Clinical Biomechanics 17 (2002) 147–151

www.elsevier.com/locate/clinbiomech

Pelvifemoral rhythm during unilateral hip flexion in standing

Richard Murray, Richard Bohannon, David Tiberio *, Michael Dewberry, Chad Zannotti
Department of Physical Therapy, School of Allied Health, University of Connecticut, Storrs, CT 06269-2101, USA
Received 24 October 2000; accepted 27 November 2001

Abstract
Objective. To quantify pelvifemoral rhythm during unilateral hip flexion in standing subjects, under different degrees of resis-
tance.
Design. Descriptive and correlational.
Background. Pelvic rotation contributes to the change in hip flexion angle while individuals are supine. Degree of contribution in
standing has not been described.
Methods. Pelvic and femoral motions, relative to vertical, were documented while 14 subjects performed three trials of standing
unilateral hip flexion, both with and without a 4.53 kg ankle weight. Pelvic and thigh segments were marked with reflective spheres,
and motion was captured by video analysis.
Results. Pelvic rotation contributed 18.1% of the change in hip flexion angle. Pelvic rotation occurred throughout the movement.
Neither presence of external resistance nor direction of movement significantly changed this contribution.
Conclusions. A strong relationship exists between pelvic rotation and hip flexion in standing subjects.

Relevance
Hip flexion, while standing, is a common element of many functional activities. Knowledge of the pelvic contribution increases
our understanding of hip and lumbar spine motion during standing activities. ! 2002 Published by Elsevier Science Ltd.

Keywords: Lower extremity; Hip; Motion analysis

1. Introduction
Distinguishing between femoral motion and pelvic
motion during hip flexion and extension is a common
clinical problem [1–3]. Certain clinical interventions (e.g.
hamstring stretching) attempt to control or eliminate
pelvic motion while flexing the hip [4]. Sagittal plane
motion of the pelvis will substitute for hip flexion/ex-
tension if there are contractures of the muscles or joint
capsule [3]. When measuring hip range of motion, cli-
nicians often erroneously identify the end of hip motion
with the initiation of pelvic movement [1,5]. This rela-
tionship has been the focus of previous research [6–10].
The contribution of pelvic motion to total motion of
the lower extremity usually has been studied in the su-
pine position during passive movements. Bohannon [11]
noted that pelvic rotation accounted for 26% of the total
straight leg raise (SLR) excursion. In a subsequent
*
Corresponding author.
E-mail address: david.tiberio@uconn.edu (D. Tiberio).
0268-0033/02/$ - see front matter ! 2002 Published by Elsevier Science Ltd.
PII: S 0 2 6 8 - 0 0 3 3 ( 0 1 ) 0 0 1 1 5 - 2
study, Bohannon et al. [12] found that 39% of the total
supine passive SLR could be attributed to pelvic rota-
tion. The pelvic contribution in those subjects ranged
from 30% to 46%. The most common assumption is that
as hamstring excursion reaches its limit the pelvis will
rotate posteriorly as the lower extremity is moved.
However, Bohannon et al. [13] demonstrated in a
follow-up study that the same pelvic rotation could be
demonstrated with the knee in a flexed position. The
flexed knee reduces the possibility of the hamstrings
causing the rotation. The substantial pelvic contribution
(26%, range 21–38%) could have resulted, at least in
part, from tension in the uniarticular muscles or joint
capsule. Accurate clinical examination requires a more
detailed understanding of this mechanism because this
pelvic motion during flexion of the hip produces lumbar
spine flexion and extension of the contralateral hip.
Pelvifemoral rhythm refers to the coordinated
movement of the pelvis and the femur during hip flex-
ion, whether the femur is moving on the pelvis or the
pelvis is moving on the femur. Pelvifemoral rhythm can
be described as the ratio of change in angle between the
148 R. Murray et al. / Clinical Biomechanics 17 (2002) 147–151

pelvis and vertical (PV) to change in angle between the
femur and vertical (FV), that is PV/FV. The concept of a
rhythm suggests that the pelvic movement does not oc-
cur only at the end of hip motion. Rather, a rhythm
would infer a continuous relationship between the two
bone segments that would occur during active motion
throughout the range, with both flexion and extension
movements of the femur.
Existing research on pelvifemoral rhythm has not
involved standing subjects. Neither the extension phase
of the hip flexion–extension cycle nor the influence of
external resistance has been examined. Hip flexion,
while standing, is a common element of many functional
activities. Actions such as walking, running, jumping
and climbing stairs all require open chain hip flexion
while individuals are in an upright position. Hip flexion
in standing is clearly followed by an extension phase and
sometimes involves resistance. Footwear, weight cuffs,
elastic tubing and other forms of resistance to weight
used during rehabilitation might alter the specifics of
this pelvifemoral rhythm, and change the indirect effect
on the lumbar spine.
The purpose of this study was to perform two-di-
mensional sagittal plane analysis in order to quantify
pelvifemoral rhythm during unilateral hip flexion by
standing subjects. Specifically, this study documented
the angular displacement of the pelvis relative to the
vertical and the angular displacement of the femur rel-
ative to vertical, and expressed them as the ratio PV/FV.
Flexion and extension phases, as well as the influence of
added external resistance, were investigated.
video tapes, 2.5 and 4 cm diameter reflective spheres, a
4.53 kg ankle weight and a metronome (Quartz Metro-
nome, Country Technologies; Gays Mill, WI, USA).
2.3. Procedures
Male subjects wore posing trunks, while female sub-
jects wore a bikini-type bathing suit, to allow unre-
stricted movement and placement of reflective markers.
To represent the pelvis, the authors applied two 2.5 cm
reflective spheres on a line connecting the PSIS and
ASIS on the right side of each subject. Four centimeter
spheres, placed 2 cm distal to the greater trochanter and
2 cm proximal to the lateral femoral condyle, repre-
sented the femur of the subject. Two 4 cm reflective
spheres were placed on the mid-axillary line of the
thorax, to measure swaying of the trunk during the
movement. Two spheres were placed on the wall next to
the subjects, along a vertical line confirmed by using a
level (Fig. 1). The video camera was placed approxi-
mately 4 m from and perpendicular to the subject’s
sagittal plane, at a height sufficient to include the sub-
ject’s entire body on film. Subjects were instructed to
maximally flex the right hip (closest to the camera), and

2. Methods

2.1. Sample

A convenience sample was used, consisting of 14

student volunteers: 9 males (mean age: 25.6 years, mean
height: 177.7 cm, mean weight: 78.5 kg) and 5 females
(mean age: 24 years, mean height: 165.4 cm, mean
weight: 54.8 kg). Provision of informed consent and the
ability to support full body weight on the left lower
extremity were prerequisites for participation. Subjects
were excluded if they had any known orthopedic or
neurological pathology of the spine, pelvis or lower ex-
tremities. In addition, subjects were disqualified if adi-
posity precluded appropriate placement and view of
reflective markers, or full range of motion needed for hip
flexion.

2.2. Instrumentation

Instrumentation included a Panasonic S-VHS

Reportex A6-450 video recorder (Matsushita Electric Fig. 1. Line drawing illustrating reflective marker placement used
Industrial; Okayama, Japan), Maxell HGX-Gold VHS during standing hip flexion.
 R. Murray et al. / Clinical Biomechanics 17 (2002) 147–151 149

then extend it back to starting position, while holding 2.5. Statistics

onto a stabilization frame. Fig. 1 illustrates a subject
during such a movement. Movement in each direction
was guided by four beats of a metronome, set at 2 beats
per second. Subjects were told to relax the ipsilateral
knee as the hip flexed. Several practice movements were
allowed before filming started, then subjects were filmed
performing the hip flexion three times. This procedure
was also performed with a 4.53 kg weight strapped to
the subject’s right ankle (both practice and filmed mo-
tions). A random drawing determined if the subject
performed weighted or non-weighted movements first.
Two specific segmental motions were documented:
angular displacement of the pelvis in relation to vertical,
and angular displacement between the femur and verti-
cal. Both of these motions were measured in degrees.
Furthermore, pelvifemoral rhythm was expressed as a
ratio of pelvi-vertical displacement (PV) to femoral-
vertical displacement (FV). Independent variables were
the presence of cuff weight and the direction of move-
ment.
The measurements associated with each interval were
entered into SPSS (Version 9.0) and Systat Statistical
Package for Windows (Version 8.0; Evanston, IL, USA)
for statistical analysis. Descriptive statistics were used to
indicate the contribution of pelvic and thigh motions to
hip excursion. Intraclass correlation coefficients (ICC)
were used to determine the reliability of the trials. As
they were found to be reliable (ICC: mean ¼ .8279,
range ¼ .6167–.9484), the mean values of the three trials
were used in all subsequent analysis. Linear regression
was used to illustrate the ratio PV/FV. Pearson product
moment correlations were used to describe the rela-
tionship between pelvic and femoral motion. Multifac-
torial analysis of variance (A N O V A ) procedures were
used to analyze the influence of weight and direction on
both the ratios and the correlations.
3. Results

2.4. Data processing
The videotape was digitized using the PEAK 5-mo-
tion analysis system (Peak Performance Technologies,
Englewood, CO, USA) at 60 Hz using automatic digi-
tizing software. Pelvic and femoral motions were
quantified. Data were entered into Microsoft Excel and
opened in spreadsheet format. Beginning, maximum and
ending FV values were selected, and femoral excursion
calculated. Excursions for both upward and downward
movements were divided into 20 equal intervals. Values
for PV and FV angles were taken from the digitized
frames corresponding to each interval.
Statistics describing pelvic and femoral motion are
presented in Table 1. A grand mean of 18.8" of pelvic
rotation relative to vertical accompanied a grand mean
of 98.9" of femoral motion relative to vertical. Table 1
shows how these values vary under each condition.
Statistics describing both the ratio and correlation
between pelvic and femoral motion are displayed in
Table 2. Rotation of the pelvis contributed a grand
mean of 18.1% (range: 8–53%) of total hip flexion angle
(femur vs. vertical). Pelvic rotation accounted for 17.0%
(range: 8–32%) of hip flexion with an unweighted lower
extremity and 19.2% (range: 9–53%) when the extremity
was loaded with a 4.53 kg cuff weight. There was a

Table 1
Mean excursions of pelvis and femur (both against vertical) during each condition
Condition Direction PV (") FV (")
No weight Up 19.1 (range: 11.7–32.2) 102.3 (88.7–117.8)
Down 17.4 (10.5–28.8) 98.8 (62.9–113.7)
Weight Up 21.1 (13.5–43.2) 98.3 (88.1–113.0)
Down 17.5 (12.3–30.0) 95.9 (80.4–108.9)
Grand mean 18.8 98.9

Table 2
Descriptive statistics for the ratio and correlation between pelvic and femoral motion during unilateral hip flexion in standing
Condition Direction Ratio Correlation
Mean SD Range Mean SD Range
No weight Up .173 .048 .102–.278 .971 .030 .888–.994
Down .168 .047 .096–.261 .979 .019 .936–.994

Weight Up .190 .050 .116–.294 .957 .071 .726–.995

Down .193 .077 .125–.416 .979 .013 .955–.996
150 R. Murray et al. / Clinical Biomechanics 17 (2002) 147–151

strong relationship seen between pelvic rotation and hip
flexion (r ¼ 0:971; range: 0.434–0.998) for all subjects.
During upward movement, the mean correlation was
0.964 (range: 0.434–0.998) and during downward
movement, the mean correlation was 0.979 (range:
0.881–0.998). Fig. 2 demonstrates the continuous and
relatively consistent contribution of the pelvis in a rep-
resentative subject performing hip flexion without the
cuff weight.
Fig. 2. Scatterplot showing the relationship of pelvic to femoral motion
for one subject performing hip flexion (unweighted).
One multifactorial A N O V A (Table 3) showed that
neither weight nor direction had a significant effect on
the PV/FV ratio. The other A N O V A (Table 4) showed
that neither weight nor direction had a significant in-
fluence on the correlation between PV and FV motion.
4. Discussion
The purpose of this study was to describe pelvife-
moral rhythm, which refers to the contribution of pelvic
rotation to hip flexion. Pelvifemoral rhythm is the an-
gular excursion between the pelvis and vertical divided
by the angular excursion between the femur and vertical,
and is expressed as the ratio PV/FV. While other au-
thors have investigated this contribution [4–13], they
have not analyzed it in the standing position. The pre-
sent study examines pelvifemoral rhythm in such an
upright position.
Statistics describing PV/FV in this study show a ratio
of 0.181 (range: 0.081–0.536) and a correlation of 0.971
(range: 0.434–0.998). The closest, previously reported
values come from Bohannon et al. [13]. Statistics de-
scribing active unilateral hip flexion (supine, knee flexed)
showed a ratio of 0.26 (range: 0.21–0.38) and correla-
tions between 0.96 and 1.00.While the correlations are
comparable, upright hip flexion in this study had less of
a contribution from the pelvis, than seen in previous
studies of supine subjects. Contributions of pelvic ro-
tation found in the present study are also less than those
found when investigating the SLR [11,12].

Table 3
Results of 2 " 2 analysis of variance examining the effect of weight and direction on ratio of pelvic to femoral motion during unilateral hip flexion in
standing
Source SS d.o.f. MS F-ratio P
Weight (W) .006 1 .006 3.862 .071
Error .021 13 .002
Direction (D) .000 1 .000 .020 .890
Error .017 13 .001
W "D .000 1 .000 .156 .699
Error .020 13 .002

Table 4
Results of 2 " 2 analysis of variance examining the effect of weight and direction on the correlation between pelvic to femoral motion during
unilateral hip flexion in standing
Source SS d.o.f. MS F-ratio P
Weight (W) .001 1 .001 .526 .481
Error .017 13 .001
Direction (D) .003 1 .003 2.337 .150
Error .017 13 .001
W "D .001 1 .001 .610 .449
Error .018 13 .001
 R. Murray et al. / Clinical Biomechanics 17 (2002) 147–151
Fig. 2 demonstrates that the PV/FV ratio is not cre-
ated by end-range pelvic rotation. The contribution of
pelvic rotation throughout the entire movement docu-
ments a true pelvifemoral rhythm during active motion.
The high correlations found in this study are evident in
this representative subject’s data. We analyzed the flex-
ion and extension phases and the influence of added
external resistance. None of these conditions had a sig-
nificant effect on the ratio or correlation of PV and FV.
This information has direct implications on func-
tional activities. Walking, climbing stairs and the step-
ping strategy to regain balance are a few of the many
activities that involve flexing the hip while in the
standing position. Also, pelvic motion will produce a
change in the lumbar spine unless the trunk flexes or
extends the same amount as the pelvis. The pelvic mo-
tion will also change the position of the stance leg hip.
Allowing the subjects to touch a support kept the trunk
relatively, but not artificially, fixed. This may have im-
proved the subject’s balance, but allows the inference to
lumbar spine motion to be stronger. This datum pro-
vides evidence that limitations in range of motion, at
either the pelvis or the hip, can change the demands
placed upon the musculoskeletal system. Lower back
pain, laminectomy or surgical fixation/fusion of the
lumbosacral region could limit pelvic movement. Indi-
viduals with such problems may have different rhythms.
The small size of the sample is a limitation of this
study. A larger sample would have given a better indi-
cation of normative values. This sample consisted of
young apparently healthy subjects. The data, therefore,
may not generalize to other populations, such as dif-
ferent age groups and those with spine or lower ex-
tremity pathology. Data on subjects with pathology
could shed light on the effects of the pathology on
pelvifemoral rhythm. The kinematic analysis was two-
dimensional, which could produce errors if the out-of-
plane motions of the hip (frontal and transverse planes)
were greater than 10" [14]. However, the nature of the
task and our observations would suggest that these
motions were minimal and would produce negligible
errors in the sagittal plane measurements. The kinetics
of hip flexion were not investigated in this study. Sta-
tistical analysis relating the kinematics of hip flexion
with data from EMG and flexibility and strength testing
might be informative.
5. Summary
The purpose of this study was to quantify pelvife-
moral rhythm, in order to objectify the contribution of
151
pelvic rotation to the change in hip flexion angle in
standing subjects. Despite this movement being very
common and functional, this was the first investigation
to focus on such a movement in an upright position.
Pelvic rotation contributed 18.1% of hip flexion excur-
sion (range: 8–53%) while subjects were standing. Di-
rection of movement, addition of external resistance and
trial number had no significant effect on this percentage.
The contribution of the pelvis to the total motion was
relatively constant throughout the range of motion.
These findings are relevant to many functional situa-
tions.
References
[1] Norkin CC, White DJ. In: Measurement of joint motion, a guide
to goniometry. second ed. New York: F.A. David Company;
1995. p. 124.
[2] Tyler T, Zook L, Brittis D, Gilbert G. A new pelvic tilt detection
device; Roentgenographic validation and application to assess-
ment of hip motion in professional ice hocky players. J Orthop
Sports Phys Ther 1996;24:303–8.
[3] Milch H. The measurement of hip motion in the sagittal and
coronal planes. J Bone Joint Surg 1959;41-A:731–6.
[4] Evjenth O, Hamberg J. In: Muscle stretching in manual therapy, a
clinical manual, vol I, The extremities. Alfta Sweden: Alfta Rehab
Forlag; 1980. p. 94.
[5] Milch H. The pelvifemoral angle: determination of hip-flexion
deformity. J Bone Joint Surg 1942;24:148–53.
[6] Bridger RS, Wilkinson D, Van Houweninge T. Hip joint mobility
and spinal angles in standing and in different sitting postures.
Hum Factors 1989;31:229–41.
[7] Esola MA, McClure PW, Fitzgerald GK, Siegler S. Analysis of
lumbar spine and hip motion during forward bending in subjects
with and without a history of low back pain. Spine 1996;21:
71–8.
[8] Porter JL, Wilkinson A. Lumbar-hip flexion motion: a compar-
ative study between asymptomatic and chronic low back pain in
18–36 year old men. Spine 1997;22:1508–14.
[9] Shirado O, Ito T, Kaneda K, Strax TE. Flexion–relaxation
phenomenon in the back muscles: a comparative study between
healthy subjects and patients with chronic low back pain. Am J
Phys Med Rehabil 1995;74:139–44.
[10] Sihvonen T. Flexion relaxation of the hamstring muscles during
lumbar-pelvic rhythm. Arch Phys Med Rehabil 1997;78:486–90.
[11] Bohannon RW. Cinematographic analysis of the passive straight-
leg-raising test for hamstring muscle length. Phys Ther
1982;62:1269–74.
[12] Bohannon RW, Gajdosik RL, LeVeau BF. Contribution of pelvic
and lower limb motion to increases in the angle of passive straight
leg raising. Phys Ther 1985;65:474–6.
[13] Bohannon RW, Gajdosik RL, LeVeau BF. Relationship of pelvic
and thigh motions during unilateral and bilateral hip flexion.
Phys Ther 1985;65:1501–4.
[14] Areblad M, Nigg BM, Ekstrand J, Olsson KO, Ekstrom H.
Three-dimensional measurement of rearfoot motion during run-
ning. J Biomech 1990;23:933–40.