JCLB-03905; No of Pages 5 Clinical Biomechanics xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis Crystal O. Kean a,b,⁎, Kim L. Bennell b, Tim V. Wrigley b, Rana S. Hinman b a b
School of Medical and Applied Sciences, CQUniversity, Rockhampton, Queensland, Australia Centre for Health, Exercise and Sports Medicine (CHESM), Department of Physiotherapy, The University of Melbourne, Victoria, Australia
a r t i c l e
i n f o
Article history: Received 16 June 2014 Accepted 27 January 2015 Keywords: Osteoarthritis Knee Hip Muscle strength Adduction moment
a b s t r a c t Background: Alterations in hip abductor strength and hip and knee adduction moments during gait are associated with knee osteoarthritis. This study examines the relationship between hip abductor strength and hip and knee adduction moments during gait in individuals with symptomatic medial knee osteoarthritis. Methods: Ninety-nine participants underwent maximal isometric hip abductor strength testing and 3D gait analysis. Pearson correlations examined relationships between non-normalized maximal hip abductor strength (Nm), peak external hip and knee adduction moments (Nm) and knee adduction moment impulse (Nm∙s). Linear regressions examined these relationships while controlling for body size (height and mass) and walking speed. Findings: Positive relationships existed between non-normalized hip abductor strength and hip and knee adduction moments (r = 0.28 and 0.37, respectively, p b 0.01) as well as the knee adduction moment impulse (r = 0.47, p b 0.01). However, after controlling for body size and walking speed, hip abductor strength was not a significant predictor of hip or knee adduction moments (B = − 0.08, and 0.04, respectively, p N 0.05) but was a significant predictor of knee adduction moment impulse (B = 0.05, p = 0.005), explaining 6% of the variance. Interpretation: While a significant relationship between hip abductor strength and knee adduction moment impulse was noted, hip abductor strength only explained a small amount of variance in the impulse. Our findings support previous research of healthy individuals and those with mild to moderate knee osteoarthritis and suggests hip abductor strength has little influence on hip and knee adduction moments during gait. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Impairments in hip abductor strength and joint kinetics exist in people with knee osteoarthritis (OA). Consistent with strength deficits in other lower extremity muscle groups (i.e., quadriceps), cross-sectional data have shown that compared to non-arthritic age-matched individuals, people with knee OA have significantly reduced isometric (~24%) and isokinetic (6-42%) hip abductor strength (Costa et al., 2010; Hinman et al., 2010). During walking, people with knee OA show alterations in parameters of the external knee adduction moment (KAM) compared to people without OA (Baliunas et al., 2002; Mundermann et al., 2004; Landry et al., 2007; Astephen et al., 2008; Mills et al., 2013). The external knee adduction moment, measured via three-dimensional gait analysis, is indicative of the medial-to-lateral joint load distribution and is often used as a surrogate for medial knee compartment loading (Zhao et al., 2007; Meyer et al., 2013). The knee adduction moment is an important ⁎ Corresponding author at: School of Medical and Applied Sciences, CQUniversity, Rockhampton, Queensland, Australia 4702. E-mail address: [email protected] (C.O. Kean).
biomechanical parameter in knee OA, as longitudinal studies show increased risk of medial compartment structural disease progression with increases in both the peak knee adduction moment and adduction moment impulse (Miyazaki et al., 2002; Bennell et al., 2011; Chang et al., 2014; Chehab et al., 2014). People with moderate to severe knee OA also demonstrate a reduced hip external adduction moment compared to healthy, age-matched controls (Mundermann et al., 2005; Astephen et al., 2008), which has also been linked to an increased risk of disease progression (Chang et al., 2005). It has been postulated that hip abductor weakness may explain the reduced peak hip external adduction moment (equivalent to the hip internal abduction moment) observed in knee OA (Chang et al., 2005; Mundermann et al., 2005) and could lead to increased medial knee compartment loading (Chang et al., 2005). Accordingly, there has been recent interest in hip muscle strengthening programs to manage knee OA symptoms and to reduce medial knee joint loads (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011). Although there has been limited research evaluating the relationships between hip abductor muscle strength and frontal plane hip and knee moments in people with knee OA, recent research suggests that hip abductor muscle
http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008 0268-0033/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
2
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
strength has little relationship with these joint moments. One small study of 22 healthy individuals (average age 46 years) showed that hip abductor strength was not related to peak external hip adduction moment normalized to body mass (Rutherford and Hubley-Kozey, 2009). Similarly, a recent study by Lewinson et al (2014) in 14 health adults (average age 25 years) found no relationship between hip abductor strength and external knee adduction moment magnitude. In contrast, a study by Rutherford et al. (2014) in 54 people with mildmoderate knee OA (mean age 58 years, predominantly male) showed that 9% of the variability in peak external knee adduction moment was explained by hip abductor strength, when data were normalized to body mass. The nature of this relationship was positive, indicating that stronger hip abductors were associated with a higher peak knee adduction moment. However, strength did not explain knee adduction moment waveform features derived from principal component analysis. It was also noted that that higher gluteus medius activity was associated with higher mid-stance knee adduction moment magnitudes relative to early and late stance and explained 16% of the variability of this knee adduction moment feature (Rutherford et al., 2014). The weak relationships observed to date may explain why hip muscle strengthening programs have been ineffective at altering parameters of the knee adduction moment in people with knee OA (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011). Although the limited research to date suggests, at best, weak relationships between hip abductor strength and peak external hip and knee adduction moments, further validation of these findings is required in larger samples of older people with symptomatic knee OA before it can be concluded that hip abductor strengthening is not a viable target for reducing medial knee load. In addition, no study has evaluated the relationship of hip abductor strength to the knee adduction moment impulse, which may be a more sensitive parameter of medial knee joint loading (Kean et al., 2012) and has also been linked to disease progression in knee OA (Bennell et al., 2011). The purpose of this study was to examine the relationship between hip abductor strength and the external hip and knee adduction moments during gait in a large sample of individuals with symptomatic medial knee OA. Based on mechanical rationale and previous studies in smaller samples, we hypothesized that hip abductor strength would be explain only a small portion of the variance in hip and knee adduction moment parameters. 2. Methods 2.1. Participants Individuals (n = 99) with medial tibiofemoral knee OA and varus malalignment participating in a clinical trial of neuromuscular training (Bennell et al., 2014) contributed data (measured at baseline from the clinical trial) to this study. Full eligibility criteria have been previously published (Bennell et al., 2014); however, the main inclusion criteria were ≥ 50 years of age, average knee pain of greater than or equal to 25 m on a 100 mm visual analog scale (0 mm = no pain; 100 mm = maximal pain), radiographic knee OA (Kellgren Lawrence grade ≥ 2) predominantly in the medial tibiofemoral compartment and varus malalignment (an anatomical axis of b181° for women and b183°for men (Kraus et al., 2005; Hinman et al., 2006). To meet the criteria of predominantly medial tibiofemoral compartment OA, the participant must have had a medial compartment osteophyte grade greater than or equal to the lateral compartment grade and medial tibiofemoral joint narrowing greater than lateral tibiofemoral joint narrowing. Major exclusion criteria were BMI N 36 kg/m2 (due to difficulties with gait analysis marker placement), history of hip or knee arthroplasty or osteotomy, knee surgery or other nonpharmacologic treatment within previous 6 months, oral corticosteroids within last 4 weeks or other condition affecting lower limb function. For participants with bilaterally eligible knee OA, the most symptomatic knee was evaluated. To be included in the present analysis, participants must have completed a
three-dimensional gait analysis and isometric hip abduction strength test at baseline of the clinical trial. Ethics approval was obtained from the institutional ethics committee and all participants provided written informed consent. 2.2. External hip adduction moment and knee adduction moment Participants underwent three-dimensional gait analysis walking barefoot at their self-selected comfortable walking speed. Thirty-seven passive reflective markers were secured to the participant according to the University of Western Australia model (Besier et al., 2003). Kinematic data were collected at 120Hz using a twelve-camera Vicon motion analysis system (Vicon, Oxford, UK). Kinetic data were collected at 1200Hz using two floor-mounted force plates (Advanced Mechanical Technology Inc., Watertown, MA) in synchrony with the cameras. Two sets of photoelectronic timing gates were positioned 4 m apart to monitor walking speed and ensure that walking trials were within ±5%. Five trials with clean, single force plate strikes were collected from the study limb. Marker trajectory and ground reaction force data were low-pass filtered at 6 Hz with a dual-pass 2nd order Butterworth filter. Joint moments were calculated using inverse dynamics programmed within Vicon Body Builder (Besier et al., 2003). The peak external hip and knee adduction moments (Nm) as well as knee adduction moment impulse (Nm∙s) were determined for each of the five trials and averaged. All moment variables were also expressed normalized to body mass (Nm/kg or Nm∙s/kg) for the purpose of comparison to hip abductor strength normalized in the same way. 2.3. Hip abductor strength Isometric hip abductor strength was tested in supine and the pelvis and contralateral thigh stabilized (Pua et al., 2008). The use of a supine position results in the legs being supported in a gravity-eliminated position and thus removes the need for limb-weight adjustment in the torque calculations. A hand-held dynamometer (Nicholas Manual Muscle Tester, Lafayette Instruments) was applied at the lateral femoral condyle using the hands of the assessor. Participants were instructed to abduct the hip in neutral rotation and given one submaximal practice trial. Three 5-s trials of maximal effort each separated by 30-s rest were then performed to minimize fatigue (Harris et al., 1976; Brown and Weir, 2001). Maximum force output (N) from the 3 trials was converted to torque (Nm) by multiplying maximum force (N) by the resistance lever arm (m). The average of the two maximal trials was recorded. The raw torque units (Nm) were additionally expressed normalized to body mass (Nm/kg). A single male assessor completed all hip abductor strength testing, and the test–retest reliability in our laboratory for this protocol is high with a reported interclass correlation coefficient of 0.84 (Pua et al., 2008). 2.4. Statistical analysis Analyses were performed using SPSS (version 20.0, SPSS, Chicago, IL). Using the non-normalized moments and strength data, Pearson correlation coefficients were calculated to examine the relationship between hip abductor strength and the hip and knee adduction moments. Since body size (height and mass) can influence both the gait joint moment and strength data and walking speed can also have a significant effect on the gait moments, we repeated the analysis and controlled for these variables through a series of hierarchical linear regression models. First, we examined the relationship between non-normalized hip adduction moment (dependent variable) and non-normalized hip abductor strength (independent variable) after controlling for the influence of body size (height and mass, independently) and walking speed. The hierarchical linear regression was then repeated with non-normalized peak knee adduction moment as the dependent variable and again with
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
non-normalized knee adduction moment impulse as the dependent variable. Pearson correlations and linear regressions were then repeated with both the moment and strength data normalized to body mass (Nm/kg or Nm∙s/kg for impulse). Since body mass was already accounted for in the normalization, these hierarchical regressions were conducted controlling for height and walking speed only. 3. Results Descriptive characteristics for participants are provided in Table 1. 3.1. Relationship between hip abductor strength and hip adduction moment A positive relationship was observed between non-normalized maximal isometric hip abductor strength and non-normalized peak hip adduction moment (r = 0.28, p N 0.01) that explained 8% of the variance (r2 = 0.08) in the data. After normalizing for body mass, this bivariate relationship was no longer significant (r = 0.002, p = 0.99). Using the non-normalized moment and strength data, step one of the hierarchical regression revealed that body size (height and mass) and walking speed accounted for 57% of the variance in peak hip adduction moment. Total variance explained by the final model with the addition of hip abductor strength was 58% (F(4, 94) = 32.93, p b 0.001). The change in total variance with the addition of hip abductor strength was not significant (R2 change = 0.02, F change (1, 94) = 3.56, p = 0.06), and hip abductor strength was not a significant predictor of peak hip adduction moment during gait (B = − 0.08 (95%CI: − 0.16, 0.004), p = 0.06). Using the normalized moment and strength data, regression results were unchanged. The addition of hip abductor strength to the model did not significantly change the total variance explained (R2 change = 0.03, F change (1, 95) = 3.819, p = 0.05), and it was not a significant predictor of hip adduction moment (B = − 0.08 (95%CI: −0.15, 0.001), p = 0.05). Table 1 Descriptive characteristics of study participants (n = 99). Presented as mean (SD) unless otherwise stated. Characteristic Age (years) Gender (n)
62.5 (7.3) Female Male
Kellgren and Lawrence grade of OA severity (n) Grade 2 Grade 3 Grade 4 Body mass index (kg/m2) Anatomical axis alignment (°) Knee Injury and Osteoarthritis Outcome Score (KOOS)a Pain Symptoms Activities of daily living Sport and recreation Quality of life Walking speed (m/s) Non-normalized units b Peak hip adduction moment Peak knee adduction moment Knee adduction moment impulse Max hip abductor strength
68.9 (18.1) 41.4 (13.8) 16.1 (6.3) 94.5 (38.2)
51 48
22 42 35 29.6 (4.1) 176.8 (3.5)
52.3 (12.9) 50.3 (19.0) 60.2 (14.1) 24.8 (17.7) 34.5 (16.0) 1.20 (0.20) Normalized units c 0.83 (0.16) 0.50 (0.15) 0.19 (0.06) 1.14 (0.41)
OA = osteoarthritis. a KOOS is scored from 0 to 100 such that 0 indicates worse pain, symptoms, limitations to activities of daily living, sport and recreation and poorer quality of life. b Non-normalized units for peak moments and strength variables are Nm, and for moment impulse is Nm∙s. c Normalized units for peak moments and strength variables are Nm/kg, and for moment impulse is Nm∙s/kg.
3
3.2. Relationship between hip abductor strength and knee adduction moment A positive relationship was observed between non-normalized maximal isometric hip abductor strength and non-normalized peak knee adduction moment (r = 0.37, p N 0.01) with 14% (r2 = 0.14) of the variance in the data explained by this relationship. However, after normalizing for body mass, the bivariate relationship was no longer significant (r = 0.163, p = 0.11). Using the non-normalized moment and strength data, the hierarchical regression revealed that body size (height and mass) and walking speed accounted for 29% of the variance in peak knee adduction moment. The addition of hip abductor strength did not significantly change the total variance explained (R2 change = 0.01, F change (1, 94) = 0.921, p = 0.34), and hip abductor strength was not a significant predictor of peak knee adduction moment during gait (B = 0.04 (95%CI: − 0.04, 0.12), p = 0.34). Using the normalized moment and strength data, findings of the regression analysis remained the same. The addition of hip abductor strength to the model did not significantly change the total variance explained (R2 change = 0.003, F change (1, 95) = 0.31, p = 0.58), and it was not a significant predictor of the knee adduction moment (B = 0.02 (95%CI: − 0.06, 0.10), p = 0.58). A positive relationship was also noted between the non-normalized hip abductor strength and non-normalized knee adduction moment impulse (r = 0.47, p b0.05), with 22% (r2 = 0.22) of variance in data explained. This relationship persisted when data were normalized (r = 0.24, p b 0.05); however, the amount of variance explained was only 6% (r2 = 0.06). Body size and walking speed accounted for 32% of the variance in non-normalized knee adduction moment impulse and the addition of non-normalized hip abductor strength to the model accounted for an additional 6% of explained variance (R2 change = 0.06, F change (1, 94) = 8.38, p = 0.005). Non-normalized hip abductor strength was also a significant predictor of non-normalized knee adduction moment impulse (B = 0.05 (95%CI: 0.02, 0.09), p = 0.005). A similar result was seen with normalized data, such that the addition of normalized hip abductor strength to the model explained an additional 5% of the variance in normalized knee adduction moment impulse (R2 change = 0.05, F change (1, 94) = 0.98, p = 0.03) and was a significant predictor of normalized knee adduction moment impulse (B = 0.04 (95%CI: 0.004, 0.07), p = 0.03). 4. Discussion As correlation coefficients between 0.25 and 0.5 denote a weak relationship (Portney and Watkins, 2009), our bivariate analyses revealed weak positive associations between non-normalized maximal isometric hip abductor strength and the hip and knee adduction moments. However, after adjusting for the influence of body size (height and mass) and walking speed, hip abductor strength did not significantly explain any variability in peak hip or knee adduction moments, nor was it a significant predictor of either the peak hip or knee adduction moments. Although the relationship between hip abductor strength and the knee adduction moment impulse persisted when data were adjusted for body size and walking speed, the relationship was weak and explained only 6% of the variance in the knee adduction moment impulse. Our study is the first study to evaluate the relationship between hip abductor strength and the hip adduction moment in a cohort with symptomatic knee OA. While a weak positive relationship did exist in our bivariate correlational analysis of non-normalized hip abductor strength and hip adduction moment, this relationship did not exist after normalizing the variables by body mass. In addition, our regression analysis revealed hip abductor strength was not a predictor of hip adduction moment after controlling for walking speed and body size thus indicating that a relationship between hip abductor strength and hip adduction moments does not exist independent of these other variables. These findings agree with those of Rutherford and Hubley-Kozey
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
4
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
(2009), who using forward stepwise regressions showed that in a small sample of younger healthy adults (mean age of 46 years), maximal isometric hip abductor strength (using different testing protocol) was not related to the early stance (0–20% gait cycle) peak external hip adduction moment. This result was consistent when non-normalized and normalized strength and moment data were examined. Similar to our findings, they showed that body mass and walking speed explained the majority of variability in the hip adduction moments (Rutherford and Hubley-Kozey, 2009). Only one other study has evaluated the relationship between hip abductor strength and the knee adduction moment in a cohort with symptomatic knee OA. In a cohort of 54 people (mean age of 58 years) with moderate knee OA, Rutherford et al. (2014) used a forward stepwise regression model and found that normalized hip abductor strength (Nm/kg) explained 9% of the variance in the peak knee adduction moment (Nm/kg), such that an increase in abductor strength was positively related to the knee adduction moment. This is in contrast to our results that showed hip abductor strength was not a predictor of the peak knee adduction moment after controlling for walking speed and body size. The differences in study findings may be related to differences in study samples (our participants were older and walked slower) as well as differences in strength testing protocols. However, we did show a weak positive relationship with knee adduction moment impulse, where hip abductor strength explained 6% of the variance, a parameter that Rutherford et al. (2014) did not measure. Taken together, these findings suggest that maximal hip abductor strength does not have a substantial influence on the magnitude features of the external knee adduction moment in people with painful medial knee OA. Although we showed hip abductor strength explains a small portion of the variance (6%) in the knee adduction moment impulse with our normalized data, this is in contrast to the peak knee adduction moment where no relationship was observed with hip abductor strength. The knee adduction moment impulse takes into account both the magnitude and the duration of loading over stance phase and is suggested to be a more sensitive parameter than the peak knee adduction moment (Kean et al., 2012). During gait, the hip abductors of the stance limb help stabilize the pelvis in the frontal plane, particularly during single leg support and ipsilateral hip abductor weakness may cause the pelvis to drop towards the contralateral swing limb. If not compensated for, this may result in a shift of the body's center of mass towards the swing limb and subsequently, increase the frontal plane lever arm of the ground reaction force at the knee and thereby increase the magnitude of the knee adduction moment throughout stance, without substantially influencing the peak. An increased knee adduction moment magnitude throughout stance would likely result in an increased knee adduction moment impulse and thus provide a possible explanation to our contrasting findings. Several intervention studies have examined the effects of hip muscle strengthening on the knee adduction moment in people with symptomatic knee OA (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011). Despite reported increases in hip abductor muscle strength ranging from 13% to 43% with such exercise programs, these studies have failed to demonstrate any significant change in the peak knee adduction moment, which lends further support to our present findings. Henriksen et al. (2009) noted that experimentally reducing gluteus medius (a primary hip abductor) function by painful injection of hypertonic saline resulted in an immediate decrease in the peak external knee adduction moment. The nature of this relationship is also consistent with our findings. Interestingly, the authors noted that timing of peak gluteus medius activity corresponded with the peak hip and knee adduction moments, suggesting that the timing of activity may be more important than maximal muscle strength (Henriksen et al., 2009). From a clinical perspective, it is important to note that significant weakness of the hip abductor muscles exists in people with knee OA (Hinman et al., 2010). Previous research has shown that hip abductor muscle weakness is a significant independent predictor of functional
limitation in people following total knee arthroplasty (Piva et al., 2011; Alnahdi et al., 2012) and that normal gait is sensitive to weakness of hip abductors (van der Krogt et al., 2012). Therefore, strengthening of the hip abductors remains a logical target for rehabilitation in people with knee OA and following surgical intervention for the disease. Strengthening of the hip abductors has been shown to significantly improve both pain and physical function (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011) in people with knee OA. Thus, assessment of hip abductor strength and provision of strengthening exercises for those with muscle weakness is an important component of clinical management of patients with knee OA. However, our data combined with the limited previous research in the area suggests that hip abductor muscle strengthening is not a logical target for reducing magnitude parameters of the external knee adduction moment, a biomechanical marker related to risk of disease progression. It is possible that hip abductor muscle strength may be more strongly related to indices of joint loading other than those measured in this study. Although the peak knee adduction moment is used as a surrogate measure of medial to lateral distribution of tibiofemoral joint loading, the relationship between medial knee contact force and peak knee adduction moment is variable, with reported R2 ranging from 0.28 to 0.87 (Zhao et al., 2007; Kutzner et al., 2013; Meyer et al., 2013), indicating that a portion of the variance in medial contact forces is often unexplained. In order to gain a greater understanding of the relationship between hip abductor strength and medial knee joint loading, future research should consider alternative measures of medial knee joint load, such as medial contact forces that can be estimated through biomechanical modeling, although such methods are still being validated (Kinney et al., 2013). One recent study through a sensitivity analysis of generically modeled total knee joint contact force in five healthy males has shown that simulated weakness of the hip abductors resulted in compensatory increases predicted for the force of some other muscles, which did lead to a predicted increase in total knee joint force (Valente et al., 2013). Whether this simulated result can also occur for medial contact force in OA patients awaits further study. In addition, examination of parameters related to the timing of hip abductor muscle activity, rather than only maximal strength, as measured in our study, may also provide additional information on relationships between muscle function and medial knee joint loading (Meyer et al., 2013; Rutherford et al., 2014). Lastly, it is possible that the use of manual resistance during the hip abductor strength testing may be a limitation to the study due to the need for the examiner to have adequate strength to resist the participant and maintain an isometric contraction. However, our past research has shown this test protocol to be reliable. 5. Conclusions In conclusion, we found weak positive relationships between hip abductor strength and the peak hip and knee adduction moments that disappeared when data were adjusted for the influence of body size and walking speed. Although a relationship persisted for the knee adduction moment impulse, hip abductor strength explained only 6% of the variance in this parameter. Competing interest The author(s) declare that they have no competing interests. Acknowledgments We wish to acknowledge the project personnel (Ben Metcalf, Georgina Morrow and Kelly Ann Bowles) who assisted with recruitment and database management. This study was supported by funding from the National Health and Medical Research Council (NHMRC No. 350297). KB is funded in part by an Australian Research Council Future Fellowship
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
(No. FT0991413). RH is funded in part by Australian Research Council Future Fellowship (No. FT130100175). References Alnahdi, A.H., Zeni, J.A., Snyder-Mackler, L., 2012. Muscle impairments in patients with knee osteoarthritis. Sports Health 4 (4), 284–292. Astephen, J.L., Deluzio, K.J., Caldwell, G.E., Dunbar, M.J., 2008. Biomechanical changes at the hip, knee, and ankle joints during gait are associated with knee osteoarthritis severity. J. Orthop. Res. 26, 332–341. Baliunas, A.J., Hurwitz, D.E., Ryals, A.B., Karrar, A., Case, J.P., Block, J.A., et al., 2002. Increased knee joint loads during walking are present in subjects with knee osteoarthritis. Osteoarthritis Cartilage 10, 573–579. Bennell, K.L., Hunt, M.A., Wrigley, T.V., Hunter, D.J., McManus, F.J., Hodges, P.W., et al., 2010. Hip strengthening reduces symptoms but not knee load in people with medial knee osteoarthritis and varus malalignment: a randomised controlled trial. Osteoarthritis Cartilage 18, 621–628. Bennell, K.L., Bowles, K.A., Wang, Y., Cicuttini, F., Davies-Tuck, M., Hinman, R.S., 2011. Higher dynamic medial knee load predicts greater cartilage loss over 12 months in medial knee osteoarthritis. Ann. Rheum. Dis. 70, 1770–1774. Bennell, K.L., Kyriakides, M., Metcalf, B., Egerton, T., Wrigley, T.V., Hodges, P.W., et al., 2014. Neuromuscular versus quadriceps strengthening exercise in people with medial knee osteoarthritis and varus malalignment: a randomised controlled trial. Arthritis Rheum. 66, 950–959. Besier, T.F., Sturnieks, D.L., Alderson, J.A., Lloyd, D.G., 2003. Repeatability of gait data using a functional hip joint centre and a mean helical knee axis. J. Biomech. 36, 1159–1168. Brown, L.E., Weir, J.P., 2001. ASEP procedures recommendation 1: accurate assessment of musclar strength and power. JEP Online 4 (3), 1–21. Chang, A., Hayes, K., Dunlop, D., Song, J., Hurwitz, D., Cahue, S., et al., 2005. Hip abduction moment and protection against medial tibiofemoral osteoarthritis progression. Arthritis Rheum. 52, 3515–3519. Chang, A.H.M., Chmie, K.C., Chmiel, K.C., Almagor, J.S., Prasad, O., Hayes, P., et al., 2014. The relationship between the external knee adduction moment during gait and progression of knee tissue damage over two years in persons with knee osteoarthritis. Osteoarthritis Cartilage 22, S44–S45. Chehab, E.F., Favre, J., Erhart-Hledik, J.C., Andriacchi, T.P., 2014. Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis. Osteoarthritis Cartilage 22, 1833–1839. Costa, R.A., Oliveira, L.M., Watanabe, S.H., Jones, A., Natour, J., 2010. Isokinetic assessment of the hip muscles in patients with osteoarthritis of the knee. Clinics 65, 1253–1259. Foroughi, N., Smith, R.M., Lange, A.K., Baker, M.K., Fiatarone Singh, M.A., Vanwanseele, B., 2011. Lower limb muscle strengthening does not change frontal plane moments in women with knee osteoarthritis: a randomized controlled trial. Clin. Biomech. 26, 167–174. Harris, R.C., Edwards, R.H., Hultman, E., Nordesjo, L.O., Nylind, B., Sahlin, K., 1976. The time course of phosphorylcreatine resynthesis during recovery of the quadriceps muscle in man. Pflugers Arch. 367, 137–142. Henriksen, M., Aaboe, J., Simonsen, E.B., Alkjaer, T., Bliddal, H., 2009. Experimentally reduced hip abductor function during walking: implications for knee joint loads. J. Biomech. 42, 1236–1240. Hinman, R.S., May, R.L., Crossley, K.M., 2006. Is there an alternative to the full-leg radiograph for determining knee joint alignment in osteoarthritis? Arthritis Rheum. 55, 306–313. Hinman, R.S., Hunt, M.A., Creaby, M.W., Wrigley, T.V., McManus, F.J., Bennell, K.L., 2010. Hip muscle weakness in individuals with medial knee osteoarthritis. Arthritis Care Res. 62, 1190–1193.
5
Kean, C.O., Hinman, R.S., Bowles, K.A., Cicuttini, F., Davies-Tuck, M., Bennell, K.L., 2012. Comparison of peak knee adduction moment and knee adduction moment impulse in distinguishing between severities of knee osteoarthritis. Clin. Biomech. 27, 520–523. Kinney, A.L., Besier, T.F., D'Lima, D.D., Fregly, B.J., 2013. Update on grand challenge competition to predict in vivo knee loads. J. Biomech. Eng. 135 (2), 021012. Kraus, V.B., Vail, T.P., Worrell, T., McDaniel, G., 2005. A comparative assessment of alignment angle of the knee by radiographic and physical examination methods. Arthritis Rheum. 52, 1730–1735. Kutzner, I., Trepczynski, A., Heller, M.O., Bergmann, G., 2013. Knee adduction moment and medial contact force—facts about their correlation during gait. PLoS One 8, e81036. Landry, S.C., McKean, K.A., Hubley-Kozey, C.L., Stanish, W.D., Deluzio, K.J., 2007. Knee biomechanics of moderate OA patients measured during gait at a self-selected and fast walking speed. J. Biomech. 40, 1754–1761. Lewinson, R.T., Worobets, J.T., Stefanyshyn, D.J., 2014. The relationship between maximal hip abductor strength and resultant loading at the knee during walking. Proc. Inst. Mech. Eng. H 228 (12), 1258–1263. Meyer, A.J., D'Lima, D.D., Besier, T.F., Lloyd, D.G., Colwell Jr., C.W., Fregly, B.J., 2013. Are external knee load and EMG measures accurate indicators of internal knee contact forces during gait? J. Orthop. Res. 31, 921–929. Mills, K., Hunt, M.A., Ferber, R., 2013. Biomechanical deviations during level walking associated with knee osteoarthritis: a systematic review and meta-analysis. Arthritis Care Res. 65, 1643–1665. Miyazaki, T., Wada, M., Kawahara, H., Sato, M., Baba, H., Shimada, S., 2002. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann. Rheum. Dis. 61, 617–622. Mundermann, A., Dyrby, C.O., Hurwitz, D.E., Sharma, L., Andriacchi, T.P., 2004. Potential strategies to reduce medial compartment loading in patients with knee osteoarthritis of varying severity: reduced walking speed. Arthritis Rheum. 50, 1172–1178. Mundermann, A., Dyrby, C.O., Andriacchi, T.P., 2005. Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arthritis Rheum. 52, 2835–2844. Piva, S.R., Teixeira, P.E., Almeida, G.J., Gil, A.B., DiGioia 3rd, A.M., Levison, T.J., et al., 2011. Contribution of hip abductor strength to physical function in patients with total knee arthroplasty. Phys. Ther. 91, 225–233. Portney, L.G., Watkins, M.P., 2009. Foundations of clinical research: applications to practice. 3rd ed. Pearson/Prentice Hall, Upper Saddle River, N.J. Pua, Y.-H., Wrigley, T.W., Cowan, S.M., Bennell, K.L., 2008. Intrarater test–retest reliability of hip range of motion and hip muscle strength measurements in persons with hip osteoarthritis. Arch. Phys. Med. Rehabil. 89, 1146. Rutherford, D.J., Hubley-Kozey, C., 2009. Explaining the hip adduction moment variability during gait: implications for hip abductor strengthening. Clin. Biomech. 24, 267–273. Rutherford, D.J., Hubley-Kozey, C., Stanish, W., 2014. Hip abductor function in individuals with medial knee osteoarthritis: implications for medial compartment loading during gait. Clin. Biomech. 29 (5), 545–550. Sled, E.A., Khoja, L., Deluzio, K.J., Olney, S.J., Culham, E.G., 2010. Effect of a home program of hip abductor exercises on knee joint loading, strength, function, and pain in people with knee osteoarthritis: a clinical trial. Phys. Ther. 90, 895–904. Valente, G., Taddei, F., Jonkers, I., 2013. Influence of weak hip abductor muscles on joint contact forces during normal walking: probabilistic modeling analysis. J. Biomech. 46, 2186–2193. van der Krogt, M.M., Delp, S.L., Schwartz, M.H., 2012. How robust is human gait to muscle weakness? Gait Posture 36, 113–119. Zhao, D., Banks, S.A., Mitchell, K.H., D'Lima, D.D., Colwell Jr., C.W., Fregly, B.J., 2007. Correlation between the knee adduction torque and medial contact force for a variety of gait patterns. J. Orthop. Res. 25, 789–797.
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis Crystal O. Kean a,b,⁎, Kim L. Bennell b, Tim V. Wrigley b, Rana S. Hinman b a b
School of Medical and Applied Sciences, CQUniversity, Rockhampton, Queensland, Australia Centre for Health, Exercise and Sports Medicine (CHESM), Department of Physiotherapy, The University of Melbourne, Victoria, Australia
a r t i c l e
i n f o
Article history: Received 16 June 2014 Accepted 27 January 2015 Keywords: Osteoarthritis Knee Hip Muscle strength Adduction moment
a b s t r a c t Background: Alterations in hip abductor strength and hip and knee adduction moments during gait are associated with knee osteoarthritis. This study examines the relationship between hip abductor strength and hip and knee adduction moments during gait in individuals with symptomatic medial knee osteoarthritis. Methods: Ninety-nine participants underwent maximal isometric hip abductor strength testing and 3D gait analysis. Pearson correlations examined relationships between non-normalized maximal hip abductor strength (Nm), peak external hip and knee adduction moments (Nm) and knee adduction moment impulse (Nm∙s). Linear regressions examined these relationships while controlling for body size (height and mass) and walking speed. Findings: Positive relationships existed between non-normalized hip abductor strength and hip and knee adduction moments (r = 0.28 and 0.37, respectively, p b 0.01) as well as the knee adduction moment impulse (r = 0.47, p b 0.01). However, after controlling for body size and walking speed, hip abductor strength was not a significant predictor of hip or knee adduction moments (B = − 0.08, and 0.04, respectively, p N 0.05) but was a significant predictor of knee adduction moment impulse (B = 0.05, p = 0.005), explaining 6% of the variance. Interpretation: While a significant relationship between hip abductor strength and knee adduction moment impulse was noted, hip abductor strength only explained a small amount of variance in the impulse. Our findings support previous research of healthy individuals and those with mild to moderate knee osteoarthritis and suggests hip abductor strength has little influence on hip and knee adduction moments during gait. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Impairments in hip abductor strength and joint kinetics exist in people with knee osteoarthritis (OA). Consistent with strength deficits in other lower extremity muscle groups (i.e., quadriceps), cross-sectional data have shown that compared to non-arthritic age-matched individuals, people with knee OA have significantly reduced isometric (~24%) and isokinetic (6-42%) hip abductor strength (Costa et al., 2010; Hinman et al., 2010). During walking, people with knee OA show alterations in parameters of the external knee adduction moment (KAM) compared to people without OA (Baliunas et al., 2002; Mundermann et al., 2004; Landry et al., 2007; Astephen et al., 2008; Mills et al., 2013). The external knee adduction moment, measured via three-dimensional gait analysis, is indicative of the medial-to-lateral joint load distribution and is often used as a surrogate for medial knee compartment loading (Zhao et al., 2007; Meyer et al., 2013). The knee adduction moment is an important ⁎ Corresponding author at: School of Medical and Applied Sciences, CQUniversity, Rockhampton, Queensland, Australia 4702. E-mail address: [email protected] (C.O. Kean).
biomechanical parameter in knee OA, as longitudinal studies show increased risk of medial compartment structural disease progression with increases in both the peak knee adduction moment and adduction moment impulse (Miyazaki et al., 2002; Bennell et al., 2011; Chang et al., 2014; Chehab et al., 2014). People with moderate to severe knee OA also demonstrate a reduced hip external adduction moment compared to healthy, age-matched controls (Mundermann et al., 2005; Astephen et al., 2008), which has also been linked to an increased risk of disease progression (Chang et al., 2005). It has been postulated that hip abductor weakness may explain the reduced peak hip external adduction moment (equivalent to the hip internal abduction moment) observed in knee OA (Chang et al., 2005; Mundermann et al., 2005) and could lead to increased medial knee compartment loading (Chang et al., 2005). Accordingly, there has been recent interest in hip muscle strengthening programs to manage knee OA symptoms and to reduce medial knee joint loads (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011). Although there has been limited research evaluating the relationships between hip abductor muscle strength and frontal plane hip and knee moments in people with knee OA, recent research suggests that hip abductor muscle
http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008 0268-0033/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
2
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
strength has little relationship with these joint moments. One small study of 22 healthy individuals (average age 46 years) showed that hip abductor strength was not related to peak external hip adduction moment normalized to body mass (Rutherford and Hubley-Kozey, 2009). Similarly, a recent study by Lewinson et al (2014) in 14 health adults (average age 25 years) found no relationship between hip abductor strength and external knee adduction moment magnitude. In contrast, a study by Rutherford et al. (2014) in 54 people with mildmoderate knee OA (mean age 58 years, predominantly male) showed that 9% of the variability in peak external knee adduction moment was explained by hip abductor strength, when data were normalized to body mass. The nature of this relationship was positive, indicating that stronger hip abductors were associated with a higher peak knee adduction moment. However, strength did not explain knee adduction moment waveform features derived from principal component analysis. It was also noted that that higher gluteus medius activity was associated with higher mid-stance knee adduction moment magnitudes relative to early and late stance and explained 16% of the variability of this knee adduction moment feature (Rutherford et al., 2014). The weak relationships observed to date may explain why hip muscle strengthening programs have been ineffective at altering parameters of the knee adduction moment in people with knee OA (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011). Although the limited research to date suggests, at best, weak relationships between hip abductor strength and peak external hip and knee adduction moments, further validation of these findings is required in larger samples of older people with symptomatic knee OA before it can be concluded that hip abductor strengthening is not a viable target for reducing medial knee load. In addition, no study has evaluated the relationship of hip abductor strength to the knee adduction moment impulse, which may be a more sensitive parameter of medial knee joint loading (Kean et al., 2012) and has also been linked to disease progression in knee OA (Bennell et al., 2011). The purpose of this study was to examine the relationship between hip abductor strength and the external hip and knee adduction moments during gait in a large sample of individuals with symptomatic medial knee OA. Based on mechanical rationale and previous studies in smaller samples, we hypothesized that hip abductor strength would be explain only a small portion of the variance in hip and knee adduction moment parameters. 2. Methods 2.1. Participants Individuals (n = 99) with medial tibiofemoral knee OA and varus malalignment participating in a clinical trial of neuromuscular training (Bennell et al., 2014) contributed data (measured at baseline from the clinical trial) to this study. Full eligibility criteria have been previously published (Bennell et al., 2014); however, the main inclusion criteria were ≥ 50 years of age, average knee pain of greater than or equal to 25 m on a 100 mm visual analog scale (0 mm = no pain; 100 mm = maximal pain), radiographic knee OA (Kellgren Lawrence grade ≥ 2) predominantly in the medial tibiofemoral compartment and varus malalignment (an anatomical axis of b181° for women and b183°for men (Kraus et al., 2005; Hinman et al., 2006). To meet the criteria of predominantly medial tibiofemoral compartment OA, the participant must have had a medial compartment osteophyte grade greater than or equal to the lateral compartment grade and medial tibiofemoral joint narrowing greater than lateral tibiofemoral joint narrowing. Major exclusion criteria were BMI N 36 kg/m2 (due to difficulties with gait analysis marker placement), history of hip or knee arthroplasty or osteotomy, knee surgery or other nonpharmacologic treatment within previous 6 months, oral corticosteroids within last 4 weeks or other condition affecting lower limb function. For participants with bilaterally eligible knee OA, the most symptomatic knee was evaluated. To be included in the present analysis, participants must have completed a
three-dimensional gait analysis and isometric hip abduction strength test at baseline of the clinical trial. Ethics approval was obtained from the institutional ethics committee and all participants provided written informed consent. 2.2. External hip adduction moment and knee adduction moment Participants underwent three-dimensional gait analysis walking barefoot at their self-selected comfortable walking speed. Thirty-seven passive reflective markers were secured to the participant according to the University of Western Australia model (Besier et al., 2003). Kinematic data were collected at 120Hz using a twelve-camera Vicon motion analysis system (Vicon, Oxford, UK). Kinetic data were collected at 1200Hz using two floor-mounted force plates (Advanced Mechanical Technology Inc., Watertown, MA) in synchrony with the cameras. Two sets of photoelectronic timing gates were positioned 4 m apart to monitor walking speed and ensure that walking trials were within ±5%. Five trials with clean, single force plate strikes were collected from the study limb. Marker trajectory and ground reaction force data were low-pass filtered at 6 Hz with a dual-pass 2nd order Butterworth filter. Joint moments were calculated using inverse dynamics programmed within Vicon Body Builder (Besier et al., 2003). The peak external hip and knee adduction moments (Nm) as well as knee adduction moment impulse (Nm∙s) were determined for each of the five trials and averaged. All moment variables were also expressed normalized to body mass (Nm/kg or Nm∙s/kg) for the purpose of comparison to hip abductor strength normalized in the same way. 2.3. Hip abductor strength Isometric hip abductor strength was tested in supine and the pelvis and contralateral thigh stabilized (Pua et al., 2008). The use of a supine position results in the legs being supported in a gravity-eliminated position and thus removes the need for limb-weight adjustment in the torque calculations. A hand-held dynamometer (Nicholas Manual Muscle Tester, Lafayette Instruments) was applied at the lateral femoral condyle using the hands of the assessor. Participants were instructed to abduct the hip in neutral rotation and given one submaximal practice trial. Three 5-s trials of maximal effort each separated by 30-s rest were then performed to minimize fatigue (Harris et al., 1976; Brown and Weir, 2001). Maximum force output (N) from the 3 trials was converted to torque (Nm) by multiplying maximum force (N) by the resistance lever arm (m). The average of the two maximal trials was recorded. The raw torque units (Nm) were additionally expressed normalized to body mass (Nm/kg). A single male assessor completed all hip abductor strength testing, and the test–retest reliability in our laboratory for this protocol is high with a reported interclass correlation coefficient of 0.84 (Pua et al., 2008). 2.4. Statistical analysis Analyses were performed using SPSS (version 20.0, SPSS, Chicago, IL). Using the non-normalized moments and strength data, Pearson correlation coefficients were calculated to examine the relationship between hip abductor strength and the hip and knee adduction moments. Since body size (height and mass) can influence both the gait joint moment and strength data and walking speed can also have a significant effect on the gait moments, we repeated the analysis and controlled for these variables through a series of hierarchical linear regression models. First, we examined the relationship between non-normalized hip adduction moment (dependent variable) and non-normalized hip abductor strength (independent variable) after controlling for the influence of body size (height and mass, independently) and walking speed. The hierarchical linear regression was then repeated with non-normalized peak knee adduction moment as the dependent variable and again with
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
non-normalized knee adduction moment impulse as the dependent variable. Pearson correlations and linear regressions were then repeated with both the moment and strength data normalized to body mass (Nm/kg or Nm∙s/kg for impulse). Since body mass was already accounted for in the normalization, these hierarchical regressions were conducted controlling for height and walking speed only. 3. Results Descriptive characteristics for participants are provided in Table 1. 3.1. Relationship between hip abductor strength and hip adduction moment A positive relationship was observed between non-normalized maximal isometric hip abductor strength and non-normalized peak hip adduction moment (r = 0.28, p N 0.01) that explained 8% of the variance (r2 = 0.08) in the data. After normalizing for body mass, this bivariate relationship was no longer significant (r = 0.002, p = 0.99). Using the non-normalized moment and strength data, step one of the hierarchical regression revealed that body size (height and mass) and walking speed accounted for 57% of the variance in peak hip adduction moment. Total variance explained by the final model with the addition of hip abductor strength was 58% (F(4, 94) = 32.93, p b 0.001). The change in total variance with the addition of hip abductor strength was not significant (R2 change = 0.02, F change (1, 94) = 3.56, p = 0.06), and hip abductor strength was not a significant predictor of peak hip adduction moment during gait (B = − 0.08 (95%CI: − 0.16, 0.004), p = 0.06). Using the normalized moment and strength data, regression results were unchanged. The addition of hip abductor strength to the model did not significantly change the total variance explained (R2 change = 0.03, F change (1, 95) = 3.819, p = 0.05), and it was not a significant predictor of hip adduction moment (B = − 0.08 (95%CI: −0.15, 0.001), p = 0.05). Table 1 Descriptive characteristics of study participants (n = 99). Presented as mean (SD) unless otherwise stated. Characteristic Age (years) Gender (n)
62.5 (7.3) Female Male
Kellgren and Lawrence grade of OA severity (n) Grade 2 Grade 3 Grade 4 Body mass index (kg/m2) Anatomical axis alignment (°) Knee Injury and Osteoarthritis Outcome Score (KOOS)a Pain Symptoms Activities of daily living Sport and recreation Quality of life Walking speed (m/s) Non-normalized units b Peak hip adduction moment Peak knee adduction moment Knee adduction moment impulse Max hip abductor strength
68.9 (18.1) 41.4 (13.8) 16.1 (6.3) 94.5 (38.2)
51 48
22 42 35 29.6 (4.1) 176.8 (3.5)
52.3 (12.9) 50.3 (19.0) 60.2 (14.1) 24.8 (17.7) 34.5 (16.0) 1.20 (0.20) Normalized units c 0.83 (0.16) 0.50 (0.15) 0.19 (0.06) 1.14 (0.41)
OA = osteoarthritis. a KOOS is scored from 0 to 100 such that 0 indicates worse pain, symptoms, limitations to activities of daily living, sport and recreation and poorer quality of life. b Non-normalized units for peak moments and strength variables are Nm, and for moment impulse is Nm∙s. c Normalized units for peak moments and strength variables are Nm/kg, and for moment impulse is Nm∙s/kg.
3
3.2. Relationship between hip abductor strength and knee adduction moment A positive relationship was observed between non-normalized maximal isometric hip abductor strength and non-normalized peak knee adduction moment (r = 0.37, p N 0.01) with 14% (r2 = 0.14) of the variance in the data explained by this relationship. However, after normalizing for body mass, the bivariate relationship was no longer significant (r = 0.163, p = 0.11). Using the non-normalized moment and strength data, the hierarchical regression revealed that body size (height and mass) and walking speed accounted for 29% of the variance in peak knee adduction moment. The addition of hip abductor strength did not significantly change the total variance explained (R2 change = 0.01, F change (1, 94) = 0.921, p = 0.34), and hip abductor strength was not a significant predictor of peak knee adduction moment during gait (B = 0.04 (95%CI: − 0.04, 0.12), p = 0.34). Using the normalized moment and strength data, findings of the regression analysis remained the same. The addition of hip abductor strength to the model did not significantly change the total variance explained (R2 change = 0.003, F change (1, 95) = 0.31, p = 0.58), and it was not a significant predictor of the knee adduction moment (B = 0.02 (95%CI: − 0.06, 0.10), p = 0.58). A positive relationship was also noted between the non-normalized hip abductor strength and non-normalized knee adduction moment impulse (r = 0.47, p b0.05), with 22% (r2 = 0.22) of variance in data explained. This relationship persisted when data were normalized (r = 0.24, p b 0.05); however, the amount of variance explained was only 6% (r2 = 0.06). Body size and walking speed accounted for 32% of the variance in non-normalized knee adduction moment impulse and the addition of non-normalized hip abductor strength to the model accounted for an additional 6% of explained variance (R2 change = 0.06, F change (1, 94) = 8.38, p = 0.005). Non-normalized hip abductor strength was also a significant predictor of non-normalized knee adduction moment impulse (B = 0.05 (95%CI: 0.02, 0.09), p = 0.005). A similar result was seen with normalized data, such that the addition of normalized hip abductor strength to the model explained an additional 5% of the variance in normalized knee adduction moment impulse (R2 change = 0.05, F change (1, 94) = 0.98, p = 0.03) and was a significant predictor of normalized knee adduction moment impulse (B = 0.04 (95%CI: 0.004, 0.07), p = 0.03). 4. Discussion As correlation coefficients between 0.25 and 0.5 denote a weak relationship (Portney and Watkins, 2009), our bivariate analyses revealed weak positive associations between non-normalized maximal isometric hip abductor strength and the hip and knee adduction moments. However, after adjusting for the influence of body size (height and mass) and walking speed, hip abductor strength did not significantly explain any variability in peak hip or knee adduction moments, nor was it a significant predictor of either the peak hip or knee adduction moments. Although the relationship between hip abductor strength and the knee adduction moment impulse persisted when data were adjusted for body size and walking speed, the relationship was weak and explained only 6% of the variance in the knee adduction moment impulse. Our study is the first study to evaluate the relationship between hip abductor strength and the hip adduction moment in a cohort with symptomatic knee OA. While a weak positive relationship did exist in our bivariate correlational analysis of non-normalized hip abductor strength and hip adduction moment, this relationship did not exist after normalizing the variables by body mass. In addition, our regression analysis revealed hip abductor strength was not a predictor of hip adduction moment after controlling for walking speed and body size thus indicating that a relationship between hip abductor strength and hip adduction moments does not exist independent of these other variables. These findings agree with those of Rutherford and Hubley-Kozey
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
4
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
(2009), who using forward stepwise regressions showed that in a small sample of younger healthy adults (mean age of 46 years), maximal isometric hip abductor strength (using different testing protocol) was not related to the early stance (0–20% gait cycle) peak external hip adduction moment. This result was consistent when non-normalized and normalized strength and moment data were examined. Similar to our findings, they showed that body mass and walking speed explained the majority of variability in the hip adduction moments (Rutherford and Hubley-Kozey, 2009). Only one other study has evaluated the relationship between hip abductor strength and the knee adduction moment in a cohort with symptomatic knee OA. In a cohort of 54 people (mean age of 58 years) with moderate knee OA, Rutherford et al. (2014) used a forward stepwise regression model and found that normalized hip abductor strength (Nm/kg) explained 9% of the variance in the peak knee adduction moment (Nm/kg), such that an increase in abductor strength was positively related to the knee adduction moment. This is in contrast to our results that showed hip abductor strength was not a predictor of the peak knee adduction moment after controlling for walking speed and body size. The differences in study findings may be related to differences in study samples (our participants were older and walked slower) as well as differences in strength testing protocols. However, we did show a weak positive relationship with knee adduction moment impulse, where hip abductor strength explained 6% of the variance, a parameter that Rutherford et al. (2014) did not measure. Taken together, these findings suggest that maximal hip abductor strength does not have a substantial influence on the magnitude features of the external knee adduction moment in people with painful medial knee OA. Although we showed hip abductor strength explains a small portion of the variance (6%) in the knee adduction moment impulse with our normalized data, this is in contrast to the peak knee adduction moment where no relationship was observed with hip abductor strength. The knee adduction moment impulse takes into account both the magnitude and the duration of loading over stance phase and is suggested to be a more sensitive parameter than the peak knee adduction moment (Kean et al., 2012). During gait, the hip abductors of the stance limb help stabilize the pelvis in the frontal plane, particularly during single leg support and ipsilateral hip abductor weakness may cause the pelvis to drop towards the contralateral swing limb. If not compensated for, this may result in a shift of the body's center of mass towards the swing limb and subsequently, increase the frontal plane lever arm of the ground reaction force at the knee and thereby increase the magnitude of the knee adduction moment throughout stance, without substantially influencing the peak. An increased knee adduction moment magnitude throughout stance would likely result in an increased knee adduction moment impulse and thus provide a possible explanation to our contrasting findings. Several intervention studies have examined the effects of hip muscle strengthening on the knee adduction moment in people with symptomatic knee OA (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011). Despite reported increases in hip abductor muscle strength ranging from 13% to 43% with such exercise programs, these studies have failed to demonstrate any significant change in the peak knee adduction moment, which lends further support to our present findings. Henriksen et al. (2009) noted that experimentally reducing gluteus medius (a primary hip abductor) function by painful injection of hypertonic saline resulted in an immediate decrease in the peak external knee adduction moment. The nature of this relationship is also consistent with our findings. Interestingly, the authors noted that timing of peak gluteus medius activity corresponded with the peak hip and knee adduction moments, suggesting that the timing of activity may be more important than maximal muscle strength (Henriksen et al., 2009). From a clinical perspective, it is important to note that significant weakness of the hip abductor muscles exists in people with knee OA (Hinman et al., 2010). Previous research has shown that hip abductor muscle weakness is a significant independent predictor of functional
limitation in people following total knee arthroplasty (Piva et al., 2011; Alnahdi et al., 2012) and that normal gait is sensitive to weakness of hip abductors (van der Krogt et al., 2012). Therefore, strengthening of the hip abductors remains a logical target for rehabilitation in people with knee OA and following surgical intervention for the disease. Strengthening of the hip abductors has been shown to significantly improve both pain and physical function (Bennell et al., 2010; Sled et al., 2010; Foroughi et al., 2011) in people with knee OA. Thus, assessment of hip abductor strength and provision of strengthening exercises for those with muscle weakness is an important component of clinical management of patients with knee OA. However, our data combined with the limited previous research in the area suggests that hip abductor muscle strengthening is not a logical target for reducing magnitude parameters of the external knee adduction moment, a biomechanical marker related to risk of disease progression. It is possible that hip abductor muscle strength may be more strongly related to indices of joint loading other than those measured in this study. Although the peak knee adduction moment is used as a surrogate measure of medial to lateral distribution of tibiofemoral joint loading, the relationship between medial knee contact force and peak knee adduction moment is variable, with reported R2 ranging from 0.28 to 0.87 (Zhao et al., 2007; Kutzner et al., 2013; Meyer et al., 2013), indicating that a portion of the variance in medial contact forces is often unexplained. In order to gain a greater understanding of the relationship between hip abductor strength and medial knee joint loading, future research should consider alternative measures of medial knee joint load, such as medial contact forces that can be estimated through biomechanical modeling, although such methods are still being validated (Kinney et al., 2013). One recent study through a sensitivity analysis of generically modeled total knee joint contact force in five healthy males has shown that simulated weakness of the hip abductors resulted in compensatory increases predicted for the force of some other muscles, which did lead to a predicted increase in total knee joint force (Valente et al., 2013). Whether this simulated result can also occur for medial contact force in OA patients awaits further study. In addition, examination of parameters related to the timing of hip abductor muscle activity, rather than only maximal strength, as measured in our study, may also provide additional information on relationships between muscle function and medial knee joint loading (Meyer et al., 2013; Rutherford et al., 2014). Lastly, it is possible that the use of manual resistance during the hip abductor strength testing may be a limitation to the study due to the need for the examiner to have adequate strength to resist the participant and maintain an isometric contraction. However, our past research has shown this test protocol to be reliable. 5. Conclusions In conclusion, we found weak positive relationships between hip abductor strength and the peak hip and knee adduction moments that disappeared when data were adjusted for the influence of body size and walking speed. Although a relationship persisted for the knee adduction moment impulse, hip abductor strength explained only 6% of the variance in this parameter. Competing interest The author(s) declare that they have no competing interests. Acknowledgments We wish to acknowledge the project personnel (Ben Metcalf, Georgina Morrow and Kelly Ann Bowles) who assisted with recruitment and database management. This study was supported by funding from the National Health and Medical Research Council (NHMRC No. 350297). KB is funded in part by an Australian Research Council Future Fellowship
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
C.O. Kean et al. / Clinical Biomechanics xxx (2015) xxx–xxx
(No. FT0991413). RH is funded in part by Australian Research Council Future Fellowship (No. FT130100175). References Alnahdi, A.H., Zeni, J.A., Snyder-Mackler, L., 2012. Muscle impairments in patients with knee osteoarthritis. Sports Health 4 (4), 284–292. Astephen, J.L., Deluzio, K.J., Caldwell, G.E., Dunbar, M.J., 2008. Biomechanical changes at the hip, knee, and ankle joints during gait are associated with knee osteoarthritis severity. J. Orthop. Res. 26, 332–341. Baliunas, A.J., Hurwitz, D.E., Ryals, A.B., Karrar, A., Case, J.P., Block, J.A., et al., 2002. Increased knee joint loads during walking are present in subjects with knee osteoarthritis. Osteoarthritis Cartilage 10, 573–579. Bennell, K.L., Hunt, M.A., Wrigley, T.V., Hunter, D.J., McManus, F.J., Hodges, P.W., et al., 2010. Hip strengthening reduces symptoms but not knee load in people with medial knee osteoarthritis and varus malalignment: a randomised controlled trial. Osteoarthritis Cartilage 18, 621–628. Bennell, K.L., Bowles, K.A., Wang, Y., Cicuttini, F., Davies-Tuck, M., Hinman, R.S., 2011. Higher dynamic medial knee load predicts greater cartilage loss over 12 months in medial knee osteoarthritis. Ann. Rheum. Dis. 70, 1770–1774. Bennell, K.L., Kyriakides, M., Metcalf, B., Egerton, T., Wrigley, T.V., Hodges, P.W., et al., 2014. Neuromuscular versus quadriceps strengthening exercise in people with medial knee osteoarthritis and varus malalignment: a randomised controlled trial. Arthritis Rheum. 66, 950–959. Besier, T.F., Sturnieks, D.L., Alderson, J.A., Lloyd, D.G., 2003. Repeatability of gait data using a functional hip joint centre and a mean helical knee axis. J. Biomech. 36, 1159–1168. Brown, L.E., Weir, J.P., 2001. ASEP procedures recommendation 1: accurate assessment of musclar strength and power. JEP Online 4 (3), 1–21. Chang, A., Hayes, K., Dunlop, D., Song, J., Hurwitz, D., Cahue, S., et al., 2005. Hip abduction moment and protection against medial tibiofemoral osteoarthritis progression. Arthritis Rheum. 52, 3515–3519. Chang, A.H.M., Chmie, K.C., Chmiel, K.C., Almagor, J.S., Prasad, O., Hayes, P., et al., 2014. The relationship between the external knee adduction moment during gait and progression of knee tissue damage over two years in persons with knee osteoarthritis. Osteoarthritis Cartilage 22, S44–S45. Chehab, E.F., Favre, J., Erhart-Hledik, J.C., Andriacchi, T.P., 2014. Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis. Osteoarthritis Cartilage 22, 1833–1839. Costa, R.A., Oliveira, L.M., Watanabe, S.H., Jones, A., Natour, J., 2010. Isokinetic assessment of the hip muscles in patients with osteoarthritis of the knee. Clinics 65, 1253–1259. Foroughi, N., Smith, R.M., Lange, A.K., Baker, M.K., Fiatarone Singh, M.A., Vanwanseele, B., 2011. Lower limb muscle strengthening does not change frontal plane moments in women with knee osteoarthritis: a randomized controlled trial. Clin. Biomech. 26, 167–174. Harris, R.C., Edwards, R.H., Hultman, E., Nordesjo, L.O., Nylind, B., Sahlin, K., 1976. The time course of phosphorylcreatine resynthesis during recovery of the quadriceps muscle in man. Pflugers Arch. 367, 137–142. Henriksen, M., Aaboe, J., Simonsen, E.B., Alkjaer, T., Bliddal, H., 2009. Experimentally reduced hip abductor function during walking: implications for knee joint loads. J. Biomech. 42, 1236–1240. Hinman, R.S., May, R.L., Crossley, K.M., 2006. Is there an alternative to the full-leg radiograph for determining knee joint alignment in osteoarthritis? Arthritis Rheum. 55, 306–313. Hinman, R.S., Hunt, M.A., Creaby, M.W., Wrigley, T.V., McManus, F.J., Bennell, K.L., 2010. Hip muscle weakness in individuals with medial knee osteoarthritis. Arthritis Care Res. 62, 1190–1193.
5
Kean, C.O., Hinman, R.S., Bowles, K.A., Cicuttini, F., Davies-Tuck, M., Bennell, K.L., 2012. Comparison of peak knee adduction moment and knee adduction moment impulse in distinguishing between severities of knee osteoarthritis. Clin. Biomech. 27, 520–523. Kinney, A.L., Besier, T.F., D'Lima, D.D., Fregly, B.J., 2013. Update on grand challenge competition to predict in vivo knee loads. J. Biomech. Eng. 135 (2), 021012. Kraus, V.B., Vail, T.P., Worrell, T., McDaniel, G., 2005. A comparative assessment of alignment angle of the knee by radiographic and physical examination methods. Arthritis Rheum. 52, 1730–1735. Kutzner, I., Trepczynski, A., Heller, M.O., Bergmann, G., 2013. Knee adduction moment and medial contact force—facts about their correlation during gait. PLoS One 8, e81036. Landry, S.C., McKean, K.A., Hubley-Kozey, C.L., Stanish, W.D., Deluzio, K.J., 2007. Knee biomechanics of moderate OA patients measured during gait at a self-selected and fast walking speed. J. Biomech. 40, 1754–1761. Lewinson, R.T., Worobets, J.T., Stefanyshyn, D.J., 2014. The relationship between maximal hip abductor strength and resultant loading at the knee during walking. Proc. Inst. Mech. Eng. H 228 (12), 1258–1263. Meyer, A.J., D'Lima, D.D., Besier, T.F., Lloyd, D.G., Colwell Jr., C.W., Fregly, B.J., 2013. Are external knee load and EMG measures accurate indicators of internal knee contact forces during gait? J. Orthop. Res. 31, 921–929. Mills, K., Hunt, M.A., Ferber, R., 2013. Biomechanical deviations during level walking associated with knee osteoarthritis: a systematic review and meta-analysis. Arthritis Care Res. 65, 1643–1665. Miyazaki, T., Wada, M., Kawahara, H., Sato, M., Baba, H., Shimada, S., 2002. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann. Rheum. Dis. 61, 617–622. Mundermann, A., Dyrby, C.O., Hurwitz, D.E., Sharma, L., Andriacchi, T.P., 2004. Potential strategies to reduce medial compartment loading in patients with knee osteoarthritis of varying severity: reduced walking speed. Arthritis Rheum. 50, 1172–1178. Mundermann, A., Dyrby, C.O., Andriacchi, T.P., 2005. Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arthritis Rheum. 52, 2835–2844. Piva, S.R., Teixeira, P.E., Almeida, G.J., Gil, A.B., DiGioia 3rd, A.M., Levison, T.J., et al., 2011. Contribution of hip abductor strength to physical function in patients with total knee arthroplasty. Phys. Ther. 91, 225–233. Portney, L.G., Watkins, M.P., 2009. Foundations of clinical research: applications to practice. 3rd ed. Pearson/Prentice Hall, Upper Saddle River, N.J. Pua, Y.-H., Wrigley, T.W., Cowan, S.M., Bennell, K.L., 2008. Intrarater test–retest reliability of hip range of motion and hip muscle strength measurements in persons with hip osteoarthritis. Arch. Phys. Med. Rehabil. 89, 1146. Rutherford, D.J., Hubley-Kozey, C., 2009. Explaining the hip adduction moment variability during gait: implications for hip abductor strengthening. Clin. Biomech. 24, 267–273. Rutherford, D.J., Hubley-Kozey, C., Stanish, W., 2014. Hip abductor function in individuals with medial knee osteoarthritis: implications for medial compartment loading during gait. Clin. Biomech. 29 (5), 545–550. Sled, E.A., Khoja, L., Deluzio, K.J., Olney, S.J., Culham, E.G., 2010. Effect of a home program of hip abductor exercises on knee joint loading, strength, function, and pain in people with knee osteoarthritis: a clinical trial. Phys. Ther. 90, 895–904. Valente, G., Taddei, F., Jonkers, I., 2013. Influence of weak hip abductor muscles on joint contact forces during normal walking: probabilistic modeling analysis. J. Biomech. 46, 2186–2193. van der Krogt, M.M., Delp, S.L., Schwartz, M.H., 2012. How robust is human gait to muscle weakness? Gait Posture 36, 113–119. Zhao, D., Banks, S.A., Mitchell, K.H., D'Lima, D.D., Colwell Jr., C.W., Fregly, B.J., 2007. Correlation between the knee adduction torque and medial contact force for a variety of gait patterns. J. Orthop. Res. 25, 789–797.
Please cite this article as: Kean, C.O., et al., Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis, Clin. Biomech. (2015), http://dx.doi.org/10.1016/j.clinbiomech.2015.01.008
