[ research report ]

OLIVERA DJORDJEVIC, MD1,2 • ALEKSANDAR DJORDJEVIC, MD1 • LJUBICA KONSTANTINOVIC, MD, DSc1,2

Interrater and Intrarater Reliability

of Transverse Abdominal and Lumbar
Multifidus Muscle Thickness in Subjects
With and Without Low Back Pain
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.

L
ow back pain (LBP) is one of the most prevalent medical and morphometry at rest and during con-
conditions, with a considerable socioeconomic impact.3-5,14,18 traction. Ultrasound imaging may be use-
ful for identifying patients with structural
Low back pain has been associated with an impaired function
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

or functional deficits of the muscles that

of deep trunk stabilizers, such as the transverse abdominal (TrA) provide spine stabilization, evaluating
and lumbar multifidus (LM) muscles.7-10,22,24,25,36,37 These muscles treatment outcomes, and providing visual
are commonly assessed by ultrasound imaging (USI), which allows feedback during therapeutic exercise.
noninvasive, real-time monitoring of changes in muscle morphology Both B-mode (brightness)13,20,26,27,29,30,32,
34,35,44,48,49,52
and M-mode (motion)8,29,38,39
USI have been used for measuring thick-
TTSTUDY DESIGN: Two-group, repeated-mea- measure thickness (LBP group, 0.73-0.84; healthy
group, 0.93-1.00). Both interrater ICC2,1 and ICC2,3
ness of TrA and LM muscles. Static, cross-
sures reliability study.
sectional images acquired from the whole
TTOBJECTIVES: To determine interrater and
were lower for the relative thickness change (0.61-
Journal of Orthopaedic & Sports Physical Therapy®

intrarater reliability of ultrasound measure-
ments of transverse abdominal (TrA) and lumbar
multifidus (LM) muscle thickness, during rest and
contraction, in subjects with low back pain (LBP)
and healthy subjects over 3 consecutive days,
performed by an experienced and a novice rater.
TTBACKGROUND: Previous reliability studies
of TrA or LM thickness did not simultaneously
account for muscle state, rater experience, and
multiday assessment in large subject samples.
TTMETHODS: The 2 raters measured TrA and
LM thickness on 3 consecutive days in 42 healthy
subjects and 56 subjects with LBP, during rest and
contraction, and calculated the percent thickness
change from rest to contraction. Intraclass correla-
tion coefficients (ICC2,k) and 95% minimal detect-
able change in thickness were derived for a single
measure (day 1) and an average measure (days 1-3).
TTRESULTS: The interrater ICC2,1 values for single-
measure thickness (LBP group, 0.71-0.87; healthy
group, 0.94-1.00) were similar to those for average-
0.96). Intrarater ICC2,1 values across 3 consecutive
days were high for both raters across the 2 groups
(LBP group, 0.95-1.00; healthy group, 0.93-1.00),
albeit lower for the relative thickness change (0.79-
0.99). The 95% minimal detectable changes were
less than 0.3 mm for the TrA and less than 2 mm
for the LM (but, in most cases, less than 10% of
average thickness).
TTCONCLUSION: Both experienced and trained
novice raters provided reliable measurements
of TrA and LM thickness in participants with
LBP and healthy participants, during rest and
contraction. One-time measurements were similar
to averaged measurements. Small absolute
errors were observed. Public trial registry:
Australian New Zealand Clinical Trials Registry
ACTRN12613001077752. J Orthop Sports Phys
Ther 2014;44(12):979-988. Epub 3 November
2014. doi:10.2519/jospt.2014.5141
TTKEY WORDS: assessment, rehabilitation,
reproducibility of measurements, trunk,
ultrasonography
surface of the transducer in B-mode are
adequate for analyzing the structure and
diameter of the muscle and its surround-
ings. The M-mode images are created
from the data received from the mid-
point of the transducer. In the continu-
ously moving image received in M-mode,
hyperechoic fascial- or bone-to-muscle
boundaries become more visible, which
allows more precise determination of the
distance between the muscle edges. Thus,
M-mode is more suitable for measuring
muscle thickness.
A comparison of USI against magnetic
resonance imaging,27 which is the gold
standard for evaluating muscle struc-
ture, showed that USI is a valid method
for assessing the morphology of the LM
muscle. In addition to validity, it is nec-
essary to establish the reliability of USI

Clinic for Rehabilitation Dr M. Zotovic, Belgrade, Serbia. 2Faculty of Medicine, University of Belgrade, Belgrade, Serbia. The research protocol was approved by the Medical
1

Research Ethic Committee of the Clinic for Rehabilitation Dr M. Zotovic. The authors certify that they have no affiliations with or financial involvement in any organization or entity
with a direct financial interest in the subject matter or materials discussed in the article. Address correspondence to Dr Olivera Djordjevic, Clinic for Rehabilitation Dr M. Zotovic,
Sokobanjska 13a, 11000 Belgrade, Serbia. E-mail: odordev@eunet.rs t Copyright ©2014 Journal of Orthopaedic & Sports Physical Therapy®

journal of orthopaedic & sports physical therapy | volume 44 | number 12 | december 2014 | 979

44-12 Djordjevic.indd 979 11/17/2014 4:44:33 PM

 [ research report
measurements to ensure proper inter-
pretation of results. The key aspects of
reliability are equivalence and stability.
Equivalence is the extent of agreement
between measurements obtained by dif-
ferent examiners at the same time (in-
terrater reliability), and stability is the
extent of agreement between the repeat-
ed measurements obtained by the same
examiner (intrarater reliability). The
study of interrater and intrarater reliabil-
ity should closely resemble the intended
use of measurements. When studying
deep trunk stabilizers in clinical settings,
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.
typically, both the TrA and LM muscles
are assessed; the assessment is performed
by clinicians with different levels of expe-
rience; the outcomes of interest are the
thickness at rest and during a contrac-
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
tion, and the relative change from rest to
contraction; and the measurements are
compared over time or between those
with and those without LBP. Thus, the
reliability study should account for all
factors that may affect the equivalence
and stability of measurements, which re-
quires larger study samples.
The reliability of USI for assessing
Journal of Orthopaedic & Sports Physical Therapy®
thickness of deep stabilizing muscles has
been investigated before as a primary
objective8,12,26,32,35,38,48,51,53 or as part of a
larger study.13,31,39,43,46,49,50 In general, these
reports indicate good to excellent reliabil-
ity of USI measurements, but they ad-
dress in isolation the factors mentioned
above, and the majority have targeted
either the TrA8,13,26,38,39,43,46,49 or LM.50,51
The comparative reliability of these 2
muscles is of interest, because they are
typically both examined in clinical and
research settings due to the controversy
as to which muscle is more important in
certain forms of LBP.22,24,36,55
Some studies have included only
a­symptomatic individuals,8,26,39,51 yet ex-
trapolation of reliability from healthy
individuals to those with LBP is not
advised, because of pain or movement
limitations that may affect USI measure-
ments. Also, previous samples typically
ranged from 10 to 20 subjects,8,13,38,39,51
which may introduce errors when deriv-
980 | december 2014 | volume 44 | number 12 | journal of orthopaedic & sports physical therapy
44-12 Djordjevic.indd 980
]
Subjects with LBP, n = 56
Healthy controls, n = 42
Experienced rater
Day 1: USI
Day 2: USI
Day 3: USI
Experienced rater
Intrarater ICC2,1
FIGURE 1. The overview of subject flow, data collection, and data analysis. Abbreviations: ICC, intraclass
correlation coefficient; LBP, low back pain; USI, ultrasound imaging.
ing multiple reliability estimates.
In terms of methodological limita-
tions, previous reliability studies have ex-
amined muscles in a relaxed state13,43,50,51
and have not reported the reliability of
relative change in thickness from rest to
contraction. Occasionally, the selected
outcomes were measures of association
(coefficient of correlation) or dispersion
(coefficient of variation)12,40 rather than
measures of agreement. In addition,
when thickness was measured more
than once, the time interval was typically
short,39,48,50,51 potentially leading to recall
bias.
Ultrasound imaging requires prop-
er education and experience.54 To our
knowledge, however, there are no specific
recommendations for USI of the TrA and
LM muscles. It is difficult to interpret
reliability without knowing the skills of
examiners, yet only a few groups have
reported the duration of training and
approximate number of USI studies per-
formed by the examiners.26,32,35,48,51
To overcome some of these limita-
tions, the objective of this study was to
determine interrater and intrarater reli-
ability of USI measurements of thickness
in the TrA and LM muscles during rest
Novice rater
Single measure
Day 1: USI
Interrater ICC2,1
Average measure
Day 2: USI
Interrater ICC2,3
Day 3: USI
Novice rater
Intrarater ICC2,1
and contraction, within a single day and
across 3 consecutive days, by an experi-
enced and a novice rater, in a sample in-
clusive of people with and without LBP.
We hypothesized that the intraclass cor-
relation coefficient (ICC) for the single-
measure (day 1) and average-measure
(days 1-3) thickness would be greater
than 0.75 across both muscles, during
rest and contraction, in both individu-
als with LBP and healthy individuals,
despite differences in rater experience,
thus indicating excellent reliability.11,21
The corresponding estimates of absolute
agreement were also calculated (stan-
dard error of measurement [SEM], 95%
minimal detectable change [MDC95]) to
indicate true change in muscle thickness
over time.
METHODS
Participants
T
he present study included 56
subjects with LBP and 42 healthy
subjects. The subjects with LBP
were recruited from patients referred for
evaluation and treatment to the Clinic
for Rehabilitation Dr M. Zotovic, and
the healthy subjects were recruited from
11/17/2014 4:44:34 PM

Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.
FIGURE 2. Examples of thickness measurements in (A) transverse abdominal and (B) lumbar multifidus muscles
from a healthy subject during relaxation (left) and contraction (right). The vertical bar spans the boundary
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
between the muscle and fascia (calibration, 1 cm between each dot). The values for the transverse abdominal are
4.7 mm (left) and 7.3 mm (right), and for the lumbar multifidus are 35.1 mm (left) and 41.9 mm (right).
the spouses, relatives, and acquaintances
of the patients with LBP and the investi-
gators. None of the subjects declined to
participate or dropped out of the study.
Low back pain was defined as pain
between the 12th rib and buttocks, with
Journal of Orthopaedic & Sports Physical Therapy®
or without radiation into one or both
legs, originating in the lumbar region.15
The exclusion criteria for all participants
were pregnancy, spinal fractures, surgery,
infectious diseases, tumors, spina bifida,
advanced forms of spinal deformity, hip
diseases, and neuromuscular disorders.
Additional exclusion criteria for partici-
pants with LBP were self-reported pain
levels of less than 3 on the numeric pain
rating scale (to account for patients com-
monly encountered in clinical practice)
and inability to remain in prone and su-
pine hook-lying positions for 10 minutes
(to ensure compliance). The participants
signed an informed-consent form, and the
rights of human subjects were protected.
The study was approved by the Medical
Research Ethics Committee of the Clinic
for Rehabilitation Dr M. Zotovic.
Raters
An experienced rater and a novice rater
independently performed USI of the
journal of orthopaedic & sports physical therapy | volume 44 | number 12 | december 2014 | 981
44-12 Djordjevic.indd 981
TrA and LM muscles on the same day.
The experienced rater was a medical
doctor board certified in physical medi-
cine and rehabilitation and in plastic
and reconstructive surgery. The expe-
rienced rater went through a 24-month
training in abdominal USI, Doppler
imaging, soft tissue and thyroid USI,
breast USI with interventional pro-
cedures, and an additional 14-month
training in musculoskeletal USI. At the
time of the study, the experienced rater
had performed USI for about 5.5 years
and averaged about 40 musculoskeletal
USI examinations per week. The novice
rater was a senior at medical school.
He was given a basic training in USI
techniques (5 hours) and performed 20
measurements of abdominal and back
muscle thickness under supervision of
the experienced rater. The 2 raters were
blind to group allocation and were not
involved in subject recruitment or clini-
cal data collection.
Procedure
A study overview is presented in FIGURE
1. Demographic data and past medical
history were acquired by a nurse, who
also administered the 11-point numeric
pain rating scale (range, 0-10)19,28 to as-
sess pain intensity and the Oswestry
Low Back Pain Disability Questionnaire
(range, 0-100) to assess self-reported dis-
ability in everyday life.23 The nurse was
not involved in the USI measures.
A Toshiba diagnostic ultrasound sys-
tem (Nemio SSA-550A; Toshiba Corpo-
ration, Tokyo, Japan) with a 3.75-MHz
curvilinear probe was used to acquire
images in B-mode. Each rater measured
TrA and LM thickness bilaterally using
the on-screen calipers once a day on 3
consecutive days. The side measured first
was counterbalanced. The raters were
unaware of each other’s measurements.
For the measure of TrA thickness, the
participant was in a supine hook-lying
position, with the transducer placed
transversely just above the iliac crest in
the mid-axillary line.30,31 The participant
was instructed to breathe normally. The
images were taken at rest and during
muscle contraction. Because the TrA is
an auxiliary respiratory muscle actively
involved in voluntary expiration,1,16 the
relaxed-TrA image was acquired at the
end of quiet inspiration39 to prevent TrA
activation. For the acquisition of the ac-
tivated-TrA image, the participant was
instructed to exhale and to slowly draw
the stomach in as much as possible, af-
ter which the image was acquired. No
instructions, tactile cues, or verbal en-
couragement were given. The thickness
of the TrA was measured as the distance
between the opposite hyperechoic fascial
layers, without taking into account fascia
thickness (FIGURE 2A).
The assessment of LM thickness was
performed with the participant in prone,
with a pillow under the participant’s
abdomen to reduce the lumbosacral
junction angle to less than 10°. The mea-
surement was taken by an electrogoni-
ometer connected to an ADU301 Angle
Display Unit (Biometrics Ltd, Newport,
UK). The transducer was positioned lon-
gitudinally at the midline of the L4 spi-
nal process, then moved slightly laterally
and angled medially to obtain the image
of the L4-5 zygapophyseal joint.30,31 The
11/17/2014 4:44:35 PM
 [ research report ]
variable was presented as mean  SD,
Characteristics of Subjects With Low calculated as the average across the 2 rat-
TABLE 1
Back Pain and Healthy Subjects* ers and 3 study days.
To evaluate agreement within and be-
Low Back Pain Healthy tween the 2 raters, a 2-way, mixed-model
Age, y 47.0  7.4 51.0  9.2 ICC2,k was selected. Specifically, ICC2,1
Sex (women, men), n 30, 26 24, 18 was used for interrater reliability of the
Body mass index, kg/m2 27.1  5.5 25.0  6.7 single measures (day 1) and ICC2,k was
ODI score 32.0  11.1 … used for interrater reliability of the aver-
NPRS last 24 h 4.1  1.8 … age measures across 3 consecutive days,
Duration of pain, d 81.4  7.1 … where k represented the number of thick-
Abbreviations: NPRS. numeric pain rating scale; ODI, Oswestry Disability Index. ness measurements averaged (ICC2,3).
*Values are mean  SD unless otherwise indicated.
Intrarater reliability was assessed with
ICC2,1 based on 3 evaluations done by
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.

each rater on 3 consecutive days. Based

Thickness of TrA and LM Muscles on ICC point estimates, reliability was
at Rest and During Contraction and considered poor (less than 0.40), mod-
TABLE 2 Relative Thickness Change From Rest erate (0.40-0.59), good (0.60-0.74), or
to Contraction by Rater in Subjects With excellent (greater than 0.75).11,21 The reli-
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Low Back Pain and Healthy Subjects* ability was considered statistically differ-
ent if the ICC 95% confidence intervals
Low Back Pain Healthy (95% CIs) did not overlap.
Experienced Rater Novice Rater Experienced Rater Novice Rater The SEM was calculated as SD ≈
TrA √1 – ICC,41 and MDC95 as 1.96 × SEM.17,45,47
Rest, mm 4.3  1.3 4.4  1.0 4.1  1.0 4.3  0.9 Bland-Altman analysis6 was used to eval-
Contracted, mm 6.0  2.7 6.0  1.5 6.5  2.1 6.7  1.9 uate a systematic bias for each dependent
Change, % 39.5  6.8 36.9  8.8 59.0  8.5 55.6  10.6 variable by plotting the average measure-
LM ment for the 2 raters (x-axis) against the
Journal of Orthopaedic & Sports Physical Therapy®

Rest, mm 27.9  5.0 28.7  5.5 28.5  5.7 29.0  5.7 difference between the 2 raters (y-axis)
Contracted, mm 33.9  5.6 34.5  6.2 37.6  5.7 38.3  5.9 for each examined muscle, muscle state
Change, % 21  1.5 20  1.3 32  3 32  3.3 during the examination, and subject
Abbreviations: LM, lumbar multifidus; TrA, transverse abdominal. group.
*Values are mean  SD averaged across 3 consecutive days for each subject.
RESULTS
distance between the dorsal edge of the Both raters performed USI of the TrA
joint and the thoracolumbar hyperechoic and LM once a day on 3 consecutive days. General Characteristics

fascia, which separates the muscle from
the subcutaneous fat tissue, was con-
sidered the thickness of the LM muscle
(fascia thickness not included) (FIGURE
2B). The contraction image was acquired
with the muscle activated, by lifting the
contralateral arm off the bed and holding
it in 120° of shoulder abduction and 90°
of elbow flexion.30,31
For the subjects with LBP, the symp-
tomatic side was included in analysis. If
the pain was evenly distributed between
the left and right paravertebral regions,
the side was chosen at random. The
side for healthy controls was selected at
random.
Thus, each rater acquired and measured
thickness on 1176 images (98 partici-
pants by 2 muscles by 2 muscle states by
3 days). Each rater was blinded to the
other’s results.
Data Analysis
Statistical analyses were conducted in
SPSS Version 21 (SPSS Inc, Chicago, IL).
The dependent variables were thickness
at rest, thickness during contraction, and
relative change in thickness from rest to
contraction (percent difference between
the contraction and resting thickness
divided by the resting thickness). For
descriptive purposes, each dependent
A
ge, body mass index, and distri-
bution by sex in subjects with LBP
and healthy subjects are shown in
TABLE 1. The subjects with LBP reported,
on average, moderate levels of pain that
lasted nearly 3 months prior to the en-
rollment and caused moderate disability
(TABLE 1). The average thicknesses of the
TrA and LM muscles measured by the ex-
perienced and novice raters during rest
and voluntary contraction are presented
in TABLE 2. On average, the thickness of
the TrA increased from rest to contrac-
tion about 40% in subjects with LBP and
60% in controls, whereas the thickness
of the LM muscle increased about 20%

982 | december 2014 | volume 44 | number 12 | journal of orthopaedic & sports physical therapy

44-12 Djordjevic.indd 982 11/17/2014 4:44:35 PM

 in subjects with LBP and 30% in healthy
subjects (TABLE 2). Interrater Reliability of Single
Measures Taken on Day 1 in Subjects
TABLE 3
Single-Measure Interrater With Low Back Pain and Healthy Subjects
Reliability (Day 1) for the TrA and LM Muscles
The single-measure interrater ICC2,1
ranged from 0.71 to 0.87 for subjects Low Back Pain Healthy
with LBP and from 0.94 to 1.00 for Single Measure ICC2,1* SEM, mm MDC95, mm ICC2,1* SEM, mm MDC95, mm
healthy subjects (TABLE 3). The ICC values TrA
for relative thickness change were also Rest 0.71 (0.50, 0.83) 0.12 0.33 0.94 (0.89, 0.97) 0.11 0.30
lower in subjects with LBP (0.69-0.71) Contracted 0.87 (0.77, 0.92) 0.19 0.52 0.97 (0.95, 0.99) 0.21 0.59
than those in healthy subjects (0.93- Change, % 0.69 (0.48, 0.82) 0.02 0.07 0.93 (0.87, 0.96) 0.04 0.11
0.96). Overall, 9 of 12 single-measure LM
interrater ICC values were greater than Rest 0.85 (0.74, 0.91) 0.50 1.37 1.00 (0.99, 1.00) 0.62 1.71
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.

0.75. The 2 ICC values less than 0.75 Contracted 0.78 (0.62, 0.87) 0.64 1.76 0.97 (0.95, 0.99) 0.62 1.72
were for the relative thickness change Change, % 0.71 (0.50, 0.83) 0.35 0.97 0.96 (0.93, 0.98) 0.02 0.04
in both the TrA and LM of subjects with Abbreviations: ICC, intraclass correlation coefficient; LM, lumbar multifidus; MDC, minimal
LBP. The 95% CIs for the TrA and LM detectable change; SEM, standard error of measurement; TrA, transverse abdominal.
muscles within each subject group and *Values in parentheses are 95% confidence interval.
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

muscle state during examination over-

lapped, but the 95% CIs for the LBP
and healthy groups did not, indicating Interrater Reliability of Average
overall significantly different ICC values Measures Over 3 Consecutive Days in
TABLE 4
between the subject groups. The inter- Subjects With Low Back Pain and Healthy
rater SEM ranged from 0.11 to 0.21 mm Subjects for the TrA and LM Muscles
for the TrA and from 0.50 to 0.64 mm
for the LM in subjects with LBP and Low Back Pain Healthy
healthy subjects during both rest and
Journal of Orthopaedic & Sports Physical Therapy®

Average
contraction. The interrater SEM for rela- Measure ICC2,3* SEM, mm MDC95, mm ICC2,3* SEM, mm MDC95, mm
tive thickness change was from 2% to 4% TrA
across both muscles and groups, except Rest 0.76 (0.59, 0.86) 0.13 0.36 0.93 (0.87, 0.96) 0.11 0.29
for the LM in subjects with LBP (35%). Contracted 0.73 (0.53, 0.84) 0.23 0.64 0.98 (0.97, 0.99) 0.21 0.60
Change, % 0.61 (0.33, 0.77) 0.03 0.09 0.94 (0.88, 0.97) 0.04 0.10
Average-Measure Interrater LM
Reliability (Days 1-3) Rest 0.84 (0.73, 0.91) 0.49 1.37 1.00 (0.99, 1.00) 0.62 1.72
The average-measure interrater ICC2,3 Contracted 0.78 (0.62, 0.87) 0.56 1.55 0.97 (0.95, 0.99) 0.63 1.73
ranged from 0.73 to 0.84 in subjects with Change, % 0.69 (0.47, 0.82) 0.01 0.04 0.96 (0.93, 0.98) 0.02 0.05
LBP and from 0.93 to 1.00 in healthy Abbreviations: ICC, intraclass correlation coefficient; LM, lumbar multifidus; MDC, minimal
subjects (TABLE 4). The ICC2,3 values for detectable change; SEM, standard error of measurement; TrA, transverse abdominal.
relative thickness change were lower in *Values in parentheses are 95% confidence interval.

subjects with LBP (0.61-0.69) than those

in healthy subjects (0.94-0.96). Overall,
9 of 12 average-measure interrater ICC2,3
values were greater than 0.75. All 3 ICC2,3
values less than 0.75 were in subjects with
LBP (TrA contracted, TrA and LM thick-
ness change). Again, the 95% CIs of ICC2,3
values overlapped between the TrA and
LM muscles within each subject group and
muscle state, but not between the LBP and
healthy subjects, indicating significantly
different ICC2,3 values between the groups.
The interrater SEM ranged from 0.11 to
0.23 mm for the TrA and from 0.49 to 0.63
mm for the LM in subjects with LBP and
healthy subjects (TABLE 4). The interrater
SEM for the relative thickness change was
from 1% to 4% for both muscles and sub-
ject groups (TABLE 4).
Intrarater Reliability (Days 1-3)
The intrarater ICC2,1 for the experienced
and novice raters during rest and con-
traction ranged from 0.95 to 1.00 in sub-
jects with LBP and from 0.93 to 1.00 in
healthy subjects (TABLE 5). The ICC values
for relative thickness change ranged from
0.79 to 0.99 in both subjects with LBP
and healthy subjects. Thus, all ICC2,1 val-
ues for intrarater reliability were greater
than 0.75, with no appreciable difference
between the 2 raters with respect to the
muscle, muscle state, or group examined.
The intrarater SEM for the experienced

journal of orthopaedic & sports physical therapy | volume 44 | number 12 | december 2014 | 983

44-12 Djordjevic.indd 983 11/17/2014 4:44:36 PM

 [ research report ]
rater ranged from 0.16 to 0.33 mm, ex-
cept that of the LM in healthy subjects Intrarater Reliability for the
(0.86-0.88 mm). The intrarater SEMs Experienced Rater and the Novice Rater
TABLE 5
for the novice rater were larger (0.69- in Subjects With Low Back Pain and Healthy
0.91 mm) than those for the experienced Subjects for the TrA and LM Muscles
rater, except those of the TrA in healthy
subjects (0.14-0.30 mm). The intrarater Low Back Pain Healthy
SEM for relative thickness change was Single Measure ICC2,1* SEM, mm MDC95, mm ICC2,1* SEM, mm MDC95, mm
between 2% and 6%. TrA, experienced rater
Rest 0.97 (0.95, 0.98) 0.16 0.43 0.94 (0.91, 0.97) 0.16 0.44
Interrater and Intrarater MDC95 Contracted 0.98 (0.95, 0.99) 0.25 0.69 0.97 (0.96, 0.99) 0.33 0.92
The interrater MDC95 values were lower Change, % 0.88 (0.82, 0.93) 0.03 0.09 0.89 (0.83, 0.94) 0.06 0.16
for the TrA (less than 0.64 mm) than LM, experienced rater
those for the LM muscle (less than 1.76
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.

Rest 0.95 (0.92, 0.97) 0.16 0.43 1.00 (0.99, 1.00) 0.88 2.44
mm), but similar between subjects with Contracted 0.97 (0.95, 0.98) 0.23 0.64 1.00 (0.99, 1.00) 0.86 2.38
LBP and healthy subjects (TABLES 3 and Change, % 0.98 (0.97, 0.99) 0.02 0.05 0.79 (0.67, 0.87) 0.03 0.07
4). Also, the single-measure interrater TrA, novice rater
MDC95 values were comparable to the Rest 0.99 (0.99, 1.00) 0.69 1.91 0.93 (0.89, 0.96) 0.14 0.39
average-measure MDC95 values. The in­ Contracted 1.00 (0.99, 1.00) 0.75 2.08 0.98 (0.97, 0.99) 0.30 0.82
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

trarater MDC95 values for both TrA and Change, % 0.79 (0.69, 0.86) 0.04 0.10 0.92 (0.87, 0.95) 0.05 0.14
LM muscles were higher for the novice LM, novice rater
rater (less than 2.35 mm) than those for Rest 1.00 (0.99, 1.00) 0.73 2.03 1.00 (0.99, 1.00) 0.88 2.44
the experienced rater (less than 0.69 Contracted 1.00 (0.99, 1.00) 0.85 2.35 0.99 (0.99, 1.00) 0.91 2.52
mm) in subjects with LBP, but compa- Change, % 0.99 (0.98, 0.99) 0.02 0.05 0.82 (0.70, 0.90) 0.02 0.06
rable in healthy subjects (TABLE 5). Abbreviations: ICC, intraclass correlation coefficient; LM, lumbar multifidus; MDC, minimal
detectable change; SEM, standard error of measurement; TrA, transverse abdominal.
Bland-Altman Analysis *Values in parentheses are 95% confidence interval.

The average difference in muscle thick-

Journal of Orthopaedic & Sports Physical Therapy®

ness between the 2 raters (bias) was less
than 1 mm, regardless of the muscle ex-
amined, muscle state during the exami-
nation, and subject group, except for the
LM muscle at rest in healthy subjects
(1.2 mm). The 95% limits of agreement
in all cases included zero. The inspection
of plots indicated that data points were
randomly distributed above and below
zero along the y-axis, with no systematic
bias along the x-axis (FIGURES 3 and 4). Ex-
cept in rare instances, all measurements
fell within the calculated 95% limits of
agreement.
DISCUSSION
reliable measurements to determine true
change in thickness over time.
Interrater and Intrarater Reliability
The ICC2,k model applied here indicates
relative agreement between measure-
ments because it is influenced by the
variability between the subjects and by
the measurement error. In general, our
study sample was representative of the
population that typically receives care
for LBP at the Clinic for Rehabilitation
Dr M. Zotovic. Also, characteristics of our
subjects with LBP were similar to those
included in previous reliability32,35 and
USI biofeedback studies,2,49 which sup-
ports generalizability of our results.
ment and an average of 3 measurements).
However, no overlap in 95% CIs between
subjects with LBP and healthy subjects
indicates significantly lower interrater
reliability in subjects with LBP, particu-
larly for the relative change in thickness
(TABLES 3 and 4). Consistently lower reli-
ability for the relative thickness change
is likely due to a compounding effect
of measurement errors at rest and dur-
ing contraction. More interaction may
be needed between the experienced
and novice raters before their measure-
ments of relative thickness change can
be considered interchangeable. Because
estimates of the interrater reliability
were similar between the single-measure

T
he overall results of this study
indicate good to excellent interrater
and intrarater reliability of USI mea-
sures of TrA and LM thickness during rest
and contraction in LBP and healthy sub-
jects for both an experienced and a novice
rater. Low measurement errors indicate
The finding that most interrater ICC
values were above 0.75 indicates excellent
agreement between the experienced and
novice raters across the examined muscles
(TrA and LM), muscle states during the
examination (at rest and during contrac-
tion), and data analyzed (for 1 measure-
(TABLE 3) and average-measure (TABLE 4)
data, we consider 1-time measurement
of TrA and LM thickness sufficient for
longitudinal follow-up.
Lower interrater reliability in subjects
with LBP may be due to disordered mo-
tor control of trunk muscles or incon-

984 | december 2014 | volume 44 | number 12 | journal of orthopaedic & sports physical therapy

44-12 Djordjevic.indd 984 11/17/2014 4:44:36 PM

 subject groups argues against large per-
A
formance variability. Other contributing
2.0
factors may be differences in position-
Difference (Rater 2 – Rater 1), mm

1.5
ing, pressure applied, or tilt angle of the
1.0
transducer between the 2 raters,35 or lack
0.5
of a clear delineation of the double con-
0.0
tour of the zygapophyseal L4-5 joint in
–0.5
subjects with LBP, due to altered muscle
–1.0
echogenicity or degenerative changes in
–1.5
the lumbosacral spine.34
–2.0
1 2 3 4 5 6 7 Our results are in general agree-
Mean Thickness, mm ment with previous reliability stud-
B
4
ies,13,26,31,32,35,38,39,42,43,46,48-51 of which only
a few included both subjects with LBP
Difference (Rater 2 – Rater 1), mm
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.

2
and healthy subjects.31,35,38 Although sub-
jects with LBP in the latter studies were
0
similar to ours, the results are not directly
comparable because those studies report-
–2
ed reliability for subjects with LBP and
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

healthy subjects combined. Thus, our re-

–4
sults expand these previous observations
0 1 2 3 4 5 6 7 8 9 10 and provide novel information that may
Mean Thickness, mm
C be useful in both clinical and research
10 settings. The overall reliability results,
Difference (Rater 2 – Rater 1), mm

however, should be viewed in the context

5 of measures of absolute agreement (SEM,
MDC).
0
Journal of Orthopaedic & Sports Physical Therapy®

Absolute Agreement
–5 of Thickness Measures
The error of measurement (SEM) is of
–10 interest because it indicates precision,
15 25 35 45
that is, how close the results of repeated
Mean Thickness, mm
D measurements actually lie. We report
10 SEM of less than 1 mm (mostly less than
Difference (Rater 2 – Rater 1), mm

0.5 mm), with no clear differences be-

5 tween the subject groups, muscle states,
or raters (TABLES 3 through 5). Interpre-
0 tation of larger SEM values in the LM
compared to the TrA muscle should take
–5 into account differences in muscle sizes,
because errors are comparable in rela-
–10 tive terms (less than 10% of the average
20 30 40 50
Mean Thickness, mm
muscle thickness). Repeating measure-
ments over 3 consecutive days did not
FIGURE 3. Bland-Altman plots of the difference between 2 raters against the mean thickness across the raters for reduce SEM values, contrary to some
the TrA at rest (A) and during contraction (B), and the LM at rest (C) and during contraction (D) in subjects with reports.26,32,33,35,51 This may be due to the
low back pain (average-measure data shown; solid horizontal line is mean difference; 2 broken lines are upper and already smaller SEM in our study (floor
lower 95% limits of agreement). Abbreviations: LM, lumbar multifidus; TrA, transverse abdominal. effect), despite the fact that the measure-
ments were performed over 3 consecutive
sistent motor performance because of trarater reliability (ICC2,1>0.9) across the days. A decrease in SEM was reported in
pain.52 However, consistently high in- examined muscles, muscle states, and the studies that repeated measurements

journal of orthopaedic & sports physical therapy | volume 44 | number 12 | december 2014 | 985

44-12 Djordjevic.indd 985 11/17/2014 4:44:37 PM

 [ research report ]
each day; repositioning the ultrasound
A
probe after identifying the landmarks;
1.5
reapplying the pressure; and repeating
Difference (Rater 2 – Rater 1), mm

1.0 the motor task, image acquisition, and

measurements. All of these procedures
0.5
may introduce variability, yet the SEM
0.0
was low for both the experienced rater
and the novice rater.
–0.5 Of primary clinical interest are the
MDC values (also known as smallest de-
–1.0
2 3 4 5 6 7 tectable change), because they reflect a
B
Mean Thickness, mm value beyond which a measured param-
2
eter may be considered truly changed (ie,
minimal change not due to error). Un-
Difference (Rater 2 – Rater 1), mm
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.

1 like ICC, which is unitless, MDC is easily

interpretable because it is a function of
0
the SEM and expressed in the same units
as the measured parameters. Although
–1 there is no consensus on an acceptable
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

MDC level, we consider MDC95 values of

–2
less than 1 mm for the TrA and less than
15 25 35 45 2.5 mm for the LM sufficiently low and
Mean Thickness, mm
C clinically acceptable. Thus, larger differ-
ences may be considered true change in
2
Difference (Rater 2 – Rater 1), mm

muscle thickness over time.

1
Implications for Clinical Research
0 The main clinical implication of the re-
Journal of Orthopaedic & Sports Physical Therapy®

sults of the present study stems from

–1 the overall high reliability and low mea-
surement error for both single-day and
–2
average measures of muscle thickness,
3 4 5 6 7 8 9 10 11 12
assessed by an experienced rater and
Mean Thickness, mm
D a novice rater, in both the TrA and LM
2 muscles, during rest and contraction, in
Difference (Rater 2 – Rater 1), mm

subjects with LBP and healthy subjects.

1 This implies high precision within and
between the 2 raters for the examined
0 muscles, muscle states, subject groups,
and data set analyzed. Because the ab-
–1 solute agreement was not lower for the
3-day average compared to the single-day
–2 measurement, we propose that a 1-time
20 30 40 50 60
Mean Thickness, mm assessment by either the experienced or
properly trained novice rater is sufficient
FIGURE 4. Bland-Altman plots of the difference between 2 raters against the mean thickness across the raters for for longitudinal evaluation of TrA and
the TrA at rest (A) and the LM at rest (B), and the TrA during contraction (C) and the LM during contraction (D) in LM thickness. Further research is needed
healthy subjects (average-measure data shown; solid horizontal line is mean difference; 2 broken lines are upper to determine a minimal clinically impor-
and lower 95% limits of agreement). Abbreviations: LM, lumbar multifidus; TrA, transverse abdominal tant difference. The change in TrA and
LM thickness of more than 1 mm and 2.5
on the same day,26,33,35,51 but these results vative and clinically more realistic ap- mm, respectively, should be considered a
may be prone to recall bias. Our conser- proach required repositioning the subject real change over time.

986 | december 2014 | volume 44 | number 12 | journal of orthopaedic & sports physical therapy

44-12 Djordjevic.indd 986 11/17/2014 4:44:38 PM

 Study Limitations
Although no participant reported pain or
difficulty while performing the task, the
pain resulting from muscle activation
might have contributed to lower reliabil-
ity in subjects with LBP. We did not as-
sess pain during USI to keep the raters
blinded and to prevent interference with
the task performance. Another limitation
is that the reliability of the relative thick-
ness change from rest to contraction was
not assessed over a longer period. The
relative thickness change may be useful
for monitoring deficits in motor control
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.
of trunk muscles in subjects with LBP
and should be studied in the future.
CONCLUSION
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
T
he USI measurements of TrA
and LM thickness show good to ex-
cellent interrater reliability for both
muscles during rest and contraction, and
regardless of the difference in rater expe-
rience, although the reliability was signif-
icantly lower in subjects with LBP than
in healthy subjects. Intrarater reliability
was nearly perfect for both the experi-
Journal of Orthopaedic & Sports Physical Therapy®
enced rater and the novice rater across
the muscles, muscle states, and subject
groups. This applied to both the single-
day measures and the average measures
over 3 consecutive days. The provided
estimates of absolute measurement er-
ror may serve as a guide for determining
true changes in TrA and LM thickness
over time. t
KEY POINTS
FINDINGS: Interrater and intrarater reli-
ability of the ultrasound measurement
of the studied trunk muscles was good
to excellent at rest and during contrac-
tion for the experienced rater and the
novice rater, whether based on a single-
day or 3-day average measurement. The
estimates of absolute errors were low.
IMPLICATIONS: Ultrasound is a reliable
method for measuring thickness of the
TrA and LM muscles in subjects with
and without LBP when performed by
either an experienced rater or a prop-
journal of orthopaedic & sports physical therapy | volume 44 | number 12 | december 2014 | 987
44-12 Djordjevic.indd 987
erly trained novice rater. An increase in
muscle thickness greater than the esti-
mated absolute error may be considered
a true change over time.
CAUTION: The reliability of the relative
change in muscle thickness from rest
to contraction was not assessed over a
longer period. Reliability of only 2 raters
was examined.
ACKNOWLEDGEMENTS: The authors thank Do-
brivoje S. Stokic, MD, DSc for his guidance
and comments on the preliminary version of
the manuscript. We also thank the staff of the
diagnostic ward at the Clinic for Rehabilita-
tion Dr M. Zotovic.
REFERENCES
1. A be T, Kusuhara N, Yoshimura N, Tomita T,
Easton PA. Differential respiratory activity of four
abdominal muscles in humans. J Appl Physiol
(1985). 1996;80:1379-1389.
2. Anderson Worth SG, Henry SM, Bunn JY. Real-
time ultrasound feedback and abdominal hol-
lowing exercises for people with low back pain.
N Z J Physiother. 2007;35:4-11.
3. Andersson GB. Epidemiological fea-
tures of chronic low-back pain. Lancet.
1999;354:581-585. http://dx.doi.org/10.1016/
S0140-6736(99)01312-4
4. Andersson GBJ. The epidemiology of spinal
disorders. In: Frymoyer JW, Ducker TB, Hadler
NM, Kostuik JP, Weinstein JN, Whitecloud TS,
3rd, eds. The Adult Spine: Principles and Prac-
tice. 2nd ed. Philadelphia, PA: Lippincott-Raven;
1997:93-141.
5. Biering-Sørensen F. Low back trouble in a gen-
eral population of 30-, 40-, 50-, and 60-year-old
men and women. Study design, represen-
tativeness and basic results. Dan Med Bull.
1982;29:289-299.
6. Bland JM, Altman DG. Measuring agreement
in method comparison studies. Stat Meth-
ods Med Res. 1999;8:135-160. http://dx.doi.
org/10.1177/096228029900800204
7. Brennan GP, Fritz JM, Hunter SJ, Thackeray A,
Delitto A, Erhard RE. Identifying subgroups of
patients with acute/subacute “nonspecific” low
back pain: results of a randomized clinical trial.
Spine (Phila Pa 1976). 2006;31:623-631. http://
dx.doi.org/10.1097/01.brs.0000202807.72292.a8
8. Bunce SM, Moore AP, Hough AD. M-mode
ultrasound: a reliable measure of transversus
abdominis thickness? Clin Biomech (Bristol,
Avon). 2002;17:315-317.
9. Campbell SE, Adler R, Sofka CM. Ultrasound
of muscle abnormalities. Ultrasound Q.
2005;21:87-94; quiz 150, 153-154.
10. C ampbell WW, Vasconcelos O, Laine FJ.
Focal atrophy of the multifidus muscle in
lumbosacral radiculopathy. Muscle Nerve.
1998;21:1350-1353. http://dx.doi.org/10.1002/
(SICI)1097-4598(199810)21:10<1350::AID-
MUS21>3.0.CO;2-4
11. Cicchetti DV, Sparrow SA. Developing criteria
for establishing interrater reliability of specific
items: applications to assessment of adaptive
behavior. Am J Ment Defic. 1981;86:127-137.
12. Costa LO, Maher CG, Latimer J, Smeets RJ.
Reproducibility of rehabilitative ultrasound
imaging for the measurement of abdominal
muscle activity: a systematic review. Phys Ther.
2009;89:756-769. http://dx.doi.org/10.2522/
ptj.20080331
13. Critchley DJ, Coutts FJ. Abdominal muscle
function in chronic low back pain patients: mea-
surement with real-time ultrasound scanning.
Physiotherapy. 2002;88:322-332.
14. Damkot DK, Pope MH, Lord J, Frymoyer JW. The
relationship between work history, work environ-
ment and low-back pain in men. Spine (Phila Pa
1976). 1984;9:395-399.
15. Dawson AP, Schluter PJ, Hodges PW, Stewart S,
Turner C. Fear of movement, passive coping,
manual handling, and severe or radiating pain
increase the likelihood of sick leave due to low
back pain. Pain. 2011;152:1517-1524. http://
dx.doi.org/10.1016/j.pain.2011.02.041
16. De Troyer A, Estenne M, Ninane V, Van Gansbeke
D, Gorini M. Transversus abdominis muscle
function in humans. J Appl Physiol (1985).
1990;68:1010-1016.
17. Eliasziw M, Young SL, Woodbury MG, Fryday-
Field K. Statistical methodology for the concur-
rent assessment of interrater and intrarater
reliability: using goniometric measurements as
an example. Phys Ther. 1994;74:777-788.
18. Engel CC, Von Korff M, Katon WJ. Back pain in
primary care: predictors of high health-care
costs. Pain. 1996;65:197-204. http://dx.doi.
org/10.1016/0304-3959(95)00164-6
19. Farrar JT, Berlin JA, Strom BL. Clinically impor-
tant changes in acute pain outcome measures:
a validation study. J Pain Symptom Manage.
2003;25:406-411. http://dx.doi.org/10.1016/
S0885-3924(03)00162-3
20. Ferreira PH, Ferreira ML, Hodges PW. Changes
in recruitment of the abdominal muscles in
people with low back pain: ultrasound mea-
surement of muscle activity. Spine (Phila Pa
1976). 2004;29:2560-2566. http://dx.doi.
org/10.1097/01.brs.0000144410.89182.f9
21. Fleiss JL. Statistical Methods for Rates and Pro-
portions. 2nd ed. New York, NY: Wiley; 1981.
22. Fritz JM, Delitto A, Erhard RE. Comparison
of classification-based physical therapy with
therapy based on clinical practice guidelines
for patients with acute low back pain: a ran-
domized clinical trial. Spine (Phila Pa 1976).
2003;28:1363-1371; discussion 1372. http://
dx.doi.org/10.1097/01.BRS.0000067115.61673.
FF
11/17/2014 4:44:38 PM
 [ research report
23. Fritz JM, Irrgang JJ. A comparison of a modified
Oswestry Low Back Pain Disability Question-
naire and the Quebec Back Pain Disability
Scale. Phys Ther. 2001;81:776-788.
24. Hicks GE, Fritz JM, Delitto A, McGill SM. Prelimi-
nary development of a clinical prediction rule for
determining which patients with low back pain
will respond to a stabilization exercise program.
Arch Phys Med Rehabil. 2005;86:1753-1762.
http://dx.doi.org/10.1016/j.apmr.2005.03.033
25. Hides J, Wilson S, Stanton W, et al. An MRI
investigation into the function of the trans-
versus abdominis muscle during “drawing-
in” of the abdominal wall. Spine (Phila Pa
1976). 2006;31:E175-E178. http://dx.doi.
org/10.1097/01.brs.0000202740.86338.df
26. Hides JA, Miokovic T, Belavý DL, Stanton WR,
Downloaded from www.jospt.org at on March 13, 2019. For personal use only. No other uses without permission.
Richardson CA. Ultrasound imaging assessment
of abdominal muscle function during drawing-in
of the abdominal wall: an intrarater reliability
study. J Orthop Sports Phys Ther. 2007;37:480-
486. http://dx.doi.org/10.2519/jospt.2007.2416
27. Hides JA, Richardson CA, Jull GA. Magnetic
Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
resonance imaging and ultrasonography of
the lumbar multifidus muscle. Comparison of
two different modalities. Spine (Phila Pa 1976).
1995;20:54-58.
28. Jensen MP, Turner JA, Romano JM,
Fisher LD. Comparative reliability and valid-
ity of chronic pain intensity measures. Pain.
1999;83:157-162. http://dx.doi.org/10.1016/
S0304-3959(99)00101-3
29. Kidd AW, Magee S, Richardson CA. Reliability
of real-time ultrasound for the assessment
of transversus abdominis function. J Gravit
Journal of Orthopaedic & Sports Physical Therapy®
Physiol. 2002;9:P131-P132.
30. Kiesel KB, Uhl TL, Underwood FB, Rodd DW, Nitz
AJ. Measurement of lumbar multifidus muscle
contraction with rehabilitative ultrasound imag-
ing. Man Ther. 2007;12:161-166. http://dx.doi.
org/10.1016/j.math.2006.06.011
31. Kiesel KB, Underwood FB, Mattacola CG, Nitz
AJ, Malone TR. A comparison of select trunk
muscle thickness change between subjects with
low back pain classified in the treatment-based
classification system and asymptomatic con-
trols. J Orthop Sports Phys Ther. 2007;37:596-
607. http://dx.doi.org/10.2519/jospt.2007.2574
32. Koppenhaver SL, Hebert JJ, Fritz JM, Parent EC,
Teyhen DS, Magel JS. Reliability of rehabilitative
ultrasound imaging of the transversus abdomi-
nis and lumbar multifidus muscles. Arch Phys
Med Rehabil. 2009;90:87-94. http://dx.doi.
org/10.1016/j.apmr.2008.06.022
33. Koppenhaver SL, Parent EC, Teyhen DS, Hebert
JJ, Fritz JM. The effect of averaging multiple
trials on measurement error during ultrasound
imaging of transversus abdominis and lumbar
multifidus muscles in individuals with low back
pain. J Orthop Sports Phys Ther. 2009;39:604-
611. http://dx.doi.org/10.2519/jospt.2009.3088
34. Langevin HM, Stevens-Tuttle D, Fox JR, et al.
Ultrasound evidence of altered lumbar con-
988 | december 2014 | volume 44 | number 12 | journal of orthopaedic & sports physical therapy
44-12 Djordjevic.indd 988
]
nective tissue structure in human subjects
with chronic low back pain. BMC Musculo-
skelet Disord. 2009;10:151. http://dx.doi.
org/10.1186/1471-2474-10-151
35. Larivière C, Gagnon D, De Oliveira E, Jr., Henry
SM, Mecheri H, Dumas JP. Ultrasound mea-
sures of the lumbar multifidus: effect of task
and transducer position on reliability. PM R.
2013;5:678-687. http://dx.doi.org/10.1016/j.
pmrj.2013.03.010
36. Long A, Donelson R, Fung T. Does it matter
which exercise? A randomized control trial
of exercise for low back pain. Spine (Phila
Pa 1976). 2004;29:2593-2602. http://dx.doi.
org/10.1097/01.brs.0000146464.23007.2a
37. MacDonald DA, Dawson AP, Hodges PW. Behav-
ior of the lumbar multifidus during lower ex-
tremity movements in people with recurrent low
back pain during symptom remission. J Orthop
Sports Phys Ther. 2011;41:155-164. http://dx.doi.
org/10.2519/jospt.2011.3410
38. Mannion AF, Pulkovski N, Gubler D, et al. Muscle
thickness changes during abdominal hollowing:
an assessment of between-day measurement
error in controls and patients with chronic low
back pain. Eur Spine J. 2008;17:494-501. http://
dx.doi.org/10.1007/s00586-008-0589-x
39. McMeeken JM, Beith ID, Newham DJ, Mil-
ligan P, Critchley DJ. The relationship between
EMG and change in thickness of transversus
abdominis. Clin Biomech (Bristol, Avon).
2004;19:337-342. http://dx.doi.org/10.1016/j.
clinbiomech.2004.01.007
40. Misuri G, Colagrande S, Gorini M, et al. In vivo
ultrasound assessment of respiratory function
of abdominal muscles in normal subjects. Eur
Respir J. 1997;10:2861-2867. http://dx.doi.org/1
0.1183/09031936.97.10122861
41. Portney LG, Watkins MP. Foundations of Clini-
cal Research: Applications to Practice. 3rd ed.
Upper Saddle River, NJ: Pearson/Prentice Hall;
2008.
42. Pressler JF, Heiss DG, Buford JA, Chidley JV.
Between-day repeatability and symmetry of
multifidus cross-sectional area measured us-
ing ultrasound imaging. J Orthop Sports Phys
Ther. 2006;36:10-18. http://dx.doi.org/10.2519/
jospt.2006.36.1.10
43. Rankin G, Stokes M, Newham DJ. Abdominal
muscle size and symmetry in normal subjects.
Muscle Nerve. 2006;34:320-326. http://dx.doi.
org/10.1002/mus.20589
44. Rho M, Spitznagle T, Van Dillen L, Maheswari
V, Oza S, Prather H. Gender differences on
ultrasound imaging of lateral abdominal muscle
thickness in asymptomatic adults: a pilot
study. PM R. 2013;5:374-380. http://dx.doi.
org/10.1016/j.pmrj.2013.03.003
45. Smidt N, van der Windt DA, Assendelft WJ, et
al. Interobserver reproducibility of the assess-
@ MORE INFORMATION
ment of severity of complaints, grip strength,
and pressure pain threshold in patients with
lateral epicondylitis. Arch Phys Med Rehabil.
2002;83:1145-1150. http://dx.doi.org/10.1053/
apmr.2002.33728
46. Springer BA, Mielcarek BJ, Nesfield TK, Teyhen
DS. Relationships among lateral abdominal
muscles, gender, body mass index, and hand
dominance. J Orthop Sports Phys Ther.
2006;36:289-297. http://dx.doi.org/10.2519/
jospt.2006.2217
47. Stratford PW, Goldsmith CH. Use of the standard
error as a reliability index of interest: an applied
example using elbow flexor strength data. Phys
Ther. 1997;77:745-750.
48. Teyhen DS, George SZ, Dugan JL, Williamson
J, Neilson BD, Childs JD. Inter-rater reliability
of ultrasound imaging of the trunk muscula-
ture among novice raters. J Ultrasound Med.
2011;30:347-356.
49. Teyhen DS, Miltenberger CE, Deiters HM,
et al. The use of ultrasound imaging of the
abdominal drawing-in maneuver in subjects
with low back pain. J Orthop Sports Phys Ther.
2005;35:346-355. http://dx.doi.org/10.2519/
jospt.2005.35.6.346
50. Van K, Hides JA, Richardson CA. The use of
real-time ultrasound imaging for biofeedback
of lumbar multifidus muscle contraction in
healthy subjects. J Orthop Sports Phys Ther.
2006;36:920-925. http://dx.doi.org/10.2519/
jospt.2006.2304
51. Wallwork TL, Hides JA, Stanton WR. Intrarater
and interrater reliability of assessment of lum-
bar multifidus muscle thickness using reha-
bilitative ultrasound imaging. J Orthop Sports
Phys Ther. 2007;37:608-612. http://dx.doi.
org/10.2519/jospt.2007.2418
52. Wallwork TL, Stanton WR, Freke M, Hides JA.
The effect of chronic low back pain on size and
contraction of the lumbar multifidus muscle.
Man Ther. 2009;14:496-500. http://dx.doi.
org/10.1016/j.math.2008.09.006
53. Wong AY, Parent EC, Kawchuk GN. Reliability of
2 ultrasonic imaging analysis methods in quan-
tifying lumbar multifidus thickness. J Orthop
Sports Phys Ther. 2013;43:251-262. http://
dx.doi.org/10.2519/jospt.2013.4478
54. World Health Organization. Training in Diagnos-
tic Ultrasound: Essentials, Principles and Stan-
dards. Report of a WHO Study Group. Geneva,
Switzerland: World Health Organization; 1998.
55. Zielinski KA, Henry SM, Ouellette-Morton
RH, DeSarno MJ. Lumbar multifidus muscle
thickness does not predict patients with
low back pain who improve with trunk sta-
bilization exercises. Arch Phys Med Rehabil.
2013;94:1132-1138. http://dx.doi.org/10.1016/j.
apmr.2012.12.001
WWW.JOSPT.ORG
11/17/2014 4:44:39 PM