Validity of Derived Measurements of Leg-Length

Differences Obtained by Use of a Tape Measure

Paul Beattie, Kale Isaacson, Dan L Riddle and Jules M
Rothstein
PHYS THER. 1990; 70:150-157.

The online version of this article, along with updated information and services, can be
found online at: http://ptjournal.apta.org/content/70/3/150

Collections This article, along with others on similar topics, appears

in the following collection(s):
Injuries and Conditions: Lower Extremity
Tests and Measurements
e-Letters To submit an e-Letter on this article, click here or click on
"Submit a response" in the right-hand menu under
"Responses" in the online version of this article.
E-mail alerts Sign up here to receive free e-mail alerts

Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014

Validity of Derived Measurements of Leg-Length
Differences Obtained by Use of a Tape Measure

Determining the dzference in the length of an individual's legs is often an impor- Paul Beattie
tant component of a musculoskeIetal emmination. Although measurements are Kale lsaacson
easily obtuined with a tape measure, the valdity of these measurements is not Dan L Riddle
known. 7 3 e p u q a e of this study was to emmine the validity of determinations of Jules M Rothstein
leg-lengthdzferences (LLDsj obtained by use of a specified tape measure method
(TMM).Leg-length dz$eences using the TMM and a radiographic technique were
determimd for 10 subjects who were candidatesfor clinical leg-length measure-
ments and for 9 healthy control subjects. Validi41of the TMM measurements was
determined by messing the degree of agreement between TMM-obtained LLDs and
those obtained by the radiographic method. Validity estimates as determined by
inh-acluss cowelation coeficients (ICG) were ,770for patients, ,359for healthy
subjects, and ,683for all subjects. W e n the means of the two values obtained by
use of the TMM were compared with the radiographic measurements, the ICCs
w e .85;?for the patient group, ,637for the healthy subjecrs, and ,793for all sub-
jects. Thk study suggests that TMM-derivedLLD measurements are tlalid indicators
of leg-length inequality and that the estimates of validity are improtled by using
the average of two determinations rather than a single determination.[Beattk P,
Isaaaon f( Riddle DL, et al: Validity of derived measurements of leg-length d z m -
ences obtained by use of a tape measure. Phys Ther 70:150-157, 19901

Key Words: Lower extremity, general; Musculoskeletal system; Radiography; Tests

and measurements,functional.

Leg-length differences (LLDs) are of LLD that is clinically significant ions regarding clinically significant
thought to contribute to the occur- remains controversial. Subotnick has LLDs are shown in Table 1. The opin-
rence or severity of many clinical reported that a dfierence of as little ion of what constitutes a "significant
syndromes.l.2Among these conditions as 3 mm is significant,"J whereas LLD" appears to be diagnosis specific.
are sc0liosis,3~4low back pain,+' sa- Anderson has stated that a difference For example, Subotnick considers a
croiliac painF.9 and a variety of run- of less than 19 mm is acceptable.12 difference of 3 mm significant enough
ning injuries.lOJ1 However, the degree Summaries of various authors' opin- to warrant a shoe lift for the treatment
of running-related injuries.10 Giles
suggests that an LLD of greater than
P Beattie, kIS, PT, is Instructor, Division of Physical Therapy, Department of Orthopedics, School of
9 mm could cause enough of a
Medicme, University of New Mexico, Albuquerque, NM 87131 (USA). change in the angle of the lumbar
facet joints to contribute to the devel-
K Isaacson, PT, is Staff Physical Therapist, Sports Physical Therapy and Rehabilitation, 2607 Wyo-
ming, Albuquerque, NM 87112. At the time this study was conducted, he was a student in the Divi- opment of back pain.13 Papaioannou
sion of Physical Therapy, University of New Mexico. et a1 report that an LLD of greater
than 22 mm causes significant com-
D Riddle, hlS, PT, is Assistant Professor, Department of Physical Therapy, School of Allied Health
Professions, Medical College of Virginia, Virginia Commonwealth University, PO Box 224, MCV Sta- pensatory scoliosis.3
tion, Richmond, VA 23298

J Rothstein, PhD, PT, is Associate Professor, Department of Physical Therapy, School of Allied
The association of LLD with many
Health Prof'essions, Medical College of Virginia, Virginia Commonwealth University. clinical syndromes has made determi-
nation of LLDs an important pan of
Thir article was s~rbtnittedApril 12, 1989; was with the authon fot. revisionfor 11 week; and was
accepted N~vember3, 1989. musculoskeletal examinations. There

Physical 'TherapyNolume 70, Number 3lMarch 1990 1501 13

Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014


Table 1 . Leg-Length~ ~ e r e n c(LLDs)
e s Considered Clinically SigniJicant
LLD
(mm) Comment
According to Subotnick,lo LLD of 3 mm or greater can cause
injury to runners.
According to Friberg,s LLD of 5 mm or greater leads to biome-
chanical compensations in the spine.
According to Brody,ll LLD of 6 mm or greater can cause inju-
ries to runners.
According to Corrigan and Maitland,' LLD of less than 7 mm
rarely causes symptoms.
According to Giles,l3 LLD of 9 mm or greater causes changes
in the angle of lumbar facets.
According to Cyriax,s LLD of 10 mm or greater contributes to
the development of back pain.
According to Gibson et al,4 LLD of 15 mm or greater can
cause compensatory scoliosis in the standing person.
According to Vogel,l LLD of greater than 20 mm requires
lower extremity compensation.
According to Papaioannou et a1,3 LLD of greater than 22 rnm
causes significant scoliosis.
According to Ingram14 LLD of greater than 40 mm often
requires surgical correction.
is, however, no universally accepted sample of five subjects.20 They com-
clinical method for measuring LLD.I4 pared radiographic and block-method
Determining leg lengths by taking measurements using an F test and a t
measurements from radiographs and test to determine whether the mean
then calculating the difference is gen- measurements obtained with both
erally considered to be the most accu- methods differed significantly. Tests of
rate method for determining leg- dfierences such as the t test or F test
length inequality.l5-l7Because of their do not indicate the degree of agree-
cost, however, radiographs are ment between repeated measure-
impractical for determining UD, and ments. Therefore, conclusions regard-
radiography exposes the subject to ing the degree of agreement of the
the adverse effects of radiation.17 measurements taken in the Woerman
Therefore, other methods are more and Binder-Macleod study cannot be
often used clinically. made.
Therapists often use simultaneous Friberg et a1 used the block method
palpation of both iliac crests of a to determine the presence and
standing subject to determine LLD. degree of LLD in 21 patients with low
The relative heights in the frontal back pain." The authors compared
plane of each crest are then measurements obtained with the
0bserved.8~~~~~9Although this method block method with those obtained
is easy to perform, intratester and from radiographs. They used only
intenester agreement has been shown descriptive statistics to repon the
to be lacking.15.18A modified version degree of agreement between radio-
of this method requires placing nar- graphic and block-method measure-
row blocks (block method) under the ments. The average intratester error
observed shoner leg until the iliac when comparing radiographic with
crest heights are le~el.1~10~~9~2~
Woer- block-method measurements was 5.8
man and Binder-Macleod studied the mm. The authors also reported that,
usefulness of the block method on a approximately 13% of the time, the
14/151 Physical TI~erapyNolume70, Number 3/March 1990
Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014
block-method measurements differed
from the radiographic measurements
in the determination of which leg was
shoner.
Other techniques have been
described for the clinical measure-
ment of LLD. Among these techniques
is the use of a measurement screen.21
The use of a level with moveable
arms to approximate the level of the
iliac crests" or the level of the ante-
rior superior iliac spines (ASISs)'3 has
also been described. There is, how-
ever, no evidence supporting the
validity and reliability of U D measure-
ments obtained by use of these meth-
ods. An additional drawback to using
these methods is that they are imprac-
tical because they require instruments
that are not usually readily available
to physical therapists.
A frequently described method for
assessing LLD requires the use of a
tape measure (TMM) to determine the
distances from the ASISs to the
medial malleoli.l5-l7.19.20.'P'7 Subjects
are usually measured while they are
positioned supine.
Eichler described several potential
sources of error when measure-
ments are obtained using the
TMM.'6 Differences in the circum-
ferences of the two legs could con-
tribute to distance differences, as
could unilateral deviations along the
long axis of the leg (eg, genu val-
gum, genu varum). In addition,
Eichler suggested that pelvic asym-
metries and difficulty identifying
bony prominences by palpation
could contribute error to these
measurements.
Beattie and colleagues conducted a
preliminary study using an operation-
ally defined TMM identical to that
used in this study to examine the reli-
ability of TMM meas~rements.~7 They
were attempting to eliminate error
attributable to the sources identified
by Eichler. Two examiners obtained
repeated measurements of UD on 50
subjects, 38 of whom were patients
who were considered clinical candi-
dates for leg-length measurement (eg,
they complained of low back or lower
extremity dysfunction). The authors
observed good intrarater reliability
(ICC[1,1] = .807) and fair interrater
reliability (ICC[l,I.] = ,668) when
comparing the first measurements
obtained by each examiner. When
they compared the mean values of
-
Table 2. Characteristics of Patient Group (n = 10)

Subject
Number
Age
(Yr) Sex
Height
(cm)
Welght
(kg) Dlagnosls

paired measurements, the interrater 1 34 F 168 61 tibia1 plateau fracture

reliability rose considerably (ICC =
2 60 F 170 68 low back pain, osteoarthritis
.910). These results agree with those
3 36 F 168 77 slipped epiphysis of femoral
of other researchers who found that
head
reliability of TMM-obtained measure-
4 28 M 203 100 tibia1 dysplasia
ments improved when the means of
reoeated measurements were 5 50 M 183 86 femoral fracture

-
6 26 M 180 95 distal tibia-fibula fracture
7 27 M 175 70 low back pain
Encouraged by the results of the pre- 8 28 M 193 86 low back pain, sacralization
liminary study, we conducted this 9 25 M 178 79 femoral fracture
study to determine the validity of LLD 27 F 178 69 low back pain
measurements obtained with the TMM
as compared with LLD measurements
obtained radiographically. In addition,
we wanted to determine whether the
mean values of paired measurements Table 3. Characteristics of ~ o m t a Group
l (n = 9)
obtained by the TMM are a more
valid indicator of LLD than are single
Subject Age Helght Weight
measurements. Number (Yr) Sex (cm) (kg)

Method

Subjects

Ten subjects, ranging in age from 25

to 60 years (X = 34.1, s = 11.2), par-
ticipated in this study as the "Patient"
Group. Each of these subjects had a
history of LLD or a recent history of
lower extremity, pelvic girdle, or spi-
nal dysfunction that required medical
care. These subjects, therefore, were
considered candidates for the clinical
assessment of LLD. A description of
the Patient Group appears in Table 2.
A second group consisting of nine
healthy subjects, ranging in age from
22 to 34 years (X = 26.5, s = 3.7),
also panicipated in this study, as the
"Normal" Group. None of these sub-
jects had a history of known LLD or
lower extremity, pelvic girdle, or spi-
nal dyshriction that required medical
care. A description of the Normal
Group appears in Table 3. These sub-
jects were included because asymptom-
atic individuals are often evaluated for
LLD during preemployment or preath-
letic screening examinations. The
validity of measurements obtained
with use of TMM on healthy subjects,
therefore, would be of value.
If a subject in either group was
known to be pregnant or was late
menstruating, she was excluded from
the study because of the potential risk
of exposure to radiation. All subjects
were instructed in the risks and bene-
fits of participating in this study and
signed a written consent statement
approved by the Human Research
Review Committee of the School of
Medicine, University of New Mexico.
Procedures
Radiographic measurements. Our
radiographic leg-length measurement
technique was based on use of the
mini-scanogram.*8Subjects were
positioned supine on standard
radiographic tables with a large radio-
opaque ruler placed under their right
lower extremity. An x-ray tube was
centered perpendicularly over the
subject's right hip, and the first x-ray
film was exposed. The tube was then
centered over the subject's knee and
finally over the subject's ankle while
the second and third x-ray films were
exposed (Fig. 1). The ruler was then
moved under the subject's left lower

Physical TherapyNolume 70, Number 3/March 1990 152/15

Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014


Exposure 1
track - I
(hip)
,
,/e
?
lead-marked ruler /
Fig. 1. Illustration of three radiographic exposures wed in mini-scanogram.X-ray
tube i s on moveable track that allous exposures to be obtained with tube positioned
above hip, knee, and ankle of supine subject.
extremity, and the procedure was
repeated for that extremity. A
licensed, boardcertified radiologist
supervised this portion of the study
and examined all radiographs for
pathology.
The mini-scanograms when placed
together show the hip, knee, and
ankle with the ruler clearly visible,
which allowed for measurements of
leg length (Fig. 2). Leg length (U)
was calculated by subtracting the
value visible on the ruler at the supe-
rior margin of the head of the femur
from the value seen at the midportion
of the joint space between the distal
tibia and the superior margin of the
talus. (For example, the marking at
the tibiotalar joint space was 101 cm
and the marking at the head of femur
was 16 cm; therefore, U = 101 - 16
= 85 cm.)
The preliminaq study indicated we
could reliably measure LLD with a
tape measure. However, because we
did not know whether we could
Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014
Exposure 2 Exposure 3
(knee)
8)- 7
'--
\
(ankle)
'y-
\1
'L7 'e
-a'
obtain reliable measurements from
the mini-scanograms, we examined
reliability of these measurements as
part of our method. Unless these
measurements were reliable, it would
not be reasonable to use the mini-
scanogram measurements as criteria
in a validity study.
Interrater reliability for measurements
from the mini-scanograms was deter-
mined from the measurements taken
by two of the authors (PB and KI).
They independently measured the leg
lengths and calculated the LLDs. Intra-
class correlation coefficients (ICC[l,l])
were used to estimate agreement
between the LLD meas~rements.~9
The ICC for the 19 paired measure-
ments taken by the two observers
from all subjects was ,993. The ICCs
revealed that the measurements of
UD obtained from the mini-
scanograrns were highly reliable
between the two examiners. The mea-
surements obtained from mini-
scanograms are commonly accepted
as valid indicators of lower extremity
Physical ThetrapyNolume 70, Number 3/March 1990
bone length.20z2eTherefore, we
believe that the measurements
obtained from the mini-scanograms
were appropriate to use as criteria to
examine the validity of measurements
obtained with the TMM.
Measurements with the tape mea-
sure. All TMM measurements were
obtained by one person (PB). During
each measurement session, subjects
wore a pair of shorts or a hospital
gown. The subjects' lower extremities
were exposed from the level of the
midthigh to their feet. Subjects were
positioned supine on a plinth. The
examiner positioned the subject's
lower extremities in neutral hip rota-
tion as determined by observation.
The examiner then placed the sub-
ject's medial malleoli together so that
they met in a plane that approximated
the midsagittal line of the body. The
subject's hip and knees were, there-
fore, in a position that closely approx-
imated the anatomical position.
The examiner stood on the same side
of the plinth as the limb he was mea-
suring. The examiner held a blank
tape measure between the thumb and
the first finger of his hand nearest the
subject's pelvis. With the same hand,
he used his thumb to palpate the sub-
ject's ASIS. One end of the tape mea-
sure was placed on the ASIS at the
site where the examiner believed he
could palpate the origin of the sano-
rius muscle on the inferior portion of
the ASIS. With the hand opposite to
that holding the tape measure on the
ASIS, the examiner gradually guided
the tape down the anteromedial
aspect of the subject's thigh, patella,
and lower leg until he made contact
with the point where the subject's
medial malleolus sloped inferiorly
and laterally (Fig. 3). The examiner
then held the tape taut and lifted it
away from the subject. Another per-
son recorded the value from the
opposite surface of the tape.
The examiner then repeated the same
procedure on the subject's opposite
lower extremity. Following this proce-
dure, the subject was asked to stand
and to move about for approximately
one minute in whatever manner was

Fig. 2. Mini-scanogram showing radio -opaque markings on ruler placed under
subject's lower limb.
comfortable and then to return to the
supine position on the plinth. The
same examiner then repeated the
entire procedure to obtain a second
pair of mt:asurements.
Data Analysis
The LLD was calculated by subtracting
the right leg-length measurement
from the left leg-length measurement.
Physical TherapyNolume 70, Number 3/March 1990
Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014
A positive value, therefore, indicated
that the left leg wa5 longer than the
right leg. A negative value indicated
that the right leg was longer than the
left leg.
We used the ICC (formula 1 , l ) to
examine the degree of agreement for
measurements taken in this study.29
This form of the ICC is typically used
when random pairs of testers take
repeated measurements. We did not
use random pairs of testers to take
measurements in this study. However,
this ICC formula does not exclude
error attributable to the testers,
whereas other forms of the ICC do
factor out error attributable to different
examiners. Based on our clinical expe-
rience, dltferent clinicians may mea-
sure LLD in the same patient; there-
fore, differences between examiners
may be a potential source of error. By
using this formula (1,l) for calculation
of the ICC, we did not exclude error
attributable to the testers. We believe
this version of the statistic provides the
most clinically meaningful estimate of
the amount of error associated with
LLD measurements.
The ICC was used to estimate agree-
ment between LLD measurements
obtained with the mini-scanograms
and the TMM. Although the absolute
values of the leg lengths obtained
using the two techniques would be
dserent, the calculated (derived)
LLDs would be in agreement if the
TMM measurements were valid for
determining the LLD.
The criterion-referenced validity of
measurements obtained with the TMM
was calculated in two ways. The first
measurements of LLD using the TMM
were compared with the measure-
ments of LLD calculated from the
mini-scanograms. The means of the
paired measurements of LLD obtained
by using the TMM were also com-
pared with the measurements of LLD
obtained from the mini-scanograms.
Separate ICCs were calculated for the
Patient Group, the Normal Group,
and the pooled data from both
groups.
Results
The values calculated for LLD from
measurements obtained by the TMM
and by the mini-scanogram method
are presented in Table 4. The ICC
values obtained by comparing the first
measurements of LLD obtained by the
TMM with the measurements of LLD
obtained using the mini-scanogram
method were ,770 for the Patient
Group, .359 for the Normal Group,
and .683 for the entire sample
(Tab. 5). The ICC values obtained by
comparing the mean values of two
measurements of LLD obtained by the medial malleolus
TMM with the measurements of LLD
obtained using the mini-scanogram I
were ,852 for the Patient Group, ,637 blank tape measure
for the Normal Group, and ,793 for
all subjects (Tab. 5).

Discussion

-
Validity of Tape Measure Method
Values of Leg-Length Difference

In all cases, the second measurements Fig. 3. Illmtration of tape measure method.for obtaining leg-length measurements
obtained using the TMM demon- of supine subject. Both legs are placed as closelv as possihle to the anatomical position.
strated considerably greater agree- Leg-length difference is calculated ly subtracting length of left 1eg.from length of right
ment with the values from the mini- leg. (ASIS = anterior superior iliac spine.)
scanogram than did the first TMM
measurements. For example, the ICC
for the first TMM measurements of Table 4. ~eg-LengthIjifferences
the Normal Group subjects compared
with the mini-scanogram measure-
ments was ,359, whereas the ICC for Calculated from
the second TMM measurements com- Calculated by Use of TMMa (cm) Mlnl-scanograms (cm)
pared with the mini-scanogram mea- Subject First Second
surements was ,786 (Tab. 5). The rea- Numberb Measurement Measurement W s
son for the difference in validity
coefficients between the first and sec-
ond measurements is unclear. The
best validity estimate (ICC = ,852) for
the Patient Group, however, was
obtained by use of the mean of the
TMM measurements.

The finding that the mean of two

measurements of LLD obtained with
the TMM was the most valid measure
should have been anticipated because
previously we demonstrated that reli-
ability of these measurements was
enhanced by use of mean values.2'
Validity is dependent o n reliability;
therefore, clinicians are advised to use
the mean of two TMM-obtained LLD
measurements.

The TMM is an indirect method for

measuring leg length and then deter-
mining LLD. The starting point of
measurement from the ASIS allows "TMM = tape measure nletllod.
iilclusion of a portion of the bony
"~ubjects1 through 10 were the Patient Group, subjects 11 through 19 were the Normal Group.
pelvis Factors such as bony asymme-
try of the pelvis o r pelvic obliquity,
which may not actually cause an LLD, tries in the surface contours of the cantly alter the position of the tape,
could influence the measurements thigh, knee, and lower leg (eg, asym- leading to meawrements that would
obtained by using the TMM. Asymme- metries caused by swelling, muscular not reflect leg length.
atrophy, or obesity) could also signifi-

IJhysical TherapyNolume 70, Number Y ~ a r c h1990

Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014

-
Table 5. Intraclass Correlatiom? for Comparisons of Leg-Length D e e n c e
Alemurements Obtained by Use of Tape Measure Method (TMM) and by Use of
Mini-scnnogrnnls
Flrst TMM
Group Measurement
Patient (n := 10) ,770
Normal (n = 9) ,359
All subjects (N = 19) ,683
"Intraclass correlation coeficients were calculated using equation 1 , l of Shrout and FIeiss.j9
'The results of the preliminary study
and of this study indicate that LLD
measurernents obtained by use of a
TMM are relatively valid when the
means of paired measurements are
used. These measurements, however,
like all measurements, have a degree
of error associated with them. In
using me:aurements in the clinical
setting, this error must be considered.
The error associated with the mea-
surements can only be meaningful
when clinicians consider the magni-
tude of LI,D that they believe warrants
treatment.
Clinicians should know whether they
are able to correctly determine which
leg is shorter in patients requiring
assessment of LLD. Examination of
our data suggests that the error asso-
ciated with LLD measurements may
be highly consequential when small
LLDs are noted. When TMM-derived
measurements of LLD were 5 mm or
less, the examiner erred in the deter-
mination of which leg was the shorter
on four out of nine subjects. This
error was determined by use of the
data from the radiographic measure-
ments. When TMM-derived measure-
ments of LLD were greater than 5
mm, the examiner never made an
error when determining which leg
was shorter. Based on these data,
therapists should be cautious when
making cllinical decisions based on
TMM-obtained LLD measurements of
5 mm or less.
The TMM provides data relative to
LLD in the supine subject. These data
do not define the degree of functional
Physical TherapyNolume 70, Number 3/March 1990
Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014
Second TMM Mean of TMM
Measurement Measurements
,803 ,852
,786 ,637
,790 ,793
impairment created by the LLD. Our
validity study showed relatively strong
agreement with a criterion measure
(measurements obtained by use of
the mini-scanograms); we did not vali-
date any other inferential use of the
TMM measurements. Our results indi-
cate that measurements of LLD
obtained using the TMM appear to
represent the same anatomical rela-
tionships that can be documented by
use of the mini-scanograms. Because
we examined LLDs with both the
TMM method and our radiographic
method in supine subjects, we cannot
be sure that our measurements reflect
functional LLD measurements. For
example, this method does not assess
structural or biomechanical asymme-
tries of the foot and ankle that could
create an LLD during such activities as
standing, walking, and running.
We suggest that in addition to using
TMM measurements, clinicians evalu-
ate the effect of LLD on patients by
analyzing specific functional activities
(eg, walking, running, stair climbing).
Therefore, although the TMM may be
considered to provide valid measure-
ments of LLD on patients, our data
represent only a portion of the infor-
mation necessary to make appropriate
clinical decisions.
The generalizability of our conclu-
sions has some limitations. The sam-
ple size was small. Perhaps most
importantly, the examiner (PB) has
used the TMM technique on many
occasions during the preparation for
this study and in previous studies. The
skill of the examiner in our study may
not be representative of many clini-
cians who may use this technique
infrequently. We believe, however,
based on the preliminary study, that
clinicians can derive fairly reliable
LLD measurements by using the TMM
defined in this article. Future research
can help clinicians by examining the
reliability and validity of LiD measure-
ments obtained on a larger sample
with a larger number of therapists
obtaining the measurements. In addi-
tion, studies examining inferential
uses of TMM-derived measurements
would be helpful.
The measurements obtained with the
TMM appear to be valid for assessing
LLDs in patients when the mean of
two measurements is used. Measure-
ments are less valid when healthy
subjects are measured. Given the indi-
rect nature of this technique and the
unresolved issue of what constitutes a
clinically significant LLD, we believe
that clinicians should not depend
solely on TMM measurements for
clinical decisions.
Acknowledgments
We thank James R Stevenson, MD, and
Gloria Gilreath, KT, for their assis-
tance with the radiographic part of
the study and Barbara Arnstadt for her
assistance in preparing this manu-
script. We would also like to express
our appreciation to the late Fred
Rutan, MS, PT, for his assistance with
this study.
References
1 Vogel F: Shon leg syndrome. Clin Podiatr
1:581-599, 1984
2 Williams M , Worthingham C: Therapeutic
Exercise for Body Function and Alignnient.
Philadelphia, PA, W B Saunders Co, 1957, pp
31-32
3 Papaioannou T, Stokes 1, Kenwright J: Scoli-
osis a~sociatedwith limb-length inequality.
J Bone Joint S u g [Am] 64:5'+62, 1982
4 Gibson PH, Papaioannou T, Kenwright J:
The influence on the spine of leg-length dis-
crepancy after femoral fracture. J Bone Joint
Surg [Br] 65:584587, 1983
5 F r i b e g 0 : Cllnical symptoms and biome-
chanics of the lumbar spine and hip joint in
leg-length inequality. Spine 8:645651, 1983
6 Cyriax J: Textbook of Orthopaedic Medicine:
Diagnosis of Soh Tissue Lesions, ed 7. London,
England, Bailliere Tindall, 1981, vol 1
7 Corrigan B, Maitland GD: Practical Ortho-
paedic Medicine. Boston, MA, Butterworth
Publishers, 1983
8 Stoddard A: Conditions of the sacroiliac
joint and their treatment. Physiotherapy 44:97-
101, 1958
9 Bourdillon JF, Day EA: Spinal Manipulation,
ed 4. East Norwalk, CT, Appleton & Lange,
1987
10 Subotnick SI: Leg-length discrepancies of
the lower extremity. Journal of Orthopaedic
Sports and Physical Therapy 311-16, 1981
11 Brody DM: Running injuries in the lower
extremity and spine in sports medicine. In
Nicholas JA, Herchman EB (eds): Lower
Extremity and Spine in Spom Medicine. St
Louis, MO, CV Mosby Co, 1986, pp 15341579
12 Anderson WV: Modern Trends in Ortho-
paedics. New York, NY, Appleton-
Century-Crofts, 1972
13 Giles LG: Lumbosacral facet "joint angles"
associated with leg length inequality. Rheuma-
to1 Rehabil 20:233-238, 1981
14 Ingram AJ: Anterior poliomyelitis. In
Edmonson AS, Crenshaw AH (eds): Campbell's
Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014
Operative Orthopaedics, ed 6. St Louis, MO,
CV Mosby Co, 1980, vol 2, pp 155G1560
15 Clarke GR: Unequal leg length: An accurate
method of detection and some clinical results.
Rheumatology and Physical Medicine 11:385-
390, 1972
16 Fisk JW, Baigent ML: Clinical and radiologi-
cal assessment of leg length. N Z Med J
81:477-480, 1975
17 Nichols PJR, Bailey Mu: The accuracy of
measuring leg-length differences. Br Med J
1247-1248, 1955
18 Mann M, Glasheen-Wray M, Nyberg R:
Therapist agreement for palpation and obser-
vation of iliac crest heights. Phys Ther 64:334-
338, 1984
19 Friberg 0,Nurminen M, Korhonen K, et al:
Accuracy and precision of clinical estimation
of leg length inequality and lumbar scoliosis:
Comparison of clinical and radiological mea-
surements. Int Disabil Stud 10:4%53, 1988
20 Woerman AL, Binder-Macleod S: Leg length
discrepancy assessment: Accuracy and preci-
sion in five clinical methods of evaluation.
Journal of Orthopaedic and Sports Physical
Therapy 5:23G239, 1984
Physical TherapyNolume 70, Number 3/March 1990
21 Jones SL: Measurement screen for leg
length and scoliosis: Suggestion from the field.
Phys Ther 56:18&189, 1976
22 Okun SJ, MorganJW, Burns MJ: Limb
length discrepancy: A new method of measure-
ment and its clinical significance.J Am Podiatr
Assoc 72:595-599, 1982
23 Hirschberg GG, Robertson KB: Device for
determining difference in leg length. Arch
Phys Med Rehabil 53:4546, 1972
24 Hoppedeld S: Physical Examination of the
Spine and Extremities. New York, NY,
Appleton-Century-Crofts, 1976
25 Kendall FP, McCreary EK: Muscle Testing
and Function, ed 3. Baltimore, MD, Williams &
Wilkins, 1983
26 Eichler J: Methodological errors in docu-
menting leg length and leg length discrepan-
cies. In: Der Orthopade. New York, NY,
Springer-Verlag New York Inc, 1972, vol 1, pp
1 4 2 0 (translated from German)
27 Beanie P, Rothstein JM, Kopriva L: The clin-
ical reliability of measuring leg length.
Abstract. Phys Ther 68:588, 1988
28 Horsefield D, Jones S: Assessment of in-
equality in length of the lower limb. Radiogra-
phy 52:233-237, 1986
29 Shrout PE, Fleiss JL: Intraclass correlations:
Uses in assessing rater reliability. Psychol Bull
86:420428, 1979
 Validity of Derived Measurements of Leg-Length
Differences Obtained by Use of a Tape Measure
Paul Beattie, Kale Isaacson, Dan L Riddle and Jules M
Rothstein
PHYS THER. 1990; 70:150-157.

Cited by This article has been cited by 3 HighWire-hosted articles:

http://ptjournal.apta.org/content/70/3/150#otherarticles
Subscription http://ptjournal.apta.org/subscriptions/
Information
Permissions and Reprints http://ptjournal.apta.org/site/misc/terms.xhtml
Information for Authors http://ptjournal.apta.org/site/misc/ifora.xhtml

Downloaded from http://ptjournal.apta.org/ by guest on May 1, 2014