Brain Activity During Experimental Knee Pain and Its Relationship With Kinesiophobia in Patients With Patellofemoral Pain: A Preliminary Functional Magnetic Resonance Imaging Investigation

Abstract

Context:
The etiology of patellofemoral pain has remained elusive, potentially due to an incomplete understanding of how pain, motor control, and kinesiophobia disrupt central nervous system functioning.

Objective:
To directly evaluate brain activity during experimental knee pain and its relationship to kinesiophobia in patients with patellofemoral pain.

Design:
Cross-sectional.

Methods:
Young females clinically diagnosed with patellofemoral pain (n = 14; 14.4 [3.3] y; body mass index = 22.4 [3.8]; height = 1.61 [0.1] m; body mass = 58.4 [12.7] kg). A modified Clarke test (experimental pain condition with noxious induction via patella pressure and quadriceps contraction) was administered to the nondominant knee (to minimize limb dominance confounds) of patients during brain functional magnetic resonance imaging (fMRI) acquisition. Patients also completed a quadriceps contraction without application of external pressure (control contraction). Kinesiophobia was measured using the Tampa Scale of Kinesiophobia. The fMRI analyses assessed brain activation during the modified Clarke test and control contraction and assessed relationships between task-induced brain activity and kinesiophobia. Standard processing for neuroimaging and appropriate cluster-wise statistical thresholds to determine significance were applied to the fMRI data (z > 3.1, P < .05).

Results:
The fMRI revealed widespread neural activation in the frontal, parietal, and occipital lobes, and cerebellum during the modified Clarke test (all zs > 4.4, all Ps < .04), whereas neural activation was localized primarily to frontal and cerebellar regions during the control contraction test (all zs > 4.4, all Ps < .01). Greater kinesiophobia was positively associated with greater activity in the cerebello-frontal network for the modified Clarke test (all zs > 5.0, all Ps < .01), but no relationships between kinesiophobia and brain activity were observed for the control contraction test (all zs < 3.1, all Ps > .05).

Conclusions:
Our novel experimental knee pain condition was associated with alterations in central nociceptive processing. These findings may provide novel complementary pathways for targeted restoration of patient function.

Full text

Brain Activity During Experimental Knee Pain and Its Relationship
With Kinesiophobia in Patients With Patellofemoral Pain:
A Preliminary Functional Magnetic Resonance Imaging Investigation
Kim D. Barber Foss, Alexis B. Slutsky-Ganesh, Jed A. Diekfuss, Dustin R. Grooms,
Janet E. Simon, Daniel K. Schneider, Neeru Jayanthi, Joseph D. Lamplot, Destin Hill,
Mathew Pombo, Philip Wong, David A. Reiter, and Gregory D. Myer
Context: The etiology of patellofemoral pain has remained elusive, potentially due to an incomplete understanding of how pain,
motor control, and kinesiophobia disrupt central nervous system functioning. Objective: To directly evaluate brain activity
during experimental knee pain and its relationship to kinesiophobia in patients with patellofemoral pain. Design: Cross-sectional.
Methods: Young females clinically diagnosed with patellofemoral pain (n = 14; 14.4 [3.3] y; body mass index = 22.4 [3.8];
height = 1.61 [0.1] m; body mass = 58.4 [12.7] kg). A modified Clarke test (experimental pain condition with noxious induction
via patella pressure and quadriceps contraction) was administered to the nondominant knee (to minimize limb dominance
confounds) of patients during brain functional magnetic resonance imaging (fMRI) acquisition. Patients also completed a
quadriceps contraction without application of external pressure (control contraction). Kinesiophobia was measured using the
Tampa Scale of Kinesiophobia. The fMRI analyses assessed brain activation during the modified Clarke test and control
contraction and assessed relationships between task-induced brain activity and kinesiophobia. Standard processing for
neuroimaging and appropriate cluster-wise statistical thresholds to determine significance were applied to the fMRI data
(z>3.1, P<.05). Results: The fMRI revealed widespread neural activation in the frontal, parietal, and occipital lobes, and
cerebellum during the modified Clarke test (all zs>4.4, all Ps<.04), whereas neural activation was localized primarily to frontal
and cerebellar regions during the control contraction test (all zs>4.4, all Ps<.01). Greater kinesiophobia was positively
associated with greater activity in the cerebello-frontal network for the modified Clarke test (all zs>5.0, all Ps<.01), but no
relationships between kinesiophobia and brain activity were observed for the control contraction test (all zs<3.1, all Ps>.05).
Conclusions: Our novel experimental knee pain condition was associated with alterations in central nociceptive processing.
These findings may provide novel complementary pathways for targeted restoration of patient function.
Keywords:fMRI, noxious stimulation, fear of movement, corticocerebellar, sensorimotor, neural activation
Patellofemoral pain (PFP) is a condition characterized by retro-
patellar or peripatellar pain during everyday activities that load the
patellofemoral joint, such as jumping or walking.
1
The PFP is one of
the most common knee conditions in the general population,
affecting females more frequently than males and approximately
1 in 4 school-aged youths.
2
The PFP can result in disabling pain
symptoms,
3
limit physical activity for health promotion,
4–8
impede
activities of daily living,
9,10
and may contribute to patellofemoral
osteoarthritis.
11
Participation in athletic activities involving running,
jumping, and cutting may increase risk of developing PFP,
12,13
with
patients often exhibiting abnormal biomechanical movement pat-
terns during functional movement, particularly in the frontal and
transverse planes.
9,10,14
However, despite a relationship between
increased number of abnormal movement patterns and lower func-
tion and greater pain,
15
the current biomechanical literature is quite
conflicting and points to a more complex etiology of PFP.
16–18
In
fact, researchers have suggested that changes in brain function and
structure—that is, alterations to the central nervous system—may
underlie chronic musculoskeletal and movement-related pain,
19
but
has only recently been considered in patients with PFP.
20
Patients with PFP present with widespread hyperalgesia,
meaning that they experience a lower pain threshold extending
beyond the localized area of maximal pain.
21,22
Two common
clinical measures of pain threshold in PFP patients include condi-
tioned pain modulation (CPM) and temporal summations of pain.
In adolescent females, individuals with both active PFP and
recovered PFP demonstrate impaired CPM and a reduced pressure
pain threshold at the knee compared with controls with no previous
knee injury history.
21,23
Indeed, a recent systematic review and
meta-analysis of quantitative sensory testing and related pain
profiling methods concluded that patients with PFP exhibit altered
Barber Foss, Diekfuss, Jayanthi, Lamplot, Hill, Pombo, and Myer are with the Emory
Sport Performance and Research Center (SPARC), Flowery Branch, GA, USA.
Slutsky-Ganesh is with the Department of Kinesiology, University of North Carolina
at Greensboro, Greensboro, NC, USA. Diekfuss, Jayanthi, Lamplot, Hill, Pombo,
Wong, Reiter, and Myer are with the Department of Orthopaedics, Emory University
School of Medicine, Atlanta, GA, USA. Grooms and Simon are with the Ohio
Musculoskeletal & Neurological Institute, Ohio University, Athens, OH, USA; and
the Division of Athletic Training, School of Applied Health Sciences and Wellness,
College of Health Sciences and Professions, Ohio University, Athens, OH, USA.
Grooms is also with the Division of Physical Therapy, School of Rehabilitation and
Communication Sciences, College of Health Sciences and Professions, Ohio Univer-
sity, Athens, OH, USA. Schneider is with the Department of Radiology, University of
Michigan, MI, USA. Wong and Reiter are also with the Department of Radiology and
Imaging Sciences, Emory University School of Medicine, Atlanta, GA, USA. Barber,
Slutsky-Ganesh, Diekfuss Foss, Jayanthi, Hill, Pombo, and Myer are also with the
Emory Sports Medicine Center, Atlanta, GA, USA. Myer is also with the Micheli
Center for Sports Injury Prevention, Waltham, MA, USA. Barber Foss (Kim.
BarberFoss@emory.edu) is corresponding author.
1
Journal of Sport Rehabilitation, (Ahead of Print)
https://doi.org/10.1123/jsr.2021-0236
© 2022 Human Kinetics, Inc. ORIGINAL RESEARCH REPORT
First Published Online: Mar. 12, 2022

pain processing and central sensitization, particularly in younger
and female patients.
24
Importantly, even after PFP has been
resolved, these patients still experience altered CPM and pain
sensitivity, suggesting that minimal nociceptive input could elicit
pain episodes after PFP recovery.
23
The chronic nature of PFP and associated central pain sensiti-
zation are likely secondary to nervous system and psychological
changes that reinforce each other to alter pain and movement
perception. Derived from the fear avoidance model,
25
pain leads
to pain catastrophizing (exaggerated and ruminating negative
cognitions toward pain)
26
and kinesiophobia, which can lead to
musculoskeletal system deconditioning and more pain.
27,28
Despite
PFP typically being considered a biomechanical/strength-related
dysfunction,
29,30
kinesiophobia is often the underlying driver
related to movement dysfunction and pain (both perceived and
actual) than classic strength assessments.
31,32
For instance, high
kinesiophobia is related to perceived physical dysfunction and pain
intensity but is not related to any measure of muscular strength or
postural control.
31
Interestingly, some research has found that
patients with PFP report reduced kinesiophobia when wearing a
knee brace during their functional activities which may aid in
compliance with exercise intervention.
33
Kinesiophobia is further
associated with disrupted neural activity when participants view
movement-evoked pain (eg, watching a video of activities associ-
ated with pain)
34
; however, the influence of kinesiophobia on
neural activity during actual knee-related pain is unknown, with
no direct assessment of brain activity for patients with PFP
previously conducted. Isolating how kinesiophobia disrupts the
neural response to pain, specifically in PFP, may provide key
neurotherapeutic targets to support the development of novel
sensorimotor rehabilitation strategies (eg, real-time visual move-
ment biofeedback
35–37
) that may promote kinesiophobia-reducing
neuroplasticity and desirable movement biomechanics.
38–42
The purpose of this study was to directly evaluate brain
activity during experimental knee pain and its relationship to
kinesiophobia in patients with PFP. To accomplish this, we
administered 2 tests during brain functional magnetic resonance
imaging (fMRI) for patients with PFP: (1) a modified Clarke test
(experimental knee pain condition; noxious induction via experi-
menter patella pressure and participant quadriceps contraction
while in knee extension) and (2) a quadriceps contraction test to
serve as a control (participant quadriceps contraction while in knee
extension with no additive experimenter patella pressure). Overall
brain activity for each task was statistically modeled with their
participants’self-reported kinesiophobia to isolate brain regions
associated with pain and fear. Given the exploratory nature of this
study, no specific hypotheses were made a priori.
Methods
Participants
This study targeted enrollment of adolescent/young adult females
due to their increased likelihood for developing PFP compared with
males and older adults.
43
All patients were recruited through sports
medicine clinics at a pediatric hospital following a clinical diag-
nosis of PFP or anterior knee pain by a board-certified sports
medicine physician (MD, DO). Diagnostic criteria for study enroll-
ment: pain duration >2 months, pain most severe during physical
activity, and no prior history of knee-related trauma. Table 1
provides patient data, detailing patient age, body mass index,
most painful knee, and the duration of PFP symptoms (in months),
as well as individual patient reported outcomes from the Anterior
Knee Pain Scale, International Knee Documentation Committee
subjective knee form, and the Tampa scale of Kinesiophobia (TSK;
17-item version). Of note, these data were collected during the
research visit and not utilized as part of the clinical evaluation and
(or) diagnosis. Study-specific inclusion criteria also required the
patient be right hand dominant, and between the ages of 7 and
25 years old (final enrollment aged 11–24). Exclusion criteria
further included: history of neurological deficit(s) or severe head
trauma; history of structural anterior compartment pathology; and
contraindications for MRI scanning including braces or permanent
metal dental implants; insulin pump; cardiac pacemaker; cochlear
Table 1 Characteristics of the Patients With Patellofemoral Pain
Patient Age, y BMI AKPS IKDC TSK Most painful side Duration, mo
1 16 21.25 53 39.08 44 Left
a
12.00
2 16 22.15 89 72.41 41 Right 24.00
3 17 29.16 83 86.21 35 Left
a
12.00
4 13 22.93 71 56.32 44 Right 12.00
5 14 19.63 74 54.22 35 Left
a
54.00
6 12 20.27 87 89.66 28 Left
a
24.00
7 24 22.27 79 66.67 31 Right
a
96.00
8 14 18.58 n/a 43.68 43 Right
a
48.00
9
b
14 28.59 62 58.62 40 Right 24.00
10 12 29.88 67 45.98 40 Right
a
6.00
11 12 24.07 60 36.78 36 Left
a
9.00
12 11 18.60 88 65.52 50 Left 2.00
13 12 20.30 72 64.37 36 Left
a
6.00
14 14 26.07 63 71.26 39 Right 5.00
15 14 18.00 72 66.67 32 Right 2.50
Abbreviations: AKPS, Anterior Knee Pain Scale; BMI, body mass index; IKDC, International Knee Documentation Committee subjective knee form; n/a, did not complete
AKPS; TSK, Tampa scale of Kinesiophobia.
a
Reported regular bilateral knee pain leading up to study visit—most painful side noted on day of testing.
b
Removed from final analyses for being left leg dominant.
2Barber Foss et al
(Ahead of Print)

implant(s); hearing aid(s); aneurysm clips; or orthopedic pins,
wires, screws, or plates inserted within the last 6 months. A total
of 15 right-hand dominant participants between the ages of 11 and
24 were enrolled. For this study, one participant was excluded due
to being left leg dominant, resulting in 14 patients included in the
final analyses. Our rationale to exclude this patient was informed
by prior literature noting the influence of limb dominance on
hemispheric lateralization
44,45
; thus, our final analyses included
14 participants whom were all right-hand and right-leg dominant.
The Institutional Review Board at Cincinnati Children’s Hospital
Medical Center approved the present study (IRB number: 2017-
5776), and the study is registered on ClinicalTrials.gov (NCT
number: 04068883). Participant assent and parental/guardian con-
sent were obtained prior to study participation.
Procedure
At the testing visit, patients completed several knee-related ques-
tionnaires, watched a video demonstration of the testing paradigm,
and were familiarized with the fMRI tasks to be completed
(described below) using a mock MR scanner. During brain imag-
ing, each participant first completed a structural MRI for fMRI
image registration and then 2 fMRI tasks with their left limb. In this
preliminary study, we controlled for limb dominance, but included
participants with left, right, or bilateral PFP pain to complete the
modified Clarke test quadriceps contraction tests using their left
leg, only. Immediately following the fMRI modified Clarke and
contraction control tasks, while the patient was still positioned in
the MR scanner, each patient rated their level of pain intensity and
pain unpleasantness using visual analog scales (VAS).
Tampa Scale of Kinesiophobia
The TSK was used as our primary method to evaluate patient
kinesiophobia.
46
The original form of the TSK possesses a high
degree of internal consistency
47
and is related to measures of fear
avoidance, pain-related disability, and pain catastrophizing.
47
The
TSK scores range from 17 to 68, with scores ≥37 indicating high
kinesiophobia (<37 considered “low”kinesiophobia).
48
Neuroimaging Acquisition
All fMRI acquisitions were conducted on a Phillips 3 T Ingenia
scanner (Philips Medical Systems, Best, the Netherlands) with a
32-channel, phased array head coil. A magnetization-prepared
rapid gradient-echo sequence was used to acquire high resolution
3D T1-weighted images (sagittal): repetition time (TR) = 8.1 ms,
echo time (TE) = 3.7 ms; field of view = 256 ×256 mm; matrix =
256 ×256; slice thickness = 1 mm; number of slices = 180. Whole
brain echo-planar imaging (transverse) was completed for the
modified Clarke and quadriceps contraction control tests:
TR = 2000 ms; TE = 35 ms; field of view = 240 ×240 mm; slice
thickness = 5 mm; number of slices = 38.
Each fMRI acquisition lasted 2 minutes and 42 seconds and
comprised five 18-second rest blocks and four 18-second task
blocks; beginning with the rest block and alternating in a traditional
block design. During the modified Clarke test blocks (experimental
knee pain condition), the experimenter applied pressure anterior–
posteriorly, while pushing the patella distally and instructed the
participant to contract their quadriceps muscle every 3 seconds in
response to a visual prompt (6 contractions per test block and 24
total contractions). During the contraction control tasks and all rest
blocks, the experimenter cradled the superior aspect of the patella
between the thumb and index finger to control for sensory stimu-
lation.
49,50
For the contraction control test, no pressure was applied
by the experimenter, and the patient engaged in the same quadri-
ceps contraction.
Visual Analog Scale
Immediately following each of the fMRI tasks, pain intensity and
pain unpleasantness was rated with a plastic VAS (Parisian Nov-
elty, Chicago, IL).
51
The minimum rating (scored “0”) was repre-
sented as “no pain sensation”or “not at all unpleasant,”whereas the
maximum rating (scored “10”) was designated with “most intense
imaginable”or “most unpleasant imaginable,”with scale incre-
ments of one tenth. Prior to administration and entering the MR
scan room, patients were provided standardized instructions to
differentiate pain intensity versus pain unpleasantness following
established methods that have reported reasonable psychometric
properties for these scales in young adult populations.
52
Neuroimaging Preprocessing and Statistical
Analyses
Neuroimaging data processing and analyses were conducted using the
fMRI of the brain (FMRIB) software library (FSL version 6.00;
Oxford Centre for Functional MRI of the Brain, Nuffield Department
of Clinical Neurosciences, University of Oxford, Oxford, United
Kingdom). Brain extraction was completed with FSL’s brain extrac-
tion tool with the robust parameters.
53
Standard preprocessing and
appropriate cluster-wise statistical thresholds to determine signifi-
cance (z>3.1, P<.05) were applied to the neuroimaging data. Further
preprocessing was completed as set forth by Pruim et al,
54
with
utilization of independent components analysis for the automatic
removal of head motion artifact (ICA-AROMA). Prior to ICA-
AROMA, images were subjected to motion correction using FMRIB's
linear image registration tool (MCFLIRT),
55
slice-timing correction
(ascending), nonbrain removal using brain extraction tool, spatial
smoothing using a Gaussian kernel of 6 mm full-width half maxi-
mum, and intensity normalization. Each functional image was regis-
tered to its corresponding high-resolution T1 structural image and to a
standard space (MNI 152) using both linear and nonlinear registration
methods.
56
Preprocessed data were then inputted into ICA-AROMA,
which utilizes FSL’s multivariate exploratory linear decomposition
into independent components (MELODIC) tool to identify and
remove motion-related components. Following ICA-AROMA,
images were subjected to a high-pass filter (cutoff = 100 s), and
first-level analyses were completed to determine neural activity for
each condition relative to rest for each patient. The time series was
modeled with a block design, with one analysis of interest for each the
modified Clarke test and the quadriceps contraction test, assessing
increased activation during test blocks compared with rest blocks
within each patient using a whole-brain approach with a cluster
threshold of Z>3.1 and cluster significance threshold to correct
for multiple comparisons in the statistical parametric mapping of
P<.05.
57
A temporal derivative was included in the model, and FILM
(FMRIB’s improved linear model) prewhitening was applied to
reduce unexplained noise and improve estimation efficiency. Proces-
sing of data with ICA-AROMA and inspection of components and
first-level results were completed with use of INFOBAR, the Interface
for Batch processing data using ICA-AROMA.
58
To assess overall group level activation for both tasks (modi-
fied Clarke test and contraction control, performed independently),
FSL’s FLAME (local analysis of mixed effects) model stage 1 and
PFP and Kinesiophobia 3
(Ahead of Print)

stage 2 was used, with mean activation as the outcome variable
of interest. Then, demeaned TSK scores were entered into the
FLAME model to identify neural correlates (positive and/or nega-
tive associations between kinesiophobia and neural activity during
each task, independently). Blood oxygen level-dependent signal
change can be affected by local gray matter volume, and thus a
voxel-wise covariate for gray matter volume was included. Seg-
mentation was carried out with FAST (FMRIB’s automated seg-
mentation tool),
59
which resulted in partial volume estimate images
of gray matter, white matter, and cerebrospinal fluid for each
patient. Each patient’s gray matter partial volume estimate image
was registered to standard space using FLIRT and then smoothed to
the same extent of the functional images (6 mm full-width half
maximum). Images were then merged into a 4D image, which was
demeaned for use as a covariate.
To identify the influence of kinesiophobia on pain processing, a
pain activation map was created by isolating activity during the
modified Clarke test that was significantly higher than during the
contraction test (isolating experimental knee pain responses from
quadriceps contraction responses). This isolated brain activity was
used to identify neural correlates of kinesiophobia via the TSK scores.
This higher level analysis was completed in 2 steps. First, a fixed-
effects model was used to determine activation that was different
during the modified Clarke test compared with the contraction control
test within each patient. The resulting comparison of parameter
estimate images were then entered into a mixed-effects model
(FLAME model stage 1 and 2) with the demeaned TSK scores, as
was done above, to assess positive or negative correlations between
the contrasted modified Clarke test >contraction test (experimental
knee pain only) and kinesiophobia. All whole-brain analyses included
a cluster threshold of Z>3.1 and a cluster significance threshold of
P<.05.
57
The statistical approach and analyses we employed fol-
lowed acceptable standards for neuroimaging data analysis and
presentation,
60
while also build upon methods previously established
with high reliability for lower-extremity specific neuroimaging.
61
Results
Kinesiophobia and Pain Ratings
Patient demographics are reported in Table 1. On average, patients
had experienced symptomatic PFP for 18.2 (26.5) months, were
14.4 (3.3) years old (range = 11–24 y), and had a body mass index
of 22.4 (3.8), height (in meters) = 1.6 (0.1), and body mass (in
kilograms) = 58.4 (12.7).
Ratings of pain unpleasantness during the modified Clarke test
ranged from 0 to 2.60, with an average of 0.93 (0.96). Ratings of
pain intensity ranged from 0 to 3.80, with an average of 0.89 (1.02).
Ratings of pain unpleasantness and intensity were significantly >0
for the entire sample, t(13) = 3.63, P= .003 and t(13) = 3.21,
P= .006, respectively.
Ratings of pain unpleasantness during the contraction control
test ranged from 0 to 3.40, with an average of 0.86 (1.07). Ratings
of pain intensity ranged from 0 to 4.60, with an average of 0.71
(1.21). Ratings of pain unpleasantness and intensity were signifi-
cantly >0 for the entire sample, t(13) = 2.99, P= .003 and
t(13) = 2.21, P= .006, respectively.
Overall Task-Evoked Activation
During the modified Clarke test, there was greater activation in 7
clusters, which had peak activation over the: left central and frontal
opercular cortices; left cerebellum I to IV; right supramarginal and
postcentral gyri; left parietal operculum and planum temporale;
right lateral occipital cortex; right cerebellum VI and crus I; and left
cerebellum VIIb, VIIIa, and crus II (Figure 1; significant clusters
colored red).
During the contraction control test, there was greater activation
in 5 clusters, which had peak activation over the: left supplemen-
tary motor cortex and superior frontal gyrus; left inferior frontal and
precentral gyri; left cerebellum I to IV; and bilateral cerebellum VI
(Figure 1; significant clusters colored blue). Table 2provides
statistical details for each significant cluster.
Neural Correlates of Kinesiophobia
During the modified Clarke test, greater kinesiophobia was posi-
tively associated with greater neural activity in 2 clusters, with peak
activity localized within the: (1) right cerebellum crus II and crus I
and (2) left paracingulate gyrus and frontal pole (Figure 1; signifi-
cant clusters colored green). There were no significant relationships
between kinesiophobia and neural activity during the contraction
control test or during the modified Clarke test once the movement
aspect was contrasted out. Table 3provides statistical details for
each significant cluster.
Discussion
In response to the modified Clarke test (experimental knee pain),
brain activity was greater in cortical and cerebellar brain regions
associated with somatosensory, sensorimotor, and cognitive func-
tion. In response to the contraction control test, there was increased
activation in primarily sensorimotor cortical and cerebellar regions.
Greater kinesiophobia was also positively associated with greater
activation during the modified Clarke test in frontal and cerebellar
regions associated with higher order cognitive function and emo-
tion. However, kinesiophobia was not associated with increased
neural activation during the contraction control test independently,
or during the modified Clarke test when contrasted with the control
contraction (isolating experimental knee pain activity [experi-
menter pressure and participant quadriceps contraction] from
contraction-related activity). Thus, these data tentatively indicate
the neural correlates to kinesiophobia may be uniquely sensitive to
experimenter-evoked knee pain/pressure with movement and not
isolated movement (ie, quadriceps contraction only without addi-
tive patella pressure). Previous studies have found that the complex
physiologic processes underlying PFP include alterations in pain
processing including pressure hyperalgesia, increased temporal
summation of pain (TSP), impaired CPM, and lower pain pressure
threshold when compared with healthy individuals,
21,23
and our
results build on these findings by providing additional insight into
central nervous system function in patients with PFP, and particu-
larly how kinesiophobia manifests when moving under potentially
painful conditions.
To date, the neural mechanisms associated with PFP are mostly
unknown, or have been limited to resting state methods.
20
Broadly,
the neural signatures of pain typically include increased activation
of the primary and secondary somatosensory cortices, primary
and secondary motor cortices, the anterior cingulate cortex, insula,
thalamus, and prefrontal cortex.
62–65
Activation within this pain
network is often greater during spontaneous pain induction para-
digms for adult patients experiencing chronic pain (eg, osteoarthritis)
relative to healthy controls, possibly due to long-term peripheral
receptor activation resulting in hypersensitivity.
66,67
However,
4Barber Foss et al
(Ahead of Print)

traditional methods of pain induction typically overlook joint-spe-
cific pain and instead focus on spontaneouspain induction, which are
limited in extrapolation to musculoskeletal disorders, as pain pre-
sents with movement of the affected joint.
68
In the modified Clarke
and contraction control tests, there was some overlap of activation
with the common pain network (as described above; eg, postcentral
gyrus), which was interesting given that the contraction control was
completed without any additional patella-specific pressure. Despite
ratings of pain intensity and unpleasantness being similar between
the tests, even minimal nociceptive input can elicit pain in those
experiencing or recovering from chronic pain. Interestingly, the
central nervous response to experimental knee pain was the only
paradigm sensitive to kinesiophobia (modified Clarke test), poten-
tially indicating a downstream fear avoidance behavior that is
associated with long-term effects on physical activity and well-
being.
The results from this study show that higher kinesiophobia was
associated with increased activation of the posterior cerebellum
(ie, cerebellum crus II) and the prefrontal cortex (ie, paracingulate
gyrus and frontal pole) during the modified Clarke test. These regions
are part of a corticocerebellar circuit involved in higher order
cognition.
69,70
The functional connectivity of the cortical structures
and cerebellum (ie, corticocerebellar circuits) provide a basis for the
importance of the posterior cerebellum in movement-evoked pain.
Sensorimotor cortical regions project onto the anterior cerebellum for
motor control; whereas, association areas that combine information
from past experiences (eg, emotion, memory) with the present
environment project onto the posterior cerebellum to further integrate
information and formulate an appropriate motor response.
69
This
higher order cognitive appraisal of pain with movement through the
posterior cerebellum provides feedback through association areas and
limbic structures that can bias attention and emotion toward pain
perception
69
; perhaps having long-term effects on the chronic pain
cycle with downstream effects of the development of kinesiophobia,
maladaptive movement patterns, and/or injury risk.
Although both the modified Clarke and contraction control test
elicited some pain-related neural activation and did not differ on
perceived pain intensity and unpleasantness, only activation during
the modified Clarke test was associated with kinesiophobia.
Despite pain ratings, the contraction control test was designed
to assess neural activity during a movement only task, and when
contrasted with the modified Clarke task, to isolate pain neural
activity by removing movement activity. However, the presence of
some pain during the contraction control may have resulted in the
null correlation to TSK for the modified Clarke contrast with
contraction control contrast. Alternatively, it could be that kine-
siophobia is uniquely sensitive to movement-related pain and; thus,
the movement and pain aspects of the modified Clarke test were
necessary to identify the relation with TSK. Behaviorally, kine-
siophobia has been shown to be increased and positively correlates
with symptomology, anxiety, and perceived physical limita-
tions.
31,71,72
We note our Clarke test was “modified”to adhere
constraints associated with neuroimaging (eg, blocked design with
multiple contractions, slow and repetitive manipulations to appro-
priately assess the hemodynamic response function). Despite the
present findings being limited to task-based regional activity and
not necessarily reflective of between-region connectivity, the
identification of fear-related alterations within corticocerebellar
circuitry provides an important first step to guide future research.
For instance, the identified regions could serve as “seed”regions to
support hypothesis-driven psychophysiological interaction analy-
ses (task-based functional connectivity) of fMRI data derived from
our novel paradigms to test this novel hypothesis.
The findings of the present study have clinical implications for
PFP rehabilitation. Current therapy protocols typically focus on
strength and local pain relieving efforts but could be enhanced by
Figure 1 —Significant cluster of increased activation, overlaid on a standard MNI 152 template, observed during the modified Clarke test (red color
[darker cluster shading relative to rest of brain]; experimental knee pain condition), quadriceps contraction test (‘contraction control’; blue color [darker
cluster shading relative to rest of brain]), and positively associated with increased kinesiophobia (‘TAMPA correlate’; green color [darker cluster shading
relative to rest of brain]). Each column are different axial slices and of the same result for visualization, with the corresponding MNI coordinate in the z
(axial) plane for each slice noted in top row. The bottom row shows all areas of activation from the analyses ‘overlaid’on each other for visualization
purposes (see online color version for precise identification of task- and kinesiophobia-related activation maps). All clusters displayed were significant at a
cluster threshold of Z>3.1 and a cluster significance threshold of P<.05 after correcting for multiple comparisons.
PFP and Kinesiophobia 5
(Ahead of Print)

addressing underlying kinesiophobia, which may have a greater
influence on addressing movement impairment.
32,73,74
Traditional
approaches focused on strengthening the knee extensors and hip/
core may be successful in reducing pain in some cases, but it is
often difficult to predict who may or may not respond to these
programs.
75–78
The PFP rehabilitation programs may benefit from
integrating sensory feedback into motor and functional movement
patterns.
19,79
One such intervention that has been shown to help
reduce kinesiophobia and may aid in compliance with rehabilita-
tion is the use of a knee brace during functional activities,
33
with
prior findings potentially secondary to knee bracing positively
modulating neural activity.
80
While the approaches above may
improve function and pain in the short term, improvement in long-
term outcomes remains a primary goal. Psychological factors, such
Table 2 Overall Activation
Cluster index Brain regions Voxel Pvalue
Peak MNI voxel
Zstat-maxxyz
Overall activation during the modified Clarke test (experimental knee pain condition)
1 Bilateral
Central opercular cortex
Frontal opercular cortex
17,519 <.0001 −46 8 −2 8.06
2 Left
Cerebellum I–IV
1667 <.0001 −2−42 −12 5.39
3 Right
Supramarginal gyrus
Postcentral gyrus
918 <.0001 54 −22 30 5.13
4 Left
Parietal operculum
Planum temporale
594 <.0001 −54 −34 18 6.23
5 Right
Lateral occipital cortex
399 <.0001 48 −72 −2 4.74
6 Right
Cerebellum VI
Cerebellum crus I
270 .0012 36 −60 −26 5.07
7 Left
Cerebellum VIIb
Cerebellum VIIIa
Cerebellum crus II
139 .039 −22 −70 −48 4.4
Overall activation during the quadriceps contraction test
1 Bilateral
Supplementary motor cortex
Superior frontal gyrus
3343 <.0001 −2 4 64 9.2
2 Left
Inferior frontal gyrus
Precentral gyrus
3167 <.0001 −54 10 4 6.37
3 Left
Cerebellum I–IV
616 <.0001 −2−44 −20 5.14
4 Left
Cerebellum VI
Cerebellum crus I
191 <.0001 −28 −60 −28 4.77
5 Right
Cerebellum VI
Cerebellum crus I
151 <.0001 28 −68 −22 4.42
Table 3 Neural Correlates
Cluster index Brain regions Voxel Pvalue
Peak MNI voxel
Zstat-maxxy z
Greater activation during the modified Clarke test was positively associated with greater kinesiophobia
1 Right
Cerebellum crus II
Cerebellum crus I
265 .0009 14 −80 −32 5.03
2 Left
Paracingulate gyrus
Frontal pole
234 .0022 10 54 6 5.08
6Barber Foss et al
(Ahead of Print)

as anxiety, are associated with physical function and pain in
patients with PFP.
31
Understanding the role of psychological
factors and the interplay with both pain and function in patients
with PFP may inform rehabilitation strategies and improve long-
term outcomes.
Despite the novel contributions of this exploratory investiga-
tion, there are some limitations that should be considered. Notably,
the Clarke test is typically a clinical assessment tool
49,50
and
applying manual pressure to the patella is inherently variable
and examiner-dependent, impacting test reliability. However,
even after modifying the frequency of experimenter pressure
and required participant contractions (more than what is performed
clinically), we aimed to control for this variability by having one
researcher perform all testing to minimize interrater variance.
Future work utilizing a mechanical device to apply quantifiable
pressure may be more effective and accurate. While the Clarke test
has limited clinical efficacy, we consider our modified methods of
this test during fMRI one of the first to isolate patella-specific
pressure/pain concurrent with brain activity. Furthermore, while
this study did not utilize electromyography/kinetics to quantify
quadriceps contractions, the same tester did verify that all con-
tractions were completed during each testing block. Furthermore,
each patient completed the instructional training session prior to the
actual test, which should have mitigated some variation in muscle
contractions. An additional limitation is the use of the nondominant
(ie, left) limb for patients with PFP only, as patients self-reported
knee pain on day of testing ranging from left knee only (n = 3),
bilateral with greater left side pain (n = 5), right knee only (n = 4),
and bilateral with greater right side pain (n = 2). Therefore, 8
participants completed testing on the side they had reported
pain and 6 completed testing on the limb with self-reported less
pain. As patients with PFP consistently present with lower pain
thresholds nonspecific to the involved knee, demonstrating central
sensitization (ie, alterations in the central nervous system is not
localized to the pain site
81,82
). Thus, we prioritized controlling for
known lateralization effects within fMRI data, using the same limb
for all participations, rather than utilizing the limb with the highest
self-reported pain. Further supporting the central sensitization
hypothesis related to PFP, significant pain >0 was reported during
the modified Clarke test (all left limb). However, we emphasize
these data, and our interpretation, should be considered cautiously
as mean self-reported VAS pain intensity and unpleasantness
during the experimental pain condition was <1.0/10. Future
research should aim to also evaluate patients’pain at baseline to
support both diagnostic evaluation and clinically meaningful inter-
pretation of paradigm-induced changes in VAS outcome mea-
sures.
1,83,84
We also did not have a control group; however,
kinesiophobia would likely not be present in a pain-free control
cohort; thus, we purposefully utilized methods optimized for
within-group and between-condition analyses. A final limitation
is related to the sample population, which consisted of a small
(n = 14) right-handed, right-leg dominant group across a wide age
spectrum (ie, adolescents—young adults) and substantial variabil-
ity in self-reported pain chronicity (mean duration of symptoms
18.2 [26.5] mo). Therefore, the generalizability of the present
results should be considered cautiously, warranting future research
with larger samples and more refined inclusion/exclusion criteria to
inform larger, randomized controlled trials. Likewise, future
research utilizing fMRI to discover neural mechanisms of pain
and fear in patients with PFP should utilize consensus reporting
standards to better synthesize outcomes across studies for this
population (eg, REPORT-PFP).
85
However, such guidelines,
specifically recommendations for results reporting (specifically
items 9–11), will require modification to align with current report-
ing standards for quantitative neuroimaging (eg, Committee on
Best Practices in Data Analysis and Sharing [COBIDAS] report).
60
Conclusions
In patients with PFP, both the modified Clarke test (experimental
knee pain) and contraction control tests induced activation in well-
established pain-related brain regions. However, kinesiophobia
was only associated with activation during our novel experimental
knee pain condition—specifically, activation of the posterior cere-
bellum and frontal regions. These regions are involved in a
corticocerebellar circuit important for emotion and higher order
cognition, and these findings indicate that experimental knee pain
elicits a feedback cycle that can bias attention and emotion toward
pain processing, resulting in negative downstream effects including
a more rapid onset of kinesiophobia. However, we emphasize our
interpretation of the findings are limited by the preliminary nature
of this study, warranting future research with more diverse sam-
pling to augment clinical translation.
References
1. Willy RW, Hoglund LT, Barton CJ, et al. Patellofemoral pain. J
Orthop Sports Phys Ther. 2019;49(9):Cpg1–Cpg95. PubMed ID:
31475628 doi:10.2519/jospt.2019.0302
2. Smith BE, Selfe J, Thacker D, et al. Incidence and prevalence of
patellofemoral pain: a systematic review and meta-analysis. PLoS
One. 2018;13(1):e0190892. PubMed ID: 29324820 doi:10.1371/
journal.pone.0190892
3. Smith BE, Moffatt F, Hendrick P, et al. The experience of living with
patellofemoral pain—loss, confusion and fear-avoidance: a UK
qualitative study. BMJ Open. 2018;8(1):e018624. PubMed ID:
29362256 doi:10.1136/bmjopen-2017-018624
4. Blond L, Hansen L. Patellofemoral pain syndrome in athletes: a 5.7-
year retrospective follow-up study of 250 athletes. Acta Orthop Belg.
1998;64(4):393–400. PubMed ID: 9922542
5. Myer GD, Ford KR, Di Stasi SL, Foss KDB, Micheli LJ, Hewett TE.
High knee abduction moments are common risk factors for patello-
femoral pain (PFP) and anterior cruciate ligament (ACL) injury in
girls: is PFP itself a predictor for subsequent ACL injury? Br J Sports
Med. 2015;49(2):118–122. PubMed ID: 24687011 doi:10.1136/
bjsports-2013-092536
6. Myer GD, Ford KR, Foss KDB, et al. The incidence and potential
pathomechanics of patellofemoral pain in female athletes. Clin
Biomech. 2010;25(7):700–707. PubMed ID: 20466469 doi:10.
1016/j.clinbiomech.2010.04.001
7. Crossley KM, Cowan SM, McConnell J, Bennell KL. Physical
therapy improves knee flexion during stair ambulation in patellofe-
moral pain. Med Sci Sports Exerc. 2005;37(2):176–183. PubMed ID:
15692311 doi:10.1249/01.MSS.0000152676.13197.49
8. Heintjes EM, Berger M, Bierma-Zeinstra SM, Bernsen RM, Verhaar
JA, Koes BW. Exercise therapy for patellofemoral pain syndrome.
Cochrane Database Syst Rev. 2008;4. doi:10.1002/14651858.
CD003470.pub2
9. Souza RB, Powers CM. Differences in hip kinematics, muscle
strength, and muscle activation between subjects with and without
patellofemoral pain. J Orthop Sports Phys Ther. 2009;39(1):12–19.
doi:10.2519/jospt.2009.2885
PFP and Kinesiophobia 7
(Ahead of Print)

10. Willson JD, Davis IS. Lower extremity mechanics of females with
and without patellofemoral pain across activities with progressively
greater task demands. Clin Biomech. 2008;23(2):203–211. PubMed
ID: 17942202 doi:10.1016/j.clinbiomech.2007.08.025
11. Utting M, Davies G, Newman J. Is anterior knee pain a predisposing
factor to patellofemoral osteoarthritis? The knee. 2005;12(5):362–
365. PubMed ID: 16146626 doi:10.1016/j.knee.2004.12.006
12. Witvrouw E, Lysens R, Bellemans J, Cambier D, Vanderstraeten G.
Intrinsic risk factors for the development of anterior knee pain in an
athletic population. Am J Sports Med. 2000;28(4):480–489. PubMed
ID: 10921638 doi:10.1177/03635465000280040701
13. Loudon JK, Gajewski B, Goist-Foley HL, Loudon KL. The effec-
tiveness of exercise in treating patellofemoral-pain syndrome. J Sport
Rehabil. 2004;13(4):323–342. doi:10.1123/jsr.13.4.323
14. Souza RB, Powers CM. Predictors of hip internal rotation during
running. Am J Sports Med. 2009;37(3):579–587. PubMed ID:
19098153 doi:10.1177/0363546508326711
15. Ferrari D, Briani RV, de Oliveira Silva D, et al. Higher pain level and
lower functional capacity are associated with the number of altered
kinematics in women with patellofemoral pain. Gait Posture.
2018;60:268–272.
16. Petersen W, Rembitzki I, Liebau C. Patellofemoral pain in athletes.
Open Access J Sports Med. 2017;8:143. PubMed ID: 28652829
doi:10.2147/OAJSM.S133406
17. Rathleff MS, Rathleff C, Crossley K, Barton CJ. Is hip strength a risk
factor for patellofemoral pain? A systematic review and meta-analy-
sis. Br J Sports Med. 2014;48(14):1088.
18. de Oliveira Silva D, Barton CJ, Pazzinatto MF, Briani RV, de
Azevedo FM. Proximal mechanics during stair ascent are more
discriminate of females with patellofemoral pain than distal mechan-
ics. Clin Biomech. 2016;35:56–61. PubMed ID: 27128766 doi:10.
1016/j.clinbiomech.2016.04.009.
19. Silfies SP, Vendemia JM, Beattie PF, Stewart JC, Jordon M. Changes
in brain structure and activation may augment abnormal movement
patterns: an emerging challenge in musculoskeletal rehabilitation.
Pain Medicine. 2017;18(11):2051–2054.
20. Diekfuss JA, Grooms DR, Nissen KS, et al. Does central nervous
system dysfunction underlie patellofemoral pain in young females?
A brain functional connectivity with patient outcomes analysis. J
Orthop Res. 2021;1–14. doi:10.1002/jor.25152
21. Holden S, Straszek CL, Rathleff MS, Petersen KK, Roos EM,
Graven-Nielsen T. Young females with long-standing patellofemoral
pain display impaired conditioned pain modulation, increased tem-
poral summation of pain, and widespread hyperalgesia. Pain.
2018;159(12):2530–2537. PubMed ID: 30074593 doi:10.1097/j.
pain.0000000000001356
22. Pazzinatto MF, de Oliveira Silva D, Barton C, Rathleff MS, Briani
RV, de Azevedo FM. Female adults with patellofemoral pain are
characterized by widespread hyperalgesia, which is not affected
immediately by patellofemoral joint loading. Pain Med. 2016;17(10):
1953–1961. PubMed ID: 27113220 doi:10.1093/pm/pnw068
23. Holden S, Rathleff M, Thorborg K, Holmich P, Graven-Nielsen T.
Mechanistic pain profiling in young adolescents with patellofemoral
pain before and after treatment: a prospective cohort study. Pain.
2020;161(5):1065–1071. doi:10.1097/j.pain.0000000000001796
24. Bartholomew C, Lack S, Neal B. Altered pain processing and
sensitisation is evident in adults with patellofemoral pain: a system-
atic review including meta-analysis and meta-regression. Scand J
Pain. 2019;20(1):11–27. PubMed ID: 31560652 doi:10.1515/sjpain-
2019-0079
25. Lethem J, Slade P, Troup J, Bentley G. Outline of a fear-avoidance
model of exaggerated pain perception—I. Behav Res Ther. 1983;
21(4):401–408. PubMed ID: 6626110 doi:10.1016/0005-7967(83)
90009-8
26. Leung L. Pain catastrophizing: an updated review. Indian J Psychol
Med. 2012;34(3):204. PubMed ID: 23441031 doi:10.4103/0253-
7176.106012
27. Leeuw M, Goossens ME, Linton SJ, Crombez G, Boersma K,
Vlaeyen JW. The fear-avoidance model of musculoskeletal pain:
current state of scientific evidence. J Behav Med. 2007;30(1):77–94.
PubMed ID: 17180640 doi:10.1007/s10865-006-9085-0
28. Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in
chronic musculoskeletal pain: a state of the art. Pain. 2000;85(3):
317–332. PubMed ID: 10781906 doi:10.1016/S0304-3959(99)
00242-0
29. Näslund J, Näslund U-B, Odenbring S, Lundeberg T. Comparison of
symptoms and clinical findings in subgroups of individuals with
patellofemoral pain. Physiother Theory Pract. 2006;22(3):105–118.
PubMed ID: 16848349 doi:10.1080/09593980600724246
30. Powers CM. The influence of altered lower-extremity kinematics on
patellofemoral joint dysfunction: a theoretical perspective. J Orthop
Sports Phys Ther. 2003;33(11):639–646. doi:10.2519/jospt.2003.33.
11.639
31. Piva SR, Fitzgerald GK, Irrgang JJ, et al. Associates of physical
function and pain in patients with patellofemoral pain syndrome. Arch
Phys Med Rehabil. 2009;90(2):285–295. PubMed ID: 19236982
doi:10.1016/j.apmr.2008.08.214
32. de Oliveira Silva D, Barton CJ, Briani RV, et al. Kinesiophobia, but
not strength is associated with altered movement in women with
patellofemoral pain. Gait Posture. 2019;68:1–5. PubMed ID:
30408709 doi:10.1016/j.gaitpost.2018.10.033
33. Priore LB, Lack S, Garcia C, Azevedo FM, de Oliveira Silva D. Two
weeks of wearing a knee brace compared with minimal intervention
on kinesiophobia at 2 and 6 weeks in people with patellofemoral pain:
a randomized controlled trial. 2020;101(4):613–623.
34. Meier ML, Stämpfli P, Vrana A, Humphreys BK, Seifritz E, Hotz-
Boendermaker S. Neural correlates of fear of movement in patients
with chronic low back pain vs. pain-free individuals. Front Hum
Neurosci. 2016;10:386. PubMed ID: 27507941
35. Bonnette S, DiCesare CA, Kiefer AW, et al. Injury risk factors
integrated into self-guided real-time biofeedback improves high-
risk biomechanics. J Sport Rehabil. 2019;28(8):831–839. doi:10.
1123/jsr.2017-0391
36. Bonnette S, DiCesare CA, Kiefer AW, et al. A technical report on the
development of a real-time visual biofeedback system to optimize
motor learning and movement deficit correction. J Sports Sci Med.
2020;19:84–94. PubMed ID: 32132831
37. Bonnette S, DiCesare CA, Diekfuss JA, et al. Advancing anterior
cruciate ligament injury prevention using real-time biofeedback for
amplified sensorimotor integration. J Athl Train. 2019;54(9):985–
986. PubMed ID: 31437016 doi:10.4085/1062-6050-54.083
38. Grooms DR, Kiefer AW, Riley MA, et al. Brain-behavior mechan-
isms for the transfer of neuromuscular training adaptions to simulated
sport: initial findings from the train the brain project. J Sport Rehabil.
2018;27(5):1–5. PubMed ID: 29584523 doi:10.1123/jsr.2017-0241
39. Diekfuss JA, Grooms DR, Bonnette S, et al. Real-time biofeedback
integrated into neuromuscular training reduces high-risk knee bio-
mechanics and increases functional brain connectivity: a preliminary
longitudinal investigation. Psychophysiology. 2020;57(5):e13545.
PubMed ID: 32052868 doi:10.1111/psyp.13545
40. Diekfuss JA, Bonnette S, Hogg JA, et al. Practical training strategies
to apply neuro-mechanistic motor learning principles to facilitate
adaptations towards injury-resistant movement in youth. J Sci Med
Sport. 2020;3(1):3–16. doi:10.1007/s42978-020-00083-0
8Barber Foss et al
(Ahead of Print)

41. Diekfuss JA, Grooms DR, Hogg JA, et al. Targeted application of
motor learning theory to leverage youth neuroplasticity for enhanced
injury-resistance and exercise performance: OPTIMAL PREP. J Sci
Med Sport. 2020;3(1):17–36. doi:10.1007/s42978-020-00085-y
42. Diekfuss JA, Hogg JA, Grooms DR, et al. Can we capitalize on
central nervous system plasticity in young athletes to inoculate
against injury? J Sci Med Sport. 2020;2(4):305–318. doi:10.1007/
s42978-020-00080-3
43. Smith BE, Selfe J, Thacker D, et al. Incidence and prevalence of
patellofemoral pain: a systematic review and meta-analysis. 2018;
13(1):e0190892. doi:10.1371/journal.pone.0190892
44. Kapreli E, Athanasopoulos S, Papathanasiou M, et al. Lateralization
of brain activity during lower limb joints movement. An fMRI study.
NeuroImage. 2006;32(4):1709–1721. doi:10.1016/j.neuroimage.
2006.05.043
45. Sainburg RLJE. Handedness: differential specializations for control
of trajectory and position. 2005;33(4):206–213.
46. Miller RP, Kori SH, Todd DD. The Tampa scale: a measure of
kinisophobia. Clin J Pain. 1991;7(1):51. doi:10.1097/00002508-
199103000-00053
47. French DJ, France CR, Vigneau F, French JA, Evans RT. Fear of
movement/(re)injury in chronic pain: a psychometric assessment of
the original English version of the Tampa scale for kinesiophobia
(TSK). Pain. 2007;127(1–2):42–51. PubMed ID: 16962238 doi:10.
1016/j.pain.2006.07.016
48. Vlaeyen JW, Kole-Snijders AM, Rotteveel AM, Ruesink R, Heuts
PHTG. The role of fear of movement/(re)injury in pain disability. J
Occup Rehabil. 1995;5(4):235–252. doi:10.1007/BF02109988
49. Doberstein ST. Validity of Clarke sign in assessing anterior knee pain.
Gundersen Lutheran Medical Journal. 2005;3(2):51.
50. Doberstein ST, Romeyn RL, Reineke DM. The diagnostic value of
the clarke sign in assessing chondromalacia patella. J Athl Train.
2008;43(2):190–196. PubMed ID: 18345345 doi:10.4085/1062-
6050-43.2.190
51. Price DD, Bush FM, Long S, Harkins SW. A comparison of pain
measurement characteristics of mechanical visual analogue and
simple numerical rating scales. Pain. 1994;56(2):217–226. PubMed
ID: 8008411 doi:10.1016/0304-3959(94)90097-3
52. Stinson JN, Kavanagh T, Yamada J, Gill N, Stevens BJP. Systematic
review of the psychometric properties, interpretability and feasibility
of self-report pain intensity measures for use in clinical trials in
children and adolescents. Pain. 2006;125(1–2):143–157. doi:10.
1016/j.pain.2006.05.006
53. Smith SM. Fast robust automated brain extraction. Hum Brain Mapp.
2002;17(3):143–155. PubMed ID: 12391568 doi:10.1002/hbm.10062
54. Pruim RH, Mennes M, van Rooij D, Llera A, Buitelaar JK, Beckmann
CF. ICA-AROMA: a robust ICA-based strategy for removing motion
artifacts from fMRI data. Neuroimage. 2015;112:267–277. PubMed
ID: 25770991 doi:10.1016/j.neuroimage.2015.02.064
55. Jenkinson M, Bannister P, Brady M, Smith S. Improved optimization
for the robust and accurate linear registration and motion correction of
brain images. Neuroimage. 2002;17(2):825–841. PubMed ID:
12377157 doi:10.1006/nimg.2002.1132
56. Andersson JL, Jenkinson M, Smith S. Non-linear registration aka
Spatial normalisation FMRIB Technial Report TR07JA2. Oxford,
UK: FMRIB Analysis Group of the University of Oxford. 2007:1–22.
57. Worsley KJ. Statistical analysis of activation images. Functional
MRI: An Introduction to Methods. 2001;14(1):251–270.
58. Anand M, Diekfuss JA, Slutsky-Ganesh AB, Bonnette S, Grooms DR,
Myer GD. Graphical interface for automated management of motion
artifact within fMRI acquisitions: INFOBAR. SoftwareX. 2020;12:
100598. PubMed ID: 33447655 doi:10.1016/j.softx.2020.100598
59. Zhang Y, Brady M, Smith S. Segmentation of brain MR images
through a hidden Markov random field model and the expectation-
maximization algorithm. IEEE Trans Med Imaging. 2001;20(1):45–
57. PubMed ID: 11293691 doi:10.1109/42.906424
60. Nichols TE, Das S, Eickhoff SB, et al. Best practices in data analysis
and sharing in neuroimaging using MRI. Nat Neurosci. 2017;20(3):
299–303. doi:10.1038/nn.4500
61. Grooms DR, Diekfuss JA, Ellis JD, et al. A novel approach to
evaluate brain activation for lower extremity motor control. J Neu-
roimaging. 2019;29(5):580–588. PubMed ID: 31270890 doi:10.
1111/jon.12645
62. Leknes S, Tracey I. A common neurobiology for pain and pleasure.
Nat Rev Neurosci. 2008;9(4):314–320. PubMed ID: 18354400
doi:10.1038/nrn2333
63. Coghill RC, McHaffie JG, Yen YF. Neural correlates of interindi-
vidual differences in the subjective experience of pain. Proc Natl
Acad Sci USA. 2003;100(14):8538–8542. doi:10.1073/pnas.14306
84100
64. Brown JE, Chatterjee N, Younger J, Mackey S. Towards a physiol-
ogy-based measure of pain: patterns of human brain activity distin-
guish painful from non-painful thermal stimulation. PLoS One.
2011;6(9):e24124. PubMed ID: 21931652 doi:10.1371/journal.
pone.0024124
65. Coghill RC, Talbot JD, Evans AC, et al. Distributed processing of
pain and vibration by the human brain. J Neurosci. 1994;14(7):4095–
4108. doi:10.1523/JNEUROSCI.14-07-04095.1994
66. Gwilym SE, Keltner JR, Warnaby CE, et al. Psychophysical and
functional imaging evidence supporting the presence of central
sensitization in a cohort of osteoarthritis patients. Arthritis Care
Res. 2009;61(9):1226–1234. PubMed ID: 19714588 doi:10.1002/
art.24837
67. Sofat N, Smee C, Hermansson M, et al. Functional MRI demonstrates
pain perception in hand osteoarthritis has features of central pain
processing. J Biomed Graph Comput. 2013;3(4):1–10. doi:10.5430/
jbgc.v3n4p20
68. Corbett DB, Simon CB, Manini TM, George SZ, Riley JLIII, Fill-
ingim RB. Movement-evoked pain: transforming the way we under-
stand and measure pain. Pain. 2019;160(4):757–761. PubMed ID:
30371555 doi:10.1097/j.pain.0000000000001431
69. Habas C. Cerebellar closed-loops. In: Essentials of cerebellum and
cerebellar disorders. Springer; 2016:343–347.
70. Bernard JA, Orr JM, Mittal VA. Differential motor and prefrontal
cerebello-cortical network development: evidence from multimodal
neuroimaging. NeuroImage. 2016;124:591–601. PubMed ID:
26391125 doi:10.1016/j.neuroimage.2015.09.022
71. Maclachlan LR, Collins NJ, Matthews ML, Hodges PW, Vicenzino
B. The psychological features of patellofemoral pain: a systematic
review. Br J Sports Med. 2017;51(9):732–742. PubMed ID:
28320733 doi:10.1136/bjsports-2016-096705
72. Priore LB, Azevedo FM, Pazzinatto MF, et al. Influence of kinesio-
phobia and pain catastrophism on objective function in women with
patellofemoral pain. Phys Ther Sport. 2019;35:116–121. PubMed ID:
30529861 doi:10.1016/j.ptsp.2018.11.013
73. Crossley KM, van Middelkoop M, Barton CJ, Culvenor AG. Rethink-
ing patellofemoral pain: prevention, management and long-term
consequences. Best Pract Res Clin Rheumatol. 2019;33(1):48–65.
doi:10.1016/j.berh.2019.02.004
74. Amin M, Esfandiarpour F, Soleimani F, Helalat Z, Derisfard F,
Neurozi S. Relationship between lower extremity kinematics and
muscle strength, pain, physical activity level, and functional status in
females with patellofemoral pain. J Rehabil Sci Res. 2019;6(3):130–
136. doi:10.30476/JRSR.2019.82432.1030
PFP and Kinesiophobia 9
(Ahead of Print)

75. Bolgla LA, Boling MC. An update for the conservative management
of patellofemoral pain syndrome: a systematic review of the literature
from 2000 to 2010. Int J Sports Phys Ther. 2011;6(2):112. PubMed
ID: 21713229
76. Clark D, Downing N, Mitchell J, Coulson L, Syzpryt E, Doherty M.
Physiotherapy for anterior knee pain: a randomised controlled trial.
Ann Rheum Dis. 2000;59(9):700–704. PubMed ID: 10976083 doi:10.
1136/ard.59.9.700
77. Kannus P, Natri A, Paakkala T, JÄrvinen M. An outcome study of
chronic patellofemoral pain syndrome. seven-year follow-up of
patients in a randomized, controlled trial. J Bone Joint Surg.
1999;81(3):355–363. doi:10.2106/00004623-199903000-00007
78. Thomeé R. A comprehensive treatment approach for patellofemoral
pain syndrome in young women. Phys Ther. 1997;77(12):1690–
1703. PubMed ID: 9413448 doi:10.1093/ptj/77.12.1690
79. Roussel NA, Nijs J, Meeus M, Mylius V, Fayt C, Oostendorp R.
Central sensitization and altered central pain processing in chronic
low back pain: fact or myth? Clin J Pain. 2013;29(7):625–638.
PubMed ID: 23739534 doi:10.1097/AJP.0b013e31826f9a71
80. Thijs Y, Vingerhoets G, Pattyn E, Rombaut L, Witvrouw EJK. Does
bracing influence brain activity during knee movement: an fMRI
study. 2010;18(8):1145–1149. doi:10.1007/s00167-009-1012-9
81. Maclachlan LR, Collins NJ, Hodges PW, Vicenzino B. Psychological
and pain profiles in persons with patellofemoral pain as the primary
symptom. Eur J Pain. 2020;24(6):1182–1196. PubMed ID:
32223042 doi:10.1002/ejp.1563
82. Bartholomew C, Lack S, Neal B. Altered pain processing and
sensitisation is evident in adults with patellofemoral pain: a system-
atic review including meta-analysis and meta-regression. Scand J
Pain. 2020;20(1):11–27. doi:10.1515/sjpain-2019-0079
83. Crossley KM, Bennell KL, Cowan SM, Green S. Analysis of outcome
measures for persons with patellofemoral pain: which are reliable and
valid? Arch Phys Med Rehabil. 2004;85(5):815–822. PubMed ID:
15129407 doi:10.1016/S0003-9993(03)00613-0
84. Crossley KM, Stefanik JJ, Selfe J, et al. 2016 Patellofemoral pain
consensus statement from the 4th International Patellofemoral Pain
Research Retreat, Manchester. Part 1: terminology, definitions, clinical
examination, natural history, patellofemoral osteoarthritis and patient-
reported outcome measures. Br J Sports Med. 2016;50(14):839–843.
PubMed ID: 27343241 doi:10.1136/bjsports-2016-096384
85. Barton CJ, Silva DDO, Morton S, et al. REPORT-PFP: a consensus
from the International Patellofemoral Research Network to improve
REPORTing of quantitative PatelloFemoral Pain studies. Br J Sports
Med. 2021;55(20):1135–1143. PubMed ID: 34127482
10 Barber Foss et al
(Ahead of Print)