospective study. Am J Sports Med 2007: 35: 2109–2117. Ardern CL, Webster KE, Taylor NF, Feller JA. Return to the preinjury level of competitive sport after anterior cruciate ligament reconstruction surgery: two-thirds of patients have not returned by 12 months after surgery. Am J Sports Med 2011: 39: 538–543. Augustsson J, Thomee R, Karlsson J. Ability of a new hop test to determine functional deficits after anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2004: 12: 350–356. Dauty M, Tortellier L, Rochcongar P. Isokinetic and anterior cruciate

ligament reconstruction with hamstrings or patella tendon graft: analysis of literature. Int J Sports Med 2005: 26: 599–606. Delince P, Ghafil D. Anterior cruciate ligament tears: conservative or surgical treatment? A critical review of the literature. Knee Surg Sports Traumatol Arthrosc 2012: 20(1): 48–51. Frobell RB, Roos HP, Roos EM, Roemer FW, Ranstam J, Lohmander LS. Treatment for acute anterior cruciate ligament tear: five year outcome of randomised trial. BMJ 2013: 346: f232. Gao B, Cordova ML, Zheng NN. Three-dimensional joint kinematics of ACL-deficient and ACL-reconstructed

Knee muscle strength 23 years after ACL injury knees during stair ascent and descent. Hum Mov Sci 2012: 31: 222–235. Gerber JP, Marcus RL, Dibble LE, Greis PE, Burks RT, LaStayo PC. Effects of early progressive eccentric exercise on muscle size and function after anterior cruciate ligament reconstruction: a 1-year follow-up study of a randomized clinical trial. Phys Ther 2009: 89: 51–59. Hirokawa S, Solomonow M, Luo Z, Lu Y, D’Ambrosia R. Muscular co-contraction and control of knee stability. J Electromyogr Kinesiol 1991: 1: 199–208. Holm I, Oiestad BE, Risberg MA Aune AK. No difference in knee function or prevalence of osteoarthritis after reconstruction of the anterior cruciate ligament with 4-strand hamstring autograft versus patellar tendon-bone autograft: a randomized study with 10-year follow-up. Am J Sports Med 2010: 38: 448–454. Keays SL, Newcombe PA, Bullock-Saxton JE, Bullock MI, Keays AC. Factors involved in the development of osteoarthritis after anterior cruciate ligament surgery. Am J Sports Med 2010: 38: 455–463. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957: 16: 494–502. Kim S, Lockhart T, Nam CS. Leg strength comparison between younger and middle-age adults. Int J Ind Ergon 2010: 40: 315–320. Knapik JJ, Bauman CL, Jones BH, Harris JM, Vaughan L. Preseason strength and flexibility imbalances associated with athletic injuries in female collegiate athletes. Am J Sports Med 1991: 19: 76–81. Kostogiannis I, Ageberg E, Neuman P, Dahlberg L, Friden T, Roos H. Activity level and subjective knee function 15 years after anterior cruciate ligament injury: a prospective, longitudinal study of nonreconstructed patients. Am J Sports Med 2007: 35: 1135–1143. Kvist J, Karlberg C, Gerdle B, Gillquist J. Anterior tibial translation during different isokinetic quadriceps torque in anterior cruciate ligament deficient and nonimpaired individuals. J Orthop Sports Phys Ther 2001: 31: 4–15. Li RC, Maffulli N, Hsu YC, Chan KM. Isokinetic strength of the quadriceps and hamstrings and functional ability of anterior cruciate deficient knees in recreational athletes. Br J Sports Med 1996: 30: 161–164. Lipscomb AB, Johnston RK, Snyder RB. The technique of cruciate ligament reconstruction. Am J Sports Med 1981: 9: 77–81.

Meunier A, Odensten M, Good L. Long-term results after primary repair or non-surgical treatment of anterior cruciate ligament rupture: a randomized study with a 15-year follow-up. Scand J Med Sci Sports 2007: 17: 230–237. Mihelic R, Jurdana H, Jotanovic Z, Madjarevic T, Tudor A. Long-term results of anterior cruciate ligament reconstruction: a comparison with non-operative treatment with a follow-up of 17–20 years. Int Orthop 2011: 35: 1093–1097. Moisala AS, Jarvela T, Kannus P, Jarvinen M. Muscle strength evaluations after ACL reconstruction. Int J Sports Med 2007: 28: 868–872. Odensten M, Gillquist J. A modified technique for anterior cruciate ligament (ACL) surgery using a new drill guide for isometric positioning of the ACL. Clin Orthop Relat Res 1986: (213): 154–158. Oiestad BE, Holm I, Aune AK, Gunderson R, Myklebust G, Engebretsen L, Fosdahl MA, Risberg MA. Knee function and prevalence of knee osteoarthritis after anterior cruciate ligament reconstruction: a prospective study with 10 to 15 years of follow-up. Am J Sports Med 2010a: 38: 2201–2210. Oiestad BE, Holm I, Gunderson R, Myklebust G, Risberg MA. Quadriceps muscle weakness after anterior cruciate ligament reconstruction: a risk factor for knee osteoarthritis? Arthritis Care Res (Hoboken) 2010b: 62: 1706–1714. Palmieri-Smith RM, Thomas AC, Wojtys EM. Maximizing quadriceps strength after ACL reconstruction. Clin Sports Med 2008: 27: 405–424. vii–ix. Pua YH, Bryant AL, Steele JR, Newton RU, Wrigley TV. Isokinetic dynamometry in anterior cruciate ligament injury and reconstruction. Ann Acad Med Singapore 2008: 37: 330–340. Roos EM, Herzog W, Block JA, Bennell KL. Muscle weakness, afferent sensory dysfunction and exercise in knee osteoarthritis. Nat Rev Rheumatol 2011: 7: 57–63. Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee Injury and Osteoarthritis Outcome Score (KOOS) – development of a self-administered outcome measure. J Orthop Sports Phys Ther 1998: 28: 88–96. Sole G, Hamren J, Milosavljevic S, Nicholson H, Sullivan SJ. Test-retest reliability of isokinetic knee extension and flexion. Arch Phys Med Rehabil 2007: 88: 626–631. Solomonow M, Baratta R, Zhou BH, Shoji H, Bose W, Beck C, D’Ambrosia

R. The synergistic action of the anterior cruciate ligament and thigh muscles in maintaining joint stability. Am J Sports Med 1987: 15: 207–213. Streich NA, Zimmermann D, Bode G, Schmitt H. Reconstructive versus non-reconstructive treatment of anterior cruciate ligament insufficiency. A retrospective matched-pair long-term follow-up. Int Orthop 2011: 35: 607–613. Tegner Y. Strength training in the rehabilitation of cruciate ligament tears. Sports Med 1990: 9: 129–136. Tegner Y, Lysholm J. Rating systems in the evaluation of knee ligament injuries. Clin Orthop Relat Res 1985: 198: 43–49. Tengman E, Brax Olofsson L, Nilsson KG, Tegner Y, Lundgren L, Häger CK. Anterior cruciate ligament injury after more than 20 years. I. Physical activity level and knee function. Scand J Med Sci Sports 2014. doi: 10.1111/sms .12212 Thelen DG, Chumanov ES, Hoerth DM, Best TM, Swanson SC, Li L, Young M, Heiderscheit BC. Hamstring muscle kinematics during treadmill sprinting. Med Sci Sports Exerc 2005: 37: 108–114. Trees AH, Howe TE, Dixon J, White L. Exercise for treating isolated anterior cruciate ligament injuries in adults. Cochrane Database Syst Rev 2005: (4): CD005316. Tsepis E, Vagenas G, Giakas G, Georgoulis A. Hamstring weakness as an indicator of poor knee function in ACL-deficient patients. Knee Surg Sports Traumatol Arthrosc 2004: 12: 22–29. Urbach D, Awiszus F. Impaired ability of voluntary quadriceps activation bilaterally interferes with function testing after knee injuries. A twitch interpolation study. Int J Sports Med 2002: 23: 231–236. von Porat A, Roos EM, Roos H. High prevalence of osteoarthritis 14 years after an anterior cruciate ligament tear in male soccer players: a study of radiographic and patient relevant outcomes. Ann Rheum Dis 2004: 63: 269–273. Westing SH, Cresswell AG, Thorstensson A. Muscle activation during maximal voluntary eccentric and concentric knee extension. Eur J Appl Physiol Occup Physiol 1991: 62: 104–108. Yoon TS, Park DS, Kang SW, Chun SI, Shin JS. Isometric and isokinetic torque curves at the knee joint. Yonsei Med J 1991: 32: 33–43.

9

Anterior cruciate ligament injury after more than 20 years. II. Concentric and eccentric knee muscle strength.

The long-term consequences on knee muscle strength some decades after rupture of the anterior cruciate ligament (ACL) are not established. The aims of...
117KB Sizes 0 Downloads 3 Views