Background

Altered foot loading during weightbearing is suggested to play a role in the development of patellofemoral pain (PFP). This study aimed to determine foot-loading characteristics associated with PFP by assessing center of pressure (COP) during single-limb loading in individuals with PFP compared with noninjured controls.

Methods

Thirty recreationally active patients with PFP and 30 noninjured control participants had barefoot plantar pressure assessed during single-limb squats (SLSs) from which COP parameters (COP velocity and COP index) were obtained. Groups were compared using independent t tests.

Results

Individuals with PFP demonstrated a greater COP index (P = .042), indicating a more lateral foot-loading pattern, and exhibited increased overall COP velocity (P = .013) and anteroposterior COP velocity during SLSs compared with control participants (P = .033).

Conclusions

Greater lateral foot loading and increased COP velocity during SLSs demonstrated by individuals with PFP may indicate reduced dynamic balance in this patient group, which may be implicated in the development of PFP.

Patellofemoral pain (PFP) is a common knee injury in athletes and in the general population.1  The typical clinical presentation includes the complaint of anterior or retropatellar pain. Symptoms are intensified by prolonged sitting, along with weightbearing exercise that requires high levels of quadriceps activity, including running, squatting, and ascending and descending stairs.2  The general consensus is that the etiology of PFP is multifactorial,3  with proximal factors, including poor neuromuscular control of the hip, and distal factors, including alterations in foot function, suggested to influence patellofemoral joint (PFJt) mechanics and play a role in the development of the condition.

As the distal segment in the lower kinetic chain, the foot acts as the main interface with the ground and plays an important role in the initial transmission of the ground reaction force vector (GRFv) through the lower kinetic chain to the knee. The center of pressure (COP) represents the instantaneous point of application of the GRFv acting on the plantar surface of the foot during weightbearing. Analysis of COP variables, including location and velocity, provides information on the temporal loading pattern of the foot.4  For example, the path of the COP has been suggested to be a direct result of foot pronation or supination during limb loading,5  with displacement of the COP hypothesized to reflect an individual's ability to maintain balance and postural stability.6,7  In addition, alterations in COP velocity have been suggested to affect force transmission through the lower kinetic chain during limb-loading tasks, including single-limb squats (SLSs).8 

It has been proposed that alterations in COP may be important in the development of lower-limb overuse injuries, including PFP.9  Prospective studies by Willems et al10,11  reported that individuals who developed exercise-related lower-limb pain displayed a more medially directed COP at forefoot flat during running.10  Further studies have also reported that patients who exhibited increased distance of the anteroposterior COP at initial contact were also at greater risk for lower-limb injury.11  Currently, only limited studies have been conducted that have assessed dynamic COP parameters in patients with PFP, with evidence indicating impaired postural control in individuals with PFP during stair negotiation.7,12  In addition, Thijs et al6  prospectively demonstrated that alterations in COP displacement were considered an important gait-related risk factor for PFP, with military recruits who went on to develop PFP displaying a more laterally displaced COP and lower maximal displacement velocity of mediolateral COP during the forefoot contact phase of walking compared with recruits who did not develop PFP.

Due to the limited number of research studies conducted, the role of COP in the development of lower-limb conditions such as PFP is still relatively unknown. Furthermore, the disparity in findings between studies may be due to the complexity of tasks in which COP is being assessed. Evaluation of COP parameters during more complex limb-loading tasks that require greater mechanical and muscular demands may provide information on dynamic foot loading that may not be observed during relatively simple motor tasks. Therefore, the purpose of this study was to assess COP parameters during SLSs in individuals with PFP compared with noninjured individuals to determine foot-loading patterns important in PFP. Based on the background literature, it was hypothesized that individuals with PFP would exhibit alterations in COP parameters during SLSs compared with noninjured individuals.

Methods

Participants

Thirty patients (18 females and 12 males) with PFP were recruited for the cross-sectional study via advertisements to university populations, local running clubs, and sports injury clinics. An a priori sample size calculation (α = 0.05; β = 0.20) based on between-group differences, and expected variability in foot-loading data collected from pilot studies suggested that 25 to 30 patients would be necessary to identify between-group differences with medium-to-large effect sizes (ESs). All of the patients were recreationally active, participating in at least 3 hours of physical activity per week. Potential participants were assessed by an experienced musculoskeletal podiatric physician for specific criteria (Table 1) to be recruited to the PFP group. Both females and males aged 18 to 40 years were recruited; although PFP is more likely to affect females, males can also be affected, with studies commonly reporting point prevalence in the general population to be 29% in females and 16% in males in this age bracket.1  Thirty individuals (15 females and 15 males) who were free of lower-extremity injury for the past 12 months acted as a control group. The PFP and control groups were similar in age, height, weight, and foot size (Table 2). The study was approved by the Cardiff Metropolitan University Research Ethics Committee, Cardiff, United Kingdom, with written informed consent obtained from all of the participants before study participation.

Table 1

Inclusion/Exclusion Criteria for the Patellofemoral Pain Group

Inclusion/Exclusion Criteria for the Patellofemoral Pain Group
Inclusion/Exclusion Criteria for the Patellofemoral Pain Group
Table 2

Participant Characteristics

Participant Characteristics
Participant Characteristics

COP Measurement

Plantar loading data were collected during barefoot SLSs to 60° of knee flexion using a plantar pressure platform (EMED; Novel GmbH, Munich, Germany). The pressure platform was embedded in the ground in a 4-m walkway and comprised 6,080 capacitive sensors in a sensing area of 475 × 320 mm (sensor resolution of 4 sensors/cm2) and had a pressure range of 10 to 1,270 kPa, accuracy of ±5%, hysteresis less than 3%, and a sampling frequency of 100 Hz.

Patients were instructed to perform SLSs to approximately 60° of knee flexion in a controlled manner and without losing balance before returning to the starting position. Participants' arms were folded in front of their body to eliminate the influence of arm swing and to assist with balance. Foot position was not standardized across participants during SLSs. Instead, participants performed each squat with their foot in its natural base and angle of stance in which they would normally position their foot when performing a single-limb loading task. Placing the foot in a contrived position may mask potential alterations in foot loading during functional activities.13  Squat depth was limited to approximately 60° to avoid higher joint forces associated with increased ranges of motion that may exacerbate knee pain symptoms in individuals with PFP.14  To standardize squat speed, each squat was performed over a 5-sec period, counted by the researcher. The first count initiated the movement, the third indicated the lowest point of the squat, and the fifth indicated the end. There was a 2-min recovery period between squats to minimize the effect of fatigue. Before testing, participants were allowed up to five practice trials to warm up and familiarize themselves with the test. Participants were given feedback on the depth of the squats (using a standard goniometer) and the speed of the squats during these trials until they could consistently and accurately perform the test movement. Trials were accepted only if the individual appeared to squat to the 60° minimum desired degree of knee flexion at a constant speed and maintained balance throughout. The mean of all kinetic data was taken from three trials because this has been shown to be sufficient for analysis.15 

The COP parameters were calculated for each participant from the maximum plantar pressure picture obtained during SLSs using Novel Gaitline software (Novel GmbH). Instantaneous COP was obtained by determining the geographic center of all of the excited sensors and assigning a weighted value based on the pressure for every frame sampled. The path of the COP during the SLSs was then plotted on the footprint by connecting each of the COP points (Fig. 1). The mediolateral path of the COP across the plantar region of the foot was quantified by calculating the COP index (COPI = area lateral/area medial).16  For each foot, the COPI was measured with respect to the longitudinal axis of the foot, defined as the line from mid-heel to forefoot, over the second metatarsal.17  A COPI of less than 1 indicated more medial weightbearing, and a COPI greater than 1 indicated more lateral weightbearing. Overall maximal velocity of the COP (vCOP) as well as mediolateral (vxCOP) and anteroposterior (vyCOP) velocity of the COP were calculated for the entire foot. The medial to lateral COPI was calculated for each participant by deviation from the longitudinal axis.

Figure 1

The X-component (mediolateral) and Y-component (anteroposterior) of the center of pressure in a typical patient with patellofemoral pain during a single-limb squat.

Figure 1

The X-component (mediolateral) and Y-component (anteroposterior) of the center of pressure in a typical patient with patellofemoral pain during a single-limb squat.

Statistical Analysis

Between-group differences in COP parameters (vCOP, vxCOP, vyCOP, COPI) were compared using independent t tests (2-tailed). The ESs were calculated to illustrate the magnitude of the difference between groups. Standard definitions were used, with a small effect equal to 0.20 to 0.50, a medium effect being between 0.50 and 0.80, and a large effect considered greater than 0.80.18  For all of the analyses, α was set at 0.05. All of the statistical analysis was conducted using IBM SPSS Statistics for Windows, Version 24.0 (IBM Corp, Armonk, New York).

Results

Individuals in both the PFP and noninjured groups were well matched for age, height, weight, and foot length (Table 2). No participants reported noteworthy pain during performance of the SLSs. Independent t tests indicated a significant difference in COPI between groups (95% CI = 0.003–0.135; P = .042; ES = 0.24), with the PFP group displaying a mean ± SD COPI of 1.50 ± 0.15 (indicating a more lateral COP) compared with the control group (1.43 ± 0.10) during SLSs (Fig. 2). Figure 2 tracks the COPI during the SLS in both groups and indicates that the noninjured group tended to start the squat with a relatively neutral COP, which gradually transferred to the lateral border of the foot during the descent phase of the SLS. Maximum lateral loading was observed at the start of the ascent phase before shifting slightly more medially in the final portion of the ascent phase. In contrast, the PFP group began the SLS with a more medial COP, which then shifted more rapidly to the lateral column of the foot throughout the descent phase. The COPI reached its maximal lateral position at the terminal portion of the descent phase, where the knee reached approximately 60° of flexion, before becoming slightly more medial during the ascent phase. Mean ± SD vCOP was also greater in the PFP group (2.14 ± 1.28 m/sec) compared with the control group (1.45 ± 0.72 m/sec) during SLSs (95% CI = 0.150–1.221; P = .013; ES = 0.34). Analysis of directional COP velocity revealed 33% greater vyCOP in patients with PFP compared with those in the control group (95% CI = 0.165–1.224; P = .033; ES = 0.33) but no significant difference in vxCOP during SLSs (Fig. 3).

Figure 2

Mean curve of the center of pressure index (COPI) in individuals with patellofemoral pain (PFP) and noninjured control participants during single-limb squats (SLSs).

Figure 2

Mean curve of the center of pressure index (COPI) in individuals with patellofemoral pain (PFP) and noninjured control participants during single-limb squats (SLSs).

Figure 3

Maximum velocity of the center of pressure (COP) in the mediolateral and anteroposterior directions during the single-limb squat. PFP, patellofemoral pain. *Significant at P < .001.

Figure 3

Maximum velocity of the center of pressure (COP) in the mediolateral and anteroposterior directions during the single-limb squat. PFP, patellofemoral pain. *Significant at P < .001.

Discussion

The etiology of PFP is considered to be multifactorial, with abnormal foot loading and associated increased PFJt stress hypothesized to be an important factor for this condition. This study aimed to determine foot-loading characteristics associated with PFP by assessing COP parameters during SLSs in individuals with PFP compared with noninjured control participants.

The present study found that patients with PFP displayed a more lateral COPI compared with participants in the control group during SLSs. Plantar force distribution that is more laterally directed may indicate that individuals with PFP exhibit decreased foot pronation and a more supinated position of the foot during limb-loading activity. Increases in the internal supination moment at the subtalar joint and resultant inversion and internal rotation of the rearfoot in the closed kinetic chain may decrease foot mobility and result in lateral COP displacement19  during performance of this complex motor task.

This finding is in agreement with previous studies that have assessed foot-loading patterns in individuals with PFP. Thijs et al6  prospectively investigated gait-related risk factors in military recruits and novice runners and found that recruits who developed PFP demonstrated a more laterally deviated COP during the forefoot contact phase of walking compared with control participants. Similarly, Duffey et al20  reported that runners who developed anterior knee pain exhibited 25% less pronation during the first 10% of the support phase compared with noninjured runners. Adequate foot pronation is necessary for appropriate shock absorption during limb-loading tasks. Decreased foot pronation may reduce force dissipation in the lower extremity, allowing a larger part of the GRFv to be transferred through the lower kinetic chain to proximal segments such as the knee,8  potentially increasing PFJt stress and etiology or exacerbation of pain symptoms.21 

The present study also found that patients with PFP displayed increased overall and vyCOP during SLSs. No previous studies have assessed the relationship between vyCOP and PFP. However, a prospective study assessing foot-loading parameters and the development of nonspecific lower-limb overuse injury found that individuals who developed injury had lower vyCOP during running compared with those who remained injury free.22  Increased vyCOP displayed by patients with PFP during SLSs in the present study may be part of a movement strategy aimed at reducing PFJt stress or pain symptoms associated with activities that require high quadriceps activity, such as SLSs.

It is suggested that the anteroposterior position of the COP is controlled by plantarflexion and dorsiflexion moments at the ankle joint in response to the location of the body's center of mass (COM) during weightbearing activity.23  Dionisio et al24  reported that anterior displacement of the COP occurs during the ascent phase of a squat when the knee is extending, in response to increased knee joint extension torque generated by strong activation of the quadriceps. Anterior COP displacement during this phase of the squat is associated with increased ankle joint torque toward plantarflexion and accompanied increased electromyography activity of the tibialis anterior. Although it has been demonstrated that the COP position does not directly correspond to the COM position,23  it is hypothesized in the present study that patients with PFP shifted their COM anteriorly during SLSs at a greater velocity than did noninjured individuals to increase the plantarflexor moment around the ankle, resulting in a more anteriorly placed COP. Consequently, this may facilitate more effective use of plantarflexor muscles in absorbing the GRFv and decreasing transmission of load through the lower kinetic chain to the knee. Increased activation of the ankle plantarflexors may also increase hip extensor (eg, hamstring muscle) contraction demand (through decreasing external knee flexor and hip flexor moments)25  during SLSs, thereby decreasing quadriceps contraction demand.

Another consideration is that altered COP parameters found in patients with PFP in this study is an adaptive response to altered proximal mechanics displayed during SLS activity. Prospective studies have demonstrated altered frontal and transverse plane hip and knee kinematics in individuals with PFP during dynamic tasks,26  thought to arise due to abnormal or poor neuromuscular control of the hip.27  The lack of adequate joint control may result in postural balance instability, with previous studies demonstrating the importance of hip and knee muscle performance in terms of posture and balance.7,28  For instance, Lee et al7  investigated hip muscle strength and COP excursion in females with PFP during a unipedal step down task. The authors found that females in the PFP group with diminished hip abductor strength also displayed increased mediolateral displacement of the COP compared with pain-free individuals with relatively stronger hip abductors. Findings indicate that the hip abductors play an important role in postural stability by controlling and minimizing the mediolateral acceleration of the COP, which keeps the COM above the support area during postural perturbations.29 

Future studies are required to increase understanding of the role of altered proximal mechanics and the influence on foot loading characteristics in individuals with PFP. If future studies demonstrate that COP is a reliable marker in PFP, then the COPI may be a useful clinical assessment criterion because of the ease with which it can be determined. Data acquisition that does not require extensive equipment or training may be attractive in clinical settings. In addition, assessment of COP during limb-loading tasks may have the potential to determine patients who may benefit from management strategies, including prescription of foot orthoses that aim to redistribute plantar pressure and correct COP displacement.

A limitation of this study is that the cross-sectional design limits the ability to establish cause and effect. It is difficult to determine whether altered COP parameters are important factors in the development of PFP or were a consequence of abnormal movement strategies exhibited as a response to symptoms associated with PFP. The results of this study are based on biomechanical data and have not assessed PFJt stress; therefore, it is not known what effect the presenting COP parameters would have on knee loading in individuals with PFP. In addition, the present study assessed COP during barefoot SLSs. Although this method permits more accurate assessment of the intrinsic foot mechanics, it does not take into account the effect of footwear on plantar-loading parameters during limb-loading tasks.

Conclusions

Patellofemoral pain may be associated with altered foot loading as assessed through COP parameters. Specifically, individuals with PFP exhibit increased COPI, consistent with increased lateral loading of the foot and greater vyCOP compared with noninjured control participants during SLSs. Changes in COP parameters may be part of a movement strategy aimed at reducing load transmission through the lower kinetic chain to the knee that may increase PFJt stress during activities such as SLSs (Collins et al, 2008). It may also be hypothesized that COP variations between groups may be related to reduced dynamic stability during SLS. Future studies are needed to provide further insight into COP parameters as a possible risk factor in the etiology of this condition.

Financial Disclosure: None reported.

Conflict of Interest: None reported.

References

References
1.
Boling
MC.
Padua
DA.
Marshall
SW.
et al:
Gender differences in the incidence and prevalence of patellofemoral pain syndrome
.
Scand J Med Sci Sports
20
:
725
,
2010
.
2.
McKenzie
K.
Galea
V.
Wessel
J.
et al:
Lower extremity kinematics of females with patellofemoral pain syndrome while stair stepping
.
J Orthop Sports Phys Ther
40
:
625
,
2010
.
3.
Crossley
KM.
van Middelkoop
M.
Callaghan
MJ.
et al:
2016 Patellofemoral pain consensus statement from the 4th international patellofemoral pain research retreat, Manchester. Part 2: recommend physical interventions (exercise, taping, bracing, foot orthoses and combined interventions)
.
Br J Sports Med
50
:
839
,
2016
.
4.
Cornwall
MW.
McPoil
TG.
Reliability and validity of center-of-pressure quantification
.
JAPMA
93
:
142
,
2003
.
5.
McPoil
TG.
Adrian
M.
Pidcoe
P.
Effects of foot orthoses on center-of-pressure patterns in women
.
Phys Ther
69
:
149
,
1989
.
6.
Thijs
Y.
Van Tiggelen
D.
Roosen
P.
et al:
A prospective study on gait-related intrinsic risk factors for patellofemoral pain
.
Clin J Sport Med
17
:
437
,
2007
.
7.
Lee
SP.
Souza
RB.
Powers
CM.
The influence of hip abductor muscle performance on dynamic postural stability in females with patellofemoral pain
.
Gait Posture
36
:
425
,
2012
.
8.
Barton
CJ.
Levinger
P.
Crossley
KM.
et al:
The relationship between rearfoot, tibial and hip kinematics in individuals with patellofemoral pain syndrome
.
Clin Biomech
27
:
702
,
2012
.
9.
Dowling
GJ.
Murley
GS.
Munteanu
SE.
et al:
Dynamic foot function as a risk factor for lower limb overuse injury: a systematic review
.
J Foot Ankle Res
7
:
53
,
2014
.
10.
Willems
TM.
De Clercq
D.
Delbaere
K.
et al:
A prospective study of gait related risk factors for exercise-related lower leg pain
.
Gait Posture
23
:
91
,
2006
.
11.
Willems
TM.
Vitvrouw
E.
De Cook
A.
et al:
Gait-related risk factors for exercise-related lower-leg pain during shod running
.
Med Sci Sports Exerc
39
:
330
,
2007
.
12.
Saad
MC.
Felicio
LR.
Masullo Cde
L.
et al:
Analysis of the center of pressure displacement, ground reaction force and muscular activity during step exercises
.
J Electromyogr Kinesiol
21
:
712
,
2011
.
13.
Gamble
F.
Yale
I
:
Clinical Foot Roentgenology
, Krieger, New York,
1975
.
14.
Wallace
DA.
GJ,
Salem
Salinas
R.
et al
:
Patellofemoral joint kinetics while squatting with and without an external load
.
J Orthop Sports Phys Ther 32: 141,
2002
.
15.
De Cock
A.
Vanrenterghem
J.
Willems
T.
et al:
The trajectory of the centre of pressure during barefoot running as a potential measure for foot function
.
Gait Posture
27
:
669
,
2008
.
16.
Scherer
PR.
Sobiesk
GA.
The center of pressure index in the evaluation of foot orthoses in shoes
.
Clin Podiatr Med Surg
11
:
355
,
1994
.
17.
Cavanagh
PR.
Rodgers
MM.
The arch index: a useful measure from footprints
.
J Biomech
20
:
547
,
1987
.
18.
Portney
L.
Watkins
M.
“Statistical Measures of Reliability,”
in
Foundations of Clinical Research: Applications to Practice
,
ed by
Cohen,
M
p
585
,
Pearson Education International
,
Upper Saddle River, NJ
,
2009
.
19.
Cornwall
MW.
McPoil
TG.
Relationship between static foot posture and foot mobility
.
J Foot Ankle Res
4
:
4
,
2011
.
20.
Duffey
MJ.
Martin
DF.
Cannon
DW.
et al:
Etiologic factors associated with anterior knee pain in distance runners
.
Med Sci Sports Exerc
32
:
1825
,
2000
.
21.
Powers
CM.
The influence of altered lower-extremity kinematics on patellofemoral joint dysfunction: a theoretical perspective
.
J Orthop Sports Phys Ther
33
:
639
,
2003
.
22.
Ghani Zadeh Hesar
N.
Van Ginckel
A.
Cools
A.
et al
:
A prospective study of the gait-related intrinsic risk factors for lower leg overuse injuries
.
Br J Sports Med
43
:
1057
,
2009
.
23.
Winter
DA.
Human balance and posture control during standing and walking
.
Gait Posture
3
:
193
,
1995
.
24.
Dionisio
VC.
Almeida
GL.
Duarte
M.
et al:
Kinematic, kinetic and EMG patterns during downward squatting
.
J Electromyogr Kinesiol
18
:
134
,
2008
.
25.
Shimokochi
Y.
Yong Lee
S.
Shultz
SJ.
et al:
The relationships among sagittal-plane lower extremity moments: implications for landing strategy in anterior cruciate ligament injury prevention
.
J Athl Train
44
:
33
,
2009
.
26.
Noehren
B.
Hamill
J.
Davis
I.
Prospective evidence for a hip etiology in patellofemoral pain
.
Med Sci Sports Exerc
45
:
1120
,
2013
.
27.
Reiman
M.
Bolgla
L.
Lorenz
D.
Hip functions influence on knee dysfunction: a proximal link to a distal problem
.
J Sport Rehabil
18
:
33
,
2009
.
28.
Citaker
S.
Kaya
D.
Yuksel
I.
et al:
Static balance in patients with patellofemoral pain syndrome
.
Sports Health
3
:
524
,
2011
.
29.
Powers
CM.
The influence of abnormal hip mechanics on knee injury: a biomechanical perspective
.
J Orthop Sports Phys Ther
40
:
42
,
2010
.