Knee Muscle Strength After Recent Partial Meniscectomy Does Not Relate to 2-year Change in Knee Adduction Moment

Abstract

Background

Knee muscle weakness and a greater external knee adduction moment are suggested risk factors for medial tibiofemoral knee osteoarthritis. Knee muscle weakness and a greater knee adduction moment may be related to each other, are potentially modifiable, and have been observed after arthroscopic partial meniscectomy.

Questions/purposes

The aim of this exploratory study was to determine if knee muscle weakness 3 months after arthroscopic partial meniscectomy (baseline) is associated with an increase in external knee adduction parameters during the subsequent 2 years.

Methods

Eighty-two participants undergoing medial arthroscopic partial meniscectomy were assessed at baseline, and 66 participants who were reassessed 2 years later were included in our study. Isokinetic muscle strength and external adduction moment parameters (peak and impulse) during normal and fast walking were measured at baseline and followup. Multiple linear regression models were used to examine the association between baseline muscle strength and 2-year change in adduction moment parameters. A post hoc power calculation showed we had 80% power to detect a correlation of 0.31 between baseline muscle strength and change in the external knee adduction, with an alpha error of 0.05 and two-sided significance.

Results

Maximal isokinetic muscle strength 3 months after arthroscopic partial meniscectomy was not associated with change in adduction moment parameters (p value range from 0.12 to 0.96).

Conclusions

No evidence was found to suggest that improving maximal knee muscle strength after a recent arthroscopic partial meniscectomy would reduce change in knee adduction moment observed during the subsequent 2 years. As muscle function is modifiable, future investigation of other aspects of muscle function that may relate to change in knee adduction moment is warranted.

Level of Evidence

Level II, prognostic study. See the Instructions for Authors for a complete description of levels of evidence.

Introduction

After meniscectomy, patients are at increased risk of having knee osteoarthritis develop [22], particularly in the medial tibiofemoral compartment [18]. Knee osteoarthritis is, in part, considered a mechanical disease [10], with increased joint loading and knee muscle weakness thought to be important in disease pathogenesis [6]. Knee muscle weakness and increased knee loading, indirectly indicated by the external knee adduction moment during walking, have been observed after arthroscopic partial meniscectomy [15, 33, 35] and may be related to each other in this population group [33]. Determining whether knee muscle weakness after a recent medial arthroscopic partial meniscectomy is associated with increases in the knee adduction moment with time may assist in developing rehabilitation interventions aimed at preventing or slowing the onset of osteoarthritis in this at-risk group.

The knee adduction moment is of particular interest in the pathogenesis of knee osteoarthritis because evidence suggests that the peak knee adduction moment is positively associated with radiographic disease progression [24] and knee pain [2]. The knee adduction moment impulse also has been associated with degenerative morphologic features of the cartilage after arthroscopic partial meniscectomy [9] and with cartilage loss in patients with established osteoarthritis [4]. A greater peak knee adduction moment [15, 33, 35] and knee adduction moment impulse [15] have been found in people after arthroscopic partial meniscectomy as compared with control subjects with intact menisci, and we recently observed that the peak knee adduction moment significantly increased during 2 years in patients after medial arthroscopic partial meniscectomy [15]. Although the clinical importance of the increase in the peak knee adduction moment remains unknown, these findings overall suggest that increased loading is likely to be an important factor to address after arthroscopic partial meniscectomy.

Muscles are considered influential in the pathogenesis of knee osteoarthritis given their ability to control load sharing between the medial and lateral tibiofemoral condyles and provide dynamic stability [6, 19, 23, 30, 38]. Biomechanical investigations have found that the quadriceps and hamstrings contribute to support the knee adduction moment [8, 21, 38]. Within 6 months of arthroscopic partial meniscectomy, patients have been reported to have weak knee musculature [14, 15, 33], which may impair their ability to control the greater knee adduction moment observed during gait after arthroscopic partial meniscectomy [15, 35]. Moreover, cross-sectional evidence suggests that 3 months after arthroscopic partial meniscectomy patients with weak knee extensors have a greater peak knee adduction moment during walking compared with patients whose strength is comparable to that of control subjects with intact menisci [33]. Although the knee adduction moment and knee muscle strength were reported previously in this cohort of patients with arthroscopic partial meniscectomy [15], to our knowledge, no longitudinal study has examined relationships between muscle weakness and knee-loading change in this population. Considering that muscle weakness can be modified with muscle-strengthening exercises, it is important to explore associations between knee muscle strength and dynamic knee load after arthroscopic partial meniscectomy.

Therefore, the aim of this exploratory study was to determine if weaker knee muscle strength at 3 months after arthroscopic partial meniscectomy is associated with an increased knee adduction moment (peak and impulse) during walking throughout the subsequent 2 years.

Patients and Methods

This is a retrospective analysis performed on a previously published prospective longitudinal study [15]. During a 33-month period starting July 2005, we recruited participants between 30 and 50 years old who had an isolated medial arthroscopic partial meniscectomy 3 months previously by one of five orthopaedic surgeons (HM, AS, JK, JF, AT) in Melbourne, Australia. Participants were excluded if they had a lateral meniscal resection; greater than ½ of the medial meniscus resected; greater than two tibiofemoral cartilage lesions observed at arthroscopy; any tibiofemoral cartilage lesion larger than approximately 10 mm in diameter as assessed at arthroscopy; previous knee or lower limb surgery (other than current arthroscopic partial meniscectomy); history of knee pain (other than that leading to arthroscopic partial meniscectomy); postoperative complications; cardiac, circulatory, or neuromuscular conditions; diabetes; stroke; multiple sclerosis; or a contraindication to MRI. Our study was approved by the Human Research Ethics Committee at The University of Melbourne and all participants provided written informed consent.

Of the 149 potentially eligible patients referred by surgeons (between 30 and 50 years old with a recent medial arthroscopic partial meniscectomy), 20 did not fit the study criteria, eight could not be contacted, and 39 were not interested in participating. Thus, 82 participants were enrolled in the 2-year longitudinal study. Sixteen participants (20%) did not return for followup gait assessments for various reasons including relocation and they no longer were interested in participating. One participant had incomplete data for normal pace walking and two different participants had incomplete data for fast pace walking. As such, all 66 participants who returned for re-assessment were included in the analyses, with 65 participants used for normal pace analyses and 64 used for fast pace walking analyses. The minimum followup was 1.5 years (average, 2.2 years; SD, 0.2; range, 1.5–2.4 years). The participants were predominantly male and many were overweight according to the World Health Organization standards (Table 1). During 2005 to 2008, participants underwent baseline strength and gait assessments, returning 2 years later for reassessment. All assessments were performed at the Centre for Health, Exercise and Sports Medicine, The University of Melbourne, Australia.

Table 1.

Participant baseline characteristics

Characteristics	Number = 66
Age (years)	41.3 ± 5.4
Male, number (%)	57 (86)
Height (m)	1.75 ± 0.09
Mass (kg)	83.8 ± 14.4
BMI (kg/m2)	27.3 ± 4.2
Normal-pace walking
Self-selected walking speed (m/second)	1.37 ± 0.15
Peak knee adduction moment (Nm/[BW × HT]%)	2.33 ± 0.89
Knee adduction moment impulse (Nm.second/[BW × HT]%)	0.87 ± 0.33
Fast-pace walking
Self-selected walking speed (m/second)	1.92 ± 0.19
Peak knee adduction moment (Nm/[BW × HT]%)	2.91 ± 1.19
Knee adduction moment impulse (Nm.second/[BW × HT]%)	0.75 ± 0.27
Isokinetic strength (Nm/kg)
Eccentric quadriceps	2.22 ± 0.71
Concentric quadriceps	1.71 ± 0.50
Eccentric hamstrings	1.37 ± 0.39
Concentric hamstrings	0.96 ± 0.26

Values are mean ± SD; BW = body weight; HT = height.

Maximal isokinetic knee extensor and flexor strength (arthroscopic partial meniscectomy leg) defined as the maximal voluntary contraction torque produced on a Kin-Com 125-AP dynamometer (Chattecx, Chattanooga, TN, USA) was assessed. Isokinetic strength at 60° per second was assessed to characterize the low-velocity, high-muscle force region of the muscle force–velocity relationship. As previously described [15], participants performed two tests of five maximal concentric–concentric contractions of knee extensors and flexors at 60° per second through a 5° to 95° range followed by reciprocal eccentric–eccentric contractions of the same knee muscles with 40 seconds separating each set. Participants received strong, standardized verbal encouragement and were instructed to push or pull as hard as possible. Peak torque for each muscle group and condition was recorded with reported values corrected for gravity and normalized to body mass (Nm/kg) [12].

Kinematic data (120 Hz) were collected using an eight-camera, Vicon M2/MX three-dimensional motion analysis system (Vicon, Oxford, UK) in synchrony with ground reaction force data (1080 Hz) recorded using three force plates (Advanced Mechanical Technology, Watertown, MA, USA). After familiarization, participants performed five walking trials along a 10-m walkway at normal and fast walking paces, respectively. Normal pace was described as a “natural and comfortable pace,” whereas fast pace walking was described as a pace that “you would walk in a hurry” [15]. In accordance with Besier et al. [7], 33 reflective markers and four three-marker clusters were placed on anatomic landmarks and body segments. Lower limb joint kinematics and kinetics were estimated using a custom seven-segment lower limb direct kinematics and inverse dynamics model written in Matlab (Mathworks, Natick, MA, USA) and BodyBuilder (Vicon) [7, 9, 15]. Hip centers and knee flexion and extension axes were individually estimated using functional movement tasks, and a foot calibration procedure, where the foot abduction and adduction and rear foot inversion and eversion angles, were measured to establish the foot alignment and coordinate system [7]. The knee adduction moment was calculated for the leg that had the arthroscopic partial meniscectomy and was expressed as external and applied to the shank segment. Similar to previous studies [4, 15], the peak knee adduction moment during the first ½ of stance and positive knee adduction moment impulse were extracted from each of five trials, averaged, and normalized to the result of body weight (N) multiplied by height (m) [29]. Change in peak knee adduction moment, knee adduction moment impulse, and walking speed were determined by subtracting the baseline from the followup scores (ie, a negative score represented a reduction at followup).

Dependent variables included the 2-year change in peak knee adduction moment and knee adduction moment impulse during normal and fast pace walking. Multiple linear regression models were adjusted for longitudinal change in walking speed (known to affect knee adduction moment [26]) and baseline peak knee adduction moment and knee adduction moment impulse (to account for potential for change). All analyses were performed using SPSS (Version 19.0; Chicago, IL, USA) and significance was set at p less than 0.05.

With a minimum of 64 people included in the analyses, post hoc power analysis confirmed that we had 80% power to detect a correlation of 0.31 between baseline muscle strength and change in the external knee adduction, with an alpha error of 0.05 and two-sided significance.

Results

There were no associations between knee muscle strength 3 months after the arthroscopic partial meniscectomy and the change in peak knee adduction moment or knee adduction moment impulse, during normal or fast pace walking in the subsequent 2 years (Table 2) after adjusting for change in walking speed and baseline parameters (p values range from 0.12 to 0.96). Results remained unchanged when knee adduction moments and strength were normalized similarly (ie, both to kg). As reported [15], the peak knee adduction moment increased with time by approximately 9% (mean ± SD peak knee adduction moment, Nm/(BW × HT)% 0.22 ± 0.71 during normal pace gait and 0.25 ± 0.91 during fast pace gait). The knee adduction moment impulse did not change during the 2-year period [15]. There was no change in normal or fast pace walking speed during the 2 years, (mean difference ± SD, 0.01 m/second ± 0.14; p = 0.665 for normal walking pace and 0.01 m/second ± 0.19, p = 0.704 for fast walking pace).

Table 2.

Relationships between baseline isokinetic knee muscle strength and change in knee adduction moment

2-year change in knee adduction moment parameters*	Normal-pace gait		Fast-pace gait
	Regression coefficient (95% CI)	p value	Regression coefficient (95% CI)	p value
Peak knee adduction moment (Nm/[BW × HT]%)
Concentric quadriceps (Nm/kg)	0.16 (−0.19 to 0.50)	0.37	−0.01 (−0.47 to 0.44)	0.96
Eccentric quadriceps (Nm/kg)	−0.04 (−0.28 to 0.20)	0.73	−0.17 (−0.49 to 0.15)	0.30
Concentric hamstrings (Nm/kg)	0.32 (−0.33 to 0.97)	0.33	0.14 (−0.75 to 1.02)	0.76
Eccentric hamstrings (Nm/kg)	−0.17 (−0.60 to 0.27)	0.45	−0.40 (−0.98 to 0.18)	0.18
Knee adduction moment impulse (Nm.second/[BW × HT]%)
Concentric quadriceps (Nm/kg)	0.04 (−0.06 to 0.15)	0.43	−0.03 (−0.12 to 0.06)	0.47
Eccentric quadriceps (Nm/kg)	0.00 (−0.08 to 0.07)	0.93	−0.05 (−0.11 to 0.01)	0.12
Concentric hamstrings (Nm/kg)	0.05 (−0.15 to 0.25)	0.62	−0.01 (−0.19 to 0.17)	0.93
Eccentric hamstrings (Nm/kg)	−0.03 (−0.16 to 0.10)	0.63	−0.08 (−0.19 to 0.03)	0.18

* Adjusting for baseline scores and change in walking speed during 2 years; BW = body weight; HT = height.

Discussion

After arthroscopic partial meniscectomy, patients have risk factors associated with knee osteoarthritis, including a greater knee adduction moment during gait and knee muscle weakness. Cross-sectional evidence obtained after a recent arthroscopic partial meniscectomy suggested that weak knee muscle strength may be related to the knee adduction moment [33]. Therefore, given that exercise can improve knee muscle strength, investigating the longitudinal relationship between these parameters is warranted. The purpose of our study was to determine if weaker knee muscle strength 3 months after arthroscopic partial meniscectomy was related to an increase in the knee adduction moment during gait during the subsequent 2 years. We found no evidence to suggest that weak knee muscle strength at 3 months after arthroscopic partial meniscectomy is associated with subsequent knee adduction moment changes.

Our study has several limitations. First, 20% of our cohort was lost owing to participant attrition by the 2-year followup. Nonetheless, no differences in baseline characteristics, including strength or knee adduction moment, were reported for participants who dropped out compared with participants who remained in the study [15]. Therefore, there is no reason to suspect that participants who dropped out differed in strength variables, and thus it is unlikely to have influenced our findings. A second limitation that may have influenced our ability to detect relationships between baseline knee muscle strength and changes in knee adduction moment was the absence of matched walking speeds between baseline and followup. Although there were no differences in walking speeds between times and walking speed was statistically accounted for, the knee adduction moment change scores may be influenced by walking speed. Nonetheless, to permit generalizable estimates of knee adduction moment measures, participants were permitted to walk at self-selected speeds.

A third limitation is that strength assessment was restricted to the knee extensors and flexors. Hip muscles also are considered to control the knee adduction moment given their prominent role in controlling the center of mass position [5, 25] and mediolateral acceleration [28]. Additionally, biomechanical modeling studies have found that the gastrocnemius contributes to stabilizing the knee adduction moment during gait [31, 38]. Given that quadriceps and hamstring weakness has been observed 3 months after arthroscopic partial meniscectomy [15, 33], it is possible that hip and gastrocnemius muscle weakness was present and may have influenced knee adduction moment changes after the arthroscopic partial meniscectomy [33]. Another limitation is the lack of information regarding physical activity levels and rehabilitation by participants immediately after surgery, because these factors may have affected baseline strength measures. However, baseline strength measures were normally distributed as skewness and kurtosis ranged between −0.093 and −0.288. This may suggest that physical activity and rehabilitation had minimal influence on baseline strength measures. Finally, the lack of information regarding the amount and exact location of the meniscus removed is a limitation because evidence suggests the amount removed alters knee contact force [3] and possibly knee adduction moment measures.

To our knowledge, this is the first study to investigate the longitudinal association between knee muscle strength and change in knee adduction moment parameters. Although biomechanical modeling studies have implicated the role of quadriceps and hamstrings in stabilizing the knee adduction moment [30, 38] during walking we found no evidence to support this. As to our knowledge no other studies have reported the association between strength and change in knee adduction moment, comparison of findings is precluded.

There are several possible explanations why no association between strength and change in knee adduction moment was observed. First, the knee adduction moment can be principally altered by changing the magnitude of the frontal plane ground reaction force vector and/or its lever arm. In turn, these two parameters are largely determined by the position of the knee center and the center of pressure under the foot (ground reaction force origin), body mass, and the body center of mass position and acceleration (vertical and mediolateral). These are potentially controlled by submaximal activations of multiple lower extremity, pelvic, and upper body muscles. Therefore, maximal isokinetic knee muscle strength may not be the most valid or sensitive approach when investigating a relationship between muscle strength and the knee adduction moment during gait. Humans do not exhibit maximal contractions of their knee extensors and flexors muscles during walking and use an unknown individual-specific proportion of their maximal strength (generally within maximum limits). Hunt et al. [16] described how estimating individual specific strength requirements during gait would require assessing muscle activation during stance for each participant and determining muscle strength at that activation level. Despite the possibility of this approach, maximal knee muscle strength was assessed because it is a clinically reliable measure and less burdensome on participants.

Another consideration is that knee muscle strength 3 months after arthroscopic partial meniscectomy was not stable, because we observed that knee muscle strength significantly increased during the 2-year period in this arthroscopic partial meniscectomy cohort [15]. Consequently, the baseline knee muscle strength used to predict change in knee load is not reflective of the whole 2-year study period. Nonetheless, baseline knee muscle strength was examined to predict change in knee load so that rehabilitation programs aiming to prevent or slow the onset of osteoarthritis after arthroscopic partial meniscectomy could be better informed and improved. Although our findings do not support a relationship, quadriceps strengthening may still be important after arthroscopic partial meniscectomy to delay symptomatic osteoarthritis and functional decline. After arthroscopic partial meniscectomy, strong quadriceps strength has been found to be positively correlated with less pain and better physical function and quality of life [11], which generally concurs with findings for patients with established knee osteoarthritis [6]. Therefore, consistent with current meniscal tear rehabilitation recommendations [27] and as previously discussed [15], we consider it clinically important for patients to regain knee muscle strength after arthroscopic partial meniscectomy.

We found that the peak knee adduction moment increased during 2 years in this cohort despite substantial improvement in the knee muscle strength during the same period. This suggests that factors other than maximal knee muscle strength may be more important in determining increases in the knee adduction moment after arthroscopic partial meniscectomy. Other interrelated aspects of muscle function, which may be influenced by pain, may provide insight into how muscles control the knee adduction moment. These aspects include muscle activity patterns and proprioceptive acuity [6, 19, 23, 33, 37, 38]. Future research should explore these aspects of muscle function, because muscle activity patterns during functional tasks [34, 36] and proprioceptive acuity [1] reportedly are altered after arthroscopic partial meniscectomy. Although the mechanism by which these aspects of muscle function would be related to increases in knee adduction moment are undefined, theoretically each could alter the knee adduction moment by affecting the main factors that determine the knee adduction moment.

After meniscus surgery patients are at high risk for knee osteoarthritis [22], which is partly thought to result from compromised ability to absorb and distribute the load across the joint [3, 17]. Our findings provide no evidence to support a relationship between knee muscle strength assessed 3 months after arthroscopic partial meniscectomy and 2-year change in knee adduction moment parameters. Our findings add to intervention studies not finding an effect of muscle strengthening on the knee adduction moment in patients with established osteoarthritis [5, 13, 20, 32]. Moreover, our findings provide an additional rationale to investigate the role of other aspects of muscle function controlling joint loading given that muscle function can be modified through conservative rehabilitation [6]. Factors other than maximal knee muscle strength such as dynamic alignment of the lower extremity and/or upper body control may influence increases in the knee adduction moment.