The pelvifemoral rhythm in cam-type femoroacetabular impingement

J. Van Houcke
a
, C. Pattyn
a
, L. Vanden Bossche
b
, C. Redant
b
, J.-W. Maes
a
, E.A. Audenaert
a,

a
Department of Orthopaedic Surgery and Traumatology, Ghent University Hospital, De Pintelaan 185, 9000 Ghent, Belgium
b
Department of Physical and Rehabilitation Medicine, Ghent University Hospital, De Pintelaan 185, 9000 Ghent, Belgium
a b s t r a c t a r t i c l e i n f o
Article history:
Received 13 May 2013
Accepted 23 October 2013
Keywords:
Hip
Femoroacetabular impingement
Pelvifemoral rhythm
Range of motion
Kinematics
Background: There is growing evidence that femoroacetabular impingement is a potentially important risk factor
for the development of early idiopathic osteoarthritis in the nondysplastic hip. Understanding of affected joint
kinematics is a basic prerequisite in the evaluation of mechanical disorders in a clinical and research oriented
setting. The aim of the present study was to compare pelvifemoral kinematics between subjects diagnosed
with femoroacetabular impingement and healthy controls.
Methods: The authors collected motion data of the femur and pelvis on a total of 43 hips 19 camimpingement
hips and 24 healthy controls using a validated electromagnetic tracking device. The pelvifemoral rhythm in
supine position was dened during both active and passive hip exion and statistically compared between
both groups.
Findings: A signicant increase in posterior pelvic rotation was observed during active hip exion in the
femoroacetabular impingement group compared with the control group (P b 0.001). During passive hip exion,
however, posterior pelvic rotation between the impingement group and the controls did not differ signicantly
(P = 0.628).
Interpretation: Posterior pelvic rotation during active high-end hip exion is increased in femoroacetabular
impingement, indicating the presence of an active compensational mechanism that decreases the extent of
harmful joint conict during high-exion activities.
2013 Elsevier Ltd. All rights reserved.
1. Introduction
Femoroacetabular impingement (FAI) is a relatively new concept,
unfolding a mechanism for the development of early labrum and carti-
lage lesions of the hip (Ganz et al., 2003). With an increased under-
standing of the condition and the recognition of FAI as a highly
prevalent pathology, especially in sports, the literature dealing with
the subject has grown exponentially. To date, FAI is recognized as a like-
ly cause of young adult osteoarthritis in Caucasian subjects (Ganz et al.,
2003, 2008; Reid et al., 2010). It is dened as a premature or repetitive
contact between the acetabular rim and proximal femur, potentially
resulting in damage to the joint. Underlying and predisposing anatom-
ical variations have been identied, in particular, an increased coverage
by the acetabulum (pincer-type impingement) and/or decreased sphe-
ricity of the femoral head (cam-type impingement) (Ganz et al., 2003;
Kassarjian et al., 2007; Lavigne et al., 2004; Tannast et al., 2007b).
Affectedsubjects usually report a history of paininthe groinor inthe
greater trochanter region, extending to the lateral side of the thigh, with
activities of daily living (e.g. sitting, stair climbing, squatting, driving)
and sporting activities (e.g. soccer, swimming, cycling, rowing) re-
quiring substantial hip exion (Clohisy et al., 2009; Ganz et al., 2003).
The range of movement in the hip joint is typically decreased, and
movements requiring high hip exion in combination with adduction
and/or internal rotation are most frequently affected among symptom-
atic patients (Clohisy et al., 2009; Ganz et al., 2003; Ito et al., 2004).
Hip exion or the approximation of the anterior thigh to the ante-
rior trunk is achieved through movement of the femur on the pelvis,
posterior tilting of the pelvis (rotation in the sagittal plane) and concur-
rent attening of the lumbar spine (Congdon et al., 2005). The relation-
ship between posterior pelvic rotation and hip exion has been studied
during active, passive and weight-loaded movements in normal sub-
jects who are supine, standing and suspended (Bohannon, 1982; R.
Bohannon et al., 1985; R.W. Bohannon et al., 1985; Dewberry et al.,
2003; Murray et al., 2002). In each of these studies, posterior pelvic
rotation has been found to clearly add to the overall hip exion. This ap-
parently synergistic relationship between femur exion and posterior
pelvic rotation during hip exion has become known as pelvifemoral
rhythm, as in the more familiar scapulohumeral rhythm, which
relates to the shoulder (Dewberry et al., 2003; Elia et al., 1996; Murray
et al., 2002). Until now, pelvifemoral rhythm remained uninvestigated
in FAI, and the question as to whether any changes in coordination
mechanisms of the body operate during hip exion remained unan-
swered. Identication of possible existing differences in pelvifemoral
kinematics is also important in light of the increasing use of patient
specic predictive models for cartilage and labrum damage based on
collision detection through motion simulation (Audenaert et al.,
2011a; Bedi et al., 2011; Tannast et al., 2007a). These protocols presume
Clinical Biomechanics 29 (2014) 6367
Corresponding author.
E-mail address: emmanuel.audenaert@ugent.be (E.A. Audenaert).
0268-0033/$ see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.clinbiomech.2013.10.019
Contents lists available at ScienceDirect
Clinical Biomechanics
j our nal homepage: www. el sevi er . com/ l ocat e/ cl i nbi omech
pelvic kinematics during open kinetic chain motions of the hip to be the
same in patients as in healthy controls.
The main purpose of this study was to investigate posterior pelvic
rotation during hip exion in FAI patients compared with controls and
thereby reveal possible differences in pelvic kinematics. Secondly, we
wanted to nd out whether potential differences in pelvifemoral
rhythmrepresent an active mechanismor, if suchis the result of passive
factors including the mechanical conict itself. To do so, we applied a ki-
nematic protocol and systemof measurement that had specically been
designed and validated for use in FAI (Audenaert et al., 2011b, 2012a).
2. Methods
2.1. Subjects
The study was designed as a casecontrol study comprising: (1) cam
impingement patients and (2) healthy controls. All subjects were men
aged 1835 years recruited between 1 January 2009 and 31 August
2012. The study was approved by the local ethics committee and all par-
ticipants signed an informed consent. In the absence of similar studies, a
preliminary pilot study of 8 cam impingement hips and 8 control hips
was performed to estimate the effect size. This allowed for a sample
size calculation based on the following parameters: effect size (=3.1),
standard deviation ( = 2.5), type II error rate ( = 0.2) and type I
error rate ( = 0.05). A minimum sample size of 9 hips per subgroup
was calculated to identify any signicant difference in pelvic kinematics
in FAI patients compared with controls. The study population com-
prised a total of 43 hips: 19 cam impingement hips (17 patients) and
24 healthy hips (12 controls). The patient group consisted of 11 right
hips (8 dominant-sided) and 8 left hips (3 dominant-sided).
Patients were recruited from cam-type FAI patients scheduled for
arthroscopic treatment. In an attempt to decrease the risk for selection
bias based on differences in training levels between patients and
healthy controls, only subjects engaged in recreational-level sport activ-
ities were included in the study. Recreational-level sport activities were
dened as a maximum of 4 h weekly sports and no competitional
activity. All patients presented with typical clinical signs of FAI, such
as limited internal rotation in 90 of exion, groin/lateral hip pain and
a painful anterior impingement test (groin pain elicited by passively
moving the hip in exion, adduction and internal rotation). The screen-
ing procedure was also aided by standard diagnostic imaging (radiogra-
phy andarthro-MRI), revealing anincreased alpha angle, and nally, the
diagnosis was conrmed during the actual arthroscopic procedure.
Healthy controls were recruited from the University Hospital per-
sonnel and from students by means of posters and emails requesting
to volunteer for the study. They were screened and selected on the
basis of a negative history of groin or lateral hip pain, the absence of
positive impingement testing, and, bilaterally, an alpha angle of b50
on anteroposterior imaging and Dunn views (45 hip exion, neutral
rotation and 20 abduction). Alpha angle measurements were per-
formed according to the original description by Notzli et al. (2002).
Patients and controls with solitary pincer-type impingement (positive
cross-over sign) or hip dysplasia (center-edge angle b 28) were ex-
cluded. Table 1 summarizes the selection criteria for both the patient
and control subgroups. Different radiographic parameters, which indi-
cated FAI or which might have an inuence on the condition, were
measured in each subject in a standardized fashion: alpha angle, caput-
collum-diaphyseal angle and lateral center-edge angle. In addition,
demographic variables (age, height, weight and BMI) were recorded.
2.2. Protocol
Kinematic measurements were performed using the Fastrak electro-
magnetic tracking system(Polhemus, Colchester, VT, United States). The
system uses magnetic eld pulses to track the position and orientation
of individual sensors relative to a satellite transmitter. A microprocessor
controls the transmitting and sensing signals andconverts theminto po-
sition and orientation data with 6 degrees of freedomrelative to a global
Cartesian coordinate systemprojected by the magnetic transmitter. The
system specications regarding measurement accuracy at close mea-
surement range of the transmitter are 0.8 mm and 0.15 degrees for po-
sition and orientation, respectively, according to the manufacturers.
Accuracy studies have conrmed that positional and orientational static
measurement errors are smaller than 2% for measurements at close
range, i.e. at a distance of 70 cm from the transmitter (Day et al., 2000;
Milne et al., 1996).
Previously, the same authors performed an in vitro validation and
reliability study of electromagnetic skin sensors for the evaluation of
end range of motion positions of the hip. Kinematic data from sensors
screwed into the bone of cadaver hips were compared to the data regis-
tered synchronously by sensors attached to the skin. This study revealed
that angular root meansquare (RMS) errors averaged 3.2 (SD3.5) and
1.8 (SD2.3) in the global reference frame for the femur and pelvic sen-
sors, respectively (Audenaert et al., 2011b). This measurement protocol
has recently been applied in a casecontrol study evaluating end range
motion in FAI (Audenaert et al., 2012a).
Patients and controls were evaluated ina supine position ona wood-
en investigation table, 1 m in height. In order to avoid any distortion of
the positional data, we ensured that there were no ferromagnetic mate-
rials nearby, and the transmitter was placed close to the test subject
(b0.2 m), parallel to the table. A supine examination position was pre-
ferred, to separate posterior pelvic rotation from other variables such
as weight shifts during hip exion, proprioceptive counterbalancing or
lumbar involvement. A unilateral femoral sensor and a contralateral
pelvic sensor were used simultaneously for all experiments. The pelvic
skin sensor was securely xed over the contralateral superioranterior
iliac spine. The femoral skin sensor was xed rigidly to a specially de-
signed distal femoral orthosis in order to decrease measurement errors
caused by skin movement. Furthermore, the sensor cables were taped
rigidly to the skin on the side of the trunk in order to prevent any inad-
vertent movements of the sensor during the course of the investigation.
Data readings were observed at a frequency rate of 40 Hz. A third sen-
sor, equipped with a stylus, was used to digitize palpable bony points
on the pelvis and femur. These digitized points were used to dene
the local coordinate system for the pelvis and femur, and comprised
the lateral and medial femoral epicondyles and the anterior and poste-
rior iliac spines. All landmarks were digitized in supine position.
The posterior iliac spines were accessed through an opening in the
wooden investigation table. Femoral and pelvic movements were then
Table 1
Selection criteria for patient and control subgroup. CE angle, center-edge angle.
Patients
(19 hips, 17 subjects)
Controls
(24 hips, 12 subjects)
Male
Aged 1835 years
CE angle between 28 and 40
Negative cross-over sign
Recretional level sports only
History of groin pain No history of groin pain
Pain on impingement testing No pain on impingement testing
Alpha angle N 55 Alpha angle b 50
Cam type impingement conrmed
on volumetric imaging
Healthy volunteer
64 J. Van Houcke et al. / Clinical Biomechanics 29 (2014) 6367
expressedinthe segmental coordinate system, as proposedby the Inter-
national Society of Biomechanics (Wu et al., 2002). The experimental
set-up and positioning of sensors is described in Fig. 1.
Initially, each subject was familiarized with the study protocol in a
training session 14 days prior to the actual measurements. These train-
ing sessions were essential for teaching the subjects howto performhip
exion without concomitant femoral rotation or abduction/adduction,
using reinforcement training based on visual feedback. All subjects
were trained and measured by the rst author. For the active exion
the subjects were instructed to ex the hip as far as they could go, there-
by ignoring any mild pain or discomfort. During passive exion the tes-
ter focusedonendfeel. Paintowards the endof the movement was often
present in patients but did not inuence the tester's decision to stop the
motion. The lumbar spine was not stabilized during hip exion to allow
for free pelvic rotation. Both active and passive exion movements were
repeated 11 times. The calculated mean values were used for statistical
analysis. During all exion movements, the knee was exed beyond
90 to eliminate hamstring involvement.
Processing of the data was carried out in a custom-developed pack-
age that provided visual feedback during the analysis (MatlabR2010a,
MathWorks, Natick, MA, United States). Measurements of posterior
pelvic rotation in relation to hip exion were recorded for both active
and passive exion; the data were then analyzed in a common global
reference frame. Segmental pelvic and femoral angles were zeroed
based on each participant's neutral position as determined by a supine
static trial where feet were parallel and facing anteriorly (Lamontagne
et al., 2009). Finally, the mean amount of posterior pelvic rotation in re-
lation to overall hip exion was graphically represented for both active
and passive hip exion in the patients and controls.
2.3. Statistical analyses
The SAS software package (SAS 9.3, SAS Institute, Cary, NC) was se-
lected for the statistical analyses. Hip exion and posterior pelvic rota-
tion, both passive and active, at the end of each subject's motion were
analyzed as dependent variables. The evaluation of differences in
range of hip exion and posterior pelvic rotation was carried out using
a mixed-design ANOVA test. This test allows for evaluating mean differ-
ences between two or more independent groups while subjecting the
subjects to repeated measures. Hip exion and posterior pelvic rotation
were the evaluated between-subject variables whereas the passive and
active movements within eachgroup served as within-subject variables.
The standard assumptions for normality of distribution were met by
means of the ShapiroWilk test and evaluation of the QQ plot. The ho-
mogeneity of variances was assessed using Levene's test. A Bonferroni
adjustment was applied for these 8 hypothesis tests, setting the level
of signicance at = 0.00625. Differences in demographic and radio-
graphic data (age, height, weight, BMI, CE-, CCD- and -angle) were
evaluated using the MannWhitney U test as the criteria required for
parametric testing had not been met. For these tests, the level of signif-
icance was set at a regular = 0.05.
3. Results
Patients and controls did not differ signicantly in terms of age,
height, weight, BMI, CE angle and CCD angle. As patients and controls
were recruited based on an alpha angle of more than 55 and less
than 50, respectively, their difference was signicant at the 0.001
level. The demographic and radiographic parameters are summarized
in Table 2.
Posterior pelvic rotation was signicantly increased in patients com-
pared with controls during active hip exion (P b 0.001; Fig. 2). During
passive hip exion, however, posterior pelvic rotation did not differ sig-
nicantly between patients and controls (P = 0.628; Fig. 3). Posterior
pelvic rotation during active hip exion signicantly increased in pa-
tients only (P = 0.003; see Table 3), not controls, compared with
rotation during passive hip exion. During both active (P = 0.005)
and passive hip exion (P = 0.003), the patients showed a signicant
decrease in the range hip exion. An overviewof the kinematic ndings
during active and passive hip exion is summarized in Table 3.
4. Discussion
This study demonstrates that posterior pelvic rotation during active
hip exion is increased in FAI patients compared with healthy controls.
As a similar increase was not observed during passive hip exion, this
nding therefore suggests the presence of an active mechanism. Similar
studies on pelvifemoral rhythm are scarce, and are usually limited to
healthy volunteers (Bohannon, 1982; Bohannon and Smutnick, 2010;
R. Bohannon et al., 1985; R.W. Bohannon et al., 1985; Congdon et al.,
2005; Dewberry et al., 2003; Murray et al., 2002). Furthermore, such
studies have been often designed to evaluate the inuence of hamstring
length on posterior pelvic rotation. Our subjects were instructed to ex
their knee progressively during hip exion, thereby eliminating ham-
string tension. The increase in posterior pelvic rotation with shorter
hamstrings in cases where the knees are locked in extension during
bilateral hip exion has been clearly demonstrated by Dewberry
et al. (2003) and Congdon et al. (2005).
Femoroacetabular impingement has been shown to occur predomi-
nantly in athletes involved in sports that require high-exion activities
such as squatting, soccer, hockey and rowing. Such sports rarely require
high hip exion coupled with extended knees. In affected patients, bony
impingement with large-hip-exion angles has been shown to occur
typically when combined with internal rotation or adduction of the
hip. With movements such as internal rotation in 90 of hip exion, sig-
nicant decreases in patients compared to controls have been reported
ranging from 11.3 to 15.6 (Audenaert et al., 2012a,b; Kubiak-Langer
et al., 2007). A relative increase in posterior pelvic rotation during active
hip exion, as observed in the present study, might therefore represent
an efcient way of decreasing or even avoiding harmful bony friction be-
tween the femur and pelvis. This increase in posterior pelvic tilt of the
cam patients was not observed in controls or during passive hip exion,
meaning active and adaptive pelvic kinematics were present. This obser-
vationimmediately raises some interesting newhypothesis andresearch
questions. Can techniques be taught and, if so, should for certain high-
risk sports preventive training programs be developed? Should more
emphasis be placed on postoperative rehabilitation programs that in-
clude exercises involving increased posterior pelvic tilt? What is the im-
pact on lumbar kinematics? Clearly, more research in this area is needed.
In our study, posterior pelvic rotation was consistently lower than
that documented in other studies, with reported ratios of posterior
pelvic rotation to hip exion ranging from 0.16 to 0.35, depending on
the specic study conditions (e.g. supine, standing, suspended from
a bar, knees exed or extended, weight loaded) (Bohannon, 1982;
R. Bohannon et al., 1985; R.W. Bohannon et al., 1985; Congdon et al.,
2005; Dewberry et al., 2003; Murray et al., 2002). In our control
group, the ratio was approximately 0.08 during both active and passive
Table 2
Demographic and radiographic parameters describing the subgroups (mean, 95% CI).
Patients differed signicantly from controls in alpha angle. CE angle, center-edge angle;
CCD angle, caput-collum-diaphyseal angle.
Patients (n = 17; 19 hips) Controls (n = 12; 24 hips) P
Age (years) 24.7 (22.227.1) 23.1 (21.624.6) 0.517
Height (m) 1.79 (1.761.83) 1.79 (1.761.82) 0.524
Weight (kg) 75.5 (71.379.7) 73.1 (68.577.7) 0.418
BMI (kg/m
2
) 23.4 (22.624.1) 22.8 (21.823.7) 0.433
CE-angle () 33.3 (31.635.1) 33.8 (31.935.6) 0.864
CCD-angle () 135.2 (133.3137.1) 135.9 (134.2137.6) 0.470
-angle () 71.8 (68.375.2) 48.1 (46.949.4) P b 0.001
P-values in bold are statistically signicant.
65 J. Van Houcke et al. / Clinical Biomechanics 29 (2014) 6367
hip exion, whereas, in patients, the ratio during active hip exion was
0.12. Besides the exclusion of elastic involvement of the hamstring mus-
cles, there are a few other methodological concepts that might add to
the observed differences reported in earlier studies. Firstly, we exclu-
sively investigated unilateral hip exion movements. R.W. Bohannon
et al. (1985) showed that unilateral hip exion movements tend to in-
volve less posterior pelvic rotation than bilateral ones, especially during
active hip exion. Secondly, the position of our pelvic skin sensors was
different to that in other studies. We specically chose to attach the
pelvic sensor to the contralateral anterior superior iliac spine instead
of the ipsilateral anterior/posterior superior iliac spine, avoiding skin
shift caused by the approaching femur. The cumulative effect of these
variables could explain the overall observed difference in pelvifemoral
rhythm compared with previous studies.
The current state of the art in pre-surgical assessment of FAI involves
ROM simulation based on collision detection methodology (Audenaert
et al., 2011a; Bedi et al., 2011; Tannast et al., 2007a). These models sim-
ulate ROMof the femur as in the open chain assessment in the coor-
dinate system of the pelvis, using a xed pelvis. Such analysis involves
two basic assumptions: Motionis bony constrained and pelvifemoral ki-
nematics are similar in FAI patients compared to normals. The goal of
the present study was not to generalize and simplify pelvic kinematics,
but to challenge the latter assumption. The validationof these predictive
models has shown troublesome, part of the problemmight in fact be at-
tributable to differences in pelvifemoral kinematics between FAI and
controls. To evaluate this, we therefore tried to decrease the number
of variables to an absolute minimum. Our results indicate indeed that
active changes in pelvic kinematics are present.
Fig. 1. Experimental set-up and positioning of sensors. Subjects lie supine with the femoral sensor (A) attached to the ipsilateral femur using a distal femoral orthosis. The pelvic sensor is
xed to the contralateral anterior superior iliac spine (B). The pointer (C) can be freely manipulated for the palpation and identication of anatomical points. The magnetic transmitter
(D) is leveled and carefully aligned with the wooden investigation table.
Fig. 2. Abar chart illustrating the relationship of posterior pelvic rotation to active hip ex-
ion for FAI patients and controls. Boxplots with mean and interquartile range, whiskers
represent the range.
Fig. 3. A bar chart illustrating the relationship of posterior pelvic rotation to passive hip
exionfor FAI patients andcontrols. Boxplots withmeanandinterquartile range, whiskers
represent the range.
66 J. Van Houcke et al. / Clinical Biomechanics 29 (2014) 6367
In strict biomechanical terms, the hip exion movement is dened
by the approximation of the femur to the pelvis (Wu et al., 2002). In
real life however, overall hip exion is achieved through the complex
combination of motion of the femur on the pelvis, posterior tilting of
the pelvis (rotation in the sagittal plane) and concurrent attening of
the lumbar spine (Congdon et al., 2005). This paper adopted the latter
denition when using the term hip exion. This reects in the study
protocol where the hip exion variable is measured as femur exion
relative to the examination table, the surrogate for the trunk.
The present study was beset with the classic limitations of in vivo ki-
nematic analysis. After all, inaccuracies in measurements resulting from
palpation errors while dening the local reference frames and errors in-
duced by skin movements are hard to avoid. This issue has been partial-
ly addressed by applying a specially designed femoral orthosis for the
femoral sensor. The current measurement methodology and protocol
have previously been validated based on acceptable RMS differences
of 1.8 in the pelvic frame and 3.2 in the femoral frame (Audenaert
et al., 2011b). The measured range of posterior pelvic rotation in this
study, however, largely exceeded this error range, and an absolute dif-
ference of 3.4 in pelvic rotation between patients and controls during
active hip exion was observed.
A second limitation refers to the limited generalizability of the nd-
ings in the present study. Clearly, open kinetic chain assessments do not
generalize to activities of daily living (ADL). This is problematic in that
joint disease occurs under loading conditions of the closed kinetic
chain. However, personalized predictive models on disease initiation
and progression in the context of FAI will need to take into account
that kinematics are different in FAI patients compared to normals, and
such cannot solely be attributed to decreased ROM in face of structural
conicts. Further work targeting pelvifemoral kinematics in ADL situa-
tions is clearly necessary for these purposes.
A nal issue is the potential for selection bias to occur in kinematic
studies involving a pathology that predominates in sportsmen/women.
This was addressed by selectively excluding subjects who participated
in sports at competition level or practiced more than 4 h of sports
weekly.
5. Conclusions
The present study reveals that posterior pelvic rotationduring active
hip exion is increased in FAI. This appears to indicate that FAI patients
actively attempt to avoid joint collision during high-exion activities by
increasing posterior pelvic rotation.
Acknowledgments
E. Audenaert is supported by the Flemish Research Foundation
(FWO) through a clinical research fellowship.
References
Audenaert, E.A., Mahieu, P., Pattyn, C., 2011a. Three-dimensional assessment of cam
engagement in femoroacetabular impingement. Arthroscopy 27, 167171.
Audenaert, E.A., Vigneron, L., Van Hoof, T., D'Herde, K., van Maele, G., Oosterlinck, D.,
Pattyn, C., 2011b. In vitro validation and reliability study of electromagnetic skin
sensors for evaluation of end range of motion positions of the hip. Med. Biol. Eng.
Comput. 49, 14051412.
Audenaert, E., Van Houcke, J., Maes, B., Vanden Bossche, L., Victor, J., Pattyn, C., 2012a.
Range of motion in femoroacetabular impingement. Acta Orthop. Belg. 78, 327332.
Audenaert, E.A., Peeters, I., Vigneron, L., Baelde, N., Pattyn, C., 2012b. Hip morphological
characteristics and range of internal rotation in femoroacetabular impingement.
Am. J. Sports Med. 40, 13291336.
Bedi, A., Dolan, M., Hetsroni, I., Magennis, E., Lipman, J., Buly, R., Kelly, B.T., 2011. Surgical
treatment of femoroacetabular impingement improves hip kinematics a computer-
assisted model. Am. J. Sports Med. 39, 43S49S.
Bohannon, R.W., 1982. Cinematographic analysis of the passive straight-leg-raising test
for the hamstring muscle length. Phys. Ther. 62, 12691274.
Bohannon, R.W., Smutnick, J., 2010. Pelvifemoral kinematics while ascending single steps
of different heights. J. Appl. Biomech. 26, 290294.
Bohannon, R., Gajdosik, R., Leveau, B.F., 1985a. Contribution of pelvic and lower-limb mo-
tion to increases in the angle of passive straight leg raising. Phys. Ther. 65, 474476.
Bohannon, R.W., Gajdosik, R.L., Leveau, B.F., 1985b. Relationship of pelvic and thigh
motions during unilateral and bilateral hip exion. Phys. Ther. 65, 15011504.
Clohisy, J.C., Knaus, E.R., Hunt, D.M., Lesher, J.M., Harris-Hayes, M., Prather, H., 2009.
Clinical presentation of patients with symptomatic anterior hip impingement. Clin.
Orthop. Relat. Res. 467, 638644.
Congdon, R., Bohannon, R., Tiberio, D., 2005. Intrinsic and imposed hamstring length
inuence posterior pelvic rotation during hip exion. Clin. Biomech. 20, 947951.
Day, J.S., Murdoch, D.J., Dumas, G.A., 2000. Calibration of position and angular data from a
magnetic tracking device. J. Biomech. 33, 10391045.
Dewberry, M.J., Bohannon, R.W., Tiberio, D., Murray, R., Zannotti, C.M., 2003. Pelvic and
femoral contributions to bilateral hip exion by subjects suspended from a bar.
Clin. Biomech. 18, 494499.
Elia, D.S., Bohannon, R.W., Cameron, D., Albro, R.C., 1996. Dynamic pelvic stabilization
during hip exion: a comparison study. J. Orthop. Sports Phys. Ther. 24, 3036.
Ganz, R., Parvizi, J., Beck, M., Leunig, M., Notzli, H., Siebenrock, K.A., 2003. Femoroacetabular
impingement: a cause for osteoarthritis of the hip. Clin. Orthop. Relat. Res. 112120.
Ganz, R., Leunig, M., Leunig-Ganz, K., Harris, W.H., 2008. The etiology of osteoarthritis of
the hip: an integrated mechanical concept. Clin. Orthop. Relat. Res. 466, 264272.
Ito, K., Leunig, M., Ganz, R., 2004. Histopathologic features of the acetabular labrum in
femoroacetabular impingement. Clin. Orthop. Relat. Res. 262271.
Kassarjian, A., Brisson, M., Palmer, W.E., 2007. Femoroacetabular impingement. Eur.
J. Radiol. 63, 2935.
Kubiak-Langer, M., Tannast, M., Murphy, S.B., Siebenrock, K.A., Langlotz, F., 2007. Range of
motion in anterior femoroacetabular impingement. Clin. Orthop. Relat. Res. 117124.
Lamontagne, M., Kennedy, M.J., Beaule, P.E., 2009. The effect of cam FAI on hip and pelvic
motion during maximum squat. Clin. Orthop. Relat. Res. 467, 645650.
Lavigne, M., Parvizi, J., Beck, M., Siebenrock, K.A., Ganz, R., Leunig, M., 2004. Anterior
femoroacetabular impingement: part I. Techniques of joint preserving surgery. Clin.
Orthop. Relat. Res. 6166.
Milne, A.D., Chess, D.G., Johnson, J.A., King, G.J.W., 1996. Accuracy of an electromagnetic
tracking device: a study of the optimal operating range and metal interference.
J. Biomech. 29, 791793.
Murray, R., Bohannon, R., Tiberio, D., Dewberry, M., Zannotti, C., 2002. Pelvifemoral
rhythm during unilateral hip exion in standing. Clin. Biomech. 17, 147151.
Notzli, H.P., Wyss, T.F., Stoecklin, C.H., Schmid, M.R., Treiber, K., Hodler, J., 2002. The con-
tour of the femoral head-neck junction as a predictor for the risk of anterior impinge-
ment. J. Bone Joint Surg. Br. 84B, 556560.
Reid, G.D., Reid, C.G., Widmer, N., Munk, P.L., 2010. Femoroacetabular impingement
syndrome: an underrecognized cause of hip pain and premature osteoarthritis?
J. Rheumatol. 37, 13951404.
Tannast, M., Kubiak-Langer, M., Langlotz, F., Puls, M., Murphy, S.B., Siebenrock, K.A., 2007a.
Noninvasive three-dimensional assessment of femoroacetabular impingement.
J. Orthop. Res. 25, 122131.
Tannast, M., Siebenrock, K.A., Anderson, S.E., 2007b. Femoroacetabular impingement:
radiographic diagnosis what the radiologist should know. Am. J. Roentgenol. 188,
15401552.
Wu, G., Siegler, S., Allard, P., Kirtley, C., Leardini, A., Rosenbaum, D., Whittle, M., D'Lima,
D.D., Cristofolini, L., Witte, H., Schmid, O., Stokes, H., 2002. ISB recommendation on
denitions of joint coordinate system of various joints for the reporting of human
joint motion part 1: ankle, hip, and spine. J. Biomech. 35, 543548.
Table 3
Kinematic data for patients and controls during active and passive hip exion movements
(mean, 95% CI). The right-hand column represents the P values of unpaired statistical
testing between patients and controls. The P values in the patient and control columns
represent the paired statistical testing between active and passive exion movements
within one subgroup.
Patients
(n = 17; 19 hips)
Controls
(n = 12; 24 hips)
P
Active hip exion ()
a
105.9 (101.3110.4) 113.4 (110.4116.5) 0.005
Passive hip exion ()
a
110.1 (106.0114.3) 118.9 (115.1122.6) 0.003
P = 0.042 P = 0.013
Active pelvic rotation ()
b
12.5 (11.513.6) 9.1 (8.29.9) P b 0.001
Passive pelvic rotation ()
b
10.5 (8.912.1) 10.0 (8.711.4) 0.628
P = 0.005 P = 0.104
P-values in bold are statistically signicant.
a
Overall hip exion in relation to the horizontal.
b
Posterior pelvic rotation in relation to the horizontal.
67 J. Van Houcke et al. / Clinical Biomechanics 29 (2014) 6367