• Bandaging kinesthesia;
  • Knee proprioception;
  • Osteoarthritis


  1. Top of page
  2. Abstract


To investigate whether osteoarthritis (OA) of the knee or bandaging of the knee affects movement detection.


Movement detection levels were determined in 10 women aged > 65 years with moderate to severe knee OA and 10 healthy women matched for age, body mass index, and activity levels. Movements were imposed at 0.5°/second, 1.0°/second, and 2.5°/second. Additionally, detection levels were compared with and without the knee bandaged at a single velocity, (0.5°/second).


Controls perceived significantly smaller movements than OA subjects at all test velocities (P < 0.01). However, the bandage did not affect movement detection (P > 0.05).


Detection of movement at the knee was impaired in subjects with severe knee OA, and a bandage did not improve detection. Thus, considering previous findings that position sense is impaired, a generalized proprioceptive deficit appears to be associated with OA. This deficit could result from loss of receptors, altered muscle function, or the consequent joint instability.


  1. Top of page
  2. Abstract

The knee is commonly affected by osteoarthritis (OA), often causing severe disability (1). This progressive degenerative disease is known to affect various changes in articular cartilage, subchondral bone, synovium, and joint capsule, and consequently these structures have been studied widely (2, 3). The effect of OA on muscles, however, has been investigated less frequently, although it is likely muscles also are involved. For example, the quadriceps muscles are weaker in elderly people with OA than in healthy, age-matched controls; the muscles are also incompletely activated (4). It is unclear whether such changes precede or result from the development of OA (2, 5). Because proprioceptive input arises from muscle, joint, and cutaneous afferents, altered input from any of these could impair proprioception (6).

Although proprioception at the OA knee has been studied extensively (4, 7–9), all studies to date have focused on one aspect, position sense. These studies have consistently found an impairment from OA in subjects with unilateral (9, 10) and bilateral (4, 7) knee OA. However, other aspects of proprioception, such as the ability to perceive movement (11), have not been studied. Because it is unknown whether the 2 sensations are affected in the same way, although preliminary work suggests no correlation between the 2 (12), conclusions cannot be drawn about the effect of OA on movement detection. The ability to perceive movement may be closely linked to stability and function (4), and therefore its integrity may be even more important than the ability to perceive a change in position. In the young healthy knee, very small movements (∼1.0° for movements imposed at 0.1°/second) can be detected (13), probably because such proprioceptive information is essential to good stability. It is possible, therefore, that even a small decrease in proprioceptive acuity has an important effect.

Many patients with OA bandage the knee to reduce symptoms of pain and swelling. However, it is also thought that application of bandages improves proprioceptive acuity by increasing cutaneous input (14). The application of a tubular bandage (e.g., Tubigrip) improved position sense in OA knees (15), but this effect has not been consistently found in other conditions (16), or other joints (17). The effect on movement detection of wearing a bandage at the knee has not been evaluated. If bandage application improves proprioception, and thereby potentially improves stability, this would be an easy clinical intervention to apply with patient-relevant benefits. The present study was therefore designed to investigate the effect of knee OA and the effect of wearing a tubular bandage on movement perception.


  1. Top of page
  2. Abstract


Twenty women aged ≥ 65 years volunteered to participate in the study. Ten subjects had a history of moderate to severe (≥ grade 3 by the Kellgren/Lawrence scale) primary or idiopathic OA in one or both knees. The Kellgren/Lawrence grading system is a 5-point scale that grades radiographs based on the presence of osteophyte formation, joint space narrowing, and sclerotic changes in the subchondral bone (18). The diagnosis of OA was based on the American College of Rheumatology (formerly the American Rheumatism Association) classification criteria (19). The severity of the disability associated with OA was determined using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) (20). The WOMAC is a self-administered questionnaire incorporating questions related to knee pain, stiffness, and difficulties performing functional activities (21). Patients were excluded from the OA group if they had any signs of inflammatory joint disease or history of knee surgery.

A control group of 10 subjects was recruited that matched the OA group for age, body mass index (BMI), and general activity level (Table 1). Subjects were excluded from both groups if they had coexisting problems that prevented them from lying flat, a requirement of testing; had neurologic disorders affecting their lower limbs; experienced any pain during testing; or suffered from attention disorders. The sample size of 10 subjects in each group provided sufficient power (0.94) to identify a difference in detection levels of 0.5° (SD 0.3°) between the groups (22).

Table 1. Demographic variables for experimental and control groups*
 Experimental groupControl group
Mean (SD)RangeMean (SD)Range
  • *

    There was no difference in age or body mass index (BMI) between the groups.

Age, years70.5 (4.0)65.0–76.069.5 (5.2)65.0–81.0
BMI, kg/m226.9 (3.0)22.4–31.027.0 (4.5)22.9–35.2

Subjects were naive to the experimental hypotheses and were not given any feedback about their performance during the testing procedure. Ethical approval for the study was obtained from the human ethics committee of each participating institution, and subjects gave informed consent prior to data collection.

Measurement of movement detection.

Experimental setup.

In each experimental condition, knee flexion and extension movements were imposed about the axis of rotation of the knee using a linear torque motor under position feedback and driven by a variable ramp generator. The test leg was supported on a carrier board below the knee joint and secured to the board with an inelastic strap (Figure 1). The axis of rotation of the apparatus was aligned with that of the knee. The hip of the test leg and the lower back were immobilized by securing them to a backboard, ensuring that test movements were confined to the knee. Subjects were unable to see the test leg or apparatus.

thumbnail image

Figure 1. Aerial view of apparatus and test position for determining detection level for knee movements. Subjects lay on their sides with the test leg uppermost and secured to the rotating platform distal to the test joint. The axis of rotation of platform and joint is marked (X). The limb segment proximal to the test joint was also secured. The knee was coupled to a carrier that was attached to a linear motor.

Download figure to PowerPoint


To determine movement detection levels, subjects were required to lie on their sides on a custom-built bed with the test knee uppermost. Subjects had bare feet and wore shorts to expose the knee. In the OA group, the knee most affected by OA was tested. In the control group, the test knee was randomly assigned. Subjects were positioned with the hip flexed to 40° from neutral and the knee flexed to 35° from the fully extended position. This knee position enabled movement in both flexion and extension directions. Movements were imposed about the knee into either flexion or extension after a random time interval. Each movement was held for 3 seconds before being reset to the initial angle, during which responses were accepted. Subjects were instructed to report the direction of any perceived movement when they could do so with certainty. To prevent fatigue, rests were allowed after every set of 20 movements.

Movement detection levels were determined at 3 velocities: 0.5°/second, 1.0°/second, and 2.5°/second, tested in pseudorandom order. The slowest velocity was never tested first because movements at the slowest velocity were difficult for subjects to detect, and because each imposed movement took more than 5 seconds, causing patients some confusion until familiar with the task. Therefore, subjects were familiar with the test procedure and apparatus before testing at 0.5°/second. The movement detection threshold was measured as the 70% detection level, determined at each velocity by imposing a random mix of 10 flexion and 10 extension movements of constant amplitude. The amplitude was adjusted until subjects correctly reported 7 of 10 movements in each direction. This test protocol has been widely used to test proprioception (23).

Application of bandage.

The 70% detection level was also determined for both subject groups at 0.5°/second with and without a 50-cm length of Tubigrip (a cotton elasticized bandage) applied to the test knee. The Tubigrip extended across the knee to cover part of the thigh above and the shank below. Care was taken to ensure that circulation was not affected by the bandage.

Relationship between detection levels, age, and WOMAC scores.

In addition to the main study, we determined whether there was a relationship between detection level, age, and WOMAC score. Relationships between age and joint position sense (8) and between disease severity and joint position sense (24) have previously been found. However, such relationships have not been investigated for movement detection. In addition, the range of age and disease severity studied has been broader than the sample recruited here. To determine whether the relationship is evident for movement detection, and within the narrow band of age and disease severity studied here, we analyzed the data for an association between these variables.

Data analysis.

Anthropometric data were examined using a Student's 2 sample t-test to ensure that the 2 groups did not differ in age or BMI (Table 1). Knee flexion and extension data were analyzed separately with a Student's paired t-test. Because there was no difference between flexion and extension, the scores were averaged for further analyses. A repeated measures 2-way analysis of variance (ANOVA) was used to compare the 70% detection level between the 2 groups and across the 3 velocities. Repeated measures 2-way ANOVA was also used to compare detection levels with and without the bandage. Linear regression was used to identify any relationships between detection levels and age, and between detection levels and WOMAC scores. The significance level was set at P < 0.05.


  1. Top of page
  2. Abstract

Movement detection.

At each of the 3 velocities tested, the detection level was significantly higher for the OA group than for the control group (F[1,54] = 14.01, P < 0.01) (Figure 2). In addition, at the fastest velocity, both groups of subjects were able to detect significantly smaller movements than at the slower velocities (F[2,57] = 55.81, P < 0.001), as previously found (13, 23). At the slowest test velocity, 0.5°/second, the OA group detected movements of 2.9 ± 0.8° (mean ± SD) and the control group detected movements of 2.4 ± 0.6°. At the fastest velocity, 2.5°/second, the OA group detected movements of 0.9 ± 0.5° and the control group detected movements of 0.4 ± 0.2°.

thumbnail image

Figure 2. Detection levels (mean ± SD) at the knee for subjects with moderate-severe osteoarthritis (OA) and healthy controls. Detection levels at the knee were significantly worse (P < 0.01, denoted by *) at all velocities for the OA group than for controls.

Download figure to PowerPoint

Application of bandage.

Subjects' ability to detect movements imposed at 0.5°/second did not alter significantly when the bandage was applied in either group (F[1,36] = 0.21, P > 0.05) (Figure 3). With the bandage applied, the 70% detection level for the OA group was 2.9 ± 1.0° and without the bandage it was 2.9 ± 0.8°. In the control group, movement detection levels were 2.1 ± 0.6° while wearing a bandage and 2.4 ± 0.6° without a bandage.

thumbnail image

Figure 3. Detection levels (mean ± SD) at the knee with and without bandage application in subjects with osteoarthritis and healthy controls. Application of a bandage did not significantly alter detection levels.

Download figure to PowerPoint

Relationship between WOMAC scores and detection levels.

Neither subjects' age (R2 = 0.20–0.32) nor WOMAC scores (R2 = 0.00–0.15) were found to be related to movement detection levels. Regression analysis revealed that age was not related to movement detection at any of the velocities tested in either group of subjects (P > 0.05). Similarly, no linear relationship existed between the WOMAC scores and movement detection levels in either group, at any test velocity (P > 0.05).


  1. Top of page
  2. Abstract

The present study found that moderate to severe OA impaired subjects' ability to perceive movements imposed at the knee compared with age-matched controls with no clinical evidence of knee OA. We also found that application of a bandage did not alter movement detection. Furthermore, within the restricted range of disease severity and age studied here, neither age nor severity measured by the WOMAC score correlated with ability to perceive knee movement.

Although joint position sense has consistently been shown to be impaired by OA (7, 8, 15), the present study is the first to investigate movement perception. Together, these findings suggest that OA causes a generalized proprioceptive deficit including both sensations, although a relationship between the position and movement sense should not be assumed. It is unclear whether this proprioceptive deficit is caused by destruction of joint structures and the associated receptors, loss of muscle function, or joint instability.

Because muscle function is more amenable to rehabilitation than articular structures, the mechanism by which proprioceptive acuity is altered is particularly relevant. It seems unlikely that loss of articular receptors, per se, is the problem, because acuity improves after joint replacement (15), i.e., after removal of joint receptors. In contrast, a number of findings suggest that OA impairs several aspects of muscle function (4), and an exercise program aimed at improving muscle function and joint stability (12) also improved proprioception. Thus, it seems more likely that either muscle function or joint stability, or both, are the critical factors.

As argued by Hurley et al (4), the explanation may not be simple, but rather may be a combination of factors. For example, destruction of articular structures by OA may produce abnormal afferent discharge (25, 26). Afferent fibers from articular mechanoreceptors also project onto gamma-motoneurones (27), which control muscle spindle sensitivity (28). If the abnormal information from the articular afferents inhibits the gamma-motoneurones, then sensitivity of muscle spindles would decrease, with a consequent reduction in proprioceptive acuity (27, 29). That is, articular damage may impair muscle function and, secondarily, proprioception in patients with OA.

Alternatively, destruction of the joint by OA with subsequent deformity may cause an altered line of action of the muscles that cross the knee, thereby changing the length–tension relationship. This would also alter the muscle spindles' sensitivity to change in muscle length, resulting in decreased proprioceptive acuity.

The lack of effect of bandage application on proprioception was unexpected. Work to date is unclear, with previous studies showing some improvement in position sense with application of a bandage to OA knees (8) and in electromyographic activity in anterior cruciate ligament-deficient knees (15). More recently, however, Birmingham et al (30) found that bandaging the knee of young healthy subjects using a neoprene sleeve had no effect on passive or active position matching, the only difference occurring during 1 active, nonaxial, loaded condition. The lack of improvement in movement detection found in the present study may be because either the appropriate receptors were not stimulated, or the additional information from cutaneous receptors did not provide coherent signals. Instead, the increased input may constitute “noise,” and therefore not be interpreted as movement signals by the central nervous system.

Unlike previous studies on joint position sense (7, 8, 24), we did not find a relationship between movement detection and either severity of OA measured by the WOMAC score or age. This may reflect the lack of variation in our subjects: all subjects had severe OA, whereas other studies included subclinical controls (7); and the age range included subjects within a decade (65–76 years), whereas differences were previously found between decades. If differences do exist within the moderate to severe range of disease severity, or within a decade in an older population, it is likely that the small subject numbers studied precluded significant findings.

In conclusion, subjects with moderate to severe OA of the knee had decreased ability to perceive movement compared with healthy controls matched for age, BMI, and activity levels. Because an impairment in joint position sense has also been found, it appears that patients with knee OA have a generalized proprioceptive deficit. However, application of a bandage did not change the ability to detect movements in the flexion-extension plane in either group.


  1. Top of page
  2. Abstract
  • 1
    Guccione AA, Felson DT, Anderson JJ, Anthony JM, Zhang Y, Wilson PW, et al. The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health 1994; 84: 3518.
  • 2
    Ettinger WH Jr, Afable RF. Physical disability from knee osteoarthritis: the role of exercise as an intervention. Med Sci Sports Exerc 1994; 26: 143540.
  • 3
    Martin DF. Pathomechanics of knee osteoarthritis. Med Sci Sports Exerc 1994; 26: 142934.
  • 4
    Hurley MV, Scott DL, Rees J, Newham DJ. Sensorimotor changes and functional performance in patients with knee osteoarthritis. Ann Rheum Dis 1997; 56: 6418.
  • 5
    Slemenda C, Brandt KD, Heilman DK, Mazzuca S, Braunstein EM, Katz BP, et al. Quadriceps weakness and osteoarthritis of the knee. Ann Intern Med 1997; 127: 97104.
  • 6
    McCloskey DI. Kinesthetic sensibility. Physiol Rev 1978; 58: 763820.
  • 7
    Pai YC, Rymer WZ, Chang RW, Sharma L. Effect of age and osteoarthritis on knee proprioception. Arthritis Rheum 1997; 40: 22605.
  • 8
    Barrack RL, Skinner HB, Cook SD, Haddad RJ Jr. Effect of articular disease and total knee arthroplasty on knee joint-position sense. J Neurophysiol 1983; 50: 6847.
  • 9
    Sharma L, Pai YC, Holtkamp K, Rymer WZ. Is knee joint proprioception worse in the arthritic knee versus the unaffected knee in unilateral knee osteoarthritis? Arthritis Rheum 1997; 40: 151825.
  • 10
    Garsden LR, Bullock-Saxton JE. Joint reposition sense in subjects with unilateral osteoarthritis of the knee. Clin Rehabil 1999; 13: 14855.
  • 11
    Gandevia SC. Kinaesthesia: roles for afferent signals and motor commands. In: RowellL, ShepherdJT, editors. Integration of motor, circulatory, respiratory and metabolic control during exercise. New York: Oxford Publisher Press; 1996. p. 12872.
  • 12
    Barrack RL, Skinner HB, Cook SD. Proprioception of the knee joint: paradoxical effect of training. Am J Phys Med 1984; 63: 17581.
  • 13
    Refshauge KM, Chan R, Taylor JL, McCloskey DI. Detection of movements imposed on human hip, knee, ankle and toe joints. J Physiol (Lond) 1995; 488: 23141.
  • 14
    Corrigan JP, Cashman WF, Brady MP. Proprioception in the cruciate deficient knee. J Bone Joint Surg Br 1992; 74: 24750.
  • 15
    Barrett DS, Cobb AG, Bentley G. Joint proprioception in normal, osteoarthritic and replaced knees. J Bone Joint Surg Br 1991; 73: 536.
  • 16
    Branch TP, Hunter R, Donath M. Dynamic EMG analysis of anterior cruciate deficient legs with and without bracing during cutting. Am J Sports Med 1989; 17: 3541.
  • 17
    Refshauge KM, Kilbreath SL, Raymond J. The effect of recurrent ankle inversion sprain and taping on proprioception at the ankle. Med Sci Sports Exerc 2000; 32: 105.
  • 18
    Kellgren JH, Lawrence JS. Radiological assessment of osteoarthritis. Ann Rheum Dis 1957; 16: 494501.
  • 19
    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: 103949.
  • 20
    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: 183340.
  • 21
    Bombardier C, Melfi CA, Paul J, Green R, Hawker G, Wright J, et al. Comparison of a generic and a disease specific measure of pain and physical function after knee replacement surgery. Med Care 1995; 33 Suppl 4: AS131–44.
  • 22
    Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale, New Jersey: Lawrence Erlbaum Associates; 1988.
  • 23
    Hall LA, McCloskey DI. Detections of movements imposed on finger, elbow and shoulder joints. J Physiol 1983; 335: 51933.
  • 24
    Marks R, Quinney HA, Wessel J. Proprioceptive sensibility in women with normal and osteoarthritic knee joints. Clin Rheumatol 1993; 12: 1705.
  • 25
    Hurley MV, Newham DJ. The influence of arthrogenous muscle inhibition on quadriceps rehabilitation of patients with early, unilateral osteoarthritic knees. Br J Rheumatol 1993; 32: 12731.
  • 26
    Schaible HG, Neugebauer V, Cervero F, Schmidt RF. Changes in tonic descending inhibition of spinal neurons with articular input during the development of acute arthritis in the cat. J Neurophysiol 1991; 66: 102132.
  • 27
    Ferrell WR, Baxendale RH, Carnachan C, Hart IK. The influence of joint afferent discharge on locomotion, proprioception and activity in conscious cats. Brain Res 1985; 347: 418.
  • 28
    Matthews PB. Proprioceptors and their contribution to somatosensory mapping: complex messages require complex processing. Can J Physiol Pharmacol 1988; 66: 4308.
  • 29
    Ferrell WR, Crighton A, Sturrock RD. Position sense at the proximal interphalangeal joint is distorted in patients with rheumatoid arthritis of finger joints. Exp Physiol 1992; 77: 67580.
  • 30
    Birmingham TB, Inglis JT, Kramer JF, Vandervoort AA. Effect of a neoprene sleeve on knee joint kinesthesis: influence of different testing procedures. Med Sci Sports Exerc 2000; 32: 3048.