See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/19114232

Relationship of Pelvic and Thigh Motions During Unilateral and Bilateral

Hip Flexion

Article in Physical Therapy · November 1985

DOI: 10.1093/ptj/65.10.1501 · Source: PubMed

CITATIONS READS

75 1,052

3 authors, including:

Richard Wallace Bohannon

Physical Therapy Consultants
536 PUBLICATIONS 36,850 CITATIONS

SEE PROFILE

All content following this page was uploaded by Richard Wallace Bohannon on 21 May 2014.

The user has requested enhancement of the downloaded file.

Relationship of Pelvic and Thigh Motions During
Unilateral and Bilateral Hip Flexion

RICHARD W. BOHANNON,
RICHARD L. GAJDOSIK,
and BARNEY F. LEVEAU

We filmed the hip flexion movement with a 16-mm motion picture camera to
determine if a synergistic relationship between the pelvis and thigh existed.
Seventeen young subjects, whose pelvises and thighs were marked with tape,
underwent active and passive, unilateral and bilateral hip flexion while in the
supine position. Analysis of the film revealed that the hip flexion movement is
composed of two components—pelvic rotation and flexion of the thigh on the
pelvis. Between one fourth and one third of the hip flexion movement was the
consequence of pelvic rotation. This rotation always occurred within the first 8
degrees of the hip flexion movement. When therapists evaluate and treat patients
with disorders of the thigh, pelvis, or lumbar spine, they should be aware that
these structures normally move in synergy with one another. We suggest that
the pelvifemoral relationship be examined further in studies with a wide range of
healthy subjects and patients with a variety of clinical disorders.
Key Words: Exercise test, Hip, Movement.

As a result of the studies by Freedman
and Munro1 and Doody et al,2 we know
that normal shoulder abduction is the
consequence of a synergistic relation-
ship involving the scapula and humerus.
More specifically, scapular rotation on
the rib cage and movement of the hu-
merus on the scapula are both known
to be prerequisites to the completion of
the shoulder abduction and flexion
movements. Whether a relationship
similar to that at the shoulder exists
between the pelvis and the femur is un-
clear. Although clinicians may use
forced flexion of either one or both of
the hips to rotate the pelvis and flatten
the back of a patient lying in the supine
position (as in the Thomas test),3 the
relationship between hip flexion and
pelvic rotation during the entire hip flex-
ion movement has not been described.
Only the presence of pelvic rotation4-7
and back flattening8 during hip flexion
Mr. Bohannon is Chief of Physical Therapy,
Southeastern Regional Rehabilitation Center, Cape
Fear Valley Medical Center, PO Box 2000, Fayette-
ville, NC 28302 (USA).
Mr. Gajdosik is Associate Professor, Physical
Therapy Program, University of Montana, Mis-
soula, MT. He is currently a graduate student at
the University of North Carolina, Chapel Hill, NC
27514.
Dr. LeVeau is Associate Professor, Division of
Physical Therapy, School of Medicine, University
of North Carolina.
This article was submitted August 29, 1984; was
with the authors for revision 29 weeks; and was
accepted March 21, 1985.
with the kneeflexedand the relationship
between pelvic rotation and hip flexion
with the knee extended during the entire
straight-leg-raising (SLR) movement
have been documented.9
Because a knowledge of the normal
hip flexion movement is important to
understanding any abnormalities of the
movement, we conducted a cinemato-
graphic investigation of the hip flexion
maneuver. The purpose of this study
was to determine and describe the rela-
tionship between the pelvis and the
thigh during unilateral and bilateral, ac-
tive and passive hip flexion movements
in supine subjects. Our expectation was
that both flexion of the thigh on the
pelvis (hip flexion) and pelvic rotation
(posterior tilting) would contribute to
increases in the angle of the thigh in
relation to the horizontal plane (thigh-
horizontal). We anticipated that most of
the pelvic rotation would occur near the
end of the hip flexion movement.
METHOD
Subjects
Thirteen female and four male vol-
unteers with height, weight, and age
means of 168.1 ± 8.6 cm, 60.1 ± 99.6
kg, and 22.2 ± 8 years, respectively,
participated as subjects. Each subject
participated with informed consent in
this study, which was approved by the
Committee on the Protection of the
Rights of Human Subjects at the Uni-
versity of North Carolina at Chapel Hill.
None of the subjects had any known
orthopedic or neurologic dysfunction
that might compromise their well-being
or the study results.
Instrumentation
A motor-driven 16-mm Bolex motion
picture camera* was used to film the
position of the pelvis and the lower
limbs during the performance of hip
flexion. The film was analyzed with a
Vanguard motionanalyzer.†
Procedure
We positioned the camera 4 m (13 ft)
from and perpendicular to the sagittal
plane of the subjects and at the height
of each subject's pelvis. We set the film
speed at 24 frames/sec.
Before beginning hip flexion, we
marked each subject's left pelvis with a
line from the anterior to the posterior
superior iliac spine, as described in a
previous study.10 In that study, we pal-
pated the iliac landmarks as several sub-
jects performed hip flexion and verified
that the landmarks remained covered
*Model HR 16, Bolex International SA, Yver-
don, Switzerland.
† Vanguard Instrument Corp, Walt Whitman
Rd, Melville, Long Island, NY 11746.

Volume 65 / Number 10, October 1985 1501

 of the pelvis-horizontal from the in-
crease in the angle of the thigh-hori-
zontal (thigh-pelvis-thigh-horizontal=
pelvis-horizontal).

Data Analysis

We performed all data analyses by

using the Stats Plus statistical software
package.11 We determined the mean,
standard deviation, and range for the
increases in the angle measurements.
Linear regression lines were fitted and
Pearson product-moment correlations
were calculated for increases in the an-
gles of thigh-horizontal and pelvis-hori-
zontal for each trial for each subject.
The slopes of these regression lines and
correlations represented the relationship
Fig. 1. Subject marked for measurement of pelvifemoral relationship showing the tape marking between increases in the two angles. A
the pelvis, proximal thigh, and distal thigh. Pearson product-moment correlation
was also used to determine test-retest
reliability by comparing the slope during
TABLE 1 the first and second trials under each
Statistics Based on the Slope of the Linear Regression Line of Increases in the Angle of condition.
the Thigh-Horizontal and Pelvis-Horizontal During Hip Flexion

Slope of Linear Correlation of Increases Results

Hip Flexion Trial 1-Trial 2
Movement Reliability
Active unilateral .94
Active bilateral .95
Passive unilateral .97
Passive bilateral .93
by the line. In this study, a mark was
also placed just distal to the left greater
trochanter and another just proximal to
the left lateral femoral condyle. After
the marking was completed, each sub-
ject was positioned supine on an unpad-
ded table with a towel placed under the
lower back and hips. Figure 1 illustrates
the method of marking and subject po-
sitioning.
Each of the hip flexion maneuvers
was practiced before filming. The ma-
neuvers were performed in the following
order: active unilateral hip flexion, ac-
tive bilateral hip flexion, passive unilat-
eral hipflexion,and passive bilateral hip
flexion. Each motion was performed
and filmed twice to determine reliabil-
ity. A metronome established the tim-
ing, and the subjects attempted to com-
plete the full range of each movement
in three seconds. The subjects began the
motions on hearing the camera motor
start. Each subject performed the active
hipflexionsindependently. The passive
hip flexions were performed on each
Regression Line in Angle of Thigh-
Horizontal and Pelvis-
s Range Horizontal
.26 .05 .21-.38 .96-1.00
.30 .06 .21-.44 .89-1.00
.27 .04 .21-.33 .97-1.00
.30 .04 .26-.37 .98-1.00
subject by one of us (R.L.G.), who
flexed each subject's hip until he felt
firm resistance to further flexion. Dur-
ing right hip flexion, the left thigh was
not stabilized.
Measurements of the angle of the
thigh-horizontal and of the angle of the
pelvis with the horizontal plane (pelvis-
horizontal) were obtained from the film.
The first measurements were recorded
before limb motion began. Every third
frame of film was analyzed until the
thigh reached its maximum displace-
ment with the horizontal plane. As a
result, 19 to 35 frames of film were
analyzed for each trial. All measure-
ments were converted to values repre-
sentative of a change in position (ie, a
subject's initial angle of the pelvis-hori-
zontal at 75.6 degrees was considered 0
degrees, and thefinalangle of the pelvis-
horizontal at 103.3 degrees was consid-
ered an increase of 27.7 degrees). We
determined the increase in the angle of
the thigh with the pelvis (thigh-pelvis)
by subtracting the increase in the angle
As can be seen in Table 1, the test-
retest reliabilities were high (range, .93-
.97). Therefore, all other results were
based solely on the data from trial 1.
Table 2 shows the range of hip flexion
in relation to the horizontal plane (in-
crease in angle of thigh-horizontal) un-
der various conditions and the contri-
bution of pelvic rotation (increase in
angle of pelvis-horizontal) to this flex-
ion. Figure 2 illustrates that pelvic ro-
tation contributed substantially to the
hipflexionmovement early in the range.
For all conditions, every subject dem-
onstrated pelvic rotation before reaching
an 8-degree increase in the angle of the
thigh-horizontal. Pelvic rotation usually
began by the time a 4-degree increase
was reached in the angle of the thigh-
horizontal. The data in Table 1 elabo-
rate on the relationship between thigh
and pelvic movement. The correlations
between increases in the angle of the
pelvis-horizontal and thigh-horizontal
ranged from .89 to 1.00 depending on
the subject and the hip flexion move-
ment measured. These high correlations
and the linear appearance of the regres-
sion lines for the increase in the angle
of the pelvis with the increase in the
angle of the thigh suggest that the rela-
tionship between thigh and pelvic move-
ments is linear. The mean slopes of the
linear regression lines for subjects under

1502 PHYSICAL THERAPY

 RESEARCH

TABLE 2
Maximum Angular Increasea in the angle of the Pelvis and Thigh in Relation to the Horizontal and in the Angle of the Thigh in Relation
to the Pelvis During Hip Flexion

Increase in Angle of Pelvis- Increase in Angle of Thigh- Increase in Angle of Thigh-

Hip Flexion Horizontal Horizontal Pelvis

s Range s Range s Range

Active unilateral 34.2 5.8 25.4-46.2 124.3 6.7 112.7-135.4 90.1 7.2 78.2-103.3
Active bilateral 43.7 7.6 33.1-60.1 138.4 7.6 126.2-152.0 94.7 7.6 80.8-105.9
Passive unilateral 35.4 4.5 30.7-45.6 125.9 5.8 116.2-140.5 90.5 7.2 82.0-109.2
Passive bilateral 37.8 4.4 31.3-44.7 123.7 6.6 113.8-139.6 85.9 7.6 72.5-102.7
a
Measured in degrees.

the various conditions ranged from .26

to .30. Therefore, on the average for the
whole range of motion, 1 degree of pel-
vic rotation in relation to the horizontal
plane accompanied 3.8,3.3,3.7, and 3.3
degrees of thigh elevation in relation to
the horizontal plane under the active
unilateral, active bilateral, passive uni-
lateral, and passive bilateral conditions,
respectively. Stated otherwise, 3.8, 3.3,
3.7, and 3.3 degrees of thigh elevation
in relation to the horizontal plane were
the consequence of 1 degree of pelvic
rotation and 2.8, 2.3, 2.7, and 2.3 de-
grees of thigh elevation in relation to the
pelvis during active unilateral, active bi-
lateral, passive unilateral, and passive
bilateral hip flexion, respectively.

DISCUSSION AND CLINICAL

IMPLICATIONS
The ability of patients to complete the
full range of hip flexion is dependent on
both pelvic and thigh movement. When
clinicians measure range of motion and
strength and evaluate patients' abilities
to perform activities that involve hip
flexion, they should consider this pelvi-
femoral relationship.
Just as a relationship for scapulohu-
meral motion exists at the shoulder, so
a relationship for pelvifemoral motion
exists at the hip. This finding was not in
itself too surprising to us. The high cor-
relation between pelvic rotation and hip
flexion suggested, however, that the two
movements were more highly related to
one another than we would have ex-
pected. To find that pelvic rotation was
responsible for about one fourth to one
third of the total hip flexion movement
and that pelvic rotation began very early
in the hip flexion movement was sur-
prising. The finding was particularly un-
expected because Ellis and Stowe re-
ported hip flexion of 70 to 130 degrees
before "pelvic tilting was observed,"4
Fig. 2. Relationship between pelvic rotation and hip flexion measured against a horizontal
reference in one subject.
and Fisk reported that "rotation of the
pelvis does not take place until the thigh
is at least at a right angle to the trunk."5
Evidently, simple observation is inade-
quate to verify the presence of pelvic
rotation during the early part of hip
flexion. As pelvic rotation and flexion
of the thigh on the pelvis both appear to
contribute to the hip flexion movement
throughout the range, we think that
measuring hip flexion to a point where
pelvic rotation begins is of doubtful
value. A better perspective may be to
accept that both movements are impor-
tant components of hip flexion and that
any abnormality affecting one compo-
nent may affect the total hip flexion
movement. This approach would be
consistent with the recommendation of
Milch that the Thomas test be discarded
as invalid and measures of pelvifemoral
motion be made in reference to the pel-
vis.7 Among the abnormalities that
might affect the pelvifemoral relation-
ship are those resulting in hypermobility
or hypomobility of the lumbosacral
spine, sacroiliac joint, or hip. Musculo-
tendinous, ligamentous, joint capsule,
or intrajoint problems may result in
such changes in mobility. We suggest
additional study to examine these pos-
sibilities.
The results of this study, though
clearly demonstrating the presence of a
pelvifemoral relationship, provide only
a limited description of this phenome-
non. A large number of subjects would
need to be tested to obtain normative

Volume 65 / Number 10, October 1985 1503

values. Moreover, the relationship dem-
onstrated by the relatively young sub-
jects in this study may not be represent-
ative of the relationship of individuals
of other age groups. The intimation by
Ellis and Stowe that pelvic rotation be-
gins earlier in older individuals implies
such a possibility.4
The relative contribution of pelvic ro-
tation to the hip flexion movement ap-
pears to be essentially the same during
passive and active movements (slopes
differ less than 4%) but greater during
bilateral than unilateral movements
(slopes differ by as much as 15%). Sta-
tistical comparisons would be necessary,
however, to establish the difference in
the relationship between pelvic rotation
and hip flexion under the various con-
ditions. Such a comparison is beyond
the scope of this descriptive study. Be-
cause of the similarity of active and
passive movement slopes, active muscle
tension is not implicated as important
to the pelvifemoral relationship in
young, healthy subjects.
The mean slope of the linear regres-
sion lines of increases in the angle of the
thigh-horizontal and pelvis-horizontal
1. Freedman L, Munro R: Abduction of the arm in
the scapular plane: Scapular and glenohumeral
movements. J Bone Joint Surg [Am] 48:1503-
1510,1966
2. Doody SG, Freedman L, Waterland JL: Shoul-
der movements during abduction in the scap-
ular plane. Arch Phys Med Rehabil 51:595-
604, 1970
3. Hoppenfeld S: Physical Examination of the
Spine and Extremities. New York, NY, Apple-
ton-Century-Crofts, 1976, pp 155-156
4. Ellis Ml, Stowe J: The hip. Clin Rheum Dis
8:655-675, 1982
5. Fisk JW: The passive hamstring stretch test:
Clinical evaluation. NZ Med J 89:209-211,
1979
1504
View publication stats
was less for all hip flexion movements
with the knees flexed (.26-.30) than the
mean slope of the linear regression lines
of increases in the angle of SLR-hori-
zontal and pelvis-horizontal during pas-
sive SLR (.39) for the same subjects.9
This difference is representative of more
pelvic rotation accompanying SLR than
hip flexion with the knee flexed. Such a
difference might be expected as the
hamstring muscles probably exert a
greater rotary influence on the pelvis
during passive SLR than during hip flex-
ion with knee flexion.
Gait studies clearly indicate that pel-
vic rotation accompanies hip flexion
and extension in the sagittal plane.1213
Because gait is such a complex recipro-
cal activity, however, the relationship
between hip flexion or extension and
pelvic rotation is difficult to establish.
The pelvifemoral relationship may,
nonetheless, have considerable impor-
tance to the performance of such activ-
ities. The relationship may also be of
substantial importance in athletic per-
formance (eg, kicking) and the adaptive
performance of various tasks by disabled
persons (eg, gait by paraplegics). The
REFERENCES
6. Milch H: The pelvifemoral angle: Determination
of hip-flexion deformity. J Bone Joint Surg
24:148-150, 1942
7. Milch H: The pelvifemoral angle of flexion and
extension at the hip joint. Clin Orthop 31:58-
64,1959
8. Moore ML: Clinical assessment of joint motion.
In Basmajian JV (ed): Therapeutic Exercise, ed
3. Baltimore, MD, Williams & Wilkins, 1978, p
171
9. Bohannon RW, Gajdosik R, LeVeau BF: Con-
tribution of pelvic and lower limb motion to
increases in the angle of passive straight leg
raising. Phys Ther 65:474-476, 1985
speed of hip flexion may affect this re-
lationship as well.
CONCLUSIONS
When examined by cinematography,
the displacement of skin marks over
bony landmarks demonstrated that the
hip flexion movement is composed of
two movements—flexion of the thigh
on the pelvis and pelvic rotation. In the
young, healthy subjects tested in the su-
pine position, more than one fourth of
the hip flexion movement was attrib-
uted to pelvic rotation. This rotation
always started before 8 degrees of hip
flexion was achieved, regardless of
whether flexion was active, passive, uni-
lateral, or bilateral. We suggest further
studies to examine the pelvifemoral re-
lationship of a wider range of healthy
subjects and of patients with problems
affecting the thigh, pelvis, or lumbar
spine.
Acknowledgment. We greatly appre-
ciate the assistance of Marian Geddie in
the preparation of the manuscript.
10. Bohannon RW: Cinematographic analysis of
the passive straight-leg-raising test for ham-
string muscle length. Phys Ther 62:1269-
1274,1982
11. Madigan S, Lawrence V: Stats Plus: A General
Statistics Package for the Apple II. Northridge,
CA, Human Systems Dynamics, 1982
12. Murray MP, Drought AB, Kory RC: Walking
patterns of normal men. J Bone Joint Surg
[Am] 46:335-360, 1964
13. Murray MP, Kory RC, Sepic SB: Walking pat-
terns of normal women. Arch Phys Med Re-
habil 51:637-650, 1970
PHYSICAL THERAPY