ARTICLE IN PRESS

Journal of Biomechanics 43 (2010) 1794–1798

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/jbiomech

Journal of Biomechanics

0021-92

doi:10.1

n Corr

Departm

00185 R

E-m

www.JBiomech.com

Kinematic and kinetic features of normal level walking in patellofemoral painsyndrome: More than a sagittal plane alteration

Marco Paoloni a,b,n, Massimiliano Mangone b, Giancarlo Fratocchi a, Massimiliano Murgia a,b,Vincenzo Maria Saraceni a,b, Valter Santilli a,b

a Physical Medicine and Rehabilitation Unit, Azienda Policlinico Umberto I, Rome, Italyb Board of Physical Medicine and Rehabilitation, Department of Orthopaedic Science, ‘‘Sapienza’’ University, Piazzale Aldo Moro 3, 00185 Rome, Italy

a r t i c l e i n f o

Article history:

Accepted 10 February 2010Patients with patellofemoral pain syndrome (PFPS) often report discomfort and pain during walking. To

date, most of the studies conducted to determine gait alterations in PFPS patients have focused on

Keywords:

Patellofemoral pain syndrome

Gait analysis

Kinetic

Kinematic

90/$ - see front matter & 2010 Elsevier Ltd. A

016/j.jbiomech.2010.02.013

esponding author at: Board of Physical M

ent of Orthopaedic Science, ‘‘Sapienza’’ Univ

ome, Italy. Tel.: +39 6 491672; fax: +39 6 49

ail address: paolonim@tin.it (M. Paoloni).

a b s t r a c t

sagittal plane alterations. Physiological and biomechanical factors, however, suggest that frontal and

transverse plane alterations may be involved in PFPS. We therefore decided to conduct a kinematic and

kinetic evaluation on all three planes in 9 PFPS subjects and 9 healthy sex- and age-matched controls.

General gait characteristics were similar in patients and controls, with the exception of swing velocity,

which was lower in PFPS patients. Patients also displayed an increased knee abductor and external

rotator moments in loading response, and reduced knee extensor moment both in loading response and

in terminal stance. We speculate that these findings may be linked both to a pain-avoiding gait pattern

and to alterations in the timing of activation of different components of the quadriceps muscle, which is

typical of PFPS. The relevance for clinicians is this gait pattern may represent a biomechanical risk factor

for future knee osteoarthritis. We therefore recommend that treatments aimed at PFPS should also

attempt to restore a correct walking pattern.

& 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Patellofemoral pain syndrome (PFPS) is a term used for avariety of pathologies or anatomical abnormalities leading to atype of anterior knee pain (Witvrouw et al., 2005) that typicallyoccurs with activity and is exacerbated by stair climbing andprolonged sitting. Alterations in knee kinetics and kinematicsremain controversial in people affected by PFPS. Reduced kneeflexion during loading response phase (LR) of walking has beenfound in some studies (Dillon et al., 1983; Nadeau et al., 1997),though not in others (Heino Brechter and Powers, 2002; Powerset al., 1996, 1997, 1999). Moreover, women with PFPS exhibitdifferent lower extremity mechanics in a variety of activities withprogressive knee loading when compared with controls (Willsonand Davis, 2008). From a kinetic point of view, PFPS patientsgenerally display a reduced knee extensor moment during the LR(Besier et al., 2009; Heino Brechter and Powers, 2002), and areduced peak vertical ground reaction force (GRF) (Powers et al.,1999). Heino Brechter and Powers (2002) found no difference in

ll rights reserved.

edicine and Rehabilitation,

ersity, Piazzale Aldo Moro 3,

914192.

the patellofemoral joint (PFJ) forces between subjects with andwithout PFPS. They did, however, establish that the contact areabetween the femur and patella was reduced in subjects in thePFPS group, who consequently displayed increased PFJ stress. Thestudy of the kinematics and kinetics involved in the frontal andtransversal planes has not been considered a major goal ofbiomechanical research in the field of PFPS, largely becausemeasurement of the sagittal plane alone may yield an indirectevaluation of patella functioning. Determination of kinetic andkinematic changes that occur in the knee joint in both the frontaland transversal planes may, however, be relevant to thebiomechanical analysis in the clinical setting. Indeed, it hasrecently been suggested that increased knee abductor and kneeexternal rotator moments during the stance phase of stairclimbing and locomotion up an incline may be considered asbiomechanical markers in the assessment of mechanisms in-volved in the development of knee osteoarthritis (OA) in theelderly (Karamanidis and Arampatzis, 2009). It is also suggestedthat kinematic and kinetic changes in the transversal plane at theknee joint play a specific role in the initiation of local degenerativecartilage alterations by shifting the mechanical load within theload-bearing regions of the knee (Andriacchi et al., 2004). This factsupports the numerous reports of an increased incidence of kneeOA in patients who have experienced abnormal joint motion and

ARTICLE IN PRESS

M. Paoloni et al. / Journal of Biomechanics 43 (2010) 1794–1798 1795

mechanical load, such as those with anterior cruciate ligamentinjury or generalized joint laxity (Andriacchi et al., 2004).

We hypothesized that PFPS subjects may display abnormalknee joint moments in the frontal and transversal planes duringgait due to a neuromuscular dysfunction that normally occurs insuch patients. Subjects with PFSP, in fact, display a dysfunction ofPFJ neuromotor control as a result of an imbalance in the timing ofvastus medialis obliquus (VMO) and vastus lateralis (VL) activity(Cowan et al., 2001; Witvrouw et al., 1996), there being asignificant delay in the electromyographic onset of the VMO whencompared with that of the VL (Cowan et al., 2001). This fact maylead to abnormal patellar tracking within the trochlear groove. Itshould be considered, furthermore, that the quadriceps muscle,whose function is altered in PFPS subjects, provides the most kneestability on the frontal plane (Shelburne et al., 2006) and mayinfluence tibial rotation (Li et al., 1999).

We therefore designed an explanatory (Bender and Lange,2001), cross-sectional trial to investigate the kinetic and kine-matic features on the transversal, frontal and sagittal planes of thewalking pattern in PFPS subjects by means of instrumental gaitanalysis and to compare these features with those of a controlgroup of healthy subjects.

2. Methods

2.1. Subjects

The experimental group (EG) was composed of 9 patients (2 men, 7 women)

affected by PFPS (mean disease duration: 9.571.5 months). In order to be

included, patients had to report anterior or retropatellar knee pain in at least two

of the following activities: prolonged sitting, climbing stairs, squatting, running,

kneeling, and hopping or jumping. In addition, they had to have pain upon

palpation of the patella, to have a pain level of 3 cm or more on a 10-cm visual

analogue scale (VAS) while stepping down from a 25-cm step or during a double

leg squat, to have had symptoms for at least 1 month, to have an average pain level

of 3 cm or more on a 10-cm VAS, and to have had an insidious onset of symptoms

unrelated to a traumatic incident. Participants were excluded if they had the

following: a recent history (within 6 months) of knee surgery or of patellar

dislocation or subluxation; clinical evidence of a meniscal lesion, ligamentous

instability, traction apophysitis around the patellofemoral complex, patellar

tendon pathology, chondral damage or osteoarthritis; pain referred to the spine.

A total of 9 age- and sex-matched healthy subjects (2 men, 7 women) were also

enrolled to form the control group (CG). The study protocol was approved by the

local ethics committee and the participants’ informed consent was obtained.

2.2. Instrumental evaluation

Gait analysis was performed using the ELITE system (BTS, Milano, Italy), with 8

infrared video cameras (TVC, BTS, Milano, Italy) for the acquisition of the

kinematic variables. Two Kistler platforms (Kistler Instruments,Winterthur,

Switzerland) were used to acquire the GRF. Kinematic and kinetic data were

acquired and digitized with a sampling rate of 100 Hz. Anthropometric data were

collected for each subject and retroreflective spherical markers were placed over

prominent bone landmarks to determine the joint centers and segment axis (Davis

et al., 1991). Subjects were then instructed to walk at a self-selected speed along a

level surface approximately 10 m in length; three valid trials were acquired for

each subject and the mean value was considered for time/distance, kinematic and

kinetic data throughout the analysis. A valid trial was defined as one in which

subjects struck the force platforms without adjusting their stride length. The

affected side was analyzed in the EG, while either the right or left side was

randomly used for the analysis in the CG.

Mean velocity (m/s) was assessed to ensure that any kinetic and/or kinematic

differences between groups were not due to differences in gait speed. Swing

velocity (distance travelled in the swing/swing duration) was also assessed.

Three-dimensional marker trajectories during walking were obtained by

means of a frame-by-frame tracking system (Tracklab, BTS, Milano, Italy) and joint

angular excursion, defined as a rotation of the distal segment relative to the

proximal segment in our biomechanical model (Vaughan et al., 1999), was

calculated; joint excursion data were normalized to the stride duration and

reduced to 100 samples over the gait cycle. The following parameters were

considered for the kinematic evaluation: (i) knee flexion angle at heel contact (HC),

(ii) knee flexion range of movement (ROM) during the LR (defined as the first sub-

phase from initial contact until contralateral toe-off, i.e., 0–12% of the gait cycle)

(Vaughan et al., 1999), (iii) hip and knee abduction peak during the LR and (iv) hip

and knee rotation ROM during the whole gait cycle.

Net internal joint moments were calculated by means of an inverse dynamics

approach. Joint moments were normalized to the subject’s body weight. We

considered the following parameters for the kinetic analysis: (i) in the LR, we

assessed the hip and knee extensor, abductor and external rotator moment peaks

and (ii) in the terminal stance sub-phase (TS), we assessed the hip flexor, abductor

and internal rotator moment peaks, as well as the knee extensor, abductor and

internal rotator moment peaks. We also considered the peak values of the vertical

GRF curve.

2.3. Statistical analysis

Statistical analysis was performed using the SAS8.2 (SAS Institute Inc., Cary,

NC, USA). Data normality was verified by means of the Shapiro–Wilk test. The

mean value7standard deviation of each parameter was calculated in both the EG

and CG. The unpaired t-test or Mann–Whitney test was used to evaluate the

significance of differences between the EG and CG. A p value of less than 0.01 was

considered statistically significant.

3. Results

No significant differences were observed between PFPSsubjects and controls in mean age [EG: 28.178.1 (range 19–45)years; CG 28.375.9 (range 21–38) years)], in mean height [EG:1.7170.09 (range 1.60–1.82) m; CG: 1.7070.09 (range 1.60–1.86) m] or in mean body weight [EG: 64.479.5 (range 55–80)kg; CG: 64.2710.8 (range 53–83) kg]. The mean velocity did notdiffer significantly between groups [EG 1.170.15 m/s; CG1.1570.16 m/s; p=0.5]. Patients in the EG, however, displayed asignificantly slower swing phase velocity than those in the CG [EG2.4470.28 m/s; CG 2.9570.46 m/s; p=0.01]. The kinematicresults are summarized in Table 1. A greater degree ofadduction was observed during the LR in the knee of PFPSpatients than in that of controls. The kinetic results are providedin Table 2. During the LR, patients in the EG exhibited an increasein knee external rotator moment associated with a markedreduction in the knee extensor moment, when compared withsubjects in the CG. Increased hip and knee abductor moment wasalso noted in EG. In the TS, knee extensor moment wassignificantly lower in the EG than in the CG. Hip abductormoment in the TS was also significantly greater in the EG than inCG. The results of the GRF analysis are provided in Table 3. Thevertical GRF peak at HC was significantly lower in the EG than inthe CG (Fig. 1), whereas no differences were observed betweenthe two groups in any of the other parameters analyzed.

4. Discussion

Subjects with PFJS generally have anterior knee pain that isexacerbated by knee loading activities. It has been hypothesizedthat the load exerted by walking upon the PFJ is not sufficient toreveal consistent biomechanical alterations, it being suggestedthat stair ambulation is a more appropriate means of elicitingsuch changes since the latter activity loads the PFJ to a greaterdegree (Costigan et al., 2002). Strategies that tend to minimizeknee loading during gait have been widely described in PFPSpatients with gait limitations (Nadeau et al., 1997; Powers et al.,1997). Moreover, as walking is the most important loadingactivity performed on a daily basis, any factors involved in gaitthat may explain joint pathology progression warrant investiga-tion.

The results of our study confirm our initial hypothesis, i.e. thataltered knee joint moments may be observed not only on thesagittal plane but also on the transversal and frontal planes. It isnoteworthy that the present study shows, to the best of our

ARTICLE IN PRESS

Table 1

Experimental group Control group p

Knee flexion angle at HC (deg.) 8.17 (5.15) 6.78 (7.59) 0.65a

Knee flex-extension ROM during LR (deg.) 6.62 (4.24) 6.62 (4.34) 0.99a

Knee adduction peak during LR (deg.) 2.12 (1.4) 0.09 (1.5) 0.009a

Knee rotation ROM (deg.) 16.86 (5.9) 18.73 (8.6) 0.60b

Hip adduction peak during LR (deg.) 6.42 (3.9) 6.53 (2.5) 0.94b

Hip rotation ROM (deg.) 13.04 (5.0) 11.82 (3.7) 0.57b

Mean values (SD) of kinematic parameters. Significant differences are in bold. ROM=range of movement; HC=heel contact; LR=loading response.

a Unpaired t-test.b Mann–Whitney test.

Table 2

Kinetic parameter (Nm/kg) Experimental group Control group p

Loading response Hip abductor moment 0.898 (0.24) 0.508 (0.17) 0.001b

Hip extensor moment 0.310 (0.19) 0.210 (0.12) 0.20a

Hip external rotator moment 0.107 (0.03) 0.09 (0.03) 0.25a

Knee abductor moment 0.555 (0.13) 0.398 (0.13) 0.01c

Knee external rotator moment 0.071 (0.02) 0.042 (0.02) 0.007a

Knee extensor moment 0.027 (0.21) 0.377 (0.25) 0.005a

Terminal stance Hip abductor moment 0.842 (0.26) 0.450 (0.18) 0.002a

Hip flexor moment �0.378 (0.20) �0.559 (0.18) 0.06a

Hip internal rotator moment �0.138 (0.04) �0.121 (0.06) 0.49a

knee abductor moment 0.477 (0.10) 0.376 (0.14) 0.10a

Knee internal rotator moment �0.075 (0.02) �0.100 (0.04) 0.11a

Knee extensor moment 0.083 (0.06) 0.329 (0.20) 0.003a

Mean values (SD) of kinetic parameters. Significant differences are in bold.

a Unpaired t-test.b Mann–Whitney test.c Unpaired t-test with Welch correction.

Table 3

Experimental group % of GC Control group % of GC p

Vertical force (N/kg)

Peak at HC (iGRF) 40.7 (15.6) 2 56.7 (9.1) 2 0.01a

Peak at LR 95.5 (25.0) 16 99.8 (6.6) 17 0.63b

Peak at TS 98.2 (23.9) 46 110.6 (8.2) 47 0.17b

Mean values (SD) of ground reaction force (GRF) parameters. For each value the percentage of gait cycle (GC) during which the event occurred is shown. HC=heel contact;

LR=loading response; TS=terminal stance; iGRF=impact GRF (see text for clarification).

a Unpaired t-test.b Mann–Whitney test.

M. Paoloni et al. / Journal of Biomechanics 43 (2010) 1794–17981796

knowledge for the first time, that such alterations are likely tooccur during walking.

Particularly, PFPS patients in our sample displayed a markedlyincreased knee external rotator moment in the LR. One possiblecausative mechanism for this increased knee external rotatormoment may be found in the altered VL and VMO function.Indeed, the delay in VMO activation often described in theliterature may determine an initial isolated VL contraction thatcreates an external rotator moment. One obvious limitation of thepresent study is that we did not perform an EMG evaluation ofVMO/VL function in our sample of patients. We may assume,however, that a delay in VMO activation did occur because it is atypical feature of PFPS patients, as has been extensively describedin the literature (Cowan et al., 2001; Witvrouw et al., 1996).Further researches, specifically aimed at analyzing the quadricepsmuscular function and its relationship with transversal planeknee kinetics, are however warranted to confirm this hypothesis.Another possible explanation may lie in the abnormal patellar

tracking observed in PFPS subjects. An in vivo evaluation duringsquatting recently demonstrated that the patellae of subjects withPFPS were subject to progressive lateral spin during knee flexion,with the difference between PFPS and control subjects becomingevident from as little as 301 knee flexion (Wilson et al., 2009). Incontrast to findings by Willson and Davis (2008), we did notdetect knee or hip joint rotation alterations in the EG. Since weconsidered gait alone in our study, we cannot rule out that suchalterations may emerge during activities requiring a greaterdegree of loading.

On the frontal plane, our patients displayed a greater kneeadduction peak than controls, which is associated with asignificantly increased knee abductor moment peak in the LR,though not in the TS. The combination of increased frontal planekinetics and kinematics is interesting from a biomechanicalperspective. It may be argued that these changes determinealterations in the direction and magnitude of the muscle forces onthe patella. Worthy of note is the fact that our PFPS patients

ARTICLE IN PRESS

0

20

40

60

80

100

120

1 10 19 28 37 46 55

*

Fig. 1. Vertical ground reaction force (GRF) in patellofemoral pain syndrome

subjects (dotted line) and in healthy controls (continuous line). The impact GRF is

highlighted for both groups.

M. Paoloni et al. / Journal of Biomechanics 43 (2010) 1794–1798 1797

displayed a marked reduction in the knee extensor moment in thesame gait cycle sub-phases. One appealing hypothesis is that aredistribution occurred among quadriceps moments from theextensor to external rotator and abductor. Indeed, in keeping withprevious findings (Heino Brechter and Powers, 2002; Powerset al., 1997, 1999), we did not find any differences in the sagittalkinematic features of the knee joint during stance, with bothpatients and controls exhibiting a small degree of knee flexion inLR, which is responsible for weight acceptance, during the firstphase of gait cycle. We believe that it would be interesting toinvestigate a detailed muscle model to test this hypothesis infuture studies. An alternative explanation for a reduced kneeextensor moment may be the need to unload the knee joint toavoid pain. This kinetic feature is associated with a significantlyreduced vertical GRF at the HC. The short spike of force in thevertical GRF vector immediately following the HC, which is calledthe impact ground reaction force peak (iGRF) (Henriksen et al.,2008), is thought to reflect the abrupt vertical deceleration of thecenter of mass immediately after foot contact with the ground.Reduced iGRF may be determined by altered neuromotor controldue to knee pathologies, as the presence of pain alone appears tobe insufficient to determine such an alteration (Henriksen et al.,2008). Impact GRF production may be influenced by both walkingspeed (Voloshin, 2000) and knee joint angle at the HC (Lafortuneet al., 1996). In our sample, however, we can exclude any suchinteraction as there was no difference between the EG and the CGin these factors. Interestingly, we observed lower swing velocityin the EG than in the CG. Grenholm et al. (2009) found thatwomen with long-term anterior knee pain, compared withhealthy controls, display a reduced knee angular velocity in thestance leg while descending stairs, even though the overallwalking speed remains similar. This reduced speed may placeless demand on the swing leg, thereby allowing the latter toabsorb the impact of foot contact when lowering the body andthus reduce joint reaction forces. One explanation for thesefindings may be that PFPS patients reduce the swing speed of theirsymptomatic limb in order to reduce the iGRF and, consequently,also reduce the extensor torque on the symptomatic knee. Futurestudies specifically designed for this purpose are needed toconfirm this hypothesis.

The overall moment redistribution may account for conse-quences in knee joint pathology progression. As extensivelyreported in the literature, an increased knee abductor moment isassociated with medial knee OA (Maly, 2008), it often beingconnected with leg axis alterations. It is noteworthy, however, thatmedications that afford pain relief in symptomatic knee OAconcurrently modify the knee abductor moment (Henriksen et al.,2006; Schnitzer et al., 1993), thereby suggesting that gaitmodifications designed to help loading are likely to occur in peoplewith knee pain. The knee abductor moment appears to be involvedin the onset of chronic knee pain (Amin et al., 2004) andradiographic knee OA (Lynn et al., 2007) and is associated with agreater risk of radiographic OA progression (Miyazaki et al., 2002).On the basis of the results of this study, PFPS patients may beconsidered as a population at risk, from a biomechanical point ofview, for the development of medial knee OA due to repetitivealtered knee loading. Lynn et al. (2007) recently described, in aprospective case study with long-term follow-up, the differences inthe biomechanical profiles of asymptomatic people who subse-quently develop medial or lateral knee OA. In particular, an initialincreased knee abductor moment and low medial–lateral shareforce are associated with future development of symptomaticmedial knee OA. The increased hip abductor moment observed inour sample of PFPS patients partially counteracts the knee abductormoment. This increased hip abductor moment was, in one study,found to effectively protect against the progression of medial kneeOA over 18 months in 57 people with knee osteoarthritis; since fewknee muscles specifically provide knee frontal-plane stability, theprotection afforded by the hip frontal-plane muscles is likely toplay an important role in regulating medial/lateral knee loaddistribution (Chang et al., 2005).

Interestingly, studies based on principal component analysis(Astephen et al., 2008; Astephen and Deluzio, 2005) demonstratedhow the abductor and the external rotator joint moments duringthe stance phase of gait are the most important discriminatoryfactors of OA gait patterns. Karamanidis and Arampatzis (2009)found that older adults had higher knee abductor and kneeexternal rotator joint moments during the stance phase of stairascending and descending. The same authors found that thekinematic and the kinetic changes they detected may be used asbiomechanical markers to assess mechanisms involved in thedevelopment of knee OA in the elderly, recommending interven-tions designed to counteract this age-related load redistributionin the knee joint before the onset of knee pain and irreversibledegenerative cartilage damage. Our findings in a relatively youngpopulation of patients suggest that gait alterations caused by PFPSmay be a risk factor for the development of chronic knee pain andknee OA.

One limitation of our study is the possible occurrence ofkinematic crosstalk when estimating knee kinematics on frontalplanes. Such crosstalk typically results from joint axis misalign-ment and may cause motion to be measured where it does notexist and vice-versa (Piazza and Cavanagh, 2000). This mistakewould have occurred particularly when measurements weremade in the presence of large knee flexion angles, e.g. duringthe swing phase. Caution should consequently be taken wheninterpreting these kinematic features in PFPS subjects.

Moreover, we cannot rule out the possibility that our study wasunder-powered to detect differences between the two groups.

5. Conclusions

The PFPS subjects we studied displayed several kinematic andkinetic alterations during gait, detected not only on the sagittalplane but also on the frontal and transverse planes. The relevance

ARTICLE IN PRESS

M. Paoloni et al. / Journal of Biomechanics 43 (2010) 1794–17981798

of our results to clinicians is that these alterations may representa biomechanical risk factor for the development of knee OA in thefuture. Further studies are warranted to shed more light on thelink between PFPS and gait pattern alterations, particularly asregards the specific role played by pain and quadriceps muscledysfunction.

Conflict of interest statement

All the authors declare that there are no relationships of afinancial nature, or other issues, that might lead to a conflict ofinterests.

References

Amin, S., Luepongsak, N., McGibbon, C.A., LaValley, M.P., Krebs, D.E., Felson, D.T.,2004. Knee adduction moment and development of chronic knee pain inelders. Arthritis and Rheumatism 51, 371–376.

Andriacchi, T.P., Mundermann, A., Smith, R.L., Alexander, E.J., Dyrby, C.O., Koo, S.,2004. A framework for the in vivo pathomechanics of osteoarthritis at theknee. Annals of Biomedical Engineering 32, 447–457.

Astephen, J.L., Deluzio, K.J., 2005. Changes in frontal plane dynamics and theloading response phase of the gait cycle are characteristic of severe kneeosteoarthritis application of a multidimensional analysis technique. ClinicalBiomechanics (Bristol, Avon) 20, 209–217.

Astephen, J.L., Deluzio, K.J., Caldwell, G.E., Dunbar, M.J., Hubley-Kozey, C.L., 2008.Gait and neuromuscular pattern changes are associated with differences inknee osteoarthritis severity levels. Journal of Biomechanics 41, 868–876.

Bender, R., Lange, S., 2001. Adjusting for multiple testing-when and how? Journalof Clinical Epidemiology 54, 343–349

Besier, T.F., Fredericson, M., Gold, G.E., Beaupre, G.S., Delp, S.L., 2009. Knee muscleforces during walking and running in patellofemoral pain patients and pain-free controls. Journal of Biomechanics 42, 898–905.

Chang, A., Hayes, K., Dunlop, D., Song, J., Hurwitz, D., Cahue, S., Sharma, L., 2005.Hip abduction moment and protection against medial tibiofemoral osteoar-thritis progression. Arthritis and Rheumatism 52, 3515–3519.

Costigan, P.A., Deluzio, K.J., Wyss, U.P., 2002. Knee and hip kinetics during normalstair climbing. Gait and Posture 16, 31–37.

Cowan, S.M., Bennell, K.L., Hodges, P.W., Crossley, K.M., McConnell, J., 2001.Delayed onset of electromyographic activity of vastus lateralis compared tovastus medialis obliquus in subjects with patellofemoral pain syndrome.Archives of Physical Medicine and Rehabilitation 82, 183–189.

Davis, R.B., Ounpuu, S., Tyburski, D., Gage, J.R., 1991. A gait analysis data collectionand reduction technique. Human Movement Science 10, 575–587.

Dillon, P.Z., Updyke, W.F., Allen, W.C., 1983. Gait analysis with reference tochondromalacia patellae. Journal of Orthopaedic and Sports Physical Therapy5, 127–131.

Grenholm, A., Stensdotter, A.K., Hager-Ross, C., 2009. Kinematic analyses duringstair descent in young women with patellofemoral pain. Clinical Biomechanics(Bristol, Avon) 24, 88–94.

Heino Brechter, J., Powers, C.M., 2002. Patellofemoral stress during walking inpersons with and without patellofemoral pain. Medicine and Science in Sportsand Exercise 34, 1582–1593.

Henriksen, M., Christensen, R., Alkjaer, T., Lund, H., Simonsen, E.B., Bliddal, H.,2008. Influence of pain and gender on impact loading during walking: arandomised trial. Clinical Biomechanics (Bristol, Avon) 23, 221–230.

Henriksen, M., Simonsen, E., Alkjaer, T., Lund, H., Graven-Nielsen, T., Danneskiold-Samsøe, B., Bliddal, H., 2006. Increased joint loads during walking: aconsequence of pain relief in knee osteoarthritis. The Knee 13, 445–450.

Karamanidis, K., Arampatzis, A., 2009. Evidence of mechanical load redistributionat the knee joint in the elderly when ascending stairs and ramps. Annals ofBiomedical Engineering 37, 467–476.

Lafortune, M.A., Lake, M.J., Hennig, E.M., 1996. Differential shock transmissionresponse of the human body to impact severity and lower limb posture.Journal of Biomechanics 29, 1531–1537.

Li, G., Rudy, T.W., Sakane, M., Kanamori, A., Ma, C.B., Woo, S.L., 1999.The importance of quadriceps and hamstring muscle loading on kneekinematics and in-situ forces in the ACL. Journal of Biomechanics 32,395–400.

Lynn, S., Reid, S., Costigan, P., 2007. The influence of gait pattern on signs of kneeosteoarthritis in older adults over a 5–11 year follow-up period: a case studyanalysis. The Knee 14, 22–28.

Maly, M.R., 2008. Abnormal and cumulative loading in knee osteoarthritis. CurrentOpinion in Rheumatology 20, 547–552.

Miyazaki, T., Wada, M., Kawahara, H., Sato, M., Baba, H., Shimada, S., 2002.Dynamic load at baseline can predict radiographic disease progression inmedial compartment knee osteoarthritis. Annals of the Rheumatic Diseases 61,617–622.

Nadeau, S., Gravel, D., Hebert, L.J., Bertrand Arsenault, A., Lepage, Y., 1997. Gaitstudy of patients with patellofemoral pain syndrome. Gait and Posture 5,21–27.

Piazza, S.J., Cavanagh, P.R., 2000. Measurement of the screw-home motion of theknee is sensitive to errors in axis alignment. Journal of Biomechanics 33,1029–1034.

Powers, C.M., Heino, J.G., Rao, S., Perry, J., 1999. The influence of patellofemoralpain on lower limb loading during gait. Clinical Biomechanics (Bristol, Avon)14, 722–728.

Powers, C.M., Landel, R., Perry, J., 1996. Timing and intensity of vastus muscleactivity during functional activities in subjects with and without patellofe-moral pain. Physical Therapy 76, 946–955.

Powers, C.M., Perry, J., Hsu, A., Hislop, H.J., 1997. Are patellofemoral pain andquadriceps strength associated with locomotor function? Physical Therapy 77,1063–1074

Schnitzer, T.J., Popovich, J.M., Andersson, G.B.J., Andriacchi, T.P., 1993. Effect ofpiroxicam on gait in patients with osteoarthritis of the knee. Arthritis andRheumatism 36, 1207–1213.

Shelburne, K.B., Torry, M.R., Pandy, M.G., 2006. Contributions of muscles,ligaments, and the ground-reaction force to tibiofemoral joint loading duringnormal gait. Journal of Orthopaedic Research 24, 1983–1990.

Vaughan, C.L., Davis, B.L., O’Connor, J.C., 1999. Dynamics of Human Gaits 2nd ed.Human Kinetics Publishers, Champaign, IL.

Voloshin, A., 2000. The influence of walking speed on dynamic loading on thehuman musculoskeletal system. Medicine and Science in Sports and Exercise32, 1156–1159.

Willson, J.D., Davis, I.S., 2008. Lower extremity mechanics of females with andwithout patellofemoral pain across activities with progressively greater taskdemands. Clinical Biomechanics (Bristol, Avon) 23, 203–211.

Wilson, N.A., Press, J.M., Koh, J.L., Hendrix, R.W., Zhang, L.Q., 2009. In vivononinvasive evaluation of abnormal patellar tracking during squatting inpatients with patellofemoral pain. Journal of Bone and Joint Surgery (Am) 91,558–566.

Witvrouw, E., Sneyers, C., Lysens, R., Victor, J., Bellemans, J., 1996. Reflex responsetimes of vastus medialis oblique and vastus lateralis in normal subjects withpatellofemoral pain syndrome. Journal of Orthopaedic and Sports PhysicalTherapy 24, 160–165.

Witvrouw, E., Werner, S., Mikkelsen, C., Van Tiggelen, D., Vanden Berghe, L., Cerulli,G., 2005. Clinical classification of patellofemoral pain syndrome: guidelines fornon-operative treatment. Knee Surgery, Sports Traumatology, Arthroscopy 13,122–130.