Arthritis & Rheumatism (Arthritis Care & Research) Vol. 59, No. 5, May 15, 2008, pp 609 – 614 DOI 10.1002/art.23578 © 2008, American College of Rheumatology ORIGINAL ARTICLE Reducing Joint Loading in Medial Knee Osteoarthritis: Shoes and Canes GEORGINA KEMP, KAY M. CROSSLEY, TIM V. WRIGLEY, BEN R. METCALF, AND RANA S. HINMAN Objective. Increased medial knee loading is associated with a much higher risk of disease progression in knee osteoarthritis (OA). Interventions that can reduce medial knee joint load have the potential to slow disease progression over time. We evaluated the effects of shoes and a cane on knee load in people with knee OA. Methods. Forty people with medial knee OA underwent 3-dimensional gait analysis to measure their peak knee adduction moment, an indicator of medial knee joint load. Results when walking in bare feet were compared with those obtained when walking in their own usual shoes. Twenty participants also underwent testing using a cane, and results were compared with walking unaided. Results. Compared with barefoot, walking in shoes was associated with a signiﬁcant increase in the peak knee adduction moment (mean ⴞ SD N ⴛ m/BW ⴛ H% 3.49 ⴞ 0.84 versus 3.77 ⴞ 0.90; P < 0.001), although there was considerable individual variation. The use of a cane resulted in a 10% decrease in the knee adduction moment (mean ⴞ SD N ⴛ m/BW ⴛ H% 3.76 ⴞ 0.95 versus 3.38 ⴞ 0.68; P ⴝ 0.001). Conclusion. Wearing shoes increases medial knee joint load compared with walking barefoot. Given the variable response to shoes observed, further research is required to ascertain which shoe types might be optimal for those with knee OA. The use of a cane signiﬁcantly reduces medial knee loading and has the potential to reduce the risk of disease progression in knee OA. INTRODUCTION Osteoarthritis (OA) is a leading cause of pain and disability in elderly people (1,2). The knee, particularly the medial tibiofemoral compartment, is commonly affected (3). Conservative management of knee OA has historically focused on relief of symptoms such as pain and impaired physical function (4). However, treatment should also aim to reduce the risk of disease progression, given that a signiﬁcant proportion of patients demonstrate worsening of disease over time and that knee OA is incurable. A biomechanical marker of knee OA progression is the knee adduction moment. The knee adduction moment, as measured by 3-dimensional (3-D) gait analysis, is an indirect measure of dynamic loading Dr. Hinman’s work was supported by an Early Career Researcher Grant provided by the University of Melbourne. Georgina Kemp, BAppSc(Physio), Kay M. Crossley, PhD, PostGradDip(Research), BAppSc(Physio), Tim V. Wrigley, BSc(Hons), MSc, Ben R. Metcalf, BSc(Hons), Rana S. Hinman, BPhysio(Hons), PhD: The University of Melbourne, Victoria, Australia. Address correspondence to Rana S. Hinman, BPhysio (Hons), PhD, Centre for Health, Exercise and Sports Medicine, School of Physiotherapy, The University of Melbourne, Victoria, Australia, 3010. E-mail: firstname.lastname@example.org. Submitted for publication April 1, 2007; accepted in revised form November 18, 2007. on the medial tibiofemoral compartment (5–7). A 20% increase in the peak knee adduction moment is associated with a ⬎6-fold increase in the risk of progression of medial knee OA over 6 years (8). A higher knee adduction moment has also been implicated in the development of chronic knee pain (9) and poorer outcome after high tibial osteotomy (10,11). Treatment strategies that reduce the knee adduction moment have the potential to slow progression of medial knee OA over time. Health practitioners frequently advise patients with knee OA regarding footwear choices and cane use to unload the affected knee joint. While these simple, inexpensive interventions have the potential to alter the peak knee adduction moment, there is little research attesting to their effects on knee load in patients with OA. To our knowledge, the one study that has evaluated the effect of a cane in people with knee OA found only a trend toward decreased knee load when using the gait aid in the contralateral hand (12). There is evidence that usual footwear can increase knee loading in healthy individuals (13–16), possibly related to the height of the shoe heel, but there is little information regarding footwear effects in patients with knee OA; to our knowledge, only 1 study has been conducted in knee OA. Its results showed that, compared with barefoot walking, wearing shoes signiﬁcantly increased knee loading (17). It is thus presently difﬁcult for health professionals to give appropriate, evidence609 610 based recommendations to patients regarding these management strategies. The aim of this study was to investigate the immediate effect of footwear and a walking cane on the peak knee adduction moment in people with knee OA. We hypothesized that the knee adduction moment would be higher when people walked in their own footwear compared with barefoot and lower when walking with a cane compared with unaided gait. PARTICIPANTS AND METHODS Participants. Forty community-dwelling volunteers were recruited, each fulﬁlling clinical and radiographic criteria for knee OA (age ⬎50 years, osteophytes, and knee pain) (18). All demonstrated medial tibiofemoral osteophytes (although concomitant lateral tibiofemoral or patellofemoral OA cannot be excluded) and had experienced knee pain averaging more than 3 points out of 10 on a numerical pain rating scale on most days of the previous month. Exclusion criteria included a history of hip or knee joint replacement, knee surgery or injection in the previous 6 months, current use of a gait aid, and any condition affecting gait or the ability to complete testing. The study was approved by the University of Melbourne Human Research Ethics Committee, and all participants provided written informed consent. Knee OA symptoms were evaluated using the Western Ontario and McMaster Universities OA Index (19) with regard to pain (scores range 0 –20, with higher scores indicating worse pain) and physical function (scores range 0 – 68, with higher scores indicating worse function). Radiographic severity of tibiofemoral OA was assessed with the Kellgren/Lawrence scale (20), where 0 ⫽ normal, 1 ⫽ possible osteophytes, 2 ⫽ minimal osteophytes and possible joint space narrowing, 3 ⫽ moderate osteophytes, some narrowing, and possible sclerosis, and 4 ⫽ large osteophytes, deﬁnite narrowing, and severe sclerosis. Participant characteristics are presented in Table 1. Gait analysis. Participants underwent 3-D gait analysis using a 6-camera VICON 612 motion analysis system (Vicon, Oxford, UK). Two force plates (Advanced Mechanical Technology, Inc., Watertown, MA) embedded in the walkway captured ground reaction force data. Reﬂective markers placed on the anterior superior iliac spine, posterior superior iliac spine, midlateral thigh, lateral knee joint, lateral shank, lateral malleolus, on the shoe over the second metatarsal head, and over the posterior calcaneus were used to capture limb movement. The Vicon Plug-inGait (V2) model (Vicon, Oxford, UK) used inverse dynamics to calculate external joint moments about an orthogonal axis system located in the distal segment of the joint (21). The model determines the hip joint center from the regression equations in Davis et al (21). It places the knee joint center at half the intercondylar width from the lateral femoral condyle marker, medially in a direction perpendicular to that from hip center to knee center, and in the plane of these joint centers and the lateral thigh marker. The ankle joint center is placed analogously, medially in a Kemp et al Table 1. Participant characteristics* Characteristic Age, years Height, meters Mass, kg Body mass index, kg/m2 Symptom duration, years Sex, no. (%) Male Female Symptoms, no. (%) Unilateral Bilateral Symptom severity† Pain Physical function Disease severity, no. (%)‡ Grade 1 Grade 2 Grade 3 Grade 4 Footwear (n ⴝ 40) Cane (n ⴝ 20) 64.7 ⫾ 9.4 1.64 ⫾ 0.08 79.1 ⫾ 12.0 29.6 ⫾ 4.2 8.9 ⫾ 7.9 65.0 ⫾ 10.2 1.60 ⫾ 0.07 75.3 ⫾ 11.2 29.6 ⫾ 4.7 7.5 ⫾ 7.7 16 (40) 24 (60) 2 (10) 18 (90) 14 (35) 26 (65) 7 (35) 13 (65) 9⫾3 29 ⫾ 11 9⫾4 28 ⫾ 13 3 (8) 10 (25) 11 (28) 16 (40) 2 (10) 6 (30) 5 (25) 7 (35) * Values are the mean ⫾ SD unless indicated otherwise. † As measured by the Western Ontario and McMaster Universities Osteoarthritis Index. Higher scores indicate worse symptoms (pain scored 0 –20 and function scored 0 – 68). ‡ Assessed via Kellgren/Lawrence disease severity scale. Higher scores indicate more severe radiographic change. direction from the lateral malleolus and perpendicular to a line from knee center to ankle center, in the plane including the lateral shank marker. The lateral thigh and shank markers were adjusted to align the abovementioned planes to include the intercondylar axis of the knee and intermalleolar axis of the ankle, respectively. The overall peak external knee adduction moment (N ⫻ m) was determined for the stance phase of the gait cycle and normalized to body weight (BW) multiplied by height (H) (22). Only the most symptomatic knee of each participant was analyzed. Test–retest reliability in our laboratory was excellent (intraclass correlation coefﬁcients 0.92– 0.97 in 11 elderly patients with knee pain tested 1 week apart). Two photoelectric beams monitored walking speed and participants were given feedback to ensure that their walking speed for each trial varied ⱕ10% from the required speed of 1 meter/second. Control of walking speed was necessary to negate a potential inﬂuence of speed on the magnitude of the peak knee adduction moment (23). A speed of 1 meter/second was selected to facilitate comparison of data across the literature (24,25). Participants were not informed about the embedded force plates in order to prevent them from altering their gait in an attempt to target the plates. Data from 5 successful trials were collected for each test condition and a mean score was used in the analyses. All participants were tested ﬁrst in bare feet, followed immediately by testing in their own shoes. Footwear was not standardized; participants were instructed to bring a pair of comfortable shoes that they would typically use for walking. Twenty consecutive participants from the cohort underwent further testing wearing their shoes and using a cane in the contralateral hand to the study knee. Effect of Shoes and Canes on Medial Knee OA 611 Table 2. Change in gait parameters by intervention* Footwear (n ⴝ 40) Walking speed, meters/second Cadence, steps/minute Stride length, meters Vertical ground reaction force, N/kg Peak adduction moment, N ⫻ m/BW ⫻ H% Cane (n ⴝ 20) Barefoot Shoes P Unaided Aided P 1.00 ⫾ 0.04 53 ⫾ 4 1.14 ⫾ 0.09 9.9 ⫾ 0.4 1.01 ⫾ 0.04 51 ⫾ 4 1.20 ⫾ 1.00 10.1 ⫾ 0.5 0.060 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 1.01 ⫾ 0.04 52 ⫾ 4 1.17 ⫾ 0.09 10.2 ⫾ 0.4 0.97 ⫾ 0.03 48 ⫾ 4 1.22 ⫾ 0.07 9.6 ⫾ 0.5 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 3.49 ⫾ 0.84 3.77 ⫾ 0.9 ⬍ 0.001 3.76 ⫾ 0.95 3.38 ⫾ 0.68 0.001 * Values are the mean ⫾ SD. The cane was adjusted so that its height corresponded to the distance from the proximal wrist crease to the ground when the participant stood erect with arms by the sides. Each participant was brieﬂy trained to walk by a physiotherapist so that the cane was on the ground during the stance phase of the study knee. Statistical analyses. Analyses were performed using the Statistical Package for the Social Sciences, version 13 (SPSS, Chicago, IL). Descriptive information was examined via means, SDs, and frequencies where appropriate. Two-tailed paired t-tests were used to compare variables between conditions, with an alpha level set at 0.05. RESULTS Changes in gait parameters across testing conditions are presented in Table 2. The peak knee adduction moment when walking with shoes was signiﬁcantly higher than walking barefoot (mean difference 0.28; 95% conﬁdence interval [95% CI] 0.18 – 0.37). Walking in shoes resulted in a 7.4% increase in the peak knee adduction moment (P ⬍ 0.001). However, the effect of footwear was not systematic. While most participants demonstrated an increased knee adduction moment in shoes, considerable individual variation was observed, with 6 participants actually demonstrating a decrease while wearing shoes (Figure 1). Changes in the peak knee adduction moment when wearing shoes ranged from a 10.8% decrease to a 30.8% increase. A mean increase in the vertical ground reaction force by 2% was observed with shoes (P ⬍ 0.001). Temporal parameters changed when participants went from walking barefoot to wearing shoes. When wearing shoes, there was a small but signiﬁcant increase in stride length (5% increase; P ⬍ 0.001) and a reduction in cadence (4% decrease; P ⬍ 0.001) compared with walking barefoot. The small increase in walking speed with shoes (1%) was not statistically signiﬁcant. The peak knee adduction moment was signiﬁcantly higher when walking unaided compared with using the cane (mean difference 0.38; 95% CI 0.13– 0.63). Using a cane reduced the peak knee adduction moment by 10.1% (P ⫽ 0.001). While the majority (75%) of participants demonstrated a decrease in the peak knee adduction moment when using a cane, 5 participants actually demonstrated an increase (Figure 2). Changes in the peak knee adduction moment when using a cane ranged from a 43.9% increase to a 34.6% decrease. A mean decrease by 5.9% in the vertical ground reaction force was observed with the cane (P ⬍ 0.001). While participants walked more slowly when using the cane (P ⬍ 0.001), they used a greater stride length (8% increase; P ⬍ 0.001) and reduced cadence (4% decrease; P ⬍ 0.001). DISCUSSION An increased peak knee adduction moment is associated with poorer clinical outcomes in knee OA, including a much higher risk of disease progression over time (8). Accordingly, management of knee OA should aim not only to reduce symptoms of the disease, but also to decrease load across the joint in order to minimize the risk of disease progression. However, there is very little research evaluating the efﬁcacy of conservative interventions in reducing knee joint load. We found that people with knee OA increased their peak knee adduction moment by 7.4% while wearing their own shoes as compared with walking barefoot. Our ﬁndings concur with the only other study we know of that has evaluated the effects of footwear on the knee adduction moment in knee OA (17). In that study, 75 participants with moderate medial knee OA were tested, and the magnitude of their peak knee adduction moment during barefoot walking was compared with that demon- Figure 1. Percentage change in peak knee adduction moment among individual participants when walking in shoes as compared with walking in bare feet (n ⫽ 40). 612 Figure 2. Percentage change in peak knee adduction moment among individual participants when using a cane as compared with walking unaided (n ⫽ 20). strated while walking in their own comfortable shoes. Findings indicated an 11.9% increase in the knee adduction moment with shoes. It is unclear why wearing shoes increases the knee adduction moment. Because the effect of shoes on the adduction moment is subject to considerable individual variation, it seems likely that individual shoe or foot characteristics (e.g., foot and ankle joint stiffness or compliance, ﬂattening of the medial longitudinal arch, rearfoot supination) may mediate the inﬂuence of footwear on the knee adduction moment. Although we did not control the type of shoe worn by our participants, most shoes demonstrated a slight heel raise. It has been shown in healthy young adults that a high or moderate heel height results in an increased knee adduction moment; thus, heel height may be an important factor that explains our results (13,15,16). Both a thicker lateral shoe sole and the insertion of a laterally-wedged orthotic into the shoe have been shown separately to decrease the knee adduction moment (26,27), probably due to an increase in the valgus moment arm of the subtalar joint, creating a lateral shift in the location of the center of pressure (27). It is believed that the lateral shift in center of pressure reduces the knee joint moment arm, thereby causing a reduction in the adduction moment magnitude. Thus, it is feasible that wearing down the lateral shoe sole, as a result of foot and leg posture or walking mechanics, may have the opposite effect. Preferential wearing down of the lateral sole of the shoe, or inbuilt medial arch supports, could move the center of pressure medially, which can increase the knee adduction moment. It is also possible that such shoe features could result in sufﬁcient subtalar supination to lead to a more varus knee alignment, which could also explain the increase we observed in the knee adduction moment. Another aspect of shoe design that may alter knee load is the stiffness of the sole. Stiffer shoe soles have been shown to increase hip joint loading and have the potential to heighten knee loads as well (26,28). Further research into the effect of different types of shoes on loading in knee OA is warranted so that appropriate clinical recommendations regarding footwear can be made. Our ﬁndings show that using a cane in the contralateral Kemp et al hand to the symptomatic knee can reduce the knee adduction moment by an average of ⬃10%. However, much greater reductions are possible, as analysis of individual results revealed that a quarter of participants demonstrated a reduction of more than 20%. Although canes are widely recommended clinically to reduce knee load for patients with knee OA, only 1 other study has investigated whether using a cane actually reduces knee loading in knee OA (12). Although the authors found that using a cane in the contralateral hand to the affected knee reduced the mean peak knee adduction moment compared with unaided gait (0.55 versus 0.51 Nm/kg), in contrast to our ﬁndings the change was not statistically signiﬁcant. The conﬂicting ﬁndings may be due to differences in the cohorts, particularly because Chan and colleagues (12) did not speciﬁcally utilize participants with medial knee OA and their sample size was only 14, which may have lead to insufﬁcient power to detect signiﬁcant differences. We did not evaluate all possible mechanisms explaining how the cane reduced the knee adduction moment. However, we did demonstrate a 5.9% reduction in the ground reaction force when using a cane compared with unaided walking. This reduction is probably due to load relief provided by the upper limb via the cane. It is also possible that improved proximal stability may explain our ﬁndings. In hip OA, contralateral cane use acts to augment the hip abduction moment on the affected limb (29). Hip abductor weakness is believed to contribute to an elevated ipsilateral knee adduction moment (30) by causing excessive pelvic drop on the contralateral swing limb during walking, resulting in a shift in the body’s center of mass toward the swing limb and thus increasing the knee joint moment arm (31). It is therefore possible that use of the cane improves pelvic control and moves the center of mass closer to the knee joint center. Alterations in gait patterns between conditions may also partially explain the observed effects of shoes and the cane on the knee adduction moment. For example, previous studies have shown that factors such as walking speed and toe-out angle may inﬂuence the magnitude of the adduction moment (23,24). As our study was not designed to evaluate mechanisms of change observed with shoes or the cane, we did not evaluate the many kinematic or kinetic gait adaptations that an individual might employ when moving between walking barefoot and in shoes, or when using a cane. However, in our participants, cane use resulted in slower walking speeds. While it is possible that such speed changes may have inﬂuenced our observed effects, Mundermann et al (23) noted that walking speed only accounted for ⬃9% of the variation in peak knee adduction moment. Considering that our differences in speed were very small (4%), it is unlikely that such temporal changes inﬂuenced our results. In the only other study that has evaluated the effects of shoes on the adduction moment in knee OA (17), the authors measured a number of gait parameters (speed, stride, cadence, toe-out angle, and ankle, hip, and knee ranges of motion) in order to evaluate their contribution to changes in the moment with shoes. Despite signiﬁcant differences in all parameters except speed between walking barefoot and in shoes, regression analyses revealed that these gait alterations did Effect of Shoes and Canes on Medial Knee OA not explain the reduction in loading observed with barefoot walking. Findings of the current study have important clinical implications. In a study of knee OA progression, Miyazaki et al (8) determined that an increase of 1 unit (N ⫻ m/BW ⫻ H%), or 20.4%, in the peak knee adduction moment increased the risk of progression of knee OA 6.5 times. Although participants in that study had a higher baseline adduction moment (mean ⫾ SD 4.9 ⫾ 1.6 BW ⫻ H%) compared with ours and a different system for data collection was used, the ﬁndings highlight the clinical relevance of our own. We showed that wearing shoes increased the peak knee adduction moment on average from 3.49 to 3.77 N ⫻ m/BW ⫻ H%, an increase of 7.4%. Therefore, based on the analysis of Miyazaki et al (8), it is possible that wearing shoes may increase the risk of knee OA progression by a factor of 2.8 on average. Furthermore, several individuals demonstrated increases of more than 20%, which may increase their risk of disease progression ⬎6-fold. Because it is potentially dangerous as well as impractical to advise patients with knee OA to walk about in bare feet, further research is needed to determine which types of shoes least increase the knee adduction moment (or, ideally, reduce it), and to evaluate the effect of wearing shoes on long-term disease progression. Using a cane in the opposite hand was associated with a mean decrease in the knee adduction moment by more than 10%. However, longitudinal, randomized controlled trials are required to establish whether using a cane leads to important clinical outcomes, such as a reduced risk of disease progression or reduction in knee symptoms in the long term. In our cohort evaluating the effect of the cane, 90% of participants were women. This under-representation of men has implications for the generalizability of results. Our ﬁnding that use of a cane reduces the adduction moment is thus only valid in women with knee OA, and future research should determine whether similar ﬁndings occur in men. Although sex differences in osteoarthritic gait have been shown by other studies (32,33), it is unclear whether differences in the response to gait interventions exist across the sexes. A number of limitations exist in this study. We evaluated only immediate changes in the knee adduction moment with shoes and the cane. It is unclear whether the immediate changes observed persist in the short or long term. Further studies are required to establish whether knee loading remains lower with ongoing use of a cane, and whether the reductions in loading translate to a reduced risk of disease progression. Also, we did not measure changes in knee pain, and thus it is unknown whether reductions in the adduction moment observed with barefoot walking and use of the cane are associated with concomitant alterations in pain. In summary, we found that wearing shoes and using a cane each affect the magnitude of the peak knee adduction moment during walking in people with knee OA. As shoes were observed to increase the adduction moment and it is impractical to recommend that patients with knee OA walk barefoot, future research should evaluate which aspects of shoe design contribute to this increase in knee load. The shoe type optimal for knee OA with regard to its 613 effects on symptoms and disease progression must be determined. Because use of a cane resulted in signiﬁcant decreases in the adduction moment, patients with knee OA should be encouraged to consider using a cane on a regular basis. AUTHOR CONTRIBUTIONS Dr. Hinman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Kemp, Crossley, Hinman. Acquisition of data. Kemp, Wrigley, Metcalf. Analysis and interpretation of data. Kemp, Crossley, Wrigley, Metcalf, Hinman. Manuscript preparation. Kemp, Crossley, Wrigley, Hinman. Statistical analysis. Kemp, Crossley, Hinman. REFERENCES 1. Dillon CF, Rasch EK, Gu Q, Hirsch R. Prevalence of knee osteoarthritis in the US: arthritis data from the Third National Health and Nutrition Examination Survey 1991–94. J Rheumatol 2006;33:2271–9. 2. Jordan JM, Helmick CG, Renner JB, Luta G, Dragomir AD, Woodard J, et al. Prevalence of knee symptoms and radiographic and symptomatic knee osteoarthritis in African Americans and Caucasians: the Johnston County Osteoarthritis Project. J Rheumatol 2007;34:172– 80. 3. Duncan RC, Hay EM, Saklatvala J, Croft PR. Prevalence of radiographic osteoarthritis: it all depends on your point of view. Rheumatology (Oxford) 2006;45:757– 60. 4. Felson DT, Lawrence RC, Hochberg MC, McAlindon T, Dieppe PA, Minor MA, et al. Osteoarthritis: new insights. II. Treatment approaches. Ann Intern Med 2000;133:726 –37. 5. Schipplein OD, Andriacchi TP. Interaction between active and passive knee stabilizers during level walking. J Orthop Res 1991;9:113–9. 6. Andriacchi TP. Dynamics of knee malalignment. Orthop Clin North Am 1994;25:395– 403. 7. Hurwitz DE, Sumner DR, Andriacchi TP, Sugar DA. Dynamic knee loads during gait predict proximal tibial bone distribution. J Biomech 1998;31:423–30. 8. Miyazaki T, Wada M, Kawahara H, Sato M, Baba H, Shimada S. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann Rheum Dis 2002;61:617–22. 9. Amin S, Luepongsak N, McGibbon CA, LaValley MP, Krebs DE, Felson DT. Knee adduction moment and development of chronic knee pain in elders. Arthritis Rheum 2004;51:371– 6. 10. Prodromos CC, Andriacchi TP, Galante JO. A relationship between gait and clinical changes following high tibial osteotomy. J Bone Joint Surg Am 1985;67:1188 –94. 11. Wang JW, Kuo KN, Andriacchi TP, Galante JO. The inﬂuence of walking mechanics and time on the results of proximal tibial osteotomy. J Bone Joint Surg Am 1990;72:905–9. 12. Chan GN, Smith AW, Kirtley C, Tsang WW. Changes in knee moments with contralateral versus ipsilateral cane usage in females with knee osteoarthritis. Clin Biomech 2005;20:396 – 404. 13. Kerrigan DC, Johansson JL, Bryant MG, Boxer JA, Croce UD, Riley PO. Moderate-heeled shoes and knee joint torques relevant to the development and progression of knee osteoarthritis. Arch Phys Med Rehabil 2005;86:871–5. 14. Kerrigan DC, Karvosky ME, Lelas JL, Riley PO. Men’s shoes and knee joint torques relevant to the development and progression of knee osteoarthritis. J Rheumatol 2003;30:529 –33. 15. Kerrigan DC, Lelas JL, Karvosky ME. Women’s shoes and knee osteoarthritis. Lancet 2001;357:1097– 8. 16. Kerrigan DC, Todd MK, Riley PO. Knee osteoarthritis and high-heeled shoes. Lancet 1998;351:1399 – 401. 614 17. Shakoor N, Block JA. Walking barefoot decreases loading on the lower extremity joints in knee osteoarthritis. Arthritis Rheum 2006;54:2923–7. 18. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classiﬁcation and reporting of osteoarthritis: classiﬁcation of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039 – 49. 19. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833– 40. 20. Kellgren JH, Lawrence JS. Radiological assessment of osteoarthrosis. Ann Rheum Dis 1957;16:494 –502. 21. Davis RB III, Ounpuu S, Tyburski D, Gage JR. A gait data collection and reduction technique. Hum Mov Sci 1991;10: 575– 87. 22. Moisio KC, Sumner DR, Shott S, Hurwitz DE. Normalization of joint moments during gait: a comparison of 2 techniques. J Biomech 2003;36:599 – 603. 23. Mundermann A, Dyrby CO, Hurwitz DE, Sharma L, Andriacchi TP. Potential strategies to reduce medial compartment loading in patients with knee osteoarthritis of varying severity: reduced walking speed. Arthritis Rheum 2004;50: 1172– 8. 24. Hurwitz DE, Ryals AB, Case JP, Block JA, Andriacchi TP. The knee adduction moment during gait in subjects with knee osteoarthritis is more closely correlated with static alignment than radiographic disease severity, toe-out angle, and pain. J Orthop Res 2002;20:101–7. 25. Bailunas A, Hurwitz D, Ryals A, Karrar A, Case J, Block J, et al. Increased knee joint loads during walking are present in sub- Kemp et al 26. 27. 28. 29. 30. 31. 32. 33. jects with knee osteoarthritis. Osteoarthritis Cartilage 2002; 10:573–9. Fisher DS, Dyrby CO, Mundermann A, Morag E, Andriacchi TP. In healthy subjects without knee osteoarthritis, the peak knee adduction moment inﬂuences the acute effect of shoe interventions designed to reduce medial compartment knee load. J Orthop Res 2007;25:540 – 6. Kakihana W, Akai M, Nakazawa K, Takashima T, Naito K, Torii S. Effects of laterally wedged insoles on knee and subtalar joint moments. Arch Phys Med Rehabil 2005;86:1465– 71. Bergmann G, Kniggendorf H, Graichen F, Rohlmann A. Inﬂuence of shoes and heel strike on the loading of the hip joint. J Biomech 1995;28:817–27. Neumann DA. Hip abductor muscle activity as subjects with hip prostheses walk with different methods of using a cane. Phys Ther 1998;78:490 –501. Chang A, Hayes K, Dunlop D, Song J, Hurwitz D, Cahue S, et al. Hip abduction moment and protection against medial tibiofemoral osteoarthritis progression. Arthritis Rheum 2005; 52:3515–9. Mundermann A, Dyrby CO, Andriacchi TP. Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arthritis Rheum 2005;52:2835– 44. McKean KA, Landry SC, Hubley-Kozey CL, Dunbar MJ, Stanish WD, Deluzio KJ. Gender differences exist in osteoarthritic gait. Clin Biomech 2007;22:400 –9. Weidenhielm L, Svensson OK, Brostrom LA, Mattson E. Adduction moment of the knee compared to radiological and clinical parameters in moderate medical osteoarthrosis of the knee. Ann Chir Gynaecol 1994;83:236 – 42.