- Email: [email protected]
Abnormal knee joint position sense in individuals with patellofemoral pain syndrome
ELSEVIER
Journal of Orthopaedic Research
Journal of Orthopaedic Research 20 (2002)208-214
www.elsevier.com/locate/orthres
Abnormal knee joint position sense in individuals with patellofemoral pain syndrome Vanessa Baker, Kim Bennell *, Barry Stillman, Sallie Cowan, Kay Crossley Centre ,for S)orts Medicine Research and Educution, School of' Phpsiutherapy. University
of Melbourne. Curlton, t k . 3010, Austrdiri
Received 17 January 2001; accepted 10 July 2001
Abstract
The purpose of this cross-sectional study was to compare knee joint position sense (JPS) in 20 individuals with and 20 without patellofemoral pain syndrome (PFPS). Five active tests with ipsilateral limb matching responses were performed at 20" and 60" flexion under non-weightbearing conditions, and at 40" flexion under uni- and bi-lateral weightbearing conditions. The response errors were calculated as the difference between each target and response position (accuracy) and the standard deviation of these differences (reliability). JPS was: (i) significantly less accurate and less consistent in the knees with PFPS during both the non-weightbearing and weightbearing tests when compared to the control subject knees; (ii) less accurate when the symptomatic and asymptomatic knees of the 13 uni-lateral PFPS subjects were compared and (iii) less accurate in the asymptomatic knees of the uni-lateral PFPS subjects and knees of the control subjects. The maximum intensity of pain experienced during each knee JPS test was not correlated to any of the JPS test results. The results confirm abnormal knee joint proprioception in individuals with PFPS. Although it cannot be determined whether the abnormality precedes or follows the development of PFPS, the results support including proprioceptive reeducation in management of PFPS. 0 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved.
Introduction
The ability to sense the position and movement of limb segments individually and relative to one another are highly specialised proprioceptive functions that are integral to motor learning and the on-going programming of complex movements. Proprioceptive afferents also contribute at the spinal level to arthrokinetic and muscular reflexes which play a large part in dynamic joint stability [ 13,181. Abnormal proprioception could predispose to musculoskeletal pathology by altering the control of movement leading to abnormal stresses on tissues. For example, in knee joint osteoarthritis sensorimotor dysfunction may cause greater impact of the leg at heel strike thereby initiating or perpetuating arthritic damage [32]. Alternatively, pathology and pain may alter proprioceptive information further compounding functional deficits. This may be the case with ankle lateral ligament sprains and anterior cruciate ligament *Corresponding author. Tel.: +61-3-8344-4171; fax: +61-3-83444188. E-muil uddrcss: [email protected] ( K . Bennell).
ruptures where injury damages mechanoreceptors in the ligaments and possibly contributes to the ensuing problem of functional instability [5,18]. Patellofemoral pain syndrome (PFPS) is a common, painful musculoskeletal condition that affects physically active young adults and adolescents [20,34]. PFPS is characterised by the presence of anterior knee pain, which is typically activity induced and aggravated by functions that increase patellofemoral compressive forces. The pain and abnormal tissue stresses arising from maltracking of the patella may lead to abnormal proprioception in patients with PFPS. While the aetiology of PFPS is not well understood and is likely to be multifactorial, abnormal lateral tracking of the patella within the femoral trochlear groove is thought to play a key role [7,26]. Tracking of the patella affects the magnitude and distribution of forces acting at the patellofemoral joint and ultimately patellofemoral joint pressures [ 101. Proprioceptive information from the active (muscular) and passive (osseous/ligamentous) systems contributes to the overall neuromuscular control of patellar tracking. Specifically, vastus medialis oblique (VMO) is believed to assist in
0736-0266/02/$- see front matter 8 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved. PII: SO7 3 6 - 0 2 6 6 ( 0 1 ) 0 0 1 0 6 - 1
I: Baker et al. I Journul
of' Orthopurdic
maintaining patella position by applying a medial force vector to counteract the lateral pull of the larger vastus lateralis (VL) [ 191. We have recently demonstrated that the onset of VMO activation relative to the VL is commonly delayed in individuals with PFPS during stair ascent and descent, in contrast to healthy controls where concurrent onset of the contraction of VMO and VL is the norm [ 3 ] . Such problems in quadriceps neuromuscular control could be initiated or augmented by abnormal proprioceptive feedback from the muscular and articular structures in and around the patellofemoral region. Only two studies have investigated proprioception in individuals with PFPS, with conflicting results [12,16]. Therefore, the aim of this cross-sectional study was to compare proprioception in PFPS and control subjects focussing on knee joint position sense (JPS). We hypothesised that subjects with PFPS would have abnormal knee JPS that may be demonstrated under both non-weightbearing and weightbearing active test conditions. Study of proprioception in PFPS patients may lead to an enhanced understanding of the mechanisms and the development of PFPS and assist with the design and implementation of appropriate prevention and treatment programmes.
Materials and methods Subjects
Twenty subjects with PFPS (15 females, five males) were recruited from the practices of medical and health professional colleagues. Additionally, advertisements were placed in gymnasiums, medical clinics and the print and radio media. The inclusion criteria were: (i) anterior or retropatellar knee pain present during at least two of the following ascending/descending stairs, hoppinghunning, squatting, kneeling and prolonged sitting; (ii) insidious onset of symptoms unrelated to a traumatic incident; (iii) presence of pain on step down from a 25 cm step or double leg squat and (iv) pain on palpation of patellar facets. Participants were excluded if they had: ( i ) symptoms present for less than one month; (ii) clinical evidence of other knee pathology (including co-existing pathology); (iii) knee surgery in the last three months; (iv) history of patellar subluxationldislocation; ( v ) current significant injury affecting other lower limb joints; (vi) current use of non-steroidal anti-inflammatory drugs or corticosteroids; and (vii) aged 2 4 0 years. Twenty control subjects were recruited from students at The University of Melbourne, and acquaintances of the investigators. Potential control subjects were excluded if there was a history or current symptoms of PFPS, or any of the above-listed exclusion criteria. As seen in Table 1, the groups were well matched on age, height, body mass and level of physical activity (as determined by each subject's recollection of work and leisure activities over the seven days before assessment "91). The right lower limb was dominant in 19 of the 20 individuals within each subject group. Twelve and seven of the subjects with PFPS had uni- and bi-lateral symptoms, respectively. For one subject the record of whether the PFPS was uni- or bi-laterdl was misplaced. The affected or mostaffected lower limb was dominant in six and four of the individuals with uni- and bi-lateral involvement, respectively. PFPS symptoms were present from 1-120 months with a mean (S.D.) of 20 months [29].
209
Research 20 (2002i ,708 21.1
Table I Characteristics of PFPS ( n = 20) and control ( n = 20) subjects -~
~
Characteristics
PFPS subjects mean (S D.)
Control Subjects mean (S.D.)
Age (years) Height ( m ) Body mass (kg) Physical activity (kcallday)
25.4 (8.5) 1.69 (0.07) 65.6 (13.8) 2964 (1011 )
25.5 (8.6) 1.68 (0.09) 65.7 (16.5) 3158 (1516)
Procedurrs
Approval from the University of Melbourne Human Ethics Committee was obtained before commencing this project, and all subjects provided informed written consent. Knee JPS was examined under non-weightbearing and uni- and bilateral weightbearing test conditions using validated and reliable protocols [33]. Four reflectivemarkers were fixed with double-sided adhesive tape to the skin of the lateral thigh and leg over the apex of the greater trochanter, iliotibial tract level with the posterior crease of the 80"-flexed knee, neck of the fibula and prominence of the lateral malleolus. These markers were subsequently used during computer digitisation of videotape records of all of the knee joint test and response positions. Non-iwighthearing JPS tests Each subject sat on a treatment couch with knees flexed and the (supported) trunk inclined backwards at an angle of 25" from the vertical (Fig. I(a)). With each subject's eyes closed the investigator lightly grasped the foot and passively extended the relaxed knee from the resting position (-80" flexion) to one of two test positions, approximately 20" and 60" flexion. The subject was instructed to hold the knee (isometrically) in the test position for -5 s whilst attempting to identify (sense) the knee position. The investigator then resupported the foot and returned the relaxed limb to the resting position. The subject was instructed to extend the knee to the perceived test position and to hold that position for -5 s. The two test positions were examined five times each in both knees of all subjects. However, in subjects with bi-lateral PFPS, measurements were only included from the most affected knee as determined by the subjects report of their most painful knee. The exact test and response positions were accurately measured from the videotape images upon completion of the assessments. The test position was defined as the knee flexion angle when the examiner initially positioned the leg while the response position was defined as the knee flexion angle when the subject returned the leg in an attempt to match the original test position. W7rigkrtheuring JPS testS For the uni- and bi-lateral weightbearing assessments, the subjects stood in bare feet with fingertip support for balance (Fig. l(b)).During the uni-lateral weightbearing assessments, the untested lower limb was flexed at the hip and knee so that it was not weightbearing. For the bilateral weightbearing assessments, the feet were shoulder-width apart. Subjects were requested to attempt to bear approximately equal body weight on both legs. During the uni- and bi-lateral weightbearing assessments, each subject with eyes closed was instructed to flex the knee(s) until told to stop when the investigator judged that the knee was in the test position (40" flexion): attempt to identify (sense) the position of the examined knee whilst holding the test position steadily for -5 s; return to the initial uni- or bi-lateral erect standing position before attempting to replicate the perceived test position. Each test and response was repeated five times. Only the affectcd or most affected knee of the PFPS subjects and a matching left or right knee of the control subjects was tested. Mrasurenzmt of km,r joint ungles The angle of the knee in each test and response position was determined using computer analysis of videotape images and the two-dimensional automatic digitising module of the Peak measurement system (Peak Motus [v 4.3.11, Peak Performance Technologies, Englewood, USA). A segment of videotape showing each position was
110
C'. Balcrr t't ul.
I Juurnul
of
Orthoparrlic Rr.seurcli 20 (2002J 208-214
60" flexion, and uni- and bi-lateral weightbearing at 40" flexion. The results were scored from 0 10 cm in increments of 1 cm. Subjects completed the VAS at the conclusion of each set of repetitions. Dutu untrljsis Three dependent variables were calculated for the non-weightbearing and weightbearing assessments at each target position:
(i) Relative error - this was defined as the arithmetic difference between the test and response positions. Relative errors represent accuracy with directional bias. If the response position was closer to the starting position than the test position, then this was called an underestimation error and given a negative sign. For example, in the non-weightbearing tests where the starting position was approximately 80" flexion, a response position of 30" flexion following a test of 10" flexion constituted an underestimation error designated -10". In the weightbearing tests where the starting position of the knee was approximately 0" flexion, a response of 10" flexion following a test of 30" flexion also constituted an underestimation error of -10". (ii) Absolute error - this was determined as the difference between the test and response position without reference to whether it constituted under- or overestimation. Accordingly it is represented by a number without any sign. Absolute errors represent accuracy without directional bias. (iii) Variable error -this was defined as the standard deviation from the mean or each set of five relative errors. Variable errors represent the ability of the subjects t o consistently sense each set of test positions. The obtained data were processed using the SPSS computer program (Norusis/SPSS,Chicago, IL, USA). After an initial examination for normality and homogeneity of variance, comparisons of JPS errors were made using factorial analysis of variance. Values o f p < 0.05 were regarded as statistically significant. Separate 7 x 2 A N O V Awere used to examine for differences in the relative, absolute and variable errors. JPS was compared between: ( i ) the affected or most-affected knees of the 20 PFPS subjects and 20 knees from the control subjects; (ii) the symptomatic and asymptomatic knec of the 11 subjects with uni-lateral PFPS: and (iii) the asymptomatic knee of the 12 uni-lateral PFPS subjects with one knee from each of the 70 control subjects. Since the results from the VAS-based pain assessmenls were not normally distributed. correlations of pain with the JPS assessment results were analysed using the Spearman rho coefficient.
Results
Non-bveightbearing test positions (b) Fig. 1. Video camera field of view and placement of reflective markers for JPS assessment. ( a ) Non-weightbearing: ( b ) uni-lateral weightbearing. automatically digitised for 0.31 s at a frequency of 50 Hz; that is 16 consecutive images. The obtained raw data representing the spatial location of the four reference markers were then filtered using a robust non-linear least-squares fourth-order (Butterworth) filter (Peak Performance Technologies, 1995). The knee angle was then calculated from the filtered data using standard trigonometric formulae, with each position being represented as the average of the 16 derived angular measurements. MLururmzeni < $ ' p i n Ten-cm horizontal visual analogue scales (VAS) were used to determine each subject's subjective estimate of worst pain experienced during each type of assessment; that is non-weightbearing at 20" and
Covripurison of afected or most-uflected knee of the PFPS subjects bvith o m knee from each of the control subjects The subjects with PFPS had significantly greater relative, absolute and variable errors than control subjects at both test angles (Table 2). There were also significantly greater relative and absolute errors at the 60" flexion target compared to 20" flexion. Cornpurison of' ,symptomatic and asjwiptouviatic knees of the uni-lateral PFPS subjects A comparison of the symptomatic and asymptomatic knees in the 13 uni-lateral PFPS subjects revealed significantly larger absolute errors in the symptomatic knees at both target angles but no difference in variable errors (Table 3). There were no significant differences between the three types of error at the two target angles.
Table 2 Comparison of non-weightbearing JPS assessment results between the affected or most-affected knees of the PFPS subjects ( n = 20) and one knee of each control subject ( n = 20) Target position
(")
Test type
PFPS subjects mean (S.D.) (") (min, max)
Control subjects mean (S.D.) (") (min, max)
Analysis of variance F (13). P
20
Relative Absolute Variable
0.9 (1.8) (-3.0, 3.4) 2.4 (1.1) (0.9, 4.6) 2.2 (1.3) (0.6, 4.9)
0.4 (1.0) (-1.8, 2.6) 1.3 (0.5) (0.6, 2.6) 1.2 (0.5) (0.3, 2.2)
4.05 <0.001 <0.01
60
Relative Absolute Variable
2.1 (2.0) (-0.9, 5.7) 2.9 (1.5) (1.0. 5.7) 2.1 (1.1) (0.6, 5.3)
0.9 (1.3) (- 1.6, 3.6) 0.6 (0.7) (-0.6, 3.6) 1.4 (0.5) (0.5, 2.7)
<0.05 <0.001 <0.01
Table 3 Comparison of non-weightbearing JPS assessment results between the symptomatic and asymptomatic knees of uni-lateral PFPS subjects ( n = 12) Target position
Test type
Uni-lateral PFPS, symptomatic knees mean (S.D.) (") (rnin. max)
Uni-lateral PFPS, asymptomatic knees mean (S.D.) (") (min, max)
Analysis of variance F (1,l I), p
20
Relative Absolute Variable
1.2 (1.5) (-1 .o, 3.4) 2.4 ( I . I ) (0.9, 4.6) 2.4 (1.3) ( I 2 , 4.9)
1.4 ( 1.2) (- 1.4, 2.7) 1.9 (0.6) (1.1, 3.1) 1.9 (0.7) (0.8, 2.9)
nsd <0.05 nsd
60
Relative Absolute Variable
2.1 (2.2) (-0.9, 5.7) 2.8 (1.6) ( I .O, 5.7) 2.0 (1.0) (0.6, 3.9)
1.3 (1.7) (-1.4, 4.0) 2.0 (1.1) (0.9, 4.0) 1.6 (0.4) (1.0, 2.4)
nsd <0.05 nsd
nsd
=
(")
No significant difference.
Comparison of asjwiptomutic knees of uni-lateral PFPS subjects with one knee from euch of rhe control subjects When the results from the asymptomatic knees of the 12 subjects with uni-lateral PFPS were compared to the results from the 20 control subjects, the asymptomatic PFPS knees demonstrated significantly larger absolute and relative errors at both target positions (Table 4). For both subject groups the relative and absolute errors were greatest at 60" knee flexion. Weightbearing assessnients Comparison of aflected or most-qffectrdkiiee of the PFPS subjects, with one knee from each of the control subjects Table 5 shows that there were no significant differences between the subject groups with respect to the
relative errors for both the uni- and bi-lateral weightbearing assessments. However, the PFPS subjects did produce significantly larger absolute and variable errors. Effect of knee pain on JPS assessment results in the PFPS subjects
The maximum levels of pain on a 10 cm VAS experienced by the PFPS subjects during the JPS assessments given as the median (interquartile range) were 2.5 (2.9) and 1.0 (3.8) for the non-weightbearing tests at 20" and 60" flexion respectively, 4.0 (5.0) and 2.5 (3.5) for the uni- and bi-lateral weightbearing tests, respectively. Calculation of the Spearman rho coefficients revealed no significant correlations between the knee JPS errors and levels of pain.
Table 4 Comparison of non-weightbearing JPS assessmet results between the asymptomatic knees of the uni-lateral PFPS subjects ( n = 12) and knees of the control subjects ( n = 20) Target position
(")
Test type
Uni-lateral PFPS, asymptoinatic knees mean (S.D.) (") (min, max)
Control subjects mean (S.D.) (") (min, max)
( 1,30), p
Analysis of variance F
20
Relative Absolute Variable
1.4 (1.2) (-1.4. 2.7) 1.9 (0.6) (1.1, 3.1) 1.9 (0.7) (0.8, 2.9)
0.4 (1.0) (-1.8, 2.6) 1.3 (0.5) (0.6, 2.6) 1.2 (0.5) (0.3, 2.2)
nsd <0.001 <0.05
60
Relative Absolute Variable
1.3 (1.7) (-1.4, 4.0) 2.0 (1.1) (0.9, 4.0) 1.6 (0.4) (1.0, 2.4)
0.9 (1.3) (-1.6, 3.6) 1.6 (1.7) (0.6, 3.6) 1.4 (0.5) (0.5, 2.7)
nsd
nsd
=
No significant difference.
I'. Bolirr er ul. I Journcll of' Orthopcrrdic Reswrch 20 (2002 1 208-214
212
Table 5 Comparison of uni- and bi-lateral weightbearing JPS assessment results between the affected or most-affected knees of the PFPS subjects (n = 20) and one knee of each control subject at 40" flexion ( n = 20) Weightbearing Uni-lateral
Hi-lateral
nsd
=
(")
PFPS subjects mean (S.D.) (min, max)
Control subjects mean (S.D.)(") (min, max)
Analysis of variance F (1,30), p
Relative Absolute Variable
0.9 (1.7) (-1.8, 3.5) 1.6 (0.8) (1.2, 3.9) 3.8 (1.1) (0.9,4.9)
0.4 (0.8) (-1.0. 1.8) 1.4 (0.4) (0.8, 2.3) 1.6 (0.6) (0.6, 2.9)
nsd
Relative Absolute Variable
0.8 (2.3) (-5.2, 5.3) 2.6 (1.2) (1.1. 5.3) 2.5 (1.4) (1.0. 7.1)
0.2 (1.0) (-2.2, 1.6) 1.5 (0.4) (0.9, 3.3) 1.6 (0.7) (0.3, 3.4)
nsd
Test type
<0.001
<0.001 <0.001 <0.001
No significant difference
Discussion The present study revealed that JPS was less accurate and less consistent in individuals with PFPS compared with controls. Abnormal results were also produced by the asymptomatic contralateral knee of the subjects with clinically uni-lateral PFPS. These findings support the presence of a proprioceptive deficit at the knee in those with PFPS. Abnormal proprioception in the PFPS group was found at both joint angles in the non-weightbearing test position suggesting that the results are not confined to a specific part in range. Non-weightbearing tests were included because although they are less functional than weightbearing tests they permit the knee joint to be assessed in isolation. However, abnormal proprioception was also found in the weightbearing tests which allow additional sources of proprioceptive information external to the tested knee [15]. In standing, proprioceptive information may be received through the compression of musculoskeletal structures in and about the knee and other joints throughout the weightbearing limb(s). Proprioceptive feedback may also be received from the compressed skin and subcutaneous structures of the sole of the weightbearing foot [ 141, through body-weight resisted muscle contractions and calf stretch [27], and through the supporting upper limbs and trunk [17]. As differences in JPS were found between PFPS and control groups in weightbearing, these results suggest that the abnormal knee proprioception in PFPS cannot be compensated by afferent feedback from outside the examined joint. It might be argued that because all tests in the present study were active, the JPS response positions, and associated errors, might have simply been the consequence of muscular weakness or attempts by the subjects with PFPS to respond in a manner that would minimise muscular effort and pain. However, if this was the case, it would be expected that the tendency would have been for the subjects to minimise muscle effort or discomfort by underestimating the test positions. In fact there was a greater tendency for the subjects to overestimate the non-weightbearing and weightbearing test positions.
Thus, we believe that results from the active tests were not compensatory, but rather were the consequence of valid proprioceptive deficits. We are aware of two previous studies of JPS in subjects with PFPS [12,16]. Jerosch et al. [12] using passive tests to study knee JPS in 43 subjects with uni-lateral PFPS and 30 control subjects, found significantly larger absolute errors in both the symptomatic and asymptomatic knees of the PFPS subjects; which is in accord with our non-weightbearing results. Conversely, Kramer et al. [16], using both active non-weightbearing tests and uni-lateral weightbearing tests, found no significant differences between the results from 24 subjects with bi-lateral PFPS, and 24 age- and gender-matched control subjects. Their protocol included one repetition at four separate test angles that were averaged to produce a single (absolute) error measurement. By measuring only once at each of a number of widely spaced test angles, the resulting average is likely to be highly variable thus potentially masking any abnormality when compared to controls. Furthermore, their non-significant results may also reflect failure to have considered relative and variable errors. In addition to the studies specifically investigating individuals with PFPS, there is one study comparing proprioception in nine individuals with recurrent subluxation/dislocation of the patella and 30 controls [I I]. The results demonstrated diminished knee JPS in both the affected and the contralateral unaffected knee compared with the controls. Since dislocation of the patella represents the ultimate example of abnormal patella tracking, it is interesting to note that the findings were similar to those of the present study. In light of our findings and those of previous studies it would appear that knee joint proprioception is abnormal in those with PFPS. PFPS could contribute to deficient knee joint proprioception by two main mechanisms; one involving abnormal tissue stresses and motor control and the other involving pain and inflammation. Abnormal tissue stresses may result from excessive lateral tracking of the patella giving rise to excessive strain on peripatellar retinacular supports
1.: Bnker et ul.
I Journnl of'Orthopnedic Rrsrardi 20 (2002i ,708-21.1
[6,30]. Alterations in motor output from the spinal cord and higher central nervous system centres may lead to changes in afferent input from active and passive systems, thus further compounding problems in the patellar tracking system. Of these propositions, the most important is likely to be altered muscle activity, as diverse research supports the dominant role that muscle receptors play in proprioception [S,21,2S]. Pain is a feature of PFPS and although our results showed no direct correlation between pain at the time of testing and errors in knee JPS, this does not necessarily preclude pain as a contributing factor to abnormal proprioception. The subjective method of pain assessment and the relative lack of variability in the pain data may account for our non-significant findings with respect to pain. There have been reports of correlations between pain and knee JPS in older individuals with osteoarthritic knees [23,33] although these did not assess pain at the time of proprioceptive testing. Other laboratory studies indicate that pain may play a role in abnormal proprioception but the mechanisms are highly complex [3].In simple terms, it is suggested that in the presence of pain and/or inflammation, a large proportion of nerve endings are sensitised by the chemical substances produced during the inflammatory or pain response. The abnormal discharge of the small diameter groups I11 and IV (pain) and large diameter group I1 (proprioceptive) afferents combine to produce abnormal sense of joint position and abnormal drive of muscle spindles in the affected region [3,9,13,311. Given the cross-sectional nature of our study, it cannot be determined whether the findings are a cause of PFPS or arise from the condition. It is also clear that PFPS and proprioceptive deficiencies may both sustain and perpetuate one another. As proprioceptive information received from the active and passive systems may contribute to the overall control of the musculature important in patellar tracking, it is possible that abnormal proprioceptive input could be responsible for the motor control problem that has been observed in patients with PFPS [3]. If the proprioceptive deficit was shown to precede the syndrome, then clinicians have a basis for developing preventative programs. Alternatively if proprioceptive deficits are a consequence of PFPS, this has implications for treatment programs. As expected, in individuals with uni-lateral PFPS, there was a significant difference in JPS between the two knees. However, it was also found that the asymptomatic knee of these individuals had worse JPS when compared with the control group participants. This agrees with the study by Jerosch et al. [12]. There are several possible mechanisms for these findings. First, the fundamental biomechanical abnormality of PFPS may commonly occur bi-laterally, although only manifesting itself uni-laterally for an indeterminate period of time. If this is the case, it might be concluded that the proprio-
'13
ceptive mechanisms at the knee appear to be susceptible to even clinically asymptomatic biomechanical changes. Second, PFPS is associated with changes in the spatiotemporal or kinematic gait variables [22,24,35]. While it is possible that these changes could lead to altered proprioceptive input in the contralateral knee no study has investigated gait kinematics changes in the contralateral knee of individuals with clinically uni-lateral PFPS symptoms. A third possible explanation is based on the fact that proprioceptive afferents associated with a proprioceptively abnormal (knee) joint have connections onto neurones in the contralateral spinal cord [ 1,4]. By this mechanism, it is possible to imagine a contralateral normal joint becoming proprioceptively abnormal. Obviously, further research will be needed to verify the above or any other possible mechanisms for proprioceptive abnormality in clinically normal contralatera1 joints associated with ipsilateral articular disorders. Based on our results, it would seem appropriate to include proprioceptive reeducation in the management of PFPS. However, it must be acknowledged that justification for proprioceptive reeducation depends on establishing the magnitude of proprioceptive abnormality required to exacerbate the pathogenesis of PFPS, and the magnitude of proprioceptive abnormality required to produce a clinically discernible disturbance of (knee) functional capacity. The difficulty of the experimentation required to resolve these questions may explain why very little is known about these matters. Suffice to say that it would be wise, and with negligible risk of harm, for clinicians and others to include proprioceptive reeducation until these matters are resolved.
Conclusions The present study has provided further evidence that proprioception is disturbed in individuals with PFPS, and in the contralateral asymptomatic knee of subjects with uni-lateral PFPS. It has also strengthened the growing conviction amongst sports practitioners and others that joint biomechanics, motor control, pain and proprioception need to be treated as co-dependents when considering the mechanisms, prevention, assessment and management of orthopaedic conditions.
Acknowledgements This study was funded by a grant from the Physiotherapy Research Foundation, Australia.
References [ I ] Burke D, Gandevia SC, Macefield G. Responses to passive movement of receptors in joint, skin and muscle of the human hand. J Physiol 1988;401:347-61.
211
I: Buker t't ul. I Journal of Orthopaedic Research 20 (2002i 108 3 1 4
[2] Capra N F , Ro JY. Experimental muscle pain produces central modulation of proprioceptive signals arising from jaw muscle spindles. Pain 2000;86:151-62. [3] Cowan S, Bennell K, Hodges P, Crossley K, McConnell J. Delayed onset of electromyographic activity of vastus lateralis compared to vastus medialis obliquus in subjects with patellofemoral pain syndrome. Arch Phys Med Rehabil 2001;82:18389. [4] Djupsjobacker M, Johansson H, Bergenlieini M, Wenngren BI. Influences on the gamma-muscle-spindle system from contralatera1 muscle afferents stimulated by KC1 and lactic acid. Neurosci Res 1995;21:301-9. [ S ] Fischer-Rasmussen T. Jensen PE. Proprioceptive sensitivity and performance in anterior cruciate ligament-deficient knee joints. Scand J Med Sci Sports 2000;10:85-9. [6] Fulkerson JP. The etiology of patellofemoral pain in young, active patients: a prospective study. Clin Orthop Re1 Res 1983:179: 129-33. [7] Fulkerson JP, Shea KP. Disorders of patellofemoral alignment. J Bone Jnt Surg [Am] 1990;72:142&9. [8] Gandevia SC, Hall LA, McCloskey DI. Potter EK. Proprioceptive sensation at the terminal joint of the middle finger. J Physiol 1983:335:507-17. [9] Hellstrom F, Thunberg J, Bergenheim M, Sjolander P, Pedersen J, et al. Elevated intramuscular concentration of bradykinin in jaw muscle increases the fusimotor drive to neck muscles in the cat. J Dental Res 200O;79:18 15-32. [lo] Hungerford DS, Barry M. Biomechanics of the patellofemoral joint. Clin Orthop Re1 Res 1979;144:9-15. [ I l l Jerosch J, Prymka M. Knee joint proprioception in patients with posttraumatic recurrent patella dislocation. Knee Surg, Sports Traumatol Arthrosc 1996;4:14-8. [I:] Jerosch J, Schmidt K, Prymka M. Proprioceptive capacities of patients with retropatellar knee pain with special reference to effectiveness of an elastic knee bandage. Unfallchirurgica 1997;100:719-23. [I 31 Johansson H, Sjnlander P, Sojka P. A sensory role for the cruciate ligaments. Clin Orthop Re1 Res 1991:268:161--78. [14] Kavounoudias A, Roll R, Roll JP. The plantar sole is a 'dynamometric map' for human balance control. Neuroreport 1998;9:3247-52. [IS] Kiefer G, Forwell L. Kramer J, Birmingham TL. Comparison of sitting and standing protocols for testing knee proprioception. Physioth Canada 1998:304. [I61 Kramer J, Handfield T, Kiefer G. Forwell L, Birmingham T. Comparisons of weight-bearing and non-weight-bearing tests of knee proprioception performed by patients with patello-femoral pain syndrome and asymptomatic individuals. Clin J Sport Med 1997;7:113 -8. [I71 Lackner J, DiZio P, Horak F, Krebs D, Rabin E. Precision contact of the fingertip reduces postural sway of individuals with bilateral vestibular loss. Exp Brain Res 1999;126:459-66.
[I81 Lephart SM, Pincivero DM, Giraldo L, Fu FH. The role of proprioception in the management and rehabilitation of athletic injuries. Am J Sports Med 1997;25:130-7. [I91 Lieb FJ, Perry J. Quadriceps function. An anatomical and mechanical study. J Bone Jnt Surg [Am] 1968;50:153548. [20] Milgrom C, Finestone A, Shlamkovitch N, Giladi M, Radin E. Anterior knee pain caused by overactivity. Clin Orthop Re1 Res 1996;331:25&60. [21] Mima T, Terada K, Maekawa M, Nagamine T, Ikeda A, et al. Somatosensory evoked potentials following proprioceptive stimulation of finger in man. Exp Brain Res 1996;111:23345. [22] Nadeau S, Gravel D, Hebert LJ, Arsenault AB, Lepage Y. Gait study of patients with patellofemoral pain syndrome. Gait Posture 1997;5:21-7. [23] Pai Y-C, Rymer WZ, Chang RW, Sharma L. Effect of age and osteoarthritis on knee proprioception. Arthrit Rheum 1997;40: 2260-5. "41 Powers CM, Perry J. Hsu A, Hislop HJ. Are patellofemoral pain and quadriceps femoris muscle torque associated with locomotor function. Phys Ther 1997;77:1063-75. [25] Powers CM, Heino JG, Rao S, Perry J. The influence of patellofemoral pain on lower limb loading during gait. Clin Biomech 1999;14:722 8. "61 Powers C. Patella kinematics, Part I: the influence of vastus medialis activity in subjects with and without patellofemoral pain. Phys Ther 2000;80:956-64. [27] Refshauge K, Chan R, Taylor J, McCloskey D. Detection of movements imposed on human hip, knee, ankle and toe joints. J Physiol 1995;488:231-41. "81 Roll J-P, Vedel J-P, Ribot-Ciscar E. Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study. Exp Brain Res 1989;76:213-21. [29] Sallis J, Haskell W, Fortmann S, Rogers T, Blair S, Pffenberger RS. Physical activity assessment methodology in the five-city project. Am J Epidemiol 1985;121:91-106. [30] Sanchis-Alfonso V, Rosello-Sastre E. Immunohistochemical analysis for neural markers of the lateral retinaculum in patients with isolated symptomatic patellofemoral malalignment. Am J Sports Med 2000;28:725-31. [31] Schaible H-G, Grubb BD. Afferent and spinal mechanisms ofjoint pain. Pain 1993;55:5-54. [32] Sharma L, Pai Y-C, Holtkamp K, Rymer W. Is knee joint proprioception worse in the arthritic knee versus the unaffected knee in unilateral knee osteoarthritis? Arthrit Rheum 1997;40:I51 8-25. [33] Stillman BC. An investigation of the clinical assessment of joint position sense. PhD Thesis, School of Physiotherapy, The University of Melbourne. http://adtl.lib.unimelb.edu.auladt-moot/ publidadt -VU200 1.OO 1Xndex.html, 2000. [34] Witvrouw E. Lysens R, Bellemans J, Peers K. Intrinsic risk factors for the development of anterior knee pain in an athletic population: a two year prospective study. Am J Sports Med 2000;28: 480-9.
Journal of Orthopaedic Research
Journal of Orthopaedic Research 20 (2002)208-214
www.elsevier.com/locate/orthres
Abnormal knee joint position sense in individuals with patellofemoral pain syndrome Vanessa Baker, Kim Bennell *, Barry Stillman, Sallie Cowan, Kay Crossley Centre ,for S)orts Medicine Research and Educution, School of' Phpsiutherapy. University
of Melbourne. Curlton, t k . 3010, Austrdiri
Received 17 January 2001; accepted 10 July 2001
Abstract
The purpose of this cross-sectional study was to compare knee joint position sense (JPS) in 20 individuals with and 20 without patellofemoral pain syndrome (PFPS). Five active tests with ipsilateral limb matching responses were performed at 20" and 60" flexion under non-weightbearing conditions, and at 40" flexion under uni- and bi-lateral weightbearing conditions. The response errors were calculated as the difference between each target and response position (accuracy) and the standard deviation of these differences (reliability). JPS was: (i) significantly less accurate and less consistent in the knees with PFPS during both the non-weightbearing and weightbearing tests when compared to the control subject knees; (ii) less accurate when the symptomatic and asymptomatic knees of the 13 uni-lateral PFPS subjects were compared and (iii) less accurate in the asymptomatic knees of the uni-lateral PFPS subjects and knees of the control subjects. The maximum intensity of pain experienced during each knee JPS test was not correlated to any of the JPS test results. The results confirm abnormal knee joint proprioception in individuals with PFPS. Although it cannot be determined whether the abnormality precedes or follows the development of PFPS, the results support including proprioceptive reeducation in management of PFPS. 0 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved.
Introduction
The ability to sense the position and movement of limb segments individually and relative to one another are highly specialised proprioceptive functions that are integral to motor learning and the on-going programming of complex movements. Proprioceptive afferents also contribute at the spinal level to arthrokinetic and muscular reflexes which play a large part in dynamic joint stability [ 13,181. Abnormal proprioception could predispose to musculoskeletal pathology by altering the control of movement leading to abnormal stresses on tissues. For example, in knee joint osteoarthritis sensorimotor dysfunction may cause greater impact of the leg at heel strike thereby initiating or perpetuating arthritic damage [32]. Alternatively, pathology and pain may alter proprioceptive information further compounding functional deficits. This may be the case with ankle lateral ligament sprains and anterior cruciate ligament *Corresponding author. Tel.: +61-3-8344-4171; fax: +61-3-83444188. E-muil uddrcss: [email protected] ( K . Bennell).
ruptures where injury damages mechanoreceptors in the ligaments and possibly contributes to the ensuing problem of functional instability [5,18]. Patellofemoral pain syndrome (PFPS) is a common, painful musculoskeletal condition that affects physically active young adults and adolescents [20,34]. PFPS is characterised by the presence of anterior knee pain, which is typically activity induced and aggravated by functions that increase patellofemoral compressive forces. The pain and abnormal tissue stresses arising from maltracking of the patella may lead to abnormal proprioception in patients with PFPS. While the aetiology of PFPS is not well understood and is likely to be multifactorial, abnormal lateral tracking of the patella within the femoral trochlear groove is thought to play a key role [7,26]. Tracking of the patella affects the magnitude and distribution of forces acting at the patellofemoral joint and ultimately patellofemoral joint pressures [ 101. Proprioceptive information from the active (muscular) and passive (osseous/ligamentous) systems contributes to the overall neuromuscular control of patellar tracking. Specifically, vastus medialis oblique (VMO) is believed to assist in
0736-0266/02/$- see front matter 8 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved. PII: SO7 3 6 - 0 2 6 6 ( 0 1 ) 0 0 1 0 6 - 1
I: Baker et al. I Journul
of' Orthopurdic
maintaining patella position by applying a medial force vector to counteract the lateral pull of the larger vastus lateralis (VL) [ 191. We have recently demonstrated that the onset of VMO activation relative to the VL is commonly delayed in individuals with PFPS during stair ascent and descent, in contrast to healthy controls where concurrent onset of the contraction of VMO and VL is the norm [ 3 ] . Such problems in quadriceps neuromuscular control could be initiated or augmented by abnormal proprioceptive feedback from the muscular and articular structures in and around the patellofemoral region. Only two studies have investigated proprioception in individuals with PFPS, with conflicting results [12,16]. Therefore, the aim of this cross-sectional study was to compare proprioception in PFPS and control subjects focussing on knee joint position sense (JPS). We hypothesised that subjects with PFPS would have abnormal knee JPS that may be demonstrated under both non-weightbearing and weightbearing active test conditions. Study of proprioception in PFPS patients may lead to an enhanced understanding of the mechanisms and the development of PFPS and assist with the design and implementation of appropriate prevention and treatment programmes.
Materials and methods Subjects
Twenty subjects with PFPS (15 females, five males) were recruited from the practices of medical and health professional colleagues. Additionally, advertisements were placed in gymnasiums, medical clinics and the print and radio media. The inclusion criteria were: (i) anterior or retropatellar knee pain present during at least two of the following ascending/descending stairs, hoppinghunning, squatting, kneeling and prolonged sitting; (ii) insidious onset of symptoms unrelated to a traumatic incident; (iii) presence of pain on step down from a 25 cm step or double leg squat and (iv) pain on palpation of patellar facets. Participants were excluded if they had: ( i ) symptoms present for less than one month; (ii) clinical evidence of other knee pathology (including co-existing pathology); (iii) knee surgery in the last three months; (iv) history of patellar subluxationldislocation; ( v ) current significant injury affecting other lower limb joints; (vi) current use of non-steroidal anti-inflammatory drugs or corticosteroids; and (vii) aged 2 4 0 years. Twenty control subjects were recruited from students at The University of Melbourne, and acquaintances of the investigators. Potential control subjects were excluded if there was a history or current symptoms of PFPS, or any of the above-listed exclusion criteria. As seen in Table 1, the groups were well matched on age, height, body mass and level of physical activity (as determined by each subject's recollection of work and leisure activities over the seven days before assessment "91). The right lower limb was dominant in 19 of the 20 individuals within each subject group. Twelve and seven of the subjects with PFPS had uni- and bi-lateral symptoms, respectively. For one subject the record of whether the PFPS was uni- or bi-laterdl was misplaced. The affected or mostaffected lower limb was dominant in six and four of the individuals with uni- and bi-lateral involvement, respectively. PFPS symptoms were present from 1-120 months with a mean (S.D.) of 20 months [29].
209
Research 20 (2002i ,708 21.1
Table I Characteristics of PFPS ( n = 20) and control ( n = 20) subjects -~
~
Characteristics
PFPS subjects mean (S D.)
Control Subjects mean (S.D.)
Age (years) Height ( m ) Body mass (kg) Physical activity (kcallday)
25.4 (8.5) 1.69 (0.07) 65.6 (13.8) 2964 (1011 )
25.5 (8.6) 1.68 (0.09) 65.7 (16.5) 3158 (1516)
Procedurrs
Approval from the University of Melbourne Human Ethics Committee was obtained before commencing this project, and all subjects provided informed written consent. Knee JPS was examined under non-weightbearing and uni- and bilateral weightbearing test conditions using validated and reliable protocols [33]. Four reflectivemarkers were fixed with double-sided adhesive tape to the skin of the lateral thigh and leg over the apex of the greater trochanter, iliotibial tract level with the posterior crease of the 80"-flexed knee, neck of the fibula and prominence of the lateral malleolus. These markers were subsequently used during computer digitisation of videotape records of all of the knee joint test and response positions. Non-iwighthearing JPS tests Each subject sat on a treatment couch with knees flexed and the (supported) trunk inclined backwards at an angle of 25" from the vertical (Fig. I(a)). With each subject's eyes closed the investigator lightly grasped the foot and passively extended the relaxed knee from the resting position (-80" flexion) to one of two test positions, approximately 20" and 60" flexion. The subject was instructed to hold the knee (isometrically) in the test position for -5 s whilst attempting to identify (sense) the knee position. The investigator then resupported the foot and returned the relaxed limb to the resting position. The subject was instructed to extend the knee to the perceived test position and to hold that position for -5 s. The two test positions were examined five times each in both knees of all subjects. However, in subjects with bi-lateral PFPS, measurements were only included from the most affected knee as determined by the subjects report of their most painful knee. The exact test and response positions were accurately measured from the videotape images upon completion of the assessments. The test position was defined as the knee flexion angle when the examiner initially positioned the leg while the response position was defined as the knee flexion angle when the subject returned the leg in an attempt to match the original test position. W7rigkrtheuring JPS testS For the uni- and bi-lateral weightbearing assessments, the subjects stood in bare feet with fingertip support for balance (Fig. l(b)).During the uni-lateral weightbearing assessments, the untested lower limb was flexed at the hip and knee so that it was not weightbearing. For the bilateral weightbearing assessments, the feet were shoulder-width apart. Subjects were requested to attempt to bear approximately equal body weight on both legs. During the uni- and bi-lateral weightbearing assessments, each subject with eyes closed was instructed to flex the knee(s) until told to stop when the investigator judged that the knee was in the test position (40" flexion): attempt to identify (sense) the position of the examined knee whilst holding the test position steadily for -5 s; return to the initial uni- or bi-lateral erect standing position before attempting to replicate the perceived test position. Each test and response was repeated five times. Only the affectcd or most affected knee of the PFPS subjects and a matching left or right knee of the control subjects was tested. Mrasurenzmt of km,r joint ungles The angle of the knee in each test and response position was determined using computer analysis of videotape images and the two-dimensional automatic digitising module of the Peak measurement system (Peak Motus [v 4.3.11, Peak Performance Technologies, Englewood, USA). A segment of videotape showing each position was
110
C'. Balcrr t't ul.
I Juurnul
of
Orthoparrlic Rr.seurcli 20 (2002J 208-214
60" flexion, and uni- and bi-lateral weightbearing at 40" flexion. The results were scored from 0 10 cm in increments of 1 cm. Subjects completed the VAS at the conclusion of each set of repetitions. Dutu untrljsis Three dependent variables were calculated for the non-weightbearing and weightbearing assessments at each target position:
(i) Relative error - this was defined as the arithmetic difference between the test and response positions. Relative errors represent accuracy with directional bias. If the response position was closer to the starting position than the test position, then this was called an underestimation error and given a negative sign. For example, in the non-weightbearing tests where the starting position was approximately 80" flexion, a response position of 30" flexion following a test of 10" flexion constituted an underestimation error designated -10". In the weightbearing tests where the starting position of the knee was approximately 0" flexion, a response of 10" flexion following a test of 30" flexion also constituted an underestimation error of -10". (ii) Absolute error - this was determined as the difference between the test and response position without reference to whether it constituted under- or overestimation. Accordingly it is represented by a number without any sign. Absolute errors represent accuracy without directional bias. (iii) Variable error -this was defined as the standard deviation from the mean or each set of five relative errors. Variable errors represent the ability of the subjects t o consistently sense each set of test positions. The obtained data were processed using the SPSS computer program (Norusis/SPSS,Chicago, IL, USA). After an initial examination for normality and homogeneity of variance, comparisons of JPS errors were made using factorial analysis of variance. Values o f p < 0.05 were regarded as statistically significant. Separate 7 x 2 A N O V Awere used to examine for differences in the relative, absolute and variable errors. JPS was compared between: ( i ) the affected or most-affected knees of the 20 PFPS subjects and 20 knees from the control subjects; (ii) the symptomatic and asymptomatic knec of the 11 subjects with uni-lateral PFPS: and (iii) the asymptomatic knee of the 12 uni-lateral PFPS subjects with one knee from each of the 70 control subjects. Since the results from the VAS-based pain assessmenls were not normally distributed. correlations of pain with the JPS assessment results were analysed using the Spearman rho coefficient.
Results
Non-bveightbearing test positions (b) Fig. 1. Video camera field of view and placement of reflective markers for JPS assessment. ( a ) Non-weightbearing: ( b ) uni-lateral weightbearing. automatically digitised for 0.31 s at a frequency of 50 Hz; that is 16 consecutive images. The obtained raw data representing the spatial location of the four reference markers were then filtered using a robust non-linear least-squares fourth-order (Butterworth) filter (Peak Performance Technologies, 1995). The knee angle was then calculated from the filtered data using standard trigonometric formulae, with each position being represented as the average of the 16 derived angular measurements. MLururmzeni < $ ' p i n Ten-cm horizontal visual analogue scales (VAS) were used to determine each subject's subjective estimate of worst pain experienced during each type of assessment; that is non-weightbearing at 20" and
Covripurison of afected or most-uflected knee of the PFPS subjects bvith o m knee from each of the control subjects The subjects with PFPS had significantly greater relative, absolute and variable errors than control subjects at both test angles (Table 2). There were also significantly greater relative and absolute errors at the 60" flexion target compared to 20" flexion. Cornpurison of' ,symptomatic and asjwiptouviatic knees of the uni-lateral PFPS subjects A comparison of the symptomatic and asymptomatic knees in the 13 uni-lateral PFPS subjects revealed significantly larger absolute errors in the symptomatic knees at both target angles but no difference in variable errors (Table 3). There were no significant differences between the three types of error at the two target angles.
Table 2 Comparison of non-weightbearing JPS assessment results between the affected or most-affected knees of the PFPS subjects ( n = 20) and one knee of each control subject ( n = 20) Target position
(")
Test type
PFPS subjects mean (S.D.) (") (min, max)
Control subjects mean (S.D.) (") (min, max)
Analysis of variance F (13). P
20
Relative Absolute Variable
0.9 (1.8) (-3.0, 3.4) 2.4 (1.1) (0.9, 4.6) 2.2 (1.3) (0.6, 4.9)
0.4 (1.0) (-1.8, 2.6) 1.3 (0.5) (0.6, 2.6) 1.2 (0.5) (0.3, 2.2)
4.05 <0.001 <0.01
60
Relative Absolute Variable
2.1 (2.0) (-0.9, 5.7) 2.9 (1.5) (1.0. 5.7) 2.1 (1.1) (0.6, 5.3)
0.9 (1.3) (- 1.6, 3.6) 0.6 (0.7) (-0.6, 3.6) 1.4 (0.5) (0.5, 2.7)
<0.05 <0.001 <0.01
Table 3 Comparison of non-weightbearing JPS assessment results between the symptomatic and asymptomatic knees of uni-lateral PFPS subjects ( n = 12) Target position
Test type
Uni-lateral PFPS, symptomatic knees mean (S.D.) (") (rnin. max)
Uni-lateral PFPS, asymptomatic knees mean (S.D.) (") (min, max)
Analysis of variance F (1,l I), p
20
Relative Absolute Variable
1.2 (1.5) (-1 .o, 3.4) 2.4 ( I . I ) (0.9, 4.6) 2.4 (1.3) ( I 2 , 4.9)
1.4 ( 1.2) (- 1.4, 2.7) 1.9 (0.6) (1.1, 3.1) 1.9 (0.7) (0.8, 2.9)
nsd <0.05 nsd
60
Relative Absolute Variable
2.1 (2.2) (-0.9, 5.7) 2.8 (1.6) ( I .O, 5.7) 2.0 (1.0) (0.6, 3.9)
1.3 (1.7) (-1.4, 4.0) 2.0 (1.1) (0.9, 4.0) 1.6 (0.4) (1.0, 2.4)
nsd <0.05 nsd
nsd
=
(")
No significant difference.
Comparison of asjwiptomutic knees of uni-lateral PFPS subjects with one knee from euch of rhe control subjects When the results from the asymptomatic knees of the 12 subjects with uni-lateral PFPS were compared to the results from the 20 control subjects, the asymptomatic PFPS knees demonstrated significantly larger absolute and relative errors at both target positions (Table 4). For both subject groups the relative and absolute errors were greatest at 60" knee flexion. Weightbearing assessnients Comparison of aflected or most-qffectrdkiiee of the PFPS subjects, with one knee from each of the control subjects Table 5 shows that there were no significant differences between the subject groups with respect to the
relative errors for both the uni- and bi-lateral weightbearing assessments. However, the PFPS subjects did produce significantly larger absolute and variable errors. Effect of knee pain on JPS assessment results in the PFPS subjects
The maximum levels of pain on a 10 cm VAS experienced by the PFPS subjects during the JPS assessments given as the median (interquartile range) were 2.5 (2.9) and 1.0 (3.8) for the non-weightbearing tests at 20" and 60" flexion respectively, 4.0 (5.0) and 2.5 (3.5) for the uni- and bi-lateral weightbearing tests, respectively. Calculation of the Spearman rho coefficients revealed no significant correlations between the knee JPS errors and levels of pain.
Table 4 Comparison of non-weightbearing JPS assessmet results between the asymptomatic knees of the uni-lateral PFPS subjects ( n = 12) and knees of the control subjects ( n = 20) Target position
(")
Test type
Uni-lateral PFPS, asymptoinatic knees mean (S.D.) (") (min, max)
Control subjects mean (S.D.) (") (min, max)
( 1,30), p
Analysis of variance F
20
Relative Absolute Variable
1.4 (1.2) (-1.4. 2.7) 1.9 (0.6) (1.1, 3.1) 1.9 (0.7) (0.8, 2.9)
0.4 (1.0) (-1.8, 2.6) 1.3 (0.5) (0.6, 2.6) 1.2 (0.5) (0.3, 2.2)
nsd <0.001 <0.05
60
Relative Absolute Variable
1.3 (1.7) (-1.4, 4.0) 2.0 (1.1) (0.9, 4.0) 1.6 (0.4) (1.0, 2.4)
0.9 (1.3) (-1.6, 3.6) 1.6 (1.7) (0.6, 3.6) 1.4 (0.5) (0.5, 2.7)
nsd
nsd
=
No significant difference.
I'. Bolirr er ul. I Journcll of' Orthopcrrdic Reswrch 20 (2002 1 208-214
212
Table 5 Comparison of uni- and bi-lateral weightbearing JPS assessment results between the affected or most-affected knees of the PFPS subjects (n = 20) and one knee of each control subject at 40" flexion ( n = 20) Weightbearing Uni-lateral
Hi-lateral
nsd
=
(")
PFPS subjects mean (S.D.) (min, max)
Control subjects mean (S.D.)(") (min, max)
Analysis of variance F (1,30), p
Relative Absolute Variable
0.9 (1.7) (-1.8, 3.5) 1.6 (0.8) (1.2, 3.9) 3.8 (1.1) (0.9,4.9)
0.4 (0.8) (-1.0. 1.8) 1.4 (0.4) (0.8, 2.3) 1.6 (0.6) (0.6, 2.9)
nsd
Relative Absolute Variable
0.8 (2.3) (-5.2, 5.3) 2.6 (1.2) (1.1. 5.3) 2.5 (1.4) (1.0. 7.1)
0.2 (1.0) (-2.2, 1.6) 1.5 (0.4) (0.9, 3.3) 1.6 (0.7) (0.3, 3.4)
nsd
Test type
<0.001
<0.001 <0.001 <0.001
No significant difference
Discussion The present study revealed that JPS was less accurate and less consistent in individuals with PFPS compared with controls. Abnormal results were also produced by the asymptomatic contralateral knee of the subjects with clinically uni-lateral PFPS. These findings support the presence of a proprioceptive deficit at the knee in those with PFPS. Abnormal proprioception in the PFPS group was found at both joint angles in the non-weightbearing test position suggesting that the results are not confined to a specific part in range. Non-weightbearing tests were included because although they are less functional than weightbearing tests they permit the knee joint to be assessed in isolation. However, abnormal proprioception was also found in the weightbearing tests which allow additional sources of proprioceptive information external to the tested knee [15]. In standing, proprioceptive information may be received through the compression of musculoskeletal structures in and about the knee and other joints throughout the weightbearing limb(s). Proprioceptive feedback may also be received from the compressed skin and subcutaneous structures of the sole of the weightbearing foot [ 141, through body-weight resisted muscle contractions and calf stretch [27], and through the supporting upper limbs and trunk [17]. As differences in JPS were found between PFPS and control groups in weightbearing, these results suggest that the abnormal knee proprioception in PFPS cannot be compensated by afferent feedback from outside the examined joint. It might be argued that because all tests in the present study were active, the JPS response positions, and associated errors, might have simply been the consequence of muscular weakness or attempts by the subjects with PFPS to respond in a manner that would minimise muscular effort and pain. However, if this was the case, it would be expected that the tendency would have been for the subjects to minimise muscle effort or discomfort by underestimating the test positions. In fact there was a greater tendency for the subjects to overestimate the non-weightbearing and weightbearing test positions.
Thus, we believe that results from the active tests were not compensatory, but rather were the consequence of valid proprioceptive deficits. We are aware of two previous studies of JPS in subjects with PFPS [12,16]. Jerosch et al. [12] using passive tests to study knee JPS in 43 subjects with uni-lateral PFPS and 30 control subjects, found significantly larger absolute errors in both the symptomatic and asymptomatic knees of the PFPS subjects; which is in accord with our non-weightbearing results. Conversely, Kramer et al. [16], using both active non-weightbearing tests and uni-lateral weightbearing tests, found no significant differences between the results from 24 subjects with bi-lateral PFPS, and 24 age- and gender-matched control subjects. Their protocol included one repetition at four separate test angles that were averaged to produce a single (absolute) error measurement. By measuring only once at each of a number of widely spaced test angles, the resulting average is likely to be highly variable thus potentially masking any abnormality when compared to controls. Furthermore, their non-significant results may also reflect failure to have considered relative and variable errors. In addition to the studies specifically investigating individuals with PFPS, there is one study comparing proprioception in nine individuals with recurrent subluxation/dislocation of the patella and 30 controls [I I]. The results demonstrated diminished knee JPS in both the affected and the contralateral unaffected knee compared with the controls. Since dislocation of the patella represents the ultimate example of abnormal patella tracking, it is interesting to note that the findings were similar to those of the present study. In light of our findings and those of previous studies it would appear that knee joint proprioception is abnormal in those with PFPS. PFPS could contribute to deficient knee joint proprioception by two main mechanisms; one involving abnormal tissue stresses and motor control and the other involving pain and inflammation. Abnormal tissue stresses may result from excessive lateral tracking of the patella giving rise to excessive strain on peripatellar retinacular supports
1.: Bnker et ul.
I Journnl of'Orthopnedic Rrsrardi 20 (2002i ,708-21.1
[6,30]. Alterations in motor output from the spinal cord and higher central nervous system centres may lead to changes in afferent input from active and passive systems, thus further compounding problems in the patellar tracking system. Of these propositions, the most important is likely to be altered muscle activity, as diverse research supports the dominant role that muscle receptors play in proprioception [S,21,2S]. Pain is a feature of PFPS and although our results showed no direct correlation between pain at the time of testing and errors in knee JPS, this does not necessarily preclude pain as a contributing factor to abnormal proprioception. The subjective method of pain assessment and the relative lack of variability in the pain data may account for our non-significant findings with respect to pain. There have been reports of correlations between pain and knee JPS in older individuals with osteoarthritic knees [23,33] although these did not assess pain at the time of proprioceptive testing. Other laboratory studies indicate that pain may play a role in abnormal proprioception but the mechanisms are highly complex [3].In simple terms, it is suggested that in the presence of pain and/or inflammation, a large proportion of nerve endings are sensitised by the chemical substances produced during the inflammatory or pain response. The abnormal discharge of the small diameter groups I11 and IV (pain) and large diameter group I1 (proprioceptive) afferents combine to produce abnormal sense of joint position and abnormal drive of muscle spindles in the affected region [3,9,13,311. Given the cross-sectional nature of our study, it cannot be determined whether the findings are a cause of PFPS or arise from the condition. It is also clear that PFPS and proprioceptive deficiencies may both sustain and perpetuate one another. As proprioceptive information received from the active and passive systems may contribute to the overall control of the musculature important in patellar tracking, it is possible that abnormal proprioceptive input could be responsible for the motor control problem that has been observed in patients with PFPS [3]. If the proprioceptive deficit was shown to precede the syndrome, then clinicians have a basis for developing preventative programs. Alternatively if proprioceptive deficits are a consequence of PFPS, this has implications for treatment programs. As expected, in individuals with uni-lateral PFPS, there was a significant difference in JPS between the two knees. However, it was also found that the asymptomatic knee of these individuals had worse JPS when compared with the control group participants. This agrees with the study by Jerosch et al. [12]. There are several possible mechanisms for these findings. First, the fundamental biomechanical abnormality of PFPS may commonly occur bi-laterally, although only manifesting itself uni-laterally for an indeterminate period of time. If this is the case, it might be concluded that the proprio-
'13
ceptive mechanisms at the knee appear to be susceptible to even clinically asymptomatic biomechanical changes. Second, PFPS is associated with changes in the spatiotemporal or kinematic gait variables [22,24,35]. While it is possible that these changes could lead to altered proprioceptive input in the contralateral knee no study has investigated gait kinematics changes in the contralateral knee of individuals with clinically uni-lateral PFPS symptoms. A third possible explanation is based on the fact that proprioceptive afferents associated with a proprioceptively abnormal (knee) joint have connections onto neurones in the contralateral spinal cord [ 1,4]. By this mechanism, it is possible to imagine a contralateral normal joint becoming proprioceptively abnormal. Obviously, further research will be needed to verify the above or any other possible mechanisms for proprioceptive abnormality in clinically normal contralatera1 joints associated with ipsilateral articular disorders. Based on our results, it would seem appropriate to include proprioceptive reeducation in the management of PFPS. However, it must be acknowledged that justification for proprioceptive reeducation depends on establishing the magnitude of proprioceptive abnormality required to exacerbate the pathogenesis of PFPS, and the magnitude of proprioceptive abnormality required to produce a clinically discernible disturbance of (knee) functional capacity. The difficulty of the experimentation required to resolve these questions may explain why very little is known about these matters. Suffice to say that it would be wise, and with negligible risk of harm, for clinicians and others to include proprioceptive reeducation until these matters are resolved.
Conclusions The present study has provided further evidence that proprioception is disturbed in individuals with PFPS, and in the contralateral asymptomatic knee of subjects with uni-lateral PFPS. It has also strengthened the growing conviction amongst sports practitioners and others that joint biomechanics, motor control, pain and proprioception need to be treated as co-dependents when considering the mechanisms, prevention, assessment and management of orthopaedic conditions.
Acknowledgements This study was funded by a grant from the Physiotherapy Research Foundation, Australia.
References [ I ] Burke D, Gandevia SC, Macefield G. Responses to passive movement of receptors in joint, skin and muscle of the human hand. J Physiol 1988;401:347-61.
211
I: Buker t't ul. I Journal of Orthopaedic Research 20 (2002i 108 3 1 4
[2] Capra N F , Ro JY. Experimental muscle pain produces central modulation of proprioceptive signals arising from jaw muscle spindles. Pain 2000;86:151-62. [3] Cowan S, Bennell K, Hodges P, Crossley K, McConnell J. Delayed onset of electromyographic activity of vastus lateralis compared to vastus medialis obliquus in subjects with patellofemoral pain syndrome. Arch Phys Med Rehabil 2001;82:18389. [4] Djupsjobacker M, Johansson H, Bergenlieini M, Wenngren BI. Influences on the gamma-muscle-spindle system from contralatera1 muscle afferents stimulated by KC1 and lactic acid. Neurosci Res 1995;21:301-9. [ S ] Fischer-Rasmussen T. Jensen PE. Proprioceptive sensitivity and performance in anterior cruciate ligament-deficient knee joints. Scand J Med Sci Sports 2000;10:85-9. [6] Fulkerson JP. The etiology of patellofemoral pain in young, active patients: a prospective study. Clin Orthop Re1 Res 1983:179: 129-33. [7] Fulkerson JP, Shea KP. Disorders of patellofemoral alignment. J Bone Jnt Surg [Am] 1990;72:142&9. [8] Gandevia SC, Hall LA, McCloskey DI. Potter EK. Proprioceptive sensation at the terminal joint of the middle finger. J Physiol 1983:335:507-17. [9] Hellstrom F, Thunberg J, Bergenheim M, Sjolander P, Pedersen J, et al. Elevated intramuscular concentration of bradykinin in jaw muscle increases the fusimotor drive to neck muscles in the cat. J Dental Res 200O;79:18 15-32. [lo] Hungerford DS, Barry M. Biomechanics of the patellofemoral joint. Clin Orthop Re1 Res 1979;144:9-15. [ I l l Jerosch J, Prymka M. Knee joint proprioception in patients with posttraumatic recurrent patella dislocation. Knee Surg, Sports Traumatol Arthrosc 1996;4:14-8. [I:] Jerosch J, Schmidt K, Prymka M. Proprioceptive capacities of patients with retropatellar knee pain with special reference to effectiveness of an elastic knee bandage. Unfallchirurgica 1997;100:719-23. [I 31 Johansson H, Sjnlander P, Sojka P. A sensory role for the cruciate ligaments. Clin Orthop Re1 Res 1991:268:161--78. [14] Kavounoudias A, Roll R, Roll JP. The plantar sole is a 'dynamometric map' for human balance control. Neuroreport 1998;9:3247-52. [IS] Kiefer G, Forwell L. Kramer J, Birmingham TL. Comparison of sitting and standing protocols for testing knee proprioception. Physioth Canada 1998:304. [I61 Kramer J, Handfield T, Kiefer G. Forwell L, Birmingham T. Comparisons of weight-bearing and non-weight-bearing tests of knee proprioception performed by patients with patello-femoral pain syndrome and asymptomatic individuals. Clin J Sport Med 1997;7:113 -8. [I71 Lackner J, DiZio P, Horak F, Krebs D, Rabin E. Precision contact of the fingertip reduces postural sway of individuals with bilateral vestibular loss. Exp Brain Res 1999;126:459-66.
[I81 Lephart SM, Pincivero DM, Giraldo L, Fu FH. The role of proprioception in the management and rehabilitation of athletic injuries. Am J Sports Med 1997;25:130-7. [I91 Lieb FJ, Perry J. Quadriceps function. An anatomical and mechanical study. J Bone Jnt Surg [Am] 1968;50:153548. [20] Milgrom C, Finestone A, Shlamkovitch N, Giladi M, Radin E. Anterior knee pain caused by overactivity. Clin Orthop Re1 Res 1996;331:25&60. [21] Mima T, Terada K, Maekawa M, Nagamine T, Ikeda A, et al. Somatosensory evoked potentials following proprioceptive stimulation of finger in man. Exp Brain Res 1996;111:23345. [22] Nadeau S, Gravel D, Hebert LJ, Arsenault AB, Lepage Y. Gait study of patients with patellofemoral pain syndrome. Gait Posture 1997;5:21-7. [23] Pai Y-C, Rymer WZ, Chang RW, Sharma L. Effect of age and osteoarthritis on knee proprioception. Arthrit Rheum 1997;40: 2260-5. "41 Powers CM, Perry J. Hsu A, Hislop HJ. Are patellofemoral pain and quadriceps femoris muscle torque associated with locomotor function. Phys Ther 1997;77:1063-75. [25] Powers CM, Heino JG, Rao S, Perry J. The influence of patellofemoral pain on lower limb loading during gait. Clin Biomech 1999;14:722 8. "61 Powers C. Patella kinematics, Part I: the influence of vastus medialis activity in subjects with and without patellofemoral pain. Phys Ther 2000;80:956-64. [27] Refshauge K, Chan R, Taylor J, McCloskey D. Detection of movements imposed on human hip, knee, ankle and toe joints. J Physiol 1995;488:231-41. "81 Roll J-P, Vedel J-P, Ribot-Ciscar E. Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study. Exp Brain Res 1989;76:213-21. [29] Sallis J, Haskell W, Fortmann S, Rogers T, Blair S, Pffenberger RS. Physical activity assessment methodology in the five-city project. Am J Epidemiol 1985;121:91-106. [30] Sanchis-Alfonso V, Rosello-Sastre E. Immunohistochemical analysis for neural markers of the lateral retinaculum in patients with isolated symptomatic patellofemoral malalignment. Am J Sports Med 2000;28:725-31. [31] Schaible H-G, Grubb BD. Afferent and spinal mechanisms ofjoint pain. Pain 1993;55:5-54. [32] Sharma L, Pai Y-C, Holtkamp K, Rymer W. Is knee joint proprioception worse in the arthritic knee versus the unaffected knee in unilateral knee osteoarthritis? Arthrit Rheum 1997;40:I51 8-25. [33] Stillman BC. An investigation of the clinical assessment of joint position sense. PhD Thesis, School of Physiotherapy, The University of Melbourne. http://adtl.lib.unimelb.edu.auladt-moot/ publidadt -VU200 1.OO 1Xndex.html, 2000. [34] Witvrouw E. Lysens R, Bellemans J, Peers K. Intrinsic risk factors for the development of anterior knee pain in an athletic population: a two year prospective study. Am J Sports Med 2000;28: 480-9.