Manual Terapy in SDFP

[ RESEARCH REPORT ]
CHRISTINE A. IVERSON, PT, DPT¹J>EC7I=$IKJB?L;"PT, PhD, OCS²C?9>7;BI$9HEM;BB"PT, DPT³

H;8;997B$CEHH;BB"PT, DPT4C7JJ>;MM$F;HA?DI"PT, DPT5C7JJ>;M8$=7H8;H"PT, DScPT, OCS, FAAOMPT6
@EI;<>$CEEH;"PT, PhD, SCS, ATC7HE8;HJI$M7?DD;H"PT, PhD, ECS, OCS, FAAOMPT8
Lumbopelvic Manipulation for the

Treatment of Patients With Patellofemoral
Pain Syndrome: Development of a
Clinical Prediction Rule
atellofemoral pain syn- iliac joint (SIJ) or lumbopelvic region knees of patients with PFPS. Similarly,
P
Downloaded from www.jospt.org at on April 12, 2022. For personal use only. No other uses without permission.
drome (PFPS) is a sig- manipulation led to a significant decrease Hillermann et al28 reported that quad-
in quadriceps inhibition in the involved riceps muscle strength increased sig-
nificant clinical problem
and one of the most com-
TIJK:O:;I?=D0 Prospective cohort/predictive if there was any change in pain ratings. An im-
mon knee disorders in young,
Copyright © 2008 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
validity study. mediate overall 50% or greater reduction in pain,

active individuals.13,52,63,64,80 De- TE8@;9J?L;0 To determine the predictive valid- or moderate or greater improvement on a global
spite its prevalence, the etiol- ity of selected clinical exam items and to develop rating of change questionnaire, was considered a
a clinical prediction rule (CPR) to determine which treatment success. Likelihood ratios (LRs) were
ogy of PFPS is not well under- calculated to determine which examination items
patients with patellofemoral pain syndrome (PFPS)
stood.12,37,41,52,59,72,79 It has been have a positive immediate response to lumbopel- were most predictive of treatment outcome.
proposed that abnormal neuro- vic manipulation. TH;IKBJI0 Data for 49 subjects were included
muscular8,9,11,22,41,54 and biome- T879A=HEKD:0 Quadriceps muscle function in the data analysis, of which 22 (45%) had a
in patients with PFPS was recently shown to successful outcome. Five predictor variables were
chanical5,23,34,40,51,62,73 factors alter identified. The most powerful predictor of treatment
Journal of Orthopaedic & Sports Physical Therapy®
improve following treatment with lumbopelvic

patellar tracking and contribute manipulation. No previous study has determined success was a side-to-side difference in hip internal
to increased patellofemoral joint if individuals with PFPS experience symptomatic rotation range of motion greater than 14° (+LR, 4.9).
relief of activity-related pain immediately following If this variable was present, the chance of experi-
contact pressures that ultimately this manipulation technique. encing a successful outcome with the lumbopelvic
lead to pain and dysfunction.16,52,72 TC;J>E:I7D:C;7IKH;I0 Fifty subjects
manipulation improved from 45% to 80%.
Because the etiology is unclear, and (26 male, 24 female; age range, 18-45 years) with T9ED9BKI?ED0 A CPR was developed to predict
due to variations in the clinical presen- PFPS underwent a standardized history and physi- an immediate successful response to lumbopelvic
cal examination. After the evaluation, each subject manipulation in patients with PFPS. However,
tation of patients with PFPS, numerous
performed 3 typically pain-producing functional in light of a limited sample size and potential
nonoperative interventions activities (squatting, stepping up a 20-cm step, omission of meaningful predictor variables, future
have been proposed for the and stepping down a 20-cm step). The pain level studies are necessary to validate the CPR.
disorder. A common goal of perceived during each activity was rated on a
TB;L;BE<;L?:;D9;0 Prognosis, level 2b.
most treatment regimens numerical pain scale (0 representing no pain and
J Orthop Sports Phys Ther 2008;38(6):297-312.
for PFPS is the restoration 10 the worst possible pain). Following the assess-
doi:10.2519/jospt.2008.2669
ment, all subjects were treated with a lumbopelvic
of quadriceps muscle strength
manipulation, which was immediately followed by TA;OMEH:I0 anterior knee pain, physical
and function.11,12,26,52,71,72 Suter and col- retesting the 3 functional activities to determine examination, rehabilitation, spinal manipulation
leagues67,68 recently showed that sacro-
1
Staff Physical Therapist, Blanchfield Army Community Hospital, Fort Campbell, KY. 2 Associate Professor, US Army-Baylor University Doctoral Program in Physical Therapy, San
Antonio, TX. 3 Staff Physical Therapist, Winn Army Community Hospital, Fort Stewart, GA. 4 Staff Physical Therapist, Reynolds Army Community Hospital, Fort Sill, OK. 5 Staff Physical
Therapist, Moncrief Army Community Hospital, Fort Jackson, SC. 6 Assistant Chief, Physical Therapy, Brooke Army Medical Center, San Antonio, TX. 7 Professor and Director, US Army-
Baylor University Doctoral Program in Physical Therapy, Fort Sam Houston, TX. 8 Associate Professor, Physical Therapy, Texas State University, San Marcos, TX. The protocol of this
study was approved by The Institutional Review Board of Brooke Army Medical Center, Fort Sam Houston, TX. Opinions or assertions herein are the private views of the authors and
are not to be construed as official or as reflecting the views of the United States Army or the Department of Defense. Address correspondence to Thomas G. Sutlive, US Army-Baylor
University Doctoral Program in Physical Therapy, 3150 Stanley Road, Room 1303, ATTN: MCCS-HMT, Fort Sam Houston, TX 78234. E-mail: thomas.sutlive@amedd.army.mil
journal of orthopaedic & sports physical therapy | volume 38 | number 6 | june 2008 | 297
[ RESEARCH REPORT ]
nificantly following SIJ manipulation compression, palpation tenderness of the assessing outcome,31 and has been shown
in patients with the disorder. However, tibiofemoral joint lines or patellar ten- to be a valid measurement of change in
the investigators did not measure the don, or a positive finding on any special patient status in other populations com-
patients’ symptomatic response to the tests aimed to identify knee ligament or plaining of pain.21 Each subject was in-
treatment in any of these previous ma- meniscal injuries. Volunteers were also structed to mark the statement that best
nipulation studies.28,67,68 excluded if they had a history of any of represented his/her status in response to
While recent evidence suggests that the following: prior surgery on the spine the intervention on a 15-point GRC.
lumbopelvic manipulation can be effec- or the symptomatic knee, osteoporosis,
tive in restoring quadriceps function in compression fracture, or a history of sys- Examination Items
patients with PFPS, there is no evidence temic, connective tissue, or neurological Each subject completed a medical history
that patients with PFPS will experience diseases. Individuals currently receiving questionnaire (7FF;D:?N 8) and received
symptomatic relief following the inter- treatment for their knee pain were also a standardized physical examination. A
vention. Furthermore, no published study excluded. Informed consent was obtained list of all clinical tests and measures per-
has identified specific clinical examina- from each subject prior to participation, formed on each subject, along with their
tion variables that are predictive of which and the rights of each subject were pro- operational definitions and measurement
patients with PFPS, if any, will respond tected. This study was approved by the properties, is shown in 7FF;D:?N 9. The
successfully to lumbopelvic manipulation. Institutional Review Board of Brooke symptomatic knee was considered the
Identification of these clinical examina- Army Medical Center, Fort Sam Houston, unit of analysis. All measurements were
tion items could lead to the development San Antonio, TX. taken on the side of the symptomatic knee
of a clinical prediction rule (CPR) to (or most symptomatic knee for those with
identify patients likely to respond posi- Instrumentation bilateral PFPS), and all angular measure-
tively to lumbopelvic manipulation. Such Numeric Pain Rating Scale The numeric ments were taken with a 17.8-cm plastic
a CPR could potentially enhance clinical pain rating scale (NPRS) was used to es- goniometer.
decision making, reduce treatment time, tablish the subject’s pain level after each
and optimize outcomes.38,44 functional test. The NPRS is an 11-point Procedures
Therefore, the purpose of our study scale that ranges from 0 (no pain) to 10 Each subject made 1 visit to the clinic.
was to (1) determine the predictive valid- (worst imaginable pain). This type of scale Following the history (completion of
ity of selected clinical examination items has been shown to be a reliable, general- questionnaire shown in 7FF;D:?N 8 and
for identifying patients with PFPS who izable, and internally consistent measure submission of general demographic
have an immediate successful response to of clinical and experimental pain sensa- data), the subject was instructed to re-
lumbopelvic manipulation, and (2) devel- tion intensity.57,58 After performing each move his or her shoes and socks and to
op a CPR derived from these same histori- of the 3 functional tests, subjects were sit on the examining table. The examiner
cal and physical examination findings. instructed to circle the number on the performed a neurological screening ex-
NPRS that best represented their knee amination of the lower extremities that
C;J>E:I pain. Subjects with bilateral PFPS were consisted of manual muscle tests, mus-
instructed to complete the NPRS based cle stretch reflex testing, and sharp/dull
Subjects on their most symptomatic knee. The sensation testing to rule out lumbosacral
most symptomatic knee was determined nerve root compression. The examiner
F
ifty subjects were recruited
from the active duty military popu- by the subject’s self-report.48 A compos- then performed selected tests and mea-
lation at Fort Sam Houston, San An- ite NPRS score (total score from the 3 sures of each subject’s low back and lower
tonio, TX. Volunteers were included in functional tests) was established for each extremities (7FF;D:?N9). The subject was
the study if they were military healthcare subject and used for data analysis. instructed to lie prone on the examining
beneficiaries between 18 and 50 years of Global Rating of Change Question- table. The symptomatic lower extremity
age and had a clinical diagnosis of PFPS. naire The global rating of change was extended so that the ipsilateral foot
Subjects were determined to have symp- questionnaire (GRC) is a single-item, hung off the end of the table and the con-
tomatic PFPS if they had a complaint of self-report instrument used to measure tralateral knee was flexed to 90°, with
anterior knee pain that was provoked by the subject’s impression of the change in the hip externally rotated (figure-four
2 or more of the following: squatting, his or her condition following an inter- position, as described by Donatelli15). Pen
stair ascent, stair descent, prolonged sit- vention (7FF;D:?N 7). A GRC measures marks were made on the following ana-
ting, kneeling, or isometric quadriceps the overall changes in the quality of life tomical landmarks of each subject’s lower
contraction.11,43,55,61 Exclusion criteria of individuals.33 The use of a GRC is a leg and foot for measurement purposes:
included pregnancy, signs of nerve root common, feasible, and useful method for the calcaneus and Achilles tendon was bi-
298 | june 2008 | volume 38 | number 6 | journal of orthopaedic & sports physical therapy
sected with a marker, and the navicular greater on the GRC was sufficiently high
tuberosity was marked with a dot. The to identify individuals who responded to
battery of clinical measurements then the intervention.
commenced as listed in 7FF;D:?N9 on the After dichotomizing the subjects into 2
side of the symptomatic knee only. A sec- outcome groups (success or nonsuccess),
ond examiner repeated the series of tests each element of the history and physical
and measures on the first 25 subjects examination was then analyzed to deter-
prior to the functional testing to assess mine if it was a predictor of treatment suc-
the interrater reliability of the measure- cess. Sensitivity (Sn), specificity (Sp), and
ments. An assistant recorded all mea- <?=KH;'$Therapist performing the lumbopelvic likelihood ratios (LR) were calculated for
manipulation technique used in this study.
surements that were taken by both the each variable. Sn of a test reflects the true
first and second examiners. To prevent positive rate, and Sp of a test is the true
order effects, as well as to expedite the side. If a cavitation was still not felt or negative rate.50 To calculate the Sn and Sp
examination process, the examiner order heard by manipulating the symptomatic for each clinical measurement item, 2-by-2
(first or second examiner) was alternated side, the examiner repeated the manipu- contingency tables were used. When a zero
sequentially. lation on the opposite side in the exact cell value was encountered, 0.5 was added
Following completion of the examina- same manner. Therefore, each subject to all cell values in the table to permit cal-
tion, subjects were asked to complete 3 received up to a maximum of 2 manipu- culation of LRs and their 95% confidence
functional activities (stepping up a 20-cm lations on each side. Following the inter- intervals (CI). Continuous variables were
step, stepping down a 20-cm step, and vention, the subject then repeated each dichotomized using a receiver operator
squatting). After each functional test, the functional test and again rated the pain characteristic (ROC) curve.50 We defined
subject immediately assessed his or her experienced during each activity on the the cut-off of a positive test to be the point
knee pain during the activity and circled NPRS. The subject concluded the visit by on the curve nearest the upper left-hand
the number that represented this pain assessing the overall change in his or her corner that maximized the area under the
on the NPRS. During the squat test, the condition on the GRC. curve, representing the value with the best
angle of knee flexion at which the sub- diagnostic accuracy.50
ject first experienced pain was measured :WjW7dWboi_i Positive likelihood ratios (+LR), nega-
with a plastic goniometer and recorded. All statistical analyses were performed tive likelihood ratios (–LR), and their as-
By doing this, the subject was able to as- using Microsoft Excel 2000 (Microsoft sociated 95% CIs were also computed for
sess the pain experienced when he or she Corp, Redmond, WA) and SPSS for Win- all clinical measurement items. LRs were
squatted to the same angle following the dows, Version 12.0 (SPSS Inc, Chicago, calculated as follows: +LR = Sn/(1 – Sp)
intervention. IL). Interrater reliability statistics were and –LR = (1 – Sn)/Sp. The LR is the sta-
After the first set of functional tests computed using intraclass correlation tistic most often used to determine the
was completed, the subject returned to coefficients (ICC2,1) for continuous vari- usefulness of a CPR.19 According to the
the examination table and was treated ables and Cohen’s kappa coefficients for User’s Guide to Medical Literature,25 a
with a lumbopelvic manipulation to the categorical measurements. +LR greater than 10.0 or a –LR less than
symptomatic side, performed as pre- For the predictive validity portion of 0.1 generate large and often conclusive
viously described3,19 and illustrated in the study, each subject was first classified changes from pretest to posttest prob-
<?=KH; '. A video demonstration of the as either a treatment success or nonsuc- ability for a given condition. LRs of 5.0
manipulation technique has previously cess. The reference criterion used to de- to 10.0 and 0.1 to 0.2 generate moderate
been published on-line.7 The second fine treatment success was either a 50% shifts in pretest-to-posttest probability.
examiner always performed the manip- or greater improvement on the composite LRs of 2.0 to 5.0 and 0.5 to 0.2 gener-
ulation if the subject was part of the in- NPRS or a score of +4 (moderately better ate small (but sometimes important)
terrater reliability study (ie, 1 of the first or improved) or higher on the GRC. It has changes in probability, and LRs of 1.0 to
25 subjects). If the subject or examiner been proposed that a 30% change on a 2.0 and 0.5 to 1.0 alter probability to a
felt a cavitation at any point during the NPRS represents a clinically meaningful small (and rarely important) degree. We
setup for the manipulation or during the reduction in pain in subjects with a vari- considered a +LR greater than 1.9 and a
first manipulation, the intervention was ety of disorders.18 Juniper et al33 proposed –LR less than 0.5 to be clinically mean-
considered complete. If no cavitation was that changes of at least 4 on the GRC in- ingful.25,31 Because several variables in our
heard or felt by the subject or examiner dicate a moderate change in the person’s study had +LRs between 1.9 and 2.0, we
after the first manipulation, the examiner condition. Therefore, we felt that a 50% included any characteristic with a +LR
repeated the manipulation on the same threshold on the NPRS or a score of +4 or greater than 1.9 to avoid discarding any
[ RESEARCH REPORT ]
potentially useful predictors of interven-
Demographics and Historical
tion success. J78B;'
Characteristics of Subjects
A binary logistic regression model
was used to develop a CPR for predicting 7bbIkX`[Yji IkYY[ii\kbEkjYec[ DedikYY[ii\kb
treatment success with the lumbopelvic LWh_WXb[ d3*/ =hekfd3(( EkjYec[=hekfd3(-
manipulation.29 A forward stepwise se- Age (y)* 24.5 6.8 (18-45) 25.3 7.1 (18-44) 23.9 6.7 (18-45)
lection procedure was used to enter the Duration of symptoms (wk)* 94.9 193.4 (1-832) 52.6 75.4 (1-260) 129.4 248.4 (2-832)
variables of those who were in the treat- Gender (% male)† 53.1 59.1 48.1
ment success group only. A liberal P value Painful side (%)†
of .15 was chosen to prevent potentially Right 8.2 9.1 7.4
useful variables from being excluded Left 26.5 18.2 33.3
from the model.20 Clinical measurement Both 65.3 72.7 59.3
items selected by the regression model Worst side (% right) †
36.7 45.5 29.6
as predictors of treatment success were Prior history of knee pain (%)† 24.5 31.8 18.5
combined into a clinical cluster for the Locking (%)† 57.1 54.5 59.3
CPR and were treated as a single test Giving way (%)† 53.1 40.9 63
item. The Sn, Sp, and LRs for the CPR Clicking (%)† 46.9 36.4 55.6
were calculated as previously described Crepitus (%)† 40.8 40.9 40.7
for other dichotomous variables. Swelling (%)† 40.8 36.4 44.4
Stiffness (%)† 46.9 27.3 63
H;IKBJI
Most painful activity (%)†

Squatting 40.8 59.1 25.9
Lifting 4.1 4.5 3.7
F
ifty subjects (26 male and 24 fe-
male) were enrolled in the study, and Running 26.5 18.2 33.3
49 were included in the data analy- Going up stairs 26.5 13.6 37
sis. One subject was excluded from the Going down stairs 2.0 4.5 0.0
study due to significant inconsistencies in * Data are mean SD (range).
†
self-reported symptoms to the 2 examin- Percentage of subjects who demonstrated this characteristic.
ers. Subject demographics and baseline

historical and physical examination items of the subjects in the success group met
16
for the entire sample, as well as for the the criteria for success on both the NPRS 14
successful and nonsuccessful outcome and the GRC. The composite NPRS score 12
groups, are shown in J78B;I' and 2. In- for each group (success, nonsuccess) at 10
NPRS
8
terrater reliability data were collected on baseline and after the manipulation is de-
6
the first 25 subjects. Nine of the 26 (35%) picted in <?=KH;(. The mean SD percent 4
measurements had moderate to good re- NPRS improvement in the success group 2
0
liability based on a threshold greater than was 80% 17%, while in the nonsuccess Success Nonsuccess
or equal to 0.40 for kappa values and group, the mean improvement was 12%
Initial Final
greater than or equal to 0.50 for ICCs.36,50 28%. The median GRC for subjects in
The ICCs and kappa coefficients for the the success group was 5 (range, 0 to 7), <?=KH;($Initial and final composite numeric pain
clinical measurement items are listed in while the median GRC score for the non- rating scale (NPRS) scores for the success and
J78B;), along with their associated stan- success group was 1 (range, –1 to 3). nonsuccess groups. Based on pain level with 3
dard error of measurement (SEMs). Based on the univariate analysis, 6 activities, each rated from 0 to 10. Mean percentage
change for the success group was 80% (SD, 17%).
Twenty-two (45%) of the 49 subjects characteristics were identified as predic-
Mean percentage change for the nonsuccess group
were considered to be a treatment suc- tors of a successful treatment outcome was 12% (SD, 28%)
cess, based on a 50% or greater improve- based on their LRs. Of these 6 clinical
ment on the final composite NPRS or a predictors, 5 were identified by the logis- accuracy of the combinations for predict-
score of at least +4 on the GRC. All 22 tic regression analysis to form a diagnos- ing treatment success are shown in J78B;
subjects were considered intervention tic test item cluster for treatment success. 5. The number of subjects within each
successes based on just the NPRS score, These 5 variables and their associated ac- group (success, nonsuccess) that exhib-
and 17 of the 22 were considered success- curacy statistics are shown in J78B;*. The ited each of the predictors is shown in
es based on the GRC score. Therefore, 17 combinations of these variables and the J78B;,.
have symptoms and a functional status
Variables From the Baseline
J78B;( that change very little following the same
Clinical Examination*
treatment. The ability to predict a priori
7bbIkX`[Yji IkYY[ii\kbEkjYec[ DedikYY[ii\kb which patients will respond favorably to a
LWh_WXb[ d3*/ =hekfd3(( EkjYec[=hekfd3(- specific intervention would be extremely
Quadriceps MMT (% 5/5) 100.0 100.0 100.0 beneficial to practicing clinicians. There-
Hamstrings MMT (% 5/5) 100.0 100.0 100.0 fore, the purpose of our study was to
Gluteus medius MMT (% 5/5) 100.0 100.0 100.0 identify clinical examination items that
Hip flexor MMT (% 5/5) 100.0 100.0 100.0 are predictive of an immediate successful
Hip extensor MMT (% 5/5) 100.0 100.0 100.0 response to a specific lumbopelvic manip-
Tibial torsion (deg) 20.7 (6.9) 20.5 (6.2) 20.8 (7.6) ulation technique. To our knowledge, no
STJN-NWB (deg varus) 3.2 (1.9) 3.0 (1.8) 3.5 (1.9) previous study has shown that lumbopel-
Forefoot alignment (deg varus) 3.5 (1.7) 3.5 (1.9) 3.6 (1.6) vic manipulation can lead to a reduc-
Relaxed calcaneal stance (deg valgus) 6.5 (3.4) 5.4 (3.1) 7.3 (3.4) tion in symptoms in patients with PFPS.
Tibial varum/valgum (deg varus) 3.8 (2.2) 4.2 (2.4) 3.4 (1.9) Forty-five percent of the subjects in our
Navicular drop (mm) 3.5 (3.4) 4.0 (3.7) 3.0 (3.0) study experienced a successful immediate
First MTP extension (deg) 91.3 (12.3) 91.4 (13.6) 91.3 (11.4) response to the manipulation, as defined
AROM, ankle DF (knee ext) (deg) 6.4 (3.9) 6.8 (3.3) 6.0 (4.4) by the criteria for success we established
AROM, ankle DF (knee flex) (deg) 16.0 (6.0) 17.1 (5.1) 15.1 (6.7) in this study. Based on this pretest prob-
Patellar glide (% with lateral glide) 46.9 59.1 37.0 ability (45%) of treatment success, if a
Lateral pull test (% positive) 6.1 0.0 11.1 subject exhibited a side-to-side hip in-
Patellar tilt (% positive) 51.0 54.5 48.1 ternal rotation asymmetry greater than
McConnell test (% positive) 51.0 54.5 48.1 14° (+LR, 4.9), the posttest probability
Thomas test (% positive) 75.5 72.7 77.8 of success increased markedly to 80%.
Hamstring 90-90 test (% positive) 65.3 63.6 66.7 If a subject exhibited at least 3 of the 5
Ober’s test (% positive) 30.6 22.7 37.0 predictors, the posttest probability of suc-
Craig’s test (deg) 13.0 (5.0) 12.7 (5.2) 13.2 (5.0) cess increased to 94%. The posttest prob-
Hip IR (deg) 43.0 (10.5) 41.3 (40.2) 44.4 (10.8) ability for treatment success was 100% if
Side-to-side difference in hip IR (deg) 9.2 (6.5) 10.2 (7.9) 8.4 (5.2) a subject exhibited at least 4 of the vari-
Q-angle (deg) 11.9 (5.6) 12.4 (6.7) 11.5 (4.6) ables constituting the CPR. However, the
Pelvic crest asymmetry (cm) –0.1 (0.6) –0.1 (0.6) 0.01 (0.6) 3 latter combinations of predictors were
Standing flexion test (% positive) 26.5 27.3 25.9 associated with extremely wide CIs and in
Sitting flexion test (% positive) 20.4 18.2 22.2 some instances crossed 1.0, creating un-
Stork test (% positive) 8.2 4.5 11.1 certainty with the predictive accuracy and
Abbreviations: AROM, active range of motion; DF, dorsiflexion; IR, internal rotation; MMT, manual clinical application of this estimate (J78B;
muscle test; MTP, metatarsophalangeal joint; NWB, non-weight bearing; STJN, subtalar joint neutral. 5). Based on this observation, we suggest
* Mean (SD) is shown for the continuous variables.
that clinicians can have greater accuracy
and confidence when determining the
The pretest probability for the like- (+LR, 18.4; 95% CI: 1.6 to 105.3) was the likelihood that a patient will exhibit a
lihood of treatment success with lum- presence of at least 3 of the 5 predictors rapid response to lumbopelvic manipula-
bopelvic manipulation was 45% (ie, 22 in J78B; +, which increased the posttest tion by relying on the hip internal rotation
out of 49 subjects responded successfully probability of success to 94%. asymmetry variable alone (+LR, 4.9; 95%
to the intervention). The most robust CI: 1.2 to 20.8). If a patient does not dem-
predictor of treatment success was a side- :?I9KII?ED onstrate hip internal rotation asymmetry
to-side difference in hip internal rotation greater than 14°, the clinician may still opt
greater than 14° (+LR, 4.9; 95% CI: 1.2 to perform the lumbopelvic manipulation
P
FPS is a prevalent condition in
to 20.8). If a subject exhibited hip inter- outpatient physical therapy practice based on the presence of at least 3 of the
nal rotation asymmetry greater than 14°, settings and management of the dis- remaining predictors.
the posttest probability of treatment suc- order is characterized by large variations One might argue that it would be more
cess increased to 80%. The combination in practice patterns. Some patients may practical for the clinician to simply try the
of variables that proved to be the most experience significant improvement with manipulation technique to see if the pa-
powerful predictor of treatment success a given intervention, while others may tient responds to the intervention, rather
[ RESEARCH REPORT ]
Interrater Reliability for Continuous
J78B;)
(ICC) and Categorical (Kappa) Variables
J[ij%C[Wikh[ ?992,1 /+9? I;C AWffW

Quadriceps MMT* 1.00
Hamstrings MMT* 1.00
Gluteus medius MMT* 1.00
Hip flexor MMT* 1.00
Hip extensor MMT* 1.00
Tibial torsion 0.20 –0.06, 0.44 5.58°
<?=KH;)$Measurement of hip internal rotation
STJN-NWB 0.04 –0.06, 0.17 4.54°
obtained with a subject in the prone position. The
Forefoot alignment 0.01 –0.22, 0.24 1.78° side-to-side difference in hip internal rotation shown
Relaxed calcaneal stance 0.11 –0.17, 0.37 3.24° in this figure is exactly 14°.
Tibial varum/valgum 0.18 –0.06, 0.41 1.99°
Navicular drop 0.11 –0.18, 0.37 2.64° pairments. Based on the patient’s success-
ful outcome, the authors concluded that
First MTP extension 0.52 0.29, 0.70 6.79°

AROM, ankle DF (ext) 0.25 0.00, 0.49 3.58° femoral or hip joint asymmetry may be
AROM, ankle DF (flex) 0.19 –0.06, 0.43 4.50° related to patellofemoral pain. Similarly,
Craig’s test 0.02 –0.28, 0.26 5.25° Powers51 reported that excessive femoral
Hip IR 0.45 –0.09, 0.75 9.57° rotation during weight-bearing activi-
Q-angle 0.30 0.04, 0.53 4.60° ties may contribute to PFPS, and rec-
Pelvic crest alignment 0.51 0.14, 0.75 0.47 cm ommended that interventions aimed at
Lateral pull test –0.03 controlling hip and pelvic motion should
Patellar tilt 0.17 be considered when treating patients with
McConnell test 0.46 PFPS.56 Several investigators have sug-
Thomas test 0.17 gested that proximal hip muscle weak-
Hamstring 90-90 test 0.84 ness, particularly of the hip abductors and
Ober’s test 0.17 external rotators, may be associated with
Patellar glide 0.003 patellofemoral pain,30,42,60 and Tyler and
Standing flexion 0.88 colleagues74 showed that improvements in

Sitting flexion –0.04 hip flexor strength and flexibility were as-
Stork test 1.00 sociated with symptomatic improvement
Abbreviations: AROM, active range of motion; CI, confidence interval; DF, dorsiflexion; ICC, intraclass
in patients with PFPS. The implication
correlation coefficient; IR, internal rotation; MMT, manual muscle test; MTP, metatarsophalangeal of the findings of these studies is that ad-
joint; NWB, non-weight bearing; STJN, subtalar joint neutral. dressing proximal impairments may be an
* All subjects had the same score, so CI and SEM could not be calculated.
important element in the successful reha-
bilitation of patients with PFPS.
than taking the time to determine if the hip range of motion was the most ro-
patient fits the CPR. However, given the bust predictor of treatment success is EkjYec[C[Wikh[i
various treatment strategies that have consistent with recent published stud- We defined treatment success as either an
been proposed for the management of pa- ies suggesting that impairment of the at least 50% improvement on the compos-
tients with PFPS, it seems impractical to lumbopelvic-hip complex is associated ite NPRS or at least a 4-point change on
use a trial-and-error approach for numer- with PFPS.5,28,30,42,57,67,68,74 In a case report, the GRC questionnaire. By setting our ref-
ous possible interventions. Goniometric Cibulka and Threlkeld-Watkins5 described erence criteria at 50%, we were confident
assessment of the most powerful predic- evaluation and intervention for a patient that we identified those subjects who ex-
tor of treatment success, hip internal rota- with a side-to-side difference in hip in- perienced a clinically meaningful positive
tion, can be obtained readily in the clinic ternal rotation of 20°. The patient also immediate response to the manipulation.
setting. Furthermore, a side-to-side dif- exhibited weakness of the hip internal While results indicate an association be-
ference in hip internal rotation of at least rotator and abductor muscles on the in- tween the manipulative intervention and
14° is clearly evident with the patient ly- volved side, and was treated with a 2-week outcome, our study design is not sufficient
ing in the prone position (<?=KH;)). regimen of stretching and strengthening to establish a cause-and-effect relation-
The fact that a measure related to exercises aimed at addressing these im- ship. In particular, the lack of a control
Sensitivity, Specificity, Likelihood Ratios (LR), and Successful
J78B;*
Cutoff Scores for the Predictors of Intervention Success*
Fh[Z_Yjehe\IkYY[ii I[di_j_l_jo If[Y_ÓY_jo !BH ÅBH 9kje÷IYeh[

Difference in hip IR (deg) 0.36 (0.16, 0.56) 0.93 (0.83, 1.02) 4.9 (1.2, 20.8) 0.7 (0.5, 1.0) 14°
Ankle dorsiflexion with knee flexed (deg) 0.59 (0.39, 0.80) 0.70 (0.53, 0.86) 2.0 (1.0, 3.9) 0.6 (0.3, 1.0) 16°
Navicular drop (mm) 0.64 (0.44, 0.84) 0.67 (0.49, 0.84) 1.91 (1.0, 3.6) 0.5 (0.3, 1.0) 3 mm
No stiffness with sitting 20 min (%) 0.73 (0.54, 0.91) 0.63 (0.45, 0.81) 2.0 (1.1, 3.4) 0.4 (0.2, 0.9) NA
Squatting is most painful activity 0.59 (0.39, 0.80) 0.74 (0.58, 0.91) 2.3 (1.1, 4.7) 0.6 (0.3, 1.0) NA
* Success defined as an immediate 50% or greater pain reduction or moderate (+4 or greater) improvement on the Global Rating of Change scale for functional
activities consisting of squatting, stepping up a 20-cm step, and stepping down a 20-cm step. 95% confidence intervals shown in parentheses.
J78B;+ Combinations of Predictor Variables and Associated Accuracy Statistics*

LWh_WXb[iFh[i[dj I[di_j_l_jo If[Y_ÓY_jo !BH ÅBH FheXWX_b_joe\IkYY[ii†

All 5 predictors 0.09 (0.02, 0.31) 1.00 (0.84, 1.00) Infinite (0.31, infinite) 0.91 (0.78, 1.1) 100% (20-100)
At least 4 predictors 0.32 (0.15, 0.55) 1.00 (0.84, 1.00) Infinite (0.90, infinite) 0.68 (0.51, 0.92) 100% (47-100)
At least 3 predictors 0.68 (0.45, 0.86) 0.96 (0.81, 1.00) 18.4 (3.6, 105.3) 0.33 (0.17, 0.55) 94% (68-99)
At least 2 predictors 0.91 (0.71, 0.99) 0.56 (0.35, 0.75) 2.05 (1.3, 2.9) 0.16 (0.04, 0.69) 63% (50-70)
At least 1 predictor 0.91 (0.71, 0.99) 0.15 (0.04, 0.35) 1.1 (0.87, 1.3) 0.61 (0.12, 3.0) 47% (41-52)
* 95% confidence intervals in parentheses.
†
The probability of success is calculated using the +LR and assumes a pretest probability of 45%.
group or competing intervention and the The 5 Variables Forming the Clinical
fact that we measured an immediate re- J78B;, Prediction Rule and the Number of Subjects
sponse to the treatment leave the possibili- in Each Group at Each Level
ty that a placebo effect took place for those
expressing a reduction of symptoms. I_Z[jei_Z[Z_÷[h[dY[_d^_f_dj[hdWbhejWj_ed14°

It is important to emphasize that we 7dab[Zehi_Ô[n_edad[[Ô[n[Z16°
developed the CPR based only on the sub- DWl_YkbWhZhef3 mm
jects’ immediate response to the interven- Dei[b\#h[fehj[Zij_÷d[iim_jî_jj_d]20 min
tion. Although duration of symptoms did IgkWjj_d]h[fehj[ZWiceijfW_d\kbWYj_l_jo
not emerge as a predictor variable in the DkcX[he\Fh[Z_Yjeh IkYY[ii\kb DedikYY[ii\kb
CPR, the participants in this study gen- LWh_WXb[iFh[i[dj EkjYec[=hekf EkjYec[=hekf
erally had chronic PFPS (mean duration 0 2 4
of symptoms, 95 weeks). Nevertheless, 1 0 11
the development of the current CPR is a 2 5 11
reasonable initial foray at identifying the 3 8 1
characteristics of patients with PFPS who 4 5 0
are most likely to experience symptomat- 5 2 0
ic improvement following manipulation

of the lumbopelvic region. Future valida- was selective and may not have included uting factor in individuals with PFPS.
tion of the proposed CPR in a random- some items that otherwise might have Therefore, assessment of the strength of
ized clinical trial is required and should emerged as predictors of intervention this muscle group should be included in
include a longer follow-up period. success. For instance, we did not include future studies aimed at validation of the
Additionally, the CPR developed in the assessment of hip external rotator proposed CPR. The clinical examination
this study was based on establishing the muscle strength in the physical examina- items included in this study were chosen
predictive validity of a limited number of tion. Several investigators30,42,60 recently because in our opinion they (1) are rou-
examination variables. We recognize that identified weakness of the hip external tinely obtained from patients with knee
the clinical examination we performed rotator muscles as a possible contrib- pain, (2) are measures that guide clinical
[ RESEARCH REPORT ]
decision making for lumbopelvic manip- ity and spine, may affect or contribute vious surgery to the lumbar spine, pelvis,
ulation, and (3) allow for comparison of to a patient’s primary complaint.4,6,14,77 or buttock.3,19
the results of the present study with our The biomechanical link between the
previous work.39,69 lumbopelvic-hip complex and the knee ?dj[hhWj[hH[b_WX_b_jo
It is possible that the predictor vari- region suggests that alterations in joint In addition to developing the CPR, we
ables that emerged in this developmen- mobility in the lumbar spine, SIJ, or hip examined the interrater reliability of the
tal study may have occurred by chance. joint may serve as an underlying con- measurements obtained in our study.
However, we believe that each of the tributor to the development of patel- Nine of the 26 (35%) clinical measure-
physical examination variables that lofemoral pain. For instance, previous ments exhibited moderate to good
emerged as predictors of a successful research has shown that femoral internal reliability.36,50 The low interrater reliabil-
response to lumbopelvic manipulation rotation during weight-bearing activities ity for some of the measures may pose a
are biologically plausible in subjects with can influence patellar alignment and ki- threat to the internal validity of our in-
PFPS, and that the predictors that sur- nematics.51,70 Furthermore, impairments vestigation and ultimately limit the in-
faced from the history are also relevant in femoral rotation have been associated terpretation and application of the CPR.
in this population. Nevertheless, future with individuals with PFPS.53 Given that Taking a mean value of multiple mea-
investigations should include all relevant hip internal rotation asymmetry was the surements may have provided improved
clinical examination variables and larger most powerful predictor of treatment reliability values,50 but we chose to obtain
sample sizes to minimize the possibility success in the current study, it is certainly each measurement just once because we
that predictors emerge by chance. plausible that the manipulation worked believe this is most representative of what
by affecting lower extremity kinemat- is done in clinical practice.
Feii_Xb[C[Y^Wd_ici\ehIkYY[ii\kbH[- ics.56 However, we did not use any imag- The ICC for the measurement of hip
ifedi[jeBkcXef[bl_YCWd_fkbWj_ed ing techniques in our study, nor did we internal rotation was 0.45, which is con-
A reduction in distal muscle inhibition repeat any of the clinical measurements sistent with reliability values reported for
following treatment with spinal ma- after the manipulation and, therefore, the same measure in previous studies of
nipulation techniques is well document- do not know if a change in hip internal healthy individuals (ICC = 0.44 to 0.51)73
ed.2,27,28,66-67 In studies of patients with rotation asymmetry occurred following and patients with unilateral hip osteoar-
PFPS, Suter and colleagues67,68 showed the treatment. In our subjects there was thritis (ICC = 0.48),10 but much lower
that SIJ manipulation led to a significant no consistent pattern with regard to hip than the goniometric and inclinometer
decrease in quadriceps inhibition of the internal rotation range of motion being values reported by Ellison and colleagues17
involved limbs, and Hillermann et al28 re- greater on the side of the symptomatic in their study of healthy individuals and
ported that quadriceps muscle strength versus the asymptomatic lower extrem- subjects with low back pain (ICC = 0.96
increased significantly following SIJ ma- ity. A current investigation in this labora- to 0.99). The method error as it relates to
nipulation in patients with the disorder. tory is directed at determining whether a the coefficient of variation (CVME) can be
The patients’ symptomatic response to successful response to the manipulation used when examining response stability
the treatment was not monitored in these is correlated with a reduction or correc- because it reflects percentage of varia-
previous studies.28,67,68 However, improve- tion of hip internal rotation asymmetry tion from trial to trial.50 The CVME for hip
ments in quadriceps strength and a re- in subjects with PFPS who meet the CPR. internal rotation was 17% (equivalent
duction in quadriceps inhibition have It is interesting that the presence of a hip to 5°-8°), which we considered to be an
been associated with symptomatic relief internal rotation measure was also a pre- acceptable level. A Cohen kappa coeffi-
and successful outcomes in other studies dictor of a successful response to manip- cient (L) was calculated to determine the
of patients with PFPS.34,45,65 Although it ulation in patients with low back pain, 3,19 chance-corrected agreement between
was not monitored in this study, an im- suggesting that there may be a common raters for this predictor (L = 0.71). A more
provement in quadriceps function imme- underlying mechanism or biomechani- clinically meaningful way to examine the
diately following the manipulation may cal link for a successful response to this reliability data for this continuous vari-
have contributed towards the successful treatment technique. None of the sub- able is in a dichotomous fashion (greater
response observed in our subjects. jects in the current study reported hav- than 14° or less than or equal to 14°),
An alternative explanation for the ing low back pain. It is important to note which is how it would be used in the con-
favorable response seen in our success that the lumbopelvic manipulation used text of this CPR.76 For instance, how often
group may be explained by the concept in this study is contraindicated in indi- did the raters agree that a subject had a
of regional interdependence. According viduals with serious spinal conditions, greater than 14° measurement of hip in-
to this theory, dysfunction and impair- with signs of nerve root compression, ternal rotation asymmetry? The percent
ments in remote regions, both extrem- who are pregnant, and who have had pre- agreement between the raters using the
cutoff score of greater than 14° was 76%, patients with PFPS who demonstrate a prediction rule to identify patients with low
back pain most likely to benefit from spinal ma-
while the kappa coefficient (L) was 0.32, side-to-side difference in hip internal ro-
nipulation: a validation study. Ann Intern Med.
which is considered poor to fair.36 tation range of motion greater than 14° 2004;141:920-928.
and are treated with a lumbopelvic ma- 4. Cibulka MT. Low back pain and its relation to
IkX`[YjiM_j^8_bWj[hWbF<FI nipulation have an 80% posttest proba- the hip and foot. J Orthop Sports Phys Ther.
1999;29:595-601.
A possible limitation of our study was the bility of experiencing a positive outcome.
5. Cibulka MT, Threlkeld-Watkins J. Patellofemoral
fact that the majority of our data were However, future studies are necessary in pain and asymmetrical hip rotation. Phys Ther.
taken from individuals with bilateral light of a limited sample size, the mar- 2005;85:1201-1207.
knee pain. Thirty-two of the 49 subjects ginal reliability of some predictors, and 6. Cleland JA, Childs JD, Fritz JM, Whitman JM,
Eberhart SL. Development of a clinical predic-
who completed the study presented with the potential omission of possibly use- tion rule for guiding treatment of a subgroup of
bilateral PFPS. Therefore, one might ful clinical examination variables in this patients with neck pain: use of thoracic spine
argue that our results were potentially study. The predictors identified in this manipulation, exercise, and patient educa-
biased due to a lack of independence of study may help investigators assemble a tion. Phys Ther. 2007;87:9-23. http://dx.doi.
org/10.2522/ptj.20060155
data. To examine this issue, we compared more homogenous subgroup of patients 7. Cleland JA, Fritz JM, Whitman JM, Childs JD,
the NPRS scores between subjects with with PFPS for a future validation study Palmer JA. The use of a lumbar spine ma-
unilateral PFPS and those with bilateral in the form of a randomized clinical trial, nipulation technique by physical therapists
pain. Descriptive statistics were calcu- which is necessary before the CPR can be in patients who satisfy a clinical prediction
rule: a case series. J Orthop Sports Phys Ther.
lated for the mean percent NPRS change used confidently in the clinic. T 2006;36:209-214.
scores of the 17 subjects with unilateral 8. Cowan SM, Bennell KL, Crossley KM,
knee pain and compared with the mean A;OFE?DJI Hodges PW, McConnell J. Physical therapy
percent NPRS change scores of 17 ran- <?D:?D=I0 A preliminary CPR was devel- alters recruitment of the vasti in patel-
lofemoral pain syndrome. Med Sci Sports
domly selected subjects with bilateral oped to predict an immediate successful Exerc. 2002;34:1879-1885. http://dx.doi.
pain. The mean NPRS change score in the response to lumbopelvic manipulation org/10.1249/01.MSS.0000038893.30443.CE
subjects with unilateral pain was 38.8% in patients with patellofemoral pain /$ Cowan SM, Bennell KL, Hodges PW, Crossley
KM, McConnell J. Delayed onset of electromyo-
(SD, 39.2%), and the mean NPRS change syndrome (PFPS). The most robust
graphic activity of vastus medialis obliquus
score in subjects with bilateral pain was predictor of a successful response to the relative to vastus lateralis in subjects with patel-
39.1% (SD, 46.1%). These results sug- manipulation was a side-to-side differ- lofemoral pain syndrome. Arch Phys Med Reha-
gested that the subjects with bilateral ence in hip internal rotation range of bil. 2001;82:183-189. http://dx.doi.org/10.1053/
apmr.2001.19022
PFPS responded similarly to those with motion of greater than 14°.
10. Croft PR, Nahit ES, Macfarlane GJ, Silman AJ.

unilateral knee pain. While the internal ?CFB?97J?ED0 The clinical prediction rule Interobserver reliability in measuring flexion,
validity of data from future studies of developed in this study may help clini- internal rotation, and external rotation of
PFPS may be strengthened by including cians identify patients with PFPS who the hip using a plurimeter. Ann Rheum Dis.
1996;55:320-323.
only subjects with unilateral knee pain, will respond successfully to lumbopelvic
11. Crossley K, Bennell K, Green S, Cowan S, McCo-
we believe that the inclusion of subjects manipulation. nnell J. Physical therapy for patellofemoral pain:
with bilateral symptoms30,46,48,49,69 is more 97KJ?ED0 Limitations of this study in- a randomized, double-blinded, placebo-con-
reflective of clinical practice and will ul- cluded a small sample size (n = 49), mar- trolled trial. Am J Sports Med. 2002;30:857-865.
12. Crossley K, Bennell K, Green S, McConnell J. A
timately increase the generalizability of ginal reliability of some predictors, and systematic review of physical interventions for
future investigations. the omission of some potentially useful patellofemoral pain syndrome. Clin J Sport Med.
clinical examination variables. Future 2001;11:103-110.
9ED9BKI?ED randomized clinical trials are needed to 13. Devereaux MD, Lachmann SM. Patello-femoral
arthralgia in athletes attending a Sports Injury
validate the findings of this study. Clinic. Br J Sports Med. 1984;18:18-21.
M
e developed a CPR that identi- 14. Deyle GD, Henderson NE, Matekel RL, Ryder MG,
fied the characteristics of patients Garber MB, Allison SC. Effectiveness of manual
with PFPS who would experience REFERENCES physical therapy and exercise in osteoarthritis
of the knee. A randomized, controlled trial. Ann
an immediate, positive response to treat- Intern Med. 2000;132:173-181.
1. Joint Motion Measurement. Vol TC 8-640.
ment with lumbopelvic manipulation, as 15. Donatelli R. The Biomechanics of the Foot and
Washington, DC: Headquarters, Department of
based on a 50% or greater reduction in Ankle. 2nd ed. Philadelphia, PA: FA Davis Com-
the Army; 1987.
the composite NPRS or a score of +4 or pany; 1996.
2. Bulbulian R, Burke J, Dishman JD. Spinal reflex
16. Dye S, Vaupel G, Dye C. Conscious neurosen-
higher on the GRC for 3 functional tests excitability changes after lumbar spine passive sory mapping of the internal structures of the
(squatting, stepping up a 20-cm step, flexion mobilization. J Manipulative Physiol Ther. human knee without intraarticular anesthesia.
and stepping down a 20-cm step). Based 2002;25:526-532. Am J Sports Med. 1998;26:773-777.
3. Childs JD, Fritz JM, Flynn TW, et al. A clinical 17. Ellison JB, Rose SJ, Sahrmann SA. Patterns
on the results of this preliminary study,
[ RESEARCH REPORT ]
of hip rotation range of motion: a comparison Working Group. JAMA. 1994;271:703-707. 47. Petrone MR, Guinn J, Reddin A, Sutlive TG, Flynn
between healthy subjects and patients with low 32. Jonson SR, Gross MT. Intraexaminer reliability, TW, Garber MP. The accuracy of the Palpation
back pain. Phys Ther. 1990;70:537-541. interexaminer reliability, and mean values Meter (PALM) for measuring pelvic crest height
18. Farrar JT, Young JP, Jr., LaMoreaux L, Werth for nine lower extremity skeletal measures in difference and leg length discrepancy. J Orthop
JL, Poole RM. Clinical importance of changes healthy naval midshipmen. J Orthop Sports Sports Phys Ther. 2003;33:319-325.
in chronic pain intensity measured on an Phys Ther. 1997;25:253-263. 48. Piva SR, Fitzgerald K, Irrgang JJ, et al. Reli-
11-point numerical pain rating scale. Pain. 33. Juniper EF, Guyatt GH, Willan A, Griffith LE. ability of measures of impairments associated
2001;94:149-158. Determining a minimal important change in a with patellofemoral pain syndrome. BMC
'/$ Flynn T, Fritz J, Whitman J, et al. A clinical disease-specific Quality of Life Questionnaire. J Musculoskelet Disord. 2006;7:33. http://dx.doi.
prediction rule for classifying patients with Clin Epidemiol. 1994;47:81-87. org/10.1186/1471-2474-7-33
low back pain who demonstrate short-term 34. Kannus P, Niittymaki S. Which factors predict */$ Piva SR, Goodnite EA, Childs JD. Strength
improvement with spinal manipulation. outcome in the nonoperative treatment of around the hip and flexibility of soft tissues in
Spine. 2002;27:2835-2843. http://dx.doi. patellofemoral pain syndrome? A prospec- individuals with and without patellofemoral
org/10.1097/01.BRS.0000035681.33747.8D tive follow-up study. Med Sci Sports Exerc. pain syndrome. J Orthop Sports Phys Ther.
20. Freedman D. A note on screening regression 1994;26:289-296. 2005;35:793-801.
equations. Am Statistician. 1983;37:152-155. 35. Kendall F. Muscles Testing and Function. Phila- 50. Portney L, Watkins M. Foundations of Clinical
21. Fritz JM, Irrgang JJ. A comparison of a modified delphia, PA: Lippincott, Williams, & Wilkins; Research: Applications to Practice. 2nd ed. Up-
Oswestry Low Back Pain Disability Question- 1993. per Saddle River, NJ: Prentice Hall Health; 2000.
naire and the Quebec Back Pain Disability Scale. 36. Landis JR, Koch GG. The measurement of ob- 51. Powers CM. The influence of altered lower-
Phys Ther. 2001;81:776-788. server agreement for categorical data. Biomet- extremity kinematics on patellofemoral joint
22. Gilleard W, McConnell J, Parsons D. The effect of rics. 1977;33:159-174. dysfunction: a theoretical perspective. J Orthop
patellar taping on the onset of vastus medialis 37. Laprade J, Culham E. Radiographic measures Sports Phys Ther. 2003;33:639-646.
obliquus and vastus lateralis muscle activity in subjects who are asymptomatic and sub- 52. Powers CM. Rehabilitation of patellofemoral
in persons with patellofemoral pain. Phys Ther. jects with patellofemoral pain syndrome. Clin joint disorders: a critical review. J Orthop Sports
1998;78:25-32. Orthop Relat Res. 2003;172-182. http://dx.doi. Phys Ther. 1998;28:345-354.
23. Grelsamer RP, Klein JR. The biomechanics of the org/10.1097/01.blo.0000079269.91782.f5 53. Powers CM, Chen PY, Reischl SF, Perry J. Com-
patellofemoral joint. J Orthop Sports Phys Ther. 38. Laupacis A, Sekar N, Stiell IG. Clinical predic- parison of foot pronation and lower extremity
1998;28:286-298. tion rules. A review and suggested modifica- rotation in persons with and without patell-
24. Gross MT. Lower quarter screening for skel- tions of methodological standards. JAMA. ofemoral pain. Foot Ankle Int. 2002;23:634-640.
etal malalignment--suggestions for orthotics 1997;277:488-494. 54. Powers CM, Landel R, Perry J. Timing and inten-
and shoewear. J Orthop Sports Phys Ther. )/$ Lesher JD, Sutlive TG, Miller GA, Chine NJ, sity of vastus muscle activity during functional
1995;21:389-405. Garber MB, Wainner RS. Development of a clini- activities in subjects with and without patel-
25. Guyatt G, Rennie D. User’s Guide to the Medical cal prediction rule for classifying patients with lofemoral pain. Phys Ther. 1996;76:946-955;
Literature: Essentials of Evidence-Based Clinical patellofemoral pain syndrome who respond discussion 956-967.
Practice. Chicago, IL: American Medical Asso- to patellar taping. J Orthop Sports Phys Ther. 55. Powers CM, Perry J, Hsu A, Hislop HJ. Are
ciation; 2002. 2006;36:854-866. http://dx.doi.org/doi:10.2519/ patellofemoral pain and quadriceps femoris
26. Herrington L, Al-Sherhi A. A controlled trial of jospt.2006.2208 muscle torque associated with locomotor func-
weight-bearing versus nonweight-bearing exer- 40. Magee D. Orthopedic Physical Assessment. 3rd tion? Phys Ther. 1997;77:1063-1075; discussion
cises for patellofemoral pain. J Orthop Sports ed. Philadelphia, PA: W.B. Saunders; 1997. 1075-1068.
Phys Ther. 2007;37:155-160. http://dx.doi. 41. Mariani PP, Caruso I. An electromyographic in- 56. Powers CM, Ward SR, Fredericson M, Guillet M,
org/10.2519/jospt.2007.2261 vestigation of subluxation of the patella. J Bone Shellock FG. Patellofemoral kinematics during
27. Herzog W, Scheele D, Conway PJ. Electromyo- Joint Surg Br. 1979;61-B:169-171. weight-bearing and nonweight-bearing knee
graphic responses of back and limb muscles 42. Mascal CL, Landel R, Powers C. Management extension in persons with lateral subluxation of
associated with spinal manipulative therapy. of patellofemoral pain targeting hip, pelvis, and the patella: a preliminary study. J Orthop Sports
Spine. 1999;24:146-152; discussion 153. trunk muscle function: 2 case reports. J Orthop Phys Ther. 2003;33:677-685.
28. Hillermann B, Gomes AN, Korporaal C, Jackson Sports Phys Ther. 2003;33:647-660. 57. Price DD, Bush FM, Long S, Harkins SW. A com-
D. A pilot study comparing the effects of spinal 43. McConnell J. The management of chondromala- parison of pain measurement characteristics of
manipulative therapy with those of extra- cia patellae: a long-term solution. Aust J Physio- mechanical visual analogue and simple numeri-
spinal manipulative therapy on quadriceps ther. 1986;32:215-223. cal rating scales. Pain. 1994;56:217-226.
muscle strength. J Manipulative Physiol Ther. 44. McGinn TG, Guyatt GH, Wyer PC, Naylor 58. Price DD, McGrath PA, Rafii A, Buckingham B.
2006;29:145-149. http://dx.doi.org/10.1016/j. CD, Stiell IG, Richardson WS. Users’ guides The validation of visual analogue scales as ratio
jmpt.2005.12.003 to the medical literature: XXII: how to use scale measures for chronic and experimental
(/$ Holleman DR, Jr., Simel DL. Quantitative as- articles about clinical decision rules. Evidence- pain. Pain. 1983;17:45-56.
sessments from the clinical examination. How Based Medicine Working Group. JAMA. +/$ Reid DC. The myth, mystic and frustration of
should clinicians integrate the numerous re- 2000;284:79-84. anterior knee pain [editorial]. Clin J Sport Med.
sults? J Gen Intern Med. 1997;12:165-171. 45. Natri A, Kannus P, Jarvinen M. Which factors 1993;3:139-143.
30. Ireland ML, Willson JD, Ballantyne BT, Davis predict the long-term outcome in chronic 60. Robinson RL, Nee RJ. Analysis of hip strength in
IM. Hip strength in females with and without patellofemoral pain syndrome? A 7-yr prospec- females seeking physical therapy treatment for
patellofemoral pain. J Orthop Sports Phys Ther. tive follow-up study. Med Sci Sports Exerc. unilateral patellofemoral pain syndrome. J Or-
2003;33:671-676. 1998;30:1572-1577. thop Sports Phys Ther. 2007;37:232-238. http://
31. Jaeschke R, Guyatt GH, Sackett DL. Users’ 46. Nijs J, Van Geel C, Van der auwera C, Van de dx.doi.org/doi:10.2519/jospt.2007.2421
guides to the medical literature. III. How to use Velde B. Diagnostic value of five clinical tests 61. Salsich GB, Brechter JH, Farwell D, Powers CM.
an article about a diagnostic test. B. What are in patellofemoral pain syndrome. Man Ther. The effects of patellar taping on knee kinetics,
the results and will they help me in caring for 2006;11:69-77. http://dx.doi.org/10.1016/j. kinematics, and vastus lateralis muscle activ-
my patients? The Evidence-Based Medicine math.2005.04.002 ity during stair ambulation in individuals with
patellofemoral pain. J Orthop Sports Phys Ther. ,/$ Sutlive TG, Mitchell SD, Maxfield SN, et al. Iden- Podiatr Med Assoc. 2002;92:317-326.
2002;32:3-10. tification of individuals with patellofemoral pain 76. Wainner RS. Reliability of the clinical examina-
62. Schutzer SF, Ramsby GR, Fulkerson JP. The whose symptoms improved after a combined tion: how close is “close enough”? J Orthop
evaluation of patellofemoral pain using com- program of foot orthosis use and modified Sports Phys Ther. 2003;33:488-491.
puterized tomography. A preliminary study. Clin activity: a preliminary investigation. Phys Ther. 77. Wainner RS, Whitman JM, Cleland JA, Flynn TW.
Orthop Relat Res. 1986;286-293. 2004;84:49-61.
Regional interdependence: a musculoskeletal
63. Shwayhat AF, Linenger JM, Hofherr LK, Slymen 70. Tennant S, Williams A, Vedi V, Kinmont C, Ge-
examination model whose time has come. J Or-
DJ, Johnson CW. Profiles of exercise history and droyc W, Hunt DM. Patello-femoral tracking in
thop Sports Phys Ther. 2007;37:658-660. http://
overuse injuries among United States Navy Sea, the weight-bearing knee: a study of asymptom-
Air, and Land (SEAL) recruits. Am J Sports Med. atic volunteers utilising dynamic magnetic reso- dx.doi.org/10.2519/jospt.2007.0110
1994;22:835-840. nance imaging: a preliminary report. Knee Surg 78. Watson CJ, Leddy HM, Dynjan TD, Parham JL.
64. Songer TJ, LaPorte RE. Disabilities due to injury Sports Traumatol Arthrosc. 2001;9:155-162. Reliability of the lateral pull test and tilt test
in the military. Am J Prev Med. 2000;18:33-40. 71. Thomee R. A comprehensive treatment ap- to assess patellar alignment in subjects with
65. Suter E, Herzog W, Bray RC. Quadriceps inhibi- proach for patellofemoral pain syndrome in symptomatic knees: student raters. J Orthop
tion following arthroscopy in patients with young women. Phys Ther. 1997;77:1690-1703. Sports Phys Ther. 2001;31:368-374.
anterior knee pain. Clin Biomech (Bristol, Avon). 72. Thomee R, Augustsson J, Karlsson J. Patel- -/$ Wilson T. The measurement of patellar align-
1998;13:314-319. lofemoral pain syndrome: a review of current ment in patellofemoral pain syndrome: are we
66. Suter E, McMorland G. Decrease in elbow flexor issues. Sports Med. 1999;28:245-262. confusing assumptions with evidence? J Orthop
inhibition after cervical spine manipulation in 73. Thomee R, Renstrom P, Karlsson J, Grimby G.
Sports Phys Ther. 2007;37:330-341. http://
patients with chronic neck pain. Clin Biomech Patellofemoral pain syndrome in young women.
dx.doi.org/doi:10.2519/jospt.2007.2281
(Bristol, Avon). 2002;17:541-544. I. A clinical analysis of alignment, pain parame-
67. Suter E, McMorland G, Herzog W, Bray R. Con- ters, common symptoms and functional activity 80. Witvrouw E, Lysens R, Bellemans J, Cambier D,
servative lower back treatment reduces inhibi- level. Scand J Med Sci Sports. 1995;5:237-244. Vanderstraeten G. Intrinsic risk factors for the
tion in knee-extensor muscles: a randomized 74. Tyler TF, Nicholas SJ, Mullaney MJ, McHugh development of anterior knee pain in an athletic
controlled trial. J Manipulative Physiol Ther. MP. The role of hip muscle function in the treat- population. A two-year prospective study. Am J
2000;23:76-80. ment of patellofemoral pain syndrome. Am J Sports Med. 2000;28:480-489.
68. Suter E, McMorland G, Herzog W, Bray R. Sports Med. 2006;34:630-636. http://dx.doi.

Decrease in quadriceps inhibition after sac- org/10.1177/0363546505281808
@
roiliac joint manipulation in patients with 75. Van Gheluwe B, Kirby KA, Roosen P, Phillips
anterior knee pain. J Manipulative Physiol Ther. RD. Reliability and accuracy of biomechanical
CEH;?D<EHC7J?ED
1999;22:149-153. measurements of the lower extremities. J Am WWW.JOSPT.ORG
7FF;D:?N7
=BE87BH7J?D=E<9>7D=;GK;IJ?EDD7?H;
Compared to your condition prior to treatment with the manipulation, which item on the scale below best describes your present condition
(choose only one):
FWj_[dj=beXWbHWj_d]IYWb[
7l[ho]h[WjZ[Wbmehi[ 7j_doX_jX[jj[hWbceijj^[iWc[
7]h[WjZ[Wbmehi[ 7b_jjb[X_jX[jj[h
Gk_j[WX_jmehi[ Iec[m^WjX[jj[h
CeZ[hWj[bomehi[ CeZ[hWj[boX[jj[h
Iec[m^Wjmehi[ Gk_j[WX_jX[jj[h
7b_jjb[X_jmehi[ 7]h[WjZ[WbX[jj[h
7j_doX_jmehi[Wbceijj^[iWc[ 7l[ho]h[WjZ[WbX[jj[h
7Xekjj^[iWc[
[ RESEARCH REPORT ]
7FF;D:?N8
GK;IJ?EDIKI;:JEE8J7?D?D<EHC7J?ED:KH?D=J>;>?IJEHO#J7A?D=I;II?ED
How long have you had your present episode of knee pain? =_l_d]mWo5
Is your knee pain on just one side or both? 9b_Ya_d]5
?\oek[nf[h_[dY[fW_d_dXejâd[[i"_ied[i_Z[mehi[j^Wdj^[ej^[h5 9h[f_jki]h_dZ_d]ehYhkdY^_d]de_i[i5
?\o[i"m^_Yî_Z[_imehi[5 Im[bb_d]5
Was there trauma associated with the onset of the present episode of pain? Does your knee get stiff after sitting still for more than 20 minutes?
Do you have a prior history of knee pain?
Which of the following activities is MOST painful for you:
Have you ever had surgery on either knee? If yes, on which knee?
IgkWjj_d]
Do you have a history of any systemic diseases (diabetes, rheumatoid ar-
thritis, heart disease, etc)? Kfij[fi
Females: Are you pregnant? :emdij[fi
With regards to your knee, do you ever experience: Hkdd_d]
BeYa_d]5 B_\j_d]
7FF;D:?N9
EF;H7J?ED7B:;<?D?J?EDI<EHJ;IJI7D:C;7IKH;I
J_X_WbJehi_ed stood relaxed with the feet 15 cm apart as the examiner measured the angle
Measured with the subject placed in the prone position, with the patellae between the 2 lines previously drawn on the leg and heel32
placed on a widthwise line on the plinth.1,15 With the knee flexed to 90°, the
J_X_Wb7d]kbWj_ed
investigator measured tibial torsion as the angle created by the line on the
table and a line bisecting the malleoli Measured with the patient standing relaxed on the 20-cm step stool with the
feet 15 cm apart. Tibial angulation was measured as the angle created by
IkXjWbWh@e_djD[kjhWbDed#M[_]^j8[Wh_d] the line on the lower leg with a line created by the surface of the step stool.
Measured with the subject in prone and the knees extended. The investigator Tibial valgum was recorded if the proximal tibia was more medial than the
drew 2 lines on the subject using a felt-tipped pen. The first was a line bisect- distal tibia, and tibial varum was recorded if the proximal tibia was more
ing the lower leg, and the second bisected the calcaneus. The subject was lateral than the distal tibia
placed by the investigator into what we believed was subtalar joint neutral
judged by palpation of the head of the talus, and measurements of subtalar DWl_YkbWh:hef
joint neutral (measured as the angle between the 2 lines)1,15 were taken Measured with the patient assuming a stance position on the floor with feet 15
cm apart (maintained by placing a wooden block between the subject’s heels
C[Wikh[c[dji\eh<eh[\eej7b_]dc[dj
before each measurement, then removing it to perform the measurement).40
Taken using the same subtalar neutral position and forefoot alignment.24 The navicular drop measurement was taken by marking a dot on the navicular
The angle created by an axis perpendicular to the calcaneal line previously
tuberosity using a felt tipped pen. The subject then shifted their weight away from
drawn and an axis along the metatarsal heads was measured as forefoot
the side to be measured. The foot was then placed by the examiner into subtalar
alignment
joint neutral, and a mark was made on an index card placed vertically next to the
7dab[:ehi_Ô[n_ed7Yj_l[ navicular tuberosity and flush with the floor. The subject was then asked to relax
Measured with the knees flexed and with the knees extended in the prone his or her stance onto both feet, and the new position of the navicular tuberosity
position with the feet and ankles off the edge of the plinth1 was marked on the same index card. The difference between the 2 markings was
measured in millimeters and reflected the navicular drop or rise40
C[Wikh[c[djie\>_f?dj[hdWbHejWj_ed
Taken with the subject in prone and the knees both flexed to 90°40 (<?=KH;)) G#7d]b[
Measured as described by Magee40
J^[9hW_]J[ij
Performed to determine if femoral retroversion or anteversion was present F[bl_YEXb_gk_jo
as described according to Magee40 Measured using the palpation meter (PALM) as described by Petrone.47 After
>_f;nj[diehCWdkWbCkiYb[J[ij_d]edW+#Fe_djIYWb[ the Q-angle measurement was taken, the subject was asked to march in
Performed with the patient prone and the knees flexed40 place 5 times, ending with the feet 15 cm apart. With the subject remaining
in the standing position with the feet 15 cm apart, the standing flexion test
H[bWn[Z9WbYWd[WbIjWdY[ and stork test were performed as described by Magee to indicate the pres-
Measured with the subject standing on a 20-cm step stool. The subject ence or absence of sacroiliac joint abnormality40
<_hijC[jWjWhief^WbWd][WbCJF@e_dj;nj[di_ed J^[FWj[bbWh=b_Z[J[ij
Measured with the subject in the tailor-sit position on the edge of the plinth Used to assess patellar position. With the subject supine and the knees
(one ankle crossed over the contralateral knee). In this position, a measure- extended, the examiner marked the center of the patella and the medial
ment of the great toe passive extension was taken. The stationary arm of the and lateral epicondyles. The test was positive if the center of the patella was
goniometer was positioned parallel to the first metatarsal, the moving arm greater than or equal to 0.5 cm from the point halfway between epicondyles,
was positioned along the proximal phalanx of the great toe, and the axis of measured by a tape measure40
motion was at the first MTP joint35
J^[FWj[bbWhJ_bjJ[ij
CY9edd[bbÊiJ[ij Used to assess patellar mobility as the examiner glided the patella later-
Used as a provocative test to reproduce PFPS pain with the subject sitting ally and attempted to lift the lateral border of the patella anteriorly. The
on the edge of the plinth with both knees flexed. The examiner placed the measurement was recorded as: no lift, lift to neutral, or lift above the hori-
knee into varying degrees of flexion (0°, 30°, 60°, 90°, and 120°). At each zontal plane78
position, the subject isometrically contracted the quadriceps against resis-
tance from the examiner. If pain was produced at one of those positions, a J^[BWj[hWbFWj[bbWhFkbbJ[ij
second isometric contraction was then performed at the same angle with Detects any excessive lateral pull of the patella by the quadriceps or from
the patella manually glided medially by the examiner. The test was positive the lateral retinaculum. The subject, in supine, isometrically contracted
if the subject’s pain was significantly reduced with the patella glided medi- the quadriceps as the examiner observes the path of the patella. A positive
ally.43 The subject then moved to the edge of the plinth so that both feet test was indicated if there was greater lateral than superior movement of
contacted the floor. From this starting position, the Thomas test was used the patella78
to measure hip flexor or quadriceps tightness. A positive test was indicated
if the angle of knee flexion was less than 80°40 EX[hÊiJ[ij
Used to assess the presence or absence of iliotibial band tightness with the
J^[/&%/&IjhW_]^j#B[]HW_i[J[ij patient in the sidelying position35
Performed with the subject in supine and the hip and knee of the limb to
be tested flexed to 90°. Normal hamstring flexibility was documented if the 7CWdkWbCkiYb[J[ije\j^[=bkj[kiC[Z_ki
subject was able to extend the knee to within 20° of full extension40 Performed as described by Kendall35
?DL?J;:9ECC;DJ7HO current study is equally limited, in that patellofemoral joint stress and, therefore,
the potential causes for the observed de- pain. For example, if a patient performed
The article by Iverson et al6 regard- crease in pain in those who responded to a knee extension maneuver with a 5-kg
ing the use of lumbopelvic manipulation manipulation were not quantified. ankle weight, and this activity reproduced
as a treatment for patellofemoral pain Unlike the spine, where a manipula- pain, would increasing the weight to 6 kg
raises some interesting questions and, tive procedure would be expected to have be expected to cause a decrease in symp-
at the same time, poses some potential a direct mechanical influence on a func- toms? While I appreciate the fact that
concerns. I appreciate the opportunity to tional spinal unit, the link between ma- improved quadriceps strength has been
comment on this study and hope to stim- nipulation and the patellofemoral joint is shown to be associated with better long-
ulate some discussion within the clinical not as obvious. In their discussion, the au- term outcomes in this population,7 one
and research communities. thors proposed 2 possible hypotheses to must be careful in using this argument
The theoretical construct underlying explain why certain patients may respond to make a case in explaining immediate
the use of lumbopelvic manipulation to favorably to lumbopelvic manipulation: changes in pain.
treat patellofemoral pain is based on 2 (1) improved quadriceps function and An alternative explanation proposed
studies that have shown that lumbopelvic (2) the concept of regional interdepen- by the authors to explain the favorable
manipulation decreases quadriceps inhi- dence. I would like to comment on both response seen in their success group is
bition11 and increases quadriceps muscle of these proposed hypotheses, as well as based on concept of regional interdepen-
strength5 in persons with patellofemoral offer some alternative explanations. dence (ie, that impairments in remote
pain. In their introduction, Iverson et al6 The premise that improved quadri- regions may contribute to the patient’s
correctly point out that a limitation of ceps force production would result in a primary complaint).12 This theory makes
these 2 previous studies was that investi- decrease in patellofemoral symptoms is more sense from a mechanistic stand-
gators did not measure the symptomatic in contrast to what is known about pa- point, particularly because recent studies
responses of their subjects. As a result, tellofemoral joint biomechanics. Given have pointed to abnormal hip function as
it could not be determined whether the that the forces generated by the quadri- being contributory to patellofemoral joint
observed change in quadriceps muscle ceps muscle group are responsible for the problems.1,6,10,14 To make a valid argument
function resulting from manipulation compressive loads experienced by the pa- for the concept of regional interdepen-
had any bearing on patellofemoral symp- tella,4 one could argue that an increase in dence, however, it would have to be shown
toms. However, one could argue that the quadriceps muscle action would increase that manipulation resulted in a change in
[ RESEARCH REPORT ]
a remote impairment that could explain strategies when asked to perform activi- Musculoskeletal Biomechanics Research
a change in symptoms. With that being ties that reproduce symptoms.14 Subtle, Laboratory
said, there is no reason to suspect that self-selected changes in foot position, University of Southern California
lumbopelvic manipulation would have an trunk position, and knee flexion angle Los Angeles, CA
influence on the most powerful predictor could explain changes in pain levels with
of treatment success found in this study repeated provocative testing. REFERENCES
(ie, side-to-side differences in hip inter- While I applaud the authors’ efforts to 1. Bolgla LA, Malone TR, Umberger BR, Uhl TL. Hip
strength and hip and knee kinematics during
nal rotation range of motion greater than propose possible treatment approaches for
stair descent in females with and without patel-
14°). However, given the recent focus on a difficult clinical problem, I would argue lofemoral pain syndrome. J Orthop Sports Phys
diminished gluteus maximus and medius that additional data are needed before lum- Ther. 2008;38:12-18. http://dx.doi.org/10.2519/
function in this population,1,10 it would be bopelvic manipulation can be considered a jospt.2008.2462
2. Childs JD, Fritz JM, Flynn TW, et al. A clinical
interesting to know if lumbopelvic ma- viable treatment option for patellofemoral prediction rule to identify patients with low
nipulation had an influence on the activa- pain. Most importantly, it needs to be es- back pain most likely to benefit from spinal ma-
tion of these muscles, along with possible tablished whether the decrease in pain as- nipulation: a validation study. Ann Intern Med.
alterations in lower-extremity alignment sociated with manipulation has any lasting 2004;141:920-928.
3. Christou EA. Patellar taping increases vastus
during functional testing. effect. I would be more persuaded of the
medialis oblique activity in the presence of

Apart from the theories put forth by value of lumbopelvic manipulation if pain patellofemoral pain. J Electromyogr Kinesiol.
the authors, one cannot discount the pos- reduction could be maintained following a 2004;14:495-504. http://dx.doi.org/10.1016/j.
jelekin.2003.10.007
sibility of a placebo effect in this study. vigorous bout of activity (eg, a 20-minute
4. Feller JA, Amis AA, Andrish JT, Arendt EA,
The placebo effect is well documented in run) or at least persisted 24 to 48 hours. Erasmus PJ, Powers CM. Surgical biomechan-
studies evaluating the immediate effects While spinal manipulation has been shown ics of the patellofemoral joint. Arthroscopy.
of pain relief using various interventions to benefit persons with acute low back 2007;23:542-553. http://dx.doi.org/10.1016/j.
arthro.2007.03.006
in persons with patellofemoral pain. For symptoms,2 I question whether this form
5. Hillermann B, Gomes AN, Korporaal C, Jackson D.
example, a study by McCrory et al8 dem- of treatment would have an enduring effect A pilot study comparing the effects of spinal ma-
onstrated that a placebo brace condition on persons with chronic symptoms (sub- nipulative therapy with those of extra-spinal ma-
resulted in a 30% decrease in patell- jects in the current study had an average nipulative therapy on quadriceps muscle strength.
J Manipulative Physiol Ther. 2006;29:145-149.
ofemoral symptoms, which was similar in length of symptoms of 95 weeks).
http://dx.doi.org/10.1016/j.jmpt.2005.12.003
magnitude to the actual brace condition. Lastly, I would encourage future in- 6. Iverson CA, Sutlive TG, Crowell MS, et al.
Similarly, Christou3 reported that a sham vestigations in this area to address some Lumbopelvic manipulation for the treatment of
tape condition reduced pain by 70%, of the methodological issues that were patients with patellofemoral pain syndrome: de-
velopment of a clinical prediction rule. J Orthop
which was statistically similar in magni- acknowledged in the current study. In Sports Phys Ther. 2008;38:297-312. doi:10.2519/
tude to the actual tape condition (80% particular, the poor reliability of predic- jospt.2008.2669.
reduction in pain). A recent systematic tor measures that were obtained, com- 7. Kannus P, Niittymaki S. Which factors predict
review and meta-analysis of studies eval- bined with the small sample size for the outcome in the nonoperative treatment of
patellofemoral pain syndrome? A prospec-
uating the effects of patella taping and number of prediction variables under tive follow-up study. Med Sci Sports Exerc.
bracing on chronic knee pain concluded consideration, casts a level of suspicion 1994;26:289-296.
that approximately 50% of the benefit of on the stability of the proposed clinical 8. McCrory JL, Quick NE, Shapiro R, Ballantyne
BT, McClay Davis I. The effect of a single treat-
medially directed tape could be explained prediction rule. I would urge the authors
ment of the Protonics system on biceps femoris
by sham treatment effects.13 Although it to consider these issues as they move for- and gluteus medius activation during gait
could be argued that a sham intervention ward in replicating these findings and/or and the lateral step up exercise. Gait Posture.
that provides local sensory input and/or validating the proposed clinical prediction 2004;19:148-153. http://dx.doi.org/10.1016/
S0966-6362(03)00055-9
proprioceptive feedback may produce a rule. Furthermore, the addition of objec-
/$ Powers CM, Ward SR, Fredericson M, Guillet M,
stronger placebo effect than a remote in- tive measures to support an underlying Shellock FG. Patellofemoral kinematics during
tervention such as lumbopelvic manipu- mechanism behind the apparent effec- weight-bearing and nonweight-bearing knee
lation, one cannot discount the possibility tiveness of lumbopelvic manipulation in extension in persons with lateral subluxation of
the patella: a preliminary study. J Orthop Sports
of a placebo effect in this study. a subgroup of patients with patellofemo-
Phys Ther. 2003;33:677-685.
Another possible explanation for the ral pain would provide a more compelling 10. Robinson RL, Nee RJ. Analysis of hip strength in
decrease in symptoms in the success argument for this treatment approach. females seeking physical therapy treatment for
group may be related to “pain avoidance” unilateral patellofemoral pain syndrome. J Or-
thop Sports Phys Ther. 2007;37:232-238. http://
behavior. For example, it has been sug- Christopher M. Powers, PT, PhD dx.doi.org/doi:10.2519/jospt.2439
gested that individuals with patellofemo- Division of Biokinesiology and 11. Suter E, McMorland G, Herzog W, Bray R.
ral pain adopt compensatory movement Physical Therapy
Decrease in quadriceps inhibition after sac- thop Sports Phys Ther. 2007;37:658-660. http:// org/10.1002/art.23242
roiliac joint manipulation in patients with dx.doi.org/10.2519/jospt.2007.0110 14. Willson JD, Davis IS. Lower extremity mechan-
anterior knee pain. J Manipulative Physiol Ther. 13. Warden SJ, Hinman RS, Watson MA, Jr., Avin KG, ics of females with and without patellofemoral
1999;22:149-153. Bialocerkowski AE, Crossley KM. Patellar taping pain across activities with progressively greater
12. Wainner RS, Whitman JM, Cleland JA, Flynn TW. and bracing for the treatment of chronic knee task demands. Clin Biomech (Bristol, Avon).
Regional interdependence: a musculoskeletal pain: a systematic review and meta-analysis. 2008;23:203-211. http://dx.doi.org/10.1016/j.
examination model whose time has come. J Or- Arthritis Rheum. 2008;59:73-83. http://dx.doi. clinbiomech.2007.08.025
7KJ>EHIÊH;IFEDI; reducing any quadriceps inhibition that following the manipulation and, there-
might have been present. We believe that fore, do not know if a change in hip in-
We appreciate Dr Powers’ comments it is possible that the manipulation could ternal rotation range of motion, or any of
in response to our paper entitled “Lum- have produced rapid neurophysiological the physical examination measures, actu-
bopelvic manipulation for the treatment effects,1,6-8,13,14 which may include (1) pain ally occurred following the treatment. A
of patients with patellofemoral pain syn- modulation or (2) a reduction in quad- current investigation in this laboratory is
drome: development of a clinical predic- riceps inhibition that improved muscle directed toward determining whether a
tion rule.” We hope that discussion of this performance and allowed the subjects to successful response to the manipulation
study will stimulate further questions, perform the functional tests following the is correlated with a reduction or correc-
ideas, and research that ultimately con- manipulation with less pain. We think tion of hip internal rotation in subjects
tribute to the successful management of that it is plausible that the manipulation who meet the CPR. Future investigations
patients with this challenging disorder. may have simply facilitated normal acti- should also determine if changes in re-
As stated in our paper, this was a pre- vation of the quadriceps muscle. mote impairments (ie, gluteal and hip ro-
liminary study and the initial step11 in the An alternative explanation that we tator muscle activation) occur following
development of a clinical prediction rule proposed for the clinical improvement lumbopelvic manipulation and if those
that identifies patients with patellofemo- seen in the success group is based on the changes are correlated with the subjects’
ral pain syndrome (PFPS) who respond concept of regional interdependence.16 response to treatment.
successfully to lumbopelvic manipula- Based on this theory, alterations in joint Dr Powers also stated that a placebo
tion. We also acknowledged that there mobility in remote regions such as the effect may have occurred in our study,
were a number of limitations in this lumbar spine, sacroiliac joint, or hip may and we certainly agree that this is possi-
developmental study. Dr Powers stated contribute to the development of PFPS. ble. As he noted, the placebo effect is well
that one of the limitations of our study Given that hip internal rotation asym- documented in investigations involving
was the fact that “the potential causes for metry was the most powerful predictor of other treatments in individuals with pa-
the observed decrease in pain in those treatment success in the current study, it tellofemoral pain. Future studies that in-
who responded to manipulation were not is possible that the manipulation worked clude control subjects and possibly sham
quantified.” However, the research design by affecting lower extremity kinematics. interventions are necessary to determine
that we employed in this study does not We agree with Dr Powers’ statement that the efficacy of the treatment that we used
allow us to elucidate or quantify the rea- this explanation makes more sense from in the current study. A future validation
sons for the observed pain relief in those a mechanistic standpoint, particularly study in the form of a randomized clini-
who responded to the manipulation. We in light of the growing body of evidence cal trial is necessary before the proposed
can only be sure that we predicted who which suggests that impairments in the clinical prediction rule can be used con-
had a clinical improvement in response hip joint and gluteal region are associ- fidently in the clinic.
to the manipulation technique. Deter- ated with PFPS.3,4,9,10,12,15 Dr Powers also We developed a preliminary clinical
mining the efficacy of a given treatment wrote that “there is no reason to suspect prediction rule and recognize that other
and the potential mechanisms for pain that lumbopelvic manipulation would limitations of the study include a short
reduction should be the goal of future have an influence on the most powerful follow-up period, a limited sample size,
randomized clinical trials and mechanis- predictor of treatment success found in and marginal reliability of some of the
tic studies. this study.” Interestingly, previous stud- predictors of treatment success, as delin-
We did speculate about possible ies in individuals with low back pain also eated in the discussion and conclusion
mechanisms for the symptomatic im- demonstrated an association between hip sections. As Dr Powers suggested, future
provement experienced by the respond- internal rotation impairments and a suc- replication or validation studies should
ers in our study. One of the possibilities cessful response to the same lumbopelvic address each of these issues. Neverthe-
that we discussed was based on previous manipulation technique that we used.2,5 less, we believe that the current study
reports by Suter et al,13,14 who showed that However, in the current study we did not was a reasonable initial investigation
the manipulation may have worked by repeat any of the clinical measurements to identify persons with PFPS who may
[ RESEARCH REPORT ]
benefit from lumbopelvic manipulation. jospt.2008.2462 and the lateral step up exercise. Gait Posture.
2. Childs JD, Fritz JM, Flynn TW, et al. A clinical 2004;19:148-153. http://dx.doi.org/10.1016/
We hope that our findings will eventually
prediction rule to identify patients with low S0966-6362(03)00055-9
contribute to the development of a useful back pain most likely to benefit from spinal ma- /$ Powers CM, Ward SR, Fredericson M, Guillet M,
clinical decision-making model for the ef- nipulation: a validation study. Ann Intern Med. Shellock FG. Patellofemoral kinematics during
fective management of patients with the 2004;141:920-928. weight-bearing and nonweight-bearing knee
3. Christou EA. Patellar taping increases vastus
condition. extension in persons with lateral subluxation of
medialis oblique activity in the presence of
the patella: a preliminary study. J Orthop Sports
patellofemoral pain. J Electromyogr Kinesiol.
Phys Ther. 2003;33:677-685.
Christine A. Iverson, PT, DPT 2004;14:495-504. http://dx.doi.org/10.1016/j.
10. Robinson RL, Nee RJ. Analysis of hip strength in
Blanchfield Army Community Hospital jelekin.2003.10.007
females seeking physical therapy treatment for
Fort Campbell, KY 4. Feller JA, Amis AA, Andrish JT, Arendt EA,
Erasmus PJ, Powers CM. Surgical biomechan- unilateral patellofemoral pain syndrome. J Or-
ics of the patellofemoral joint. Arthroscopy. thop Sports Phys Ther. 2007;37:232-238. http://
Thomas G. Sutlive, PT, PhD 2007;23:542-553. http://dx.doi.org/10.1016/j. dx.doi.org/doi:10.2519/jospt.2439
US Army-Baylor University Doctoral arthro.2007.03.006 11. Suter E, McMorland G, Herzog W, Bray R.
5. Hillermann B, Gomes AN, Korporaal C, Jackson Decrease in quadriceps inhibition after sac-
Program in Physical Therapy
D. A pilot study comparing the effects of spinal roiliac joint manipulation in patients with
San Antonio, TX manipulative therapy with those of extra- anterior knee pain. J Manipulative Physiol Ther.
spinal manipulative therapy on quadriceps 1999;22:149-153.

Josef H. Moore, PT, PhD muscle strength. J Manipulative Physiol Ther. 12. Wainner RS, Whitman JM, Cleland JA, Flynn TW.
US Army-Baylor University Doctoral 2006;29:145-149. http://dx.doi.org/10.1016/j. Regional interdependence: a musculoskeletal
jmpt.2005.12.003 examination model whose time has come. J Or-
Program in Physical Therapy 6. Iverson CA, Sutlive TG, Crowell MS, et al. thop Sports Phys Ther. 2007;37:658-660. http://
Fort Sam Houston, TX Lumbopelvic manipulation for the treatment of dx.doi.org/10.2519/jospt.2007.0110
patients with patellofemoral pain syndrome: de- 13. Warden SJ, Hinman RS, Watson MA, Jr., Avin KG,
Robert S. Wainner, PT, PhD velopment of a clinical prediction rule. J Orthop
Bialocerkowski AE, Crossley KM. Patellar taping
Sports Phys Ther. 2008;38:297-312.doi:10.2519/
Texas State University and bracing for the treatment of chronic knee
jospt.2008.2669.
San Marcos, TX pain: a systematic review and meta-analysis.
7. Kannus P, Niittymaki S. Which factors predict
Arthritis Rheum. 2008;59:73-83. http://dx.doi.
outcome in the nonoperative treatment of
patellofemoral pain syndrome? A prospec- org/10.1002/art.23242
REFERENCES 14. Willson JD, Davis IS. Lower extremity mechan-
tive follow-up study. Med Sci Sports Exerc.
1. Bolgla LA, Malone TR, Umberger BR, Uhl TL. Hip ics of females with and without patellofemoral
1994;26:289-296.
strength and hip and knee kinematics during pain across activities with progressively greater
8. McCrory JL, Quick NE, Shapiro R, Ballantyne
stair descent in females with and without patel- BT, McClay Davis I. The effect of a single treat- task demands. Clin Biomech (Bristol, Avon).
lofemoral pain syndrome. J Orthop Sports Phys ment of the Protonics system on biceps femoris 2008;23:203-211. http://dx.doi.org/10.1016/j.
Ther. 2008;38:12-18. http://dx.doi.org/10.2519/ and gluteus medius activation during gait clinbiomech.2007.08.025

Manual Terapy in SDFP

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Manual Terapy in SDFP

Uploaded by

Copyright:

Available Formats

[ RESEARCH REPORT ]

CHRISTINE A. IVERSON, PT, DPT¹J>EC7I=$IKJB?L;"PT, PhD, OCS²C?9>7;BI$9HEM;BB"PT, DPT³

Lumbopelvic Manipulation for the

validity study. mediate overall 50% or greater reduction in pain,

improve following treatment with lumbopelvic

Most painful activity (%)†

ers. Subject demographics and baseline

J[ij%C[Wikh[ ?992,1 /+9? I;C AWffW

First MTP extension 0.52 0.29, 0.70 6.79°

Standing ﬂexion 0.88 colleagues74 showed that improvements in

Fh[Z_Yjehe\IkYY[ii I[di_j_l_jo If[Y_ÓY_jo !BH ÅBH 9kje÷IYeh[

J78B;+ Combinations of Predictor Variables and Associated Accuracy Statistics*

LWh_WXb[iFh[i[dj I[di_j_l_jo If[Y_ÓY_jo !BH ÅBH FheXWX_b_joe\IkYY[ii†

expressing a reduction of symptoms.  I_Z[jei_Z[Z_÷[h[dY[_d^_f_dj[hdWbhejWj_ed14°

erally had chronic PFPS (mean duration 0 2 4

of symptoms, 95 weeks). Nevertheless, 1 0 11

the development of the current CPR is a 2 5 11

reasonable initial foray at identifying the 3 8 1

characteristics of patients with PFPS who 4 5 0

are most likely to experience symptomat- 5 2 0

ic improvement following manipulation

10. Croft PR, Nahit ES, Macfarlane GJ, Silman AJ.

68. Suter E, McMorland G, Herzog W, Bray R. Sports Med. 2006;34:630-636. http://dx.doi.

medialis oblique activity in the presence of

spinal manipulative therapy on quadriceps 1999;22:149-153.

You might also like

CHRISTINE A. IVERSON, PT, DPT¹J>EC7I=$IKJB?L;"PT, PhD, OCS²C?9>7;BI$9HEM;BB"PT, DPT³

J[ij%C[Wikh[ ?992,1 /+9? I;C AWffW

Fh[Z_Yjehe\IkYY[ii I[di_j_l_jo If[Y_ÓY_jo !BH ÅBH 9kje÷IYeh[

J78B;+ Combinations of Predictor Variables and Associated Accuracy Statistics*

LWh_WXb[iFh[i[dj I[di_j_l_jo If[Y_ÓY_jo !BH ÅBH FheXWX_b_joe\IkYY[ii†

expressing a reduction of symptoms. I_Z[jei_Z[Z_÷[h[dY[_d^_f_dj[hdWbhejWj_ed14°