- Email: [email protected]
A new biomechanical model for the functional assessment of knee osteoarthritis
The Knee 11 (2004) 225–231
A new biomechanical model for the functional assessment of knee osteoarthritis W.Y. Kima,*, J. Richardsb, Richard K. Jonesb, A. Hegabc a North West Orthopaedic Higher Surgical Rotation, Fairfield General Hospital, Bury, UK Centre for Rehabilitation and Human Performance Research, University of Salford, Salford M6 6PU, UK c Fairfield General Hospital, Bury, UK
b
Received 20 February 2003; accepted 14 April 2003
Abstract The severity of symptoms in osteoarthritis (OA) of the knee can be difficult to assess. A new method for assessing medial compartment OA of the knee is proposed. The purpose of this study was to determine if a relationship exists between the severity of OA and single limb stance knee adduction moment (SSKAM). Fourteen patients with medial compartment OA of the knee were tested and compared with 14 age-matched volunteers. WOMAC OA index scores were documented. Biomechanical data were collected from a single Kistler force platform and a six-camera ProReflex motion analysis system, and the knee adduction moments were calculated. A significant difference in the maximum, minimum and mean adduction moments was found between the two groups, and a correlation of rs0.66 for WOMAC physical function subscale, rs0.63 for the WOMAC pain subscale and rs0.63 for the sum of the subscales and mean SSKAM was found. Single limb stance adduction moments provide additional, objective information in the assessment of medial compartment OA of the knee. It may be useful in the selection of patients for surgery and for evaluating various treatment modalities for medial compartment OA of the knee. 䊚 2003 Elsevier Science B.V. All rights reserved. Keywords: Knee osteoarthritis; Assessment; Biomechanical model
1. Introduction The severity of symptomatic osteoarthritis (OA) of the knee can be difficult to assess. Various methods are available including plain radiographs, knee arthroscopy and magnetic resonance imaging w1x. However, these methods may not correlate with the severity of symptoms reported by patients w2,3x, which may be variable w4x. In clinical trials of OA of the knee, well-validated questionnaires w5,6x are available. However, in theory, the use of questionnaires may allow patients to overestimate symptoms to achieve secondary gains, for instance, to expedite surgical management or employment compensation issues. It is the lack of accurate, objective, non-invasive clinical tests to assess medial *Corresponding author. 14 Croftleigh Close, Whitefield, Manchester M45 7DL, UK. Tel.: q44-161-766-5384; fax: q44-161-3316300. E-mail address: [email protected] (W.Y. Kim).
compartment OA of the knee that is the basis of this work. Previous similar studies w7–9x documented knee adduction moment during gait, rather than single limb stance, in patients with knee OA to estimate external loads acting on the knee joint. However, there is debate as to whether the severity of OA correlates with the magnitude of knee adduction moment during gait. Weidenhielm et al. w9x failed to demonstrate any correlation among clinical symptoms, radiographic parameters and knee adduction moment during gait. In contrast, Sharma et al. w8x found that adduction moment during gait in patients with OA correlated with the severity of OA radiologically. However, as discussed above, radiographic severity of OA may not have any significant relationship with patient symptoms. Adaptive mechanisms occur during gait that can reduce the load on the knee, which result in a decreased knee adduction moment during gait in patients with medial compartment OA of the knee. These mechanisms
0968-0160/04/$ - see front matter 䊚 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0968-0160Ž03.00068-1
226
W.Y. Kim et al. / The Knee 11 (2004) 225–231
Fig. 1. Diagram illustrating knee adduction moment during single limb stance.
include increased toe out angle of the foot w10x, displacing the trunk towards the affected limb w11x and decreasing stride length w12x. Single limb stance support would prevent these adaptive mechanisms from occurring, while allowing measurement of adduction moment at the knee. Knee adduction moment may be described as the turning effect due to the resultant ground reaction force acting on the foot as it passes medial to the centre of the knee joint (Fig. 1). The perpendicular distance from the line of action of the ground reaction force is the lever arm for this force. The product of this force with this lever arm produces a moment tending to adduct the knee joint. This knee adduction moment is significant and is the main determinant of medial to lateral load distribution in the knee w12x. The knee adduction moment and other external forces acting on the knee may be measured using a motion analysis system and a force platform to measure the ground reaction forces. Knee adduction moment provides information regarding the dynamic loading conditions within the knee during gait, which may be 2–3 times the body weight compared with static conditions w13x. An increase in the knee adduction moment during gait has been directly related to an increase in the medial compartment loading at the knee joint w14x. Therefore, evaluation of the knee adduction moments during gait
provides an indication of the degree of medial compartment loading at the knee joint. The knee adduction moment during gait has been shown to provide information regarding biomechanical prognosis of surgical procedures. Prodromos et al. w15x classified 21 patients who had high tibial osteotomy for varus OA of the knee into a high knee adduction moment group or low knee adduction moment group based on gait analysis performed preoperatively. Following surgery, gait analysis was repeated. The knee adduction moment was reduced in both groups of patients. However, 3.2 years after surgery, patients in the low adduction moment group had substantially better results compared with patients in the high adduction moment group. It was concluded that knee adduction moment during gait preoperatively was predictive of postoperative clinical results. However, conflicting results have been reported in the literature. Wada et al. w16x studied 32 patients who had high tibial osteotomy for primary OA of the knee. Gait analysis was performed preoperatively and repeated at 1, 3 and 6 years after high tibial osteotomy. The peak adduction moment at the knee correlated with alignment before and 6 years after surgery, but preoperative peak adduction moment of the knee did not correlate with clinical or radiographic outcomes of high tibial osteotomy, provided sufficient valgus was obtained at surgery. Knee adduction moment during gait may also predict radiological progression of knee OA. Miyazaki et al. w17x assessed knee adduction moment at baseline and 6 years later. Of 106 patients initially referred for medial compartment OA of the knee, 74 patients were reassessed 6 years later. In the 32 patients showing disease progression, which is defined as more than one grade of narrowing of minimum joint space, pain was more severe and the knee adduction moment at baseline was higher compared with the 42 patients without evidence of radiologic disease progression. 2. Objectives of the study 1. To determine if adduction moment during single limb stance is significantly different in patients with medial compartment OA of the knee compared with normal knees. 2. To determine if correlation exists between the severity of the knee OA as assessed by a validated patient questionnaire and single limb stance adduction moment. 3. Materials and methods The subjects with knee OA were recruited from the outpatient clinics of an orthopaedic surgeon. Subjects underwent clinical, radiographic and single limb stance analysis. Only patients who were able to perform a
W.Y. Kim et al. / The Knee 11 (2004) 225–231
227
Table 1 Demographics of the two subject groups Control subjects
Subjects with OA
Age (years) Knee range of movement (degrees) Knee fixed flexion deformity Knee varus deformity (degrees)
54.6 (range 39–73) 135.3 (range 125–144) Nil Nil
60.4 (range 47–72) 115.9 (range 96–130) Nil 3.7 (range 0–15)
Grade of knee OA (Kellgren–Lawrence)
Not applicable
Grade 2 (ns5) Grade 3 (ns7) Grade 4 (ns2)
single limb stance for at least 3 s were included in the study. This was confirmed in the orthopaedic clinic prior to assessment in the gait laboratory. The study had ethical committee approval and all patients granted informed consent. The study was restricted to patients who had isolated medial compartment OA of the knee. This was confirmed on weight-bearing anterior–posterior, lateral and skyline radiographs of the knee. Only patients without clinical evidence of anterior cruciate ligament deficiency were included in the study. None of the subjects had hip or lumbosacral spine OA. Only patients who had a Kellgren–Lawrence w18x grade of at least grade 2, indicating definite osteophyte and possible joint space narrowing on weight-bearing radiographs, were included in the study. Fourteen patients fulfilled these criteria and made up the OA subjects group (Table 1). The subjects with OA were assessed for pain and function using the visual analog version 3.0 of the Western Ontario and McMaster University Rating (WOMAC-VA3.0) w5x. The pain (maximum 500 points), stiffness (maximum 200 points) and physical function (maximum 1700 points) subscales and the sum of these three subscales (maximum 2400 points) were assessed. Fourteen asymptomatic age-matched control subjects were also recruited into the study. All control subjects were able to perform single limb stance for at least 3 s. This was confirmed prior to assessment in the gait laboratory. All control subjects had no clinical diagnosis of OA, pain, rheumatoid arthritis or history of knee trauma. They also had no significant history of OA of the hip or lumbosacral spine. All control subjects had a stable knee clinically. None of the control subjects, neither the OA subjects, had a fixed flexion deformity of the knee. They also had no significant medical history that would restrict their activities. 3.1. Method of data collection Force and motion data were collected using a single Kistler Force platform and a six-camera ProReflex motion analysis system. Retroreflective markers were
placed on the centre of the greater trochanter, superio aspect of patella, lateral joint line of the knee, tibial tuberosity, lateral aspect of the lateral maleolus and at the base of the fifth metatarsal. Three-dimensional spatial positions of the markers were measured and the joint centres were calculated. The external moments about the knee were calculated using inverse dynamics and by modelling the leg as a collection of rigid links or segments. The link model included the assumption that no axial rotation occurred about the long axis of each segment. During the single limb stance test, the subjects were asked to stand on the affected limb for approximately 3 s. Five trials of single limb stance were obtained. The maximum, minimum and mean external adduction moments at the knee were found for the five trials. The moments were then normalised to body mass (N mykg). The investigator performing the analysis of the adduction moment data was blinded to the clinical status of the subjects tested. 3.2. Statistical methods Differences in the maximum, minimum and mean moments found during single-legged stance were investigated between the subjects with and without OA, with a two-tailed, two-sampled equal variance t-test. The relationship between the mean knee adduction normalised moment (N mykg) and the subscales of the WOMAC OA index was investigated with a Pearson’s correlation coefficient test. 4. Results The adduction moments at the knee showed a significant difference in the maximum, minimum and mean moments, with the OA subjects experiencing greater moments than the age-matched volunteers (Ps0.0122, 0.0013, 0.0017) (Table 2). There was no significant difference in the range of values found. Using the WOMAC OA index for severity of OA, a correlation of rs0.63 was found for WOMAC pain subscale, rs0.66 for WOMAC physical function subscale, rs0.10 for the WOMAC stiffness subscale and
228
W.Y. Kim et al. / The Knee 11 (2004) 225–231
rs0.63 for the WOMAC sum of the subscales and single limb stance adduction moment (Table 3). Data from one OA subject had to be excluded owing to equipment failure.
Table 3 Relationship between WOMAC OA scores and SSKAM Pain score Correlation coefficients, r R2 Correlation (P-value)
5. Discussion Radiographic assessment of the knee is inexpensive, simple, non-invasive and provides information regarding the alignment of the knee in OA. In severe OA, it provides a quick diagnosis and further management is easily instituted. However, in a 1-year follow-up of patients with knee OA, who had radiologic assessment, radiographic deterioration did not correlate with clinical symptoms of worsening pain and loss of function w2x. Patients with progressive deterioration of OA of the knee may have symptoms that precede radiographic evidence of worsening OA. Varus alignment of the knee appears to be associated with a higher knee adduction moment compared with neutral or valgus aligned knees. In an analysis of 11 healthy, normal subjects, Andrews et al. w19x demonstrated that patients who had radiographic evidence indicative of varus alignment of the lower extremity had statistically higher peaks in knee adduction moment during the early stance phase of gait. Furthermore, Hurwitz et al. w20x demonstrated that, among patients with knee OA, the single best predictor of high peak adduction moment of the knee was mechanical axis, with patients with varus alignment demonstrating higher peak adduction moments compared with valgus aligned knees. However, different results of the relationship between the knee adduction moment and the static alignment of the knee have been reported. In a comparison of static and dynamic assessments of OA patients, Harrington w21x failed to demonstrate a direct relationship between angular deformity at the knee and loads in the knee during walking. It was hypothesised that the load on the knee depended on the centre of gravity of body mass and forces acting on it. It was demonstrated that the location of joint force was not clearly predictable during dynamic loading. Similarly, Weidenhielm et al. w9x investigated the relationship among clinical symptoms, radiographs and
Physical Stiffness Total function score score
0.634 0.663 0.401 0.440 0.020 0.013
0.106 0.011 0.731
0.633 0.400 0.020
knee adduction moments during gait in 54 patients and found no significant correlations between any of these parameters. Weidenhielm et al. assessed function and pain by a modified British Orthopaedic Association knee function chart, which is not very well validated or widely used. Weidenhielm et al. concluded that compensatory gait mechanisms might have occurred in the subjects with knee OA, as otherwise the knee adduction moments during gait would have been greater. Factors other than mechanical alignment contribute to variations in external moments about the knee joint. Rotating the foot outward (toe out) moves the position of the ground reaction force closer to the centre of the knee joint, reducing the effective tendency of the ground reaction force to adduct the knee w10x. Other factors include displacing the trunk over the affected limb while walking w11x and decreasing stride length w12x. The adoption of these compensatory gait mechanisms by subjects with knee OA may account for the conflicting relationship between knee adduction moments during gait and knee malalignment (varus malalignment in medial compartment knee OA) reported in the literature. Knee adduction moments provide biomechanical data regarding actual loading at the knee during gait. Compensatory mechanisms are probably adopted by patients with knee OA to reduce pain in the affected knee. However, the values measured in gait analysis will, therefore, not reflect the severity of symptoms attributable to OA. Furthermore, the use of gait analysis is time-consuming and complex as it requires the measurement of several biomechanical parameters in the assessment of knee OA. This study tested the hypothesis that the use of single limb support prevented the ability to reduce peak knee adduction moment by increasing the foot progression (toe out) angle, decreasing the stride length and displac-
Table 2 Comparison of SSKAMs for subjects with and without OA (control ns14, OA ns13) Moments (N mykg)
Subjects with OA Mean (S.D.)
Maximum moment Minimum moment Mean moment Range of values
0.7137 0.4962 0.6273 0.2377
(0.2003) (0.1674) (0.1826) (0.1443)
Subjects with no OA Mean (S.D.) 0.5516 0.2867 0.4134 0.2704
(0.1873) (0.1334) (0.1306) (0.1476)
Unpaired t-test P-value 0.0122 0.0013 0.0017 0.6401
W.Y. Kim et al. / The Knee 11 (2004) 225–231
229
Fig. 2. Graph of relationship between pain subscale of WOMAC and SSKAM with 95% confidence intervals.
ing the trunk towards the affected side; all of which may occur during gait. Patients with medial compartment knee OA who have progressive clinical symptoms and who do not have corresponding deterioration in knee radiographs is a common scenario encountered in clinical practice. These patients would benefit from an assessment method that would accurately document functional impairment. The method of using single limb stance knee adduction moment (SSKAM) may therefore provide objective criteria in the selection of these patients for nonoperative or surgical intervention. In the present study, a correlation was found between the severity of OA using a specific, purpose-built highperformance instrument for evaluative research in OA w5x and SSKAM. Pain and functional impairment are the principal complaints of patients with increasing disability from knee OA. A strong and significant correlation with single limb stance adduction moment was obtained with these two parameters ((rs0.634, rs 0.663), Table 3). The intercept of the regression model of mean adduction moment and the WOMAC pain, function and total score varied very little (0.41–0.43 N mykg, Figs. 2 and 3). It is also interesting to note that this is close to the value of mean adduction moment recorded from the control group (0.413 N mykg, Table 2). This demonstrates that as the WOMAC score tends to zero (normal) the regression models predict a similar value to that recorded by the control subjects. It could be stated that the greater moment may be due to the OA subjects having a more unstable single-
legged support. However, the range of values obtained by the control subjects was not significantly different from the subjects with OA (Table 2), thus indicating a similar degree of balance and stability during single limb stance was obtained by both groups. The results of this work are comparable with previous studies. Sharma et al. w8x reported a correlation between peak knee adduction moment during gait and the severity of OA radiologically. However, the study by Sharma et al. w8x was not limited to patients with isolated medial compartment knee OA, and functional impairment as a result of knee OA, apart from pain, was not documented. Miyazaki et al. w17x reported that the knee adduction moment during gait was higher in patients who later demonstrated disease progression radiologically. Functional impairment due to knee OA, apart from pain, was also not documented. SSKAM provides objective information in the assessment of patients with medial compartment OA of the knee. The authors believe that this may be a useful adjunct in the selection of patients for surgery, particularly patients with progressively worsening knee symptoms. It may also provide useful baseline biomechanical information for a patient with equivocal knee symptoms. It may aid in the evaluation of various treatment modalities for medial compartment knee OA such as unicompartmental knee arthroplasty or high tibial osteotomy. Further research is warranted to determine if it may be used as a marker of progression of medial compartment knee OA.
230
W.Y. Kim et al. / The Knee 11 (2004) 225–231
Fig. 3. Graph of relationship between sum of the subscales WOMAC and SSKAM with 95% confidence intervals.
References w1x Burgkart R, Glaser C, Hyhlik-Durr A, Englmeier KH, Reiser M, Eckstein F. Magnetic resonance imaging-based assessment of cartilage loss in severe osteoarthritis: accuracy, precision and diagnostic value. Arthritis Rheum 2001;44(9):2072 –2077. w2x Dougados M, Gueguen A, Nguyen M, et al. Longitudinal radiologic evaluation of osteoarthritis of the knee. J Rheum 1992;19(3):378 –384. w3x Dieppe PA, Cushnaghan J, Shepstone L. The Bristol ‘OA500’ study: progression of osteoarthritis over 3 years and the relationship between clinical and radiographic changes at the knee joint. Osteoarthr Cartilage 1997;5:87 –97. w4x Summers MN, Haley WE, Reveille JD, Alarcon GS. Radiographic assessment and psychologic variables as predictors of pain and functional impairment in osteoarthritis of the knee or hip. Arthritis Rheum 1988;31(2):204 –209. w5x 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(12):1833 –1840. w6x Lequesne MG, Mery C, Samson M, et al. Indexes of severity for osteoarthritis of the hip and knee: validation-value in comparison with other assessment tests. Scand J Rheumatol 1987;65:85 –89. w7x Hurwitz DE, Ryals AR, Block JA, Sharma L, Schnitzer TJ, Andriacchi TP. Knee pain and joint loading in subjects with osteoarthritis of the knee. J Orthop Res 2000;18:572 –579. w8x Sharma L, Hurwitz DE, Thonar EJM, et al. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum 1998;41(7):1233 – 1240.
w9x Weidenhielm L, Svensson OK, Brostrom LA, Mattsson E. Adduction moment of the knee compared to radiological and clinical parameters in moderate medial osteoarthrosis of the knee. Ann Chir Gynaecol 1994;83:236 –242. w10x Wang JW, Kuo KN, Andriacchi TP, Galante JO. The influence of walking mechanics and time on the results on proximal tibial osteotomy. J Bone Joint Surg wAmx 1990;72:905 –909. w11x Maquet PGJ. Biomechanics of the knee. With application to the pathogenesis and surgical treatment of osteoarthritis. 2nd ed.. New York: Springer, 1984. w12x Andriacchi TP. Dynamics of knee malalignment. Orthop Clin North Am 1994;25(3):395 –403. w13x Morrison JB. The mechanics of the knee joint in relation to normal walking. J Biomech 1970;3(1):51 –61. w14x Schipplein OD, Andriacchi TP. Interaction between active and passive knee stabilizers during level walking. J Orthop Res 1991;9:113 –119. w15x Prodromos CC, Andriacchi TP, Galante JO. A relationship between gait and clinical changes following high tibial osteotomy. J Bone Joint Surg wAmx 1985;67:1188 –1194. w16x Wada M, Imura S, Nagatani K, Baba H, Shimada S, Sasaki S. Relationship between gait and clinical results after high tibial osteotomy. Clin Orthop 1998;354:180 –188. w17x 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 –622. w18x Kellgren JH, Lawrence JS. Atlas of standard radiographs. Oxford: Department of Rheumatology and Medical illustrations, University of Manchester, Blackwell, 1963. w19x Andrews M, Noyes FR, Hewett TE, Andriacchi TP. Lower limb alignment and foot angle are related to stance phase knee
W.Y. Kim et al. / The Knee 11 (2004) 225–231 adduction in normal subjects: a critical analysis of the reliability of gait analysis data. J Orthop Res 1996;14:289 –295. w20x 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
231
than radiographic disease severity, toe out angle and pain. J Orthop Res 2002;20:101 –107. w21x Harrington IJ. Static and dynamic loading patterns in knee joints with deformities. J Bone Joint Surg wAmx 1983;65(2):247 –259.
A new biomechanical model for the functional assessment of knee osteoarthritis W.Y. Kima,*, J. Richardsb, Richard K. Jonesb, A. Hegabc a North West Orthopaedic Higher Surgical Rotation, Fairfield General Hospital, Bury, UK Centre for Rehabilitation and Human Performance Research, University of Salford, Salford M6 6PU, UK c Fairfield General Hospital, Bury, UK
b
Received 20 February 2003; accepted 14 April 2003
Abstract The severity of symptoms in osteoarthritis (OA) of the knee can be difficult to assess. A new method for assessing medial compartment OA of the knee is proposed. The purpose of this study was to determine if a relationship exists between the severity of OA and single limb stance knee adduction moment (SSKAM). Fourteen patients with medial compartment OA of the knee were tested and compared with 14 age-matched volunteers. WOMAC OA index scores were documented. Biomechanical data were collected from a single Kistler force platform and a six-camera ProReflex motion analysis system, and the knee adduction moments were calculated. A significant difference in the maximum, minimum and mean adduction moments was found between the two groups, and a correlation of rs0.66 for WOMAC physical function subscale, rs0.63 for the WOMAC pain subscale and rs0.63 for the sum of the subscales and mean SSKAM was found. Single limb stance adduction moments provide additional, objective information in the assessment of medial compartment OA of the knee. It may be useful in the selection of patients for surgery and for evaluating various treatment modalities for medial compartment OA of the knee. 䊚 2003 Elsevier Science B.V. All rights reserved. Keywords: Knee osteoarthritis; Assessment; Biomechanical model
1. Introduction The severity of symptomatic osteoarthritis (OA) of the knee can be difficult to assess. Various methods are available including plain radiographs, knee arthroscopy and magnetic resonance imaging w1x. However, these methods may not correlate with the severity of symptoms reported by patients w2,3x, which may be variable w4x. In clinical trials of OA of the knee, well-validated questionnaires w5,6x are available. However, in theory, the use of questionnaires may allow patients to overestimate symptoms to achieve secondary gains, for instance, to expedite surgical management or employment compensation issues. It is the lack of accurate, objective, non-invasive clinical tests to assess medial *Corresponding author. 14 Croftleigh Close, Whitefield, Manchester M45 7DL, UK. Tel.: q44-161-766-5384; fax: q44-161-3316300. E-mail address: [email protected] (W.Y. Kim).
compartment OA of the knee that is the basis of this work. Previous similar studies w7–9x documented knee adduction moment during gait, rather than single limb stance, in patients with knee OA to estimate external loads acting on the knee joint. However, there is debate as to whether the severity of OA correlates with the magnitude of knee adduction moment during gait. Weidenhielm et al. w9x failed to demonstrate any correlation among clinical symptoms, radiographic parameters and knee adduction moment during gait. In contrast, Sharma et al. w8x found that adduction moment during gait in patients with OA correlated with the severity of OA radiologically. However, as discussed above, radiographic severity of OA may not have any significant relationship with patient symptoms. Adaptive mechanisms occur during gait that can reduce the load on the knee, which result in a decreased knee adduction moment during gait in patients with medial compartment OA of the knee. These mechanisms
0968-0160/04/$ - see front matter 䊚 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0968-0160Ž03.00068-1
226
W.Y. Kim et al. / The Knee 11 (2004) 225–231
Fig. 1. Diagram illustrating knee adduction moment during single limb stance.
include increased toe out angle of the foot w10x, displacing the trunk towards the affected limb w11x and decreasing stride length w12x. Single limb stance support would prevent these adaptive mechanisms from occurring, while allowing measurement of adduction moment at the knee. Knee adduction moment may be described as the turning effect due to the resultant ground reaction force acting on the foot as it passes medial to the centre of the knee joint (Fig. 1). The perpendicular distance from the line of action of the ground reaction force is the lever arm for this force. The product of this force with this lever arm produces a moment tending to adduct the knee joint. This knee adduction moment is significant and is the main determinant of medial to lateral load distribution in the knee w12x. The knee adduction moment and other external forces acting on the knee may be measured using a motion analysis system and a force platform to measure the ground reaction forces. Knee adduction moment provides information regarding the dynamic loading conditions within the knee during gait, which may be 2–3 times the body weight compared with static conditions w13x. An increase in the knee adduction moment during gait has been directly related to an increase in the medial compartment loading at the knee joint w14x. Therefore, evaluation of the knee adduction moments during gait
provides an indication of the degree of medial compartment loading at the knee joint. The knee adduction moment during gait has been shown to provide information regarding biomechanical prognosis of surgical procedures. Prodromos et al. w15x classified 21 patients who had high tibial osteotomy for varus OA of the knee into a high knee adduction moment group or low knee adduction moment group based on gait analysis performed preoperatively. Following surgery, gait analysis was repeated. The knee adduction moment was reduced in both groups of patients. However, 3.2 years after surgery, patients in the low adduction moment group had substantially better results compared with patients in the high adduction moment group. It was concluded that knee adduction moment during gait preoperatively was predictive of postoperative clinical results. However, conflicting results have been reported in the literature. Wada et al. w16x studied 32 patients who had high tibial osteotomy for primary OA of the knee. Gait analysis was performed preoperatively and repeated at 1, 3 and 6 years after high tibial osteotomy. The peak adduction moment at the knee correlated with alignment before and 6 years after surgery, but preoperative peak adduction moment of the knee did not correlate with clinical or radiographic outcomes of high tibial osteotomy, provided sufficient valgus was obtained at surgery. Knee adduction moment during gait may also predict radiological progression of knee OA. Miyazaki et al. w17x assessed knee adduction moment at baseline and 6 years later. Of 106 patients initially referred for medial compartment OA of the knee, 74 patients were reassessed 6 years later. In the 32 patients showing disease progression, which is defined as more than one grade of narrowing of minimum joint space, pain was more severe and the knee adduction moment at baseline was higher compared with the 42 patients without evidence of radiologic disease progression. 2. Objectives of the study 1. To determine if adduction moment during single limb stance is significantly different in patients with medial compartment OA of the knee compared with normal knees. 2. To determine if correlation exists between the severity of the knee OA as assessed by a validated patient questionnaire and single limb stance adduction moment. 3. Materials and methods The subjects with knee OA were recruited from the outpatient clinics of an orthopaedic surgeon. Subjects underwent clinical, radiographic and single limb stance analysis. Only patients who were able to perform a
W.Y. Kim et al. / The Knee 11 (2004) 225–231
227
Table 1 Demographics of the two subject groups Control subjects
Subjects with OA
Age (years) Knee range of movement (degrees) Knee fixed flexion deformity Knee varus deformity (degrees)
54.6 (range 39–73) 135.3 (range 125–144) Nil Nil
60.4 (range 47–72) 115.9 (range 96–130) Nil 3.7 (range 0–15)
Grade of knee OA (Kellgren–Lawrence)
Not applicable
Grade 2 (ns5) Grade 3 (ns7) Grade 4 (ns2)
single limb stance for at least 3 s were included in the study. This was confirmed in the orthopaedic clinic prior to assessment in the gait laboratory. The study had ethical committee approval and all patients granted informed consent. The study was restricted to patients who had isolated medial compartment OA of the knee. This was confirmed on weight-bearing anterior–posterior, lateral and skyline radiographs of the knee. Only patients without clinical evidence of anterior cruciate ligament deficiency were included in the study. None of the subjects had hip or lumbosacral spine OA. Only patients who had a Kellgren–Lawrence w18x grade of at least grade 2, indicating definite osteophyte and possible joint space narrowing on weight-bearing radiographs, were included in the study. Fourteen patients fulfilled these criteria and made up the OA subjects group (Table 1). The subjects with OA were assessed for pain and function using the visual analog version 3.0 of the Western Ontario and McMaster University Rating (WOMAC-VA3.0) w5x. The pain (maximum 500 points), stiffness (maximum 200 points) and physical function (maximum 1700 points) subscales and the sum of these three subscales (maximum 2400 points) were assessed. Fourteen asymptomatic age-matched control subjects were also recruited into the study. All control subjects were able to perform single limb stance for at least 3 s. This was confirmed prior to assessment in the gait laboratory. All control subjects had no clinical diagnosis of OA, pain, rheumatoid arthritis or history of knee trauma. They also had no significant history of OA of the hip or lumbosacral spine. All control subjects had a stable knee clinically. None of the control subjects, neither the OA subjects, had a fixed flexion deformity of the knee. They also had no significant medical history that would restrict their activities. 3.1. Method of data collection Force and motion data were collected using a single Kistler Force platform and a six-camera ProReflex motion analysis system. Retroreflective markers were
placed on the centre of the greater trochanter, superio aspect of patella, lateral joint line of the knee, tibial tuberosity, lateral aspect of the lateral maleolus and at the base of the fifth metatarsal. Three-dimensional spatial positions of the markers were measured and the joint centres were calculated. The external moments about the knee were calculated using inverse dynamics and by modelling the leg as a collection of rigid links or segments. The link model included the assumption that no axial rotation occurred about the long axis of each segment. During the single limb stance test, the subjects were asked to stand on the affected limb for approximately 3 s. Five trials of single limb stance were obtained. The maximum, minimum and mean external adduction moments at the knee were found for the five trials. The moments were then normalised to body mass (N mykg). The investigator performing the analysis of the adduction moment data was blinded to the clinical status of the subjects tested. 3.2. Statistical methods Differences in the maximum, minimum and mean moments found during single-legged stance were investigated between the subjects with and without OA, with a two-tailed, two-sampled equal variance t-test. The relationship between the mean knee adduction normalised moment (N mykg) and the subscales of the WOMAC OA index was investigated with a Pearson’s correlation coefficient test. 4. Results The adduction moments at the knee showed a significant difference in the maximum, minimum and mean moments, with the OA subjects experiencing greater moments than the age-matched volunteers (Ps0.0122, 0.0013, 0.0017) (Table 2). There was no significant difference in the range of values found. Using the WOMAC OA index for severity of OA, a correlation of rs0.63 was found for WOMAC pain subscale, rs0.66 for WOMAC physical function subscale, rs0.10 for the WOMAC stiffness subscale and
228
W.Y. Kim et al. / The Knee 11 (2004) 225–231
rs0.63 for the WOMAC sum of the subscales and single limb stance adduction moment (Table 3). Data from one OA subject had to be excluded owing to equipment failure.
Table 3 Relationship between WOMAC OA scores and SSKAM Pain score Correlation coefficients, r R2 Correlation (P-value)
5. Discussion Radiographic assessment of the knee is inexpensive, simple, non-invasive and provides information regarding the alignment of the knee in OA. In severe OA, it provides a quick diagnosis and further management is easily instituted. However, in a 1-year follow-up of patients with knee OA, who had radiologic assessment, radiographic deterioration did not correlate with clinical symptoms of worsening pain and loss of function w2x. Patients with progressive deterioration of OA of the knee may have symptoms that precede radiographic evidence of worsening OA. Varus alignment of the knee appears to be associated with a higher knee adduction moment compared with neutral or valgus aligned knees. In an analysis of 11 healthy, normal subjects, Andrews et al. w19x demonstrated that patients who had radiographic evidence indicative of varus alignment of the lower extremity had statistically higher peaks in knee adduction moment during the early stance phase of gait. Furthermore, Hurwitz et al. w20x demonstrated that, among patients with knee OA, the single best predictor of high peak adduction moment of the knee was mechanical axis, with patients with varus alignment demonstrating higher peak adduction moments compared with valgus aligned knees. However, different results of the relationship between the knee adduction moment and the static alignment of the knee have been reported. In a comparison of static and dynamic assessments of OA patients, Harrington w21x failed to demonstrate a direct relationship between angular deformity at the knee and loads in the knee during walking. It was hypothesised that the load on the knee depended on the centre of gravity of body mass and forces acting on it. It was demonstrated that the location of joint force was not clearly predictable during dynamic loading. Similarly, Weidenhielm et al. w9x investigated the relationship among clinical symptoms, radiographs and
Physical Stiffness Total function score score
0.634 0.663 0.401 0.440 0.020 0.013
0.106 0.011 0.731
0.633 0.400 0.020
knee adduction moments during gait in 54 patients and found no significant correlations between any of these parameters. Weidenhielm et al. assessed function and pain by a modified British Orthopaedic Association knee function chart, which is not very well validated or widely used. Weidenhielm et al. concluded that compensatory gait mechanisms might have occurred in the subjects with knee OA, as otherwise the knee adduction moments during gait would have been greater. Factors other than mechanical alignment contribute to variations in external moments about the knee joint. Rotating the foot outward (toe out) moves the position of the ground reaction force closer to the centre of the knee joint, reducing the effective tendency of the ground reaction force to adduct the knee w10x. Other factors include displacing the trunk over the affected limb while walking w11x and decreasing stride length w12x. The adoption of these compensatory gait mechanisms by subjects with knee OA may account for the conflicting relationship between knee adduction moments during gait and knee malalignment (varus malalignment in medial compartment knee OA) reported in the literature. Knee adduction moments provide biomechanical data regarding actual loading at the knee during gait. Compensatory mechanisms are probably adopted by patients with knee OA to reduce pain in the affected knee. However, the values measured in gait analysis will, therefore, not reflect the severity of symptoms attributable to OA. Furthermore, the use of gait analysis is time-consuming and complex as it requires the measurement of several biomechanical parameters in the assessment of knee OA. This study tested the hypothesis that the use of single limb support prevented the ability to reduce peak knee adduction moment by increasing the foot progression (toe out) angle, decreasing the stride length and displac-
Table 2 Comparison of SSKAMs for subjects with and without OA (control ns14, OA ns13) Moments (N mykg)
Subjects with OA Mean (S.D.)
Maximum moment Minimum moment Mean moment Range of values
0.7137 0.4962 0.6273 0.2377
(0.2003) (0.1674) (0.1826) (0.1443)
Subjects with no OA Mean (S.D.) 0.5516 0.2867 0.4134 0.2704
(0.1873) (0.1334) (0.1306) (0.1476)
Unpaired t-test P-value 0.0122 0.0013 0.0017 0.6401
W.Y. Kim et al. / The Knee 11 (2004) 225–231
229
Fig. 2. Graph of relationship between pain subscale of WOMAC and SSKAM with 95% confidence intervals.
ing the trunk towards the affected side; all of which may occur during gait. Patients with medial compartment knee OA who have progressive clinical symptoms and who do not have corresponding deterioration in knee radiographs is a common scenario encountered in clinical practice. These patients would benefit from an assessment method that would accurately document functional impairment. The method of using single limb stance knee adduction moment (SSKAM) may therefore provide objective criteria in the selection of these patients for nonoperative or surgical intervention. In the present study, a correlation was found between the severity of OA using a specific, purpose-built highperformance instrument for evaluative research in OA w5x and SSKAM. Pain and functional impairment are the principal complaints of patients with increasing disability from knee OA. A strong and significant correlation with single limb stance adduction moment was obtained with these two parameters ((rs0.634, rs 0.663), Table 3). The intercept of the regression model of mean adduction moment and the WOMAC pain, function and total score varied very little (0.41–0.43 N mykg, Figs. 2 and 3). It is also interesting to note that this is close to the value of mean adduction moment recorded from the control group (0.413 N mykg, Table 2). This demonstrates that as the WOMAC score tends to zero (normal) the regression models predict a similar value to that recorded by the control subjects. It could be stated that the greater moment may be due to the OA subjects having a more unstable single-
legged support. However, the range of values obtained by the control subjects was not significantly different from the subjects with OA (Table 2), thus indicating a similar degree of balance and stability during single limb stance was obtained by both groups. The results of this work are comparable with previous studies. Sharma et al. w8x reported a correlation between peak knee adduction moment during gait and the severity of OA radiologically. However, the study by Sharma et al. w8x was not limited to patients with isolated medial compartment knee OA, and functional impairment as a result of knee OA, apart from pain, was not documented. Miyazaki et al. w17x reported that the knee adduction moment during gait was higher in patients who later demonstrated disease progression radiologically. Functional impairment due to knee OA, apart from pain, was also not documented. SSKAM provides objective information in the assessment of patients with medial compartment OA of the knee. The authors believe that this may be a useful adjunct in the selection of patients for surgery, particularly patients with progressively worsening knee symptoms. It may also provide useful baseline biomechanical information for a patient with equivocal knee symptoms. It may aid in the evaluation of various treatment modalities for medial compartment knee OA such as unicompartmental knee arthroplasty or high tibial osteotomy. Further research is warranted to determine if it may be used as a marker of progression of medial compartment knee OA.
230
W.Y. Kim et al. / The Knee 11 (2004) 225–231
Fig. 3. Graph of relationship between sum of the subscales WOMAC and SSKAM with 95% confidence intervals.
References w1x Burgkart R, Glaser C, Hyhlik-Durr A, Englmeier KH, Reiser M, Eckstein F. Magnetic resonance imaging-based assessment of cartilage loss in severe osteoarthritis: accuracy, precision and diagnostic value. Arthritis Rheum 2001;44(9):2072 –2077. w2x Dougados M, Gueguen A, Nguyen M, et al. Longitudinal radiologic evaluation of osteoarthritis of the knee. J Rheum 1992;19(3):378 –384. w3x Dieppe PA, Cushnaghan J, Shepstone L. The Bristol ‘OA500’ study: progression of osteoarthritis over 3 years and the relationship between clinical and radiographic changes at the knee joint. Osteoarthr Cartilage 1997;5:87 –97. w4x Summers MN, Haley WE, Reveille JD, Alarcon GS. Radiographic assessment and psychologic variables as predictors of pain and functional impairment in osteoarthritis of the knee or hip. Arthritis Rheum 1988;31(2):204 –209. w5x 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(12):1833 –1840. w6x Lequesne MG, Mery C, Samson M, et al. Indexes of severity for osteoarthritis of the hip and knee: validation-value in comparison with other assessment tests. Scand J Rheumatol 1987;65:85 –89. w7x Hurwitz DE, Ryals AR, Block JA, Sharma L, Schnitzer TJ, Andriacchi TP. Knee pain and joint loading in subjects with osteoarthritis of the knee. J Orthop Res 2000;18:572 –579. w8x Sharma L, Hurwitz DE, Thonar EJM, et al. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum 1998;41(7):1233 – 1240.
w9x Weidenhielm L, Svensson OK, Brostrom LA, Mattsson E. Adduction moment of the knee compared to radiological and clinical parameters in moderate medial osteoarthrosis of the knee. Ann Chir Gynaecol 1994;83:236 –242. w10x Wang JW, Kuo KN, Andriacchi TP, Galante JO. The influence of walking mechanics and time on the results on proximal tibial osteotomy. J Bone Joint Surg wAmx 1990;72:905 –909. w11x Maquet PGJ. Biomechanics of the knee. With application to the pathogenesis and surgical treatment of osteoarthritis. 2nd ed.. New York: Springer, 1984. w12x Andriacchi TP. Dynamics of knee malalignment. Orthop Clin North Am 1994;25(3):395 –403. w13x Morrison JB. The mechanics of the knee joint in relation to normal walking. J Biomech 1970;3(1):51 –61. w14x Schipplein OD, Andriacchi TP. Interaction between active and passive knee stabilizers during level walking. J Orthop Res 1991;9:113 –119. w15x Prodromos CC, Andriacchi TP, Galante JO. A relationship between gait and clinical changes following high tibial osteotomy. J Bone Joint Surg wAmx 1985;67:1188 –1194. w16x Wada M, Imura S, Nagatani K, Baba H, Shimada S, Sasaki S. Relationship between gait and clinical results after high tibial osteotomy. Clin Orthop 1998;354:180 –188. w17x 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 –622. w18x Kellgren JH, Lawrence JS. Atlas of standard radiographs. Oxford: Department of Rheumatology and Medical illustrations, University of Manchester, Blackwell, 1963. w19x Andrews M, Noyes FR, Hewett TE, Andriacchi TP. Lower limb alignment and foot angle are related to stance phase knee
W.Y. Kim et al. / The Knee 11 (2004) 225–231 adduction in normal subjects: a critical analysis of the reliability of gait analysis data. J Orthop Res 1996;14:289 –295. w20x 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
231
than radiographic disease severity, toe out angle and pain. J Orthop Res 2002;20:101 –107. w21x Harrington IJ. Static and dynamic loading patterns in knee joints with deformities. J Bone Joint Surg wAmx 1983;65(2):247 –259.