0196-601l/86/0804-Ol80$02.

00/0
THEJOURNAL OF ORTHOPAEO~C AND SPORTS PHYSICAL THERAPY
Copyright O 1986 by The Orthopaedic and Sports Physical Therapy Sections of the
American Physical Therapy Association

Reliability of a Noninvasive Method for

Measuring the Lumbar Curve*
DENNIS L. HART, PT, PhD.t STEVEN J. ROSE, PT, PhD*

The purposes of this paper were to describe a clinically useful and noninvasive
method of characterizing the shape of the lumbar spine and to evaluate the reliability
and validity of this measurement technique. A flexible ruler was applied to the skin
over the lumbar spines of 23 normal adults and an angle in degrees between two
spinous processes (L 1-S2) was calculated. lntratester test-retest reliability was good
(ICC = 0.97, N = 89) for two separate measures of two spinal postures. The validity
of the flexible ruler measurements was also good when compared to two different
measurement techniques from a limited number of patient roentgenographs. The
Downloaded from www.jospt.org at on November 21, 2021. For personal use only. No other uses without permission.

flexible ruler was determined to be a reliable and valid measurement technique for
the shape of the lumbar spine and may prove helpful in quantifying lumbar postures
and the effectiveness of clinical treatments designed to affect lumbar postures.
Copyright © 1986 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Quantification of the sagittal plane positions of ment of skin markers over spinous proc-
the lumbar vertebrae requires invasive measure- esSeS,lo.~5, 17 2) lateral plumb line measurements

ment techniques. Examples of various roentgen-
ographic techniques for these measurements are
prevalent. Three of the more common utilize: 1)
lines drawn through the center of the disc space;5
2) lines drawn tangentially to the inferior lips of
vertebral b o d i e ~ ,or~ ,the
~ ~cranial end plates of
Journal of Orthopaedic & Sports Physical Therapy®
(pendulum goniometer),153) photography with ex-
ternal marker^,'^^^' 4 ) goniometry (protrac-
tor),1°r215) flexible ruler measurement^,'^^ 6) com-
bined flexible ruler and hydrog6niometer meas-
urements,' 7) h y d r o g o n i ~ m e t e r s , ' ~and ~ ~a ~
~ ' ~8)
hand-held stylus with a multiturn potentiometer.lg
Obviously, external obstructions, e.g., tumors,

the vertebral b ~ d i e sand

; ~ 3) composite measure-
ments of two roentgenographs placed on top of
one another with one common reference point."
All of these methods require radiological exposure
which is not without risk. Other invasive methods
include the insertion of Kirschner wires into the
spinous processes16or posterior iliac spines4 and
photography instead of roentgenographs used for
the quantification of spinal posture. Although no
radiologic exposure is necessary, these proce-
dures are also not without risk.
Noninvasive techniques for sagittal plane mo-
tion attempt to characterize the shape of the spine
but do not offer the precision of radiologic meas-
urements. Several techniques utilize: 1 ) the move-
* Adapted from a paper at the InternationalSociety for the Study of the
Lumbar Spine Meeting in Toronto, 1982 and the American Physical
Therapy Association Meeting, Anahiem. 1982.
t President, Assessment Center Technology, Arlington, VA 22207;and
Research Fellow. Program in Physical Therapy. Washington University
Medical School, St. Louis, MO 631 10.
$ Director, Program in Physical Therapy, Washington University Medical
Center. St. Louis, MO 631 10.
adipose tissue, etc., negate the effectiveness of
any measurement tool that is placed on the sur-
face of the skin. However, all of these measure-
ments produce a quantified description of the
shape of the lumbar spine or the total movement
allowed during a task, e.g., forward bending.
None of the measures quantifies the vertebral
body positions, nor generates an angular meas-
urement similar to the angular measures made on
roentgenographs. Therefore, the purposes of this
paper were to: 7 ) describe a flexible ruler which
may be used to quantify the shape of the lumbar
spine in an angular measurement similar to those
that quantify vertebral positions, and 2) test the
reliability and validity of the flexible ruler for the
determination of the shape of the lumbar spine.
Preliminary analyses have been reported else-
where.'
METHODS
The flexible ruler (or curve) used in this study
was a 61 cm long, 0.8 cm wide, pliable metal
 JOSPT October 1986
band encased in a supple nonelastic plastic
(Brookstone Co., Peterborough, NY 03458). The
flexible ruler measurements for each of the testing
positions were made by firmly contouring the ruler
to the skin overlying the spinous processes of L1-
S2 inclusively. Skin marks denoting S2 (directly
between the posterior superior iliac spines) and
L1 were transferred to the ruler with a pencil (Fig.
1). The ruler was then placed on a piece of poster
Downloaded from www.jospt.org at on November 21, 2021. For personal use only. No other uses without permission.
Copyright © 1986 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Journal of Orthopaedic & Sports Physical Therapy®
Fig. 1 . Contouring the flexible ruler.
Fig. 2. Transferringthe shape of the lumbar spine to the poster
board.
FLEXIBLE RULER
Fig. 3. Trigonometric derivation of the angle representing the
shape of the lumbar pine.^*'^
Given: 1. d = R cos (012); 2. d = R - h; 3. / = 2R sin (012); 4.
cot ( 4 2 ) = (s~na)/(l - cos a) (for any angle a).
Derivation: 1. R - h = R cos (012); 2. R - R cos (012) = h; 3. R
[1 - cos (0/2)] = h; 4. R = h l j l - C O S ( ~ / ~5.) ]/; = 2h ( h / [ l -
cos (0/2)]) sin (812); 6. k' = 2h (sin (0/2)/[1 - cos (0/2)]; 7. / =
2h cot (014); 8. tan (014) = 2h//; 9. 014 = arctan (2h//); 10. 0
= 4 arctan (2h//).
TABLE 1
Flexible ruler validation
Flexible Vertebral Vertebral
Subject Level* ruler axis angle
8 8' 8"
* Roentgenographically confirmed levels.
t Pearson product moment correlation.
board, and the curve representing the shape of
the back with the L1 and S1 markers was trans-
ferred to the poster board (Fig. 2).
The calculation of the angle of the shape of the
lumbar spine proceded as described by Shoun''
(Fig. 3). The cord (f) connecting the L1 and S2
points on the arc representing the shape of the
lumbar spine was measured. A perpendicular (h)
was drawn from the arc to the cord f at its center
(f/2) by using a compass and a straight edge and
measured. Using the formula 0 = 4 arctan (2h/e),
the shape of the lumbar spine was reduced to an
 182 HART AND ROSE JOSPT Vol. 8, No. 4

angle in degrees. If the angle was posterior (lor-

dosis or backward bending), the angle was la-
beled negatively. If f3 was anterior (forward bend-
ing), the angle was considered positive.
Two measurements were made for each posi-
tion studied to establish the intratester test-retest
reliability of the flexible ruler measurements. The
positions studied were: 1 ) normal quiet standing,
and at the extremes of 2) forward bending in a
subject preferred manner, 3) forward bendirlg
with a trunk curl, and 4 ) forward bending while
keeping the back straight. The subjects returned
to upright posture between forward bending mo-
tions. All measurements were performed by one
of the authors (D. L. H.). The degree of agreement
between the test-retest measurements was ana-
Downloaded from www.jospt.org at on November 21, 2021. For personal use only. No other uses without permission.

lyzed.*
To begin to validate the measurement tech-
nique, six patients who were scheduled for a
lateral roentgenograph of their lumbar spines
were measured with the flexible ruler immediately
prior to their roentgenogram. Multiple small metal
Copyright © 1986 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

markers were placed over several lumbar spinous

processes to confirm the anatomical level of the
skin markers (Table 1). The flexible ruler measure-
ments of the shape of the back were compared
to the radiological shape of the roentgenographi-
cally confirmed intervertebral segments. To meas-
ure the shape of the back from the roentgeno-
graphs, the center of each vertebral body was
located and connected. The intervertebral seg-
Journal of Orthopaedic & Sports Physical Therapy®

ments between the points defined a curve on the

Fig. 4. Derivat~onof the vertebral axis angle 0'.
t i o n ~ . Lines
roentgenograph similar to the arc of the shape of
the back made by the flexible ruler (Fig. 4). Once
the arc from the roentgenograph was drawn, the
same method of derivation of 0 was used to
calculate 8'. The angle of the flexible ruler meas-
urements and the level confirmed roentgeno-
graphic measurements were compared.
0 and 8' were also compared to a traditional
measure of sagittal plane vertebral body posi-
~ ~ ~were
. ~ ~drawn tangentially to the
anterior and posterior inferior lips of the roentgen-
ographically confirmed vertebral bodies. This
angle was labeled 8" (Fig. 5). The degree of as-
sociation between all three measures of 0 was
analyzed with a Pearson Product Moment Corre-
lation.

RESULTS
The total range of motion (forward bending
minus backward bending) of the three movements
Fig. 5. Derivation of the vertebral angle 0". were analyzed by means of a Bartko intraclass
 JOSPT October 1986
correlation coefficient (ICC)~(Table 2). As this
number approaches 1.O, the difference between
the movements approaches zero and the meas-
urements are said to be in more agreement. The
test-retest measurements of 23 pairs of standing
measurements and 66 pairs of complete forward
bending (intratester reliability) correlated well (ICC
= 0.97, N = 89).
The flexible ruler measures (8) on six patients
(8 measures) were compared to the arc measures
of the vertebral axes (8') and to the vertebral
angles (8"). A good correlation was found be-
tween the flexible ruler measure and the vertebral
angle 8" (r = 0.87, N = 8) and between the
vertebral axis 8' and the vertebral angle 8" (r =
0.79, N = 8). A poor relation was found between
Downloaded from www.jospt.org at on November 21, 2021. For personal use only. No other uses without permission.
the flexible ruler measure 8 and the vertebral axis
measure 8' (r = 0.51, N = 8).
DISCUSSION
The flexible ruler was determined to be a relia-
Copyright © 1986 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
ble clinical measure for the quantification of the
shape of the lumbar spine for intratester test-
retest reliability (ICC = 0.97, N = 89). The meas-
ure was determined to have good clinical validity
as well when ,compared to a limited number of
roentgenographs using the angle between verte-
bral bodies (r = 0.87, N = 8).639320The latter
correlation is not in conformity with Shoun," who
found a poor relation between flexible ruler meas-
Journal of Orthopaedic & Sports Physical Therapy®
ure (8) and the vertebral angle measure (8") (r =
0.78, N = 6). However, Shoun found a fair relation
between flexible ruler measure (8) and vertebral
axis measure (8') (r = 0.78, N = 6), and a poor
relation between the two radiologic methods (r =
0.42, N = 6) where this study determined a fair
relation (r = 0.77, N = 8). One possible reason
for the discrepancy between the results of the
two studies was the limited number of patients
studied in each group.
Flexible rulershave been described beforeln8in
an attempt to estimate the movement of the spine
without radiologic exposure. Israel8utilized a flex-
ible ruler to characterize spinal flexion and exten-
sion, both static posture and at the extremes of
TABLE 2
Total range of forward bending (forward bending minus
backward bending) Mean f SD
Forward bending Forward bending with Fir:s.$g
preferred a curl back
+
58.6 11.2 59.6 & 12.0 +
58.4 11.9
N = 89 Bartko ICC3 = 0.97 F = 2.19
FLEXIBLE RULER 183
dynamic movements, but his trigonometric meas-
ures could not be easily compared to goniometric
measurements. In comparison, the trigonometric
method by Shoun18could be compared to gonio-
metric measurements. Anderson and Sweetman'
used a combination of a flexible ruler with two
hydrogoniometers to measure the contours of the
lumbar spine, while Mayer et aI.l4 used two hy-
drogoniometers to quantify lumbar motion. How-
ever, Anderson1 recommended the use of flexible
ruler measurements without the hydrogoniome-
ter, since the same hydrogoniometer readings
could be recorded in subjects with different spinal
postures.
Correlations between skin mark distraction and
vertebral mobility for normal subjects and patients
with reduced mobility have been reported.13-15.17
However, their measures do not allow a quantifi-
cation of the static posture nor equate to the
"shape" of the lumbar spine. The flexible ruler
allows the clinician to record the relative curvature
of the patient's lumbar spine without the con-
founding factor of hip mobility and the flexible
ruler measurement may be compared directly to
the measures from a roentgenograph. In a less
time consuming approach, clinicians may elect to
compare only the shapes of the curves between
treatments to determine any patient changes
without the need for mathematical calculations.
The ease of application and the modest cost of
instrument should allow any clinical department
to conduct between therapists reliability studies
and develop normative data for their specific pa-
tient populations. However, in order to expand
the clinical usefulness and scientific credibility of
the flexible ruler measurement technique, intertes-
ter reliability and expanded validity studies should
be performed.
CONCLUSION
The application of this simple clinical tool to
quantitate the shape of the lumbar spine was
presented. The intratester reliability and the valid-
ity of the tool were tested and found to be good.
The authors recommended several uses for the
flexible ruler and urged that further testing be
conducted.
REFERENCES
1. Anderson JHD, Sweetman BJ: A combined flexirule/hydrogoniom-
eter for measurement of lumbar spine and its sagittal movement.
Rheumatol Rehabil 14:173-179, 1975
Downloaded from www.jospt.org at on November 21, 2021. For personal use only. No other uses without permission.
HART AND ROSE
2. Bartko JJ, Carpenter WT: On the methods and theory of reliability.
J New Ment Dis 163(5):307-317, 1976
3. Beyer WH (ed): CRC Handbook of Mathematical Sciences, Ed 5,
pp 192,217. West Palm Beach, FL: CRC Press, 1975
4. Colachis SC, Worden RE, Bechtol CO, Strohm BR: Movement of
the sacroiliac joint in the adult male: a preliminary report. Arch
Copyright © 1986 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Phys Med Rehabil44:490-499,1963
5. Farfan HF, Huberdeau RM, Dubow HI: Lumbar intervertebral disc
degeneration. J Bone Joint Surg (Am) 54(3):492-510, 1972
6. Hansson T, Bigos S, Beecher P, Wortfey M: The lumbar lordosis
in acute and chronic low-back pain. Spine 10:154-155, 1985
7. Hart DL. Rose SJ: Reliability of a non-invasive method for meas-
uring the lumbar curve. Phys Ther 62:642. 1982
8. Israel M: A quantitative method of estimating flexion and extension
of the spine-a preliminary report. Milit Med 124(3):181-186,1959
9. Johnson RM, Hart DL. Simmons EF, Ramsby GR. Southwick WO:
Cervical orthoses: a study comparingtheir effectivenessin restrict-
ing cervical motion in normal subjects. J Bone Joint Surg (Am)
Journal of Orthopaedic & Sports Physical Therapy®
59(3):492-510, 1972
10. Joint Motion: Method of Measuring and Recording, pp 48-49.
Chicago: American Academy of Orthopedic Surgeons, 1965
11. Keegan JJ: Alterations of the lumbar curve related to posture and
seating. J Bone Joint Surg (Am) 35(3):589-603, 1953
12. Loebl WY: Measurements of spinal postures and range of spinal
movement. Ann Phys Med 9:103-110,1967
JOSPT Vol. 8, No. 4
13. MaCrae IF, Wright V: Measurement of back movement. Ann
Rheum Dis 28:584-589,1969
14. Mayer TG, Tencer AF, Kristoferson S, Mooney V: Use of noninva-
sive techniques for quantification of spinal range-of-motion in nor-
mal subjects and chronic low-back dysfunction patients. Spine
9:588-595,1984
15. Moll JMH, Wright V: Normal range of spinal mobility. Ann Rheum
Dis 30:381-386, 1971
16. Norton PL. Brown T: The immobilizing efficiency of back braces:
their effect on the posture and motion of the lumbosacral spine. J
Bone Joint Surg (Am) 39(1):111-138, 1957
17. Rae PS, Waddell G, Venner RM: A simple technique for measuring
lumbar spinal flexion. J R Coll Surg Edinb 29:281-284, 1984
18. Shoun C, Birmingham, AL (unpublished paper)
19. Stokes IAF, Abery JM: Influence of the hamstring muscles on
lumbar spine curvature in sitting. Spine 5(6):525-528, 1980
20. Torgenson WR, Dotter WE: Compqrative roentgenograpic study
of the asymptomatic and symptomatic lumbar spine. J Bone Joint
Surg (Am) 58(6):850-853, 1976
21. Troup JDG, Hood CA, Chapman AE: Measurements of the sagittal
mobility of the lumbar spine and hips. Ann Phys Med 9:308-321,
1968