TECHNICAL INNOVATION
Ultrasound Elastographic Measurement
of Sciatic Nerve Displacement and
Shear Strain During Active and Passive
Knee Extension
Richard Ellis, PhD , Maheswaran Rohan, PhD, James Fox, PhD, Juvena Hitt, BSc, Helene Langevin, MD,
Sharon Henry, PT, PhD
Received September 5, 2017, from the Depart-
ment of Physiotherapy, School of Clinical Scien-
ces, Faculty of Health and Environmental
Sciences (R.E.), Health and Rehabilitation
Research Institute, School of Clinical Sciences
(R.E), and Department of Biostatistics and Epi-
demiology, Faculty of Health and Environmen-
tal Sciences (M.R.), Auckland University of
Technology, Auckland, New Zealand; Depart-
ment of Neurological Sciences, University of Ver-
mont College of Medicine, Burlington, Vermont
USA (J.F., H.L.); Department of Medicine,
University of Vermont Larner College of Medi-
cine, Burlington, Vermont USA (J.H.); Osher
Center for Integrative Medicine, Division of Pre-
ventive Medicine, Brigham and Women’s Hos-
pital, Harvard Medical School, Boston,
Massachusetts USA (H.L.); and Department of
Rehabilitation Science, University of Vermont
College of Nursing and Health Science, Burling-
ton, Vermont USA (S.H.). Manuscript
accepted for publication November 24, 2017.
We thank Lindsey Tulipani, MS, DPT,
for providing assistance with testing and partici-
pant recruitment and Monkia Kessling for pro-
viding the illustrations for Figures 1 and 2.
Address correspondence to Richard Ellis,
PhD, Department of Physiotherapy, School of
Clinical Sciences, Faculty of Health and
Environmental Sciences, Auckland University of
Technology, Private Bag 92006, Auckland
1020, New Zealand.
E-mail: rellis@aut.ac.nz
Abbreviations
EMG, electromyographic; RF, radiofrequency;
ROI, region of interest; US, ultrasound
doi:10.1002/jum.14560
C 2018 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2018; 00:00–00 | 0278-4297 | www.aium.org
V
There is current need for objective measures of sciatic nerve mobility in patients
with sciatic-type pain. The objective of the study was to assess the feasibility and reli-
ability of ultrasound elastography to quantify sciatic nerve displacement and shear
strain at the sciatic nerve–hamstring muscle interface during active and passive knee
extension-flexion exercises performed while sitting in healthy people. Ultrasound
elastography showed excellent intrarater within-session reliability for assessing sciatic
nerve displacement and sciatic nerve–hamstring muscle interface shear strain during
active knee extension-flexion exercises. These findings will inform similar future
work conducted in patients with sciatic-type pain.
Key Words—peripheral nerve; peripheral nerve biomechanics; sciatic nerve;
sciatic nerve displacement; shear strain; ultrasound elastography
S
ciatica is characterized by radicular pain that radiates into the
leg and is a common variant of low-back pain.1 Approximately
70% of the population have low-back pain at some point in
their lives.2 An estimated 25%3 to 60%1 of cases of low-back pain
have associated sciatica. Although the most common cause for nerve-
related low-back or leg pain is from lumbar nerve root compression
or irritation,4 other less well-understood etiologies, such as deep glu-
teal syndrome (including sciatic nerve entrapment),5,6 could account
for a substantial subset of patients with nerve-related leg pain or
sciatica-type pain. In conditions such as deep gluteal syndrome, it has
been postulated that fibrosis and entrapment of the sciatic nerve at
the muscular interfaces of either the gluteal or hamstring muscles
may be key etiologic factors.5,6 However, there is currently a lack of
objective methods that can accurately assess the impact of peripheral
nerve irritation through these muscular interfaces.
As the body moves, peripheral nerves are constantly being
exposed to mechanical forces and stress from the surrounding tis-
sues that they come into contact with. These surrounding tissues,
otherwise known as the “mechanical interface,”7,8 are varied and
can include tissues such as bone, muscle, fascia, ligament, interverte-
bral disk, etc. For nerves to maintain optimal impulse conduction,
intraneural blood flow, and axoplasmic flow, they must be able to
independently move in response to these interfacing tissues to dissi-
pate mechanical forces.9,10
Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
There are several clinical conditions in which the
movement of nerves against their mechanical interface is
impaired. For example, there is compelling evidence that
impaired nerve excursion is a fundamental etiologic fac-
tor in compressive and entrapment peripheral neuropa-
thies such as carpal tunnel syndrome,11–13 lumbo-sacral
nerve root adhesion,14,15 and deep gluteal syndrome
(including sciatic nerve entrapment).5,6 Furthermore,
aberrant relationships between the peripheral nerve and
its mechanical interface will impose mechanical stress
on nerves.16 Several examples have been reported,
including fibrosis (or scar tissue) formation between
the nerve and surrounding muscles,5,6,17,18 increased
protective muscle spasms in the presence of neural
pain,19–21 and an impaired ratio of flexor tendon excur-
sion to median nerve excursion for people with carpal
tunnel syndrome.11,16,22
An important mechanical interface for the sciatic
nerve is that of the hamstring muscles. It has been sug-
gested that nerve mechanics may be altered depending
on muscle activity, which may be in response to a disor-
der or pain.23 In vivo research that examined the electro-
myographic (EMG) activity of the gluteal and hamstring
muscles showed significantly earlier onset of EMG activ-
ity during the straight-leg raise test for participants with
lumbar spine–related leg pain (radicular pain) compared
to healthy participants.20 The interpretation from these
authors was that this earlier onset of hamstring and glu-
teal EMG activity in the pain group indicated protective
muscle contraction due to increased neural mechanosen-
sitivity.20 Although not directly examined within this
study, the possibility remains that muscle contraction
(generally or as a protective response) may have an
influence on nerve mechanics,20,23 in this instance the
sciatic nerve, due to increased extrinsic pressures placed
on it from contracting muscles.
The use of ultrasound (US) imaging to examine
peripheral nerve mechanics is becoming increasingly
popular. From in vivo research using US imaging, it is
clear that nerves can move independently from their sur-
rounding interfacing tissues.16,24 Furthermore, studies
that have examined nerve excursion in response to active
joint movements25–27 and functional, weight-bearing
movements28,29 have also shown independent nerve
movement. The common feature across most studies
that have used US imaging to assess nerve movement
has been the examination of nerve movement in isola-
tion, without examining the potential influence of the
2 J Ultrasound Med 2018; 00:00–00
mechanical interface (ie, connective tissues, muscles,
etc) surrounding the nerve of interest. There are very
few studies that have looked at the biomechanical inter-
action at the interface between peripheral nerves and
their surrounding tissues. Liao et al16 examined median
nerve, flexor tendon, and flexor retinaculum displace-
ment and strain (axial and shear strain) using US imag-
ing. One of key findings of that research was the
significant decrease in shear strain between the median
nerve and flexor retinaculum in people with carpal tun-
nel syndrome compared to controls.16 Liao et al16 postu-
lated that this decrease in shear strain was due to a
decrease in median nerve displacement in people with
carpal tunnel syndrome.
A further use of US imaging has been to use US
elastographic techniques to examine biomechanical fea-
tures such as shear strain and passive stiffness of different
tissues.30 In the basic elastographic method, cross-
correlations of individual voxels between successive US
frames are used to calculate displacement in the axial
(along the US beam) and lateral (perpendicular to the
US beam) directions.31,32 This technique (using lateral
cross-correlations) has been previously adapted to mea-
sure shear strain between 2 adjacent regions of interest
(ROIs) within the thoracolumbar fascia in the low
back33 but so far has not been applied to measure the
motion of a nerve relative to surrounding tissue. Shear
wave elastography has also been used to examine sci-
atic,34 median,30,35 and tibial30 nerve passive stiffness in
response to different limb movements and postures.
A greater knowledge of the biomechanical influence
that surrounding tissues have on nerve excursion is criti-
cal to better understand the pathogenesis of compressive
and entrapment neuropathies. Furthermore, therapeutic
interventions such as neural mobilization have been
advocated to improve nerve excursion for several condi-
tions such as nerve-related low-back and leg pain and
nerve-related neck and arm pain.36 The design and
implementation of neural mobilization exercises would
be greatly enhanced with an advanced understanding of
the influence that the surrounding interfacing tissues
have on nerve mechanics. The primary aim of this study
was to assess the reliability of quantifying sciatic nerve–
hamstring muscle interface shear strain and sciatic nerve
displacement using US elastographic techniques similar
to those used in our previous study.33 The secondary
aim was to compare active and passive knee flexion-
extension exercises, used to induce movement of the
 Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
sciatic nerve, for the outcome measures of sciatic nerve–
hamstring muscle interface shear strain and sciatic nerve
displacement in healthy people, with the ultimate goal of
applying these measurements to patients with lumbar
spine–related leg pain.
Materials and Methods
Participant Recruitment
Healthy participants between the ages of 18 and 65 years
were recruited for this study. Participants were recruited
by advertisements and notices placed at the University
of Vermont within the College of Nursing and Health
Sciences.
Participants were excluded from the study if they
had current lumbar spine pain or lumbar spine–related
leg pain (ie, sciatica) or had a history of surgery to the
lumbar spine, thigh, or hip regions. On recruitment into
the study, each participant was provided with a research
information and summary sheet. The researchers pro-
vided a verbal summary of the protocol and answered
any remaining questions. All participants provided
informed and written consent. After written consent to
participate in this study was received, the following base-
line demographic details were recorded from all partici-
pants: sex, age, height, weight, and body mass index.
Ethical approval for this research was sought from and
Figure 1. Timing and phases of test movements.
J Ultrasound Med 2018; 00:00–00
approved by the University of Vermont Institutional
Review Board (Committee on Human Research in the
Medical Sciences No. 15-063) and was conducted in
compliance with the Declaration of Helsinki.
Testing Procedure
The leg that was tested was randomly chosen by rolling
a 6-sided die (even numbers, right leg; odd numbers, left
leg). There were 2 exercise conditions performed in a
standard order: (1) upright sitting with passive knee
extension/flexion; and (2) upright sitting with active
knee extension/flexion.
The knee joint range of movement was standar-
dized for both exercise conditions, from 80 8 flexion to
20 8 flexion (knee extension) and back again (knee flex-
ion) from 20 8 flexion to 80 8 flexion. This cycle of move-
ment occurred over 9 seconds: 4 seconds to extend the
knee, a 1-second pause at 20 8, and 4 seconds to flex the
knee (Figure 1). The movement cycle was performed in
time with a metronome for consistent timing and
standardization.
During the active condition, the participants moved
their own knees. During the passive condition, a research
assistant moved the participants’ knees for them. Before
the start of each of the new conditions (active versus
passive), there were up to 2 familiarization trials so that
3
Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
the participants and research assistant could become
synchronous with the metronome beat.
Participant Setup
Participants were seated upright over the end of a plinth,
with their spines resting against a rigid back support
(Figure 2). Both the left and right thighs were sup-
ported, with the posterior knees resting over a foam sup-
port. This held the knees in the starting position of 80 8
flexion and maintained the posterior thigh unsupported
to allow US imaging. For the leg that was randomized
for testing, the knee was extended passively to 20 8
extension. At this point, the position of the distal aspect
of the first toe was marked against an upright ruler. This
mark was used as a target by the researcher and partici-
pant for standardization during both the passive and
active movements.
The ankle and foot of the tested leg were secured in
a rigid foot/ankle orthosis, which held the ankle joint in
neutral (plantargrade). This standardized foot/ankle
position was used to ensure that there was no contribu-
tion from ankle joint movement to sciatic nerve excur-
sion, via the tibial nerve, during the respective exercises.
Electromyographic Measures
Surface EMG recordings were taken to quantify the
activity of the lateral and medial hamstring muscles. As
the hamstring muscles are a substantial mechanical inter-
face for the sciatic nerve, it was important to monitor
the activity of the hamstrings during both active and pas-
sive exercise conditions. During the passive exercise
Figure 2. Participant setup.
4 J Ultrasound Med 2018; 00:00–00
condition, it was not anticipated that there would be any
relevant EMG activity. Considering that the hamstrings
are not responsible for knee extension, it was also antici-
pated that there would be no relevant EMG activity of
the hamstrings during the active exercise condition.
For each participant, the area of skin where the elec-
trodes were placed was prepared by initially shaving the
skin, then abrading the skin lightly, and finally cleaning
the skin with isopropyl alcohol. This preparation was
used to facilitate the best conditions for conductance of
electrical signals between the skin and EMG electrodes.
Two wireless EMG electrodes were placed (with adhe-
sive strips) parallel to the muscle fibers over the muscle
belly of the medial and lateral hamstring muscles respec-
tively. Electromyographic data were recorded from
Trigno surface electrodes (Delsys, Inc, Natick, MA)
through Nexus software (Vicon, Denver, CO).
The EMG signals were collected over a 15-second
period divided into 5 epochs (Figure 1): a 5-second
baseline period (phase 1), 4 seconds during the knee
extension phase (phase 2), a 1-second pause, 4 seconds
during the knee flexion phase (phase 3), and a final 3-
second period on completion of the exercise (phase 4).
Electromyographic signals were collected at a sampling
rate of 1000 Hz and were bandpass filtered at 30 to
200 Hz with a Butterworth filter.
Ultrasound Imaging Procedure
B-mode US imaging with a Terason 3000 US machine
(Teratech Corporation, Burlington, MA) equipped with
a 10-MHz (12L5) linear array transducer was used to
collect all of the raw radiofrequency (RF) US signals. A
small amount of water-soluble gel was used as a coupling
medium. The US transducer was held by hand, with
light contact made against the skin, using the minimum
amount of pressure that ensured contact of the trans-
ducer with the skin. The location used for all US imaging
of the sciatic nerve was the posterior midthigh, at or
immediately distal to halfway between the gluteal and
popliteal creases. This location has been successfully
used in previous studies that have quantified sciatic
nerve movement.37–39 Care was taken to correctly iden-
tify the sciatic nerve and not the free tendon of the
biceps femoris. The free tendon of the biceps femoris is
found more distal to the midthigh location used in this
study, located at a distance of approximately 70% to
80% of the length of the biceps femoris muscle.40,41
 Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
Three separate US cine loops were recorded for each 9-
second trial at a sampling rate of 20 Hz.
The location of the sciatic nerve was visualized ini-
tially in the transverse plane at the posterior midthigh
location (Figure 3). Once located, the EMG sensors
were secured in place (see above). This action allowed
for placement of the EMG electrodes slightly proximal
to the midthigh location to keep them clear from the US
transducer and related coupling gel.
All US cine loops were visually inspected for image
quality. For a cine loop to be selected, the sciatic nerve
must have remained clearly visible in the field of view
during the entire 9-second trial. Those trials in which
the sciatic nerve was not clearly visualized through the 9-
second capture period were excluded from analysis.
Ultrasound Cine Loop Analysis
An offline analysis of the US cine loops was used to
quantify the sciatic nerve–hamstring muscle interactions.
The “raw” US RF data were processed by a custom pro-
gram33 written in MATLAB (The MathWorks, Natick,
MA). From the RF data, tissue displacements between
successive US frames (equivalent to the RF data acquired
at this specific time point) were estimated by cross-
correlation techniques31–33 with a 1-mm window incre-
mented with a 90% overlap. This technique analyzes the
RF data for both axial (parallel to the US beam) and lat-
eral (perpendicular to the US beam) displacements
by correlating the position of individual voxels respec-
tively along and across lines of US data using a cross-
correlation. The axial displacement measure is time
Figure 3. Ultrasound appearance of the sciatic nerve in the longitudinal plane (A) and transverse plane (B).
J Ultrasound Med 2018; 00:00–00
matched with the axial location of the ROI used to calcu-
late the lateral displacement. Thus, the axial location of
the ROI was adjusted from its starting position to com-
pensate for the axial displacement of the tissue before
determining the lateral displacement (perpendicular to
the US beam) at each time point using the same cross-
correlation technique. Analyses of tissue displacement
and shear strain were conducted in response to the fol-
lowing conditions: (1) as the knee extended from 80 8
flexion to 20 8 flexion (phase 2, knee extension); and (2)
as the knee flexed from 20 8 flexion to 80 8 flexion (phase
3, knee flexion; Figure 1).
Measurement of Tissue Displacement
Tissue displacement was calculated within a 10 3 25-
mm ROI centered within the sciatic nerve in both axial
and lateral directions relative to the US beam: either
along or perpendicular to the US beam, respectively.
While centered within the sciatic nerve, this 10 3 25-
mm ROI overlapped the sciatic nerve–hamstring muscle
interface. On B-mode US imaging, peripheral nerves are
identified in the longitudinal plane as tubular structures
delineated by 2 hyperechoic borders, consistent with the
external epineurium.42,43 The superficial sciatic nerve–
hamstring muscle interface (superficial epineurial border
of the sciatic nerve in contact with the adjacent ham-
string muscle) was used as a standardized location from
which to calculate relative displacement and shear strain
between the 2 interfacing tissues. This superficial inter-
face was chosen for 2 reasons. First, the thickness of the
sciatic nerve can vary considerably between individuals.
5
Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
From observations made during this study, sciatic nerve
thickness ranged from 4.0 to 8.1mm. Furthermore, US
identification of the superficial epineurial border is more
consistent because of the potential decrease in resolution
clarity that can be apparent when viewing the deeper
epineurial border.
Quantification of Sciatic Nerve–Hamstring Muscle
Interface Shear Strain
Shear strain represents the percentage of shear deforma-
tion between, in this instance, the sciatic nerve and ham-
string muscle. The calculation of shear strain between
these adjacent tissue layers allows an assessment of rela-
tive motion between these layers.
For the purpose of calculating shear strain between
nerve and muscle, 2 bands of tissue (sub-ROIs, each
3 3 10 mm) were identified: (1) within the hamstring
muscles, immediately above the superficial epineurial
layer (superficial ROI); and (2) within the sciatic nerve,
immediately below the superficial epineurial layer (deep
ROI; Figure 4).
Shear strain between the sub-ROIs was calculated
by methods previously reported.33 The absolute differ-
ence in lateral displacement between the superficial ROI
(hamstrings) and deep ROI (sciatic nerve), divided by
the vertical distance (3 mm) between the centers of the
2 sub-ROIs, accounted for the shear strain at the sciatic
Figure 4. Placement of ROIs.
6 J Ultrasound Med 2018; 00:00–00
nerve–hamstring muscle interface and was expressed as
a percentage.
Correction for Axial Tissue Displacement
At the region of the posterior midthigh, the movement
of the sciatic nerve is predominantly in a longitudinal
direction38 (lateral displacement perpendicular to the
US beam). However, a small amount of superficial-deep
movement of the sciatic nerve will also occur38 (axial dis-
placement along the US beam). To correct for this axial
displacement, an automated tracking system was used,
as within the MATLAB program.33 This system deter-
mined the axial displacement of the ROIs between each
successive frame. The location of the ROIs at each time
point was then offset according to the mean axial dis-
placement of the ROIs relative to the start position,
therefore allowing this correction for the final lateral dis-
placement calculation.33
Statistical Methods
Four primary outcome measures were assessed during
both active and passive exercise conditions: (1) sciatic
nerve–hamstring muscle interface shear strain (percent)
during knee extension; (2) sciatic nerve–hamstring mus-
cle interface shear strain (percent) during knee flexion;
(3) sciatic nerve displacement (millimeters) during knee
extension; and (4) sciatic nerve displacement (milli-
meters) during knee flexion.
All statistical analyses were conducted with R ver-
sion 3.2.4 statistical software (R Core Team, Vienna,
Austria).44 Participant demographic characteristics and
anthropometric measures (sex, age, height, weight, and
body mass index) were reported as mean 6 standard
deviation or frequency. The raw data for all primary out-
come measures (see above) versus the 2 exercise condi-
tions (passive and active knee movements) were
presented graphically. A Shapiro-Wilk test was used to
examine the data for normality. Statistical significance
was defined as P < .05.
To determine the test-retest reliability of all primary
outcome measurements with 2 exercise conditions, 2-
way mixed intraclass correlation coefficients with 95%
confidence intervals were calculated. Hallgren45 and Cic-
chetti46 discussed the strength of agreement based on
intraclass correlation coefficient values, with “perfect”
agreement for 1, “excellent” for values between 0.75 and
1, “good” for values between 0.6 and 0.74, “fair” for val-
ues between 0.4 and 0.59, “poor” for values of less than
 Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
0.4, 0 indicating only random agreement, and negative
indicating systematic disagreement.
On confirming data normality, the nlme package47
was used in the R statistical software to perform a linear
mixed-effects model. To compare the EMG data during
the passive and active exercise conditions, a repeated-
measures analysis of variance was performed, with the
results presented graphically with P values. To investi-
gate the relationship between the primary measurements
and exercise conditions of active and passive knee move-
ments, a linear mixed-effects model was used. As they
were fixed effects, the testing conditions (active and pas-
sive exercises) were entered into the model with the
addition of random effects for subjects into the model,
which treated the trials as nested within subjects.
Results
Twelve healthy participants (2 male and 10 female;
mean age, 23.9 6 4.74 years) volunteered for this cross-
sectional observational laboratory study. Two partici-
pants were excluded from the study (both female), as
the US footage was poor quality, which did not allow an
analysis. The most common cause of poor image quality
was movement of the sciatic nerve in the coronal plane,
which took the nerve beyond the field of view of the
transducer. The remaining 10 participants (2 male and 8
female; age, 24 6 5 years; height, 169 6 7 cm; weight,
65 6 9 kg; body mass index, 23 6 3 kg/m2) completed
the full study protocol.
Figure 5 shows all 4 primary outcome measures ver-
sus testing conditions. It displays the variability between
subjects and variation between trials within subjects.
However, no considerable difference between testing
conditions of active and passive movements was found
with respect to all outcome measures.
After the Shapiro-Wilk test, it was apparent that data
for sciatic nerve displacement during both knee flexion
and extension exercises, for this sample, were normally
distributed. Furthermore, sciatic nerve–hamstring muscle
interface shear strain during knee flexion and extension,
in a log scale, was normally distributed (P > .05).
The reliability of measuring all primary outcomes
with exercise conditions ranged from excellent to fair
(Table 1). For the active exercise condition, the reliabil-
ity for measuring shear strain and sciatic nerve displace-
ment, during both flexion and extension, was excellent.
For the passive exercise condition, the reliability of
J Ultrasound Med 2018; 00:00–00
measuring shear strain and sciatic nerve displacement
was fair aside from measurement of sciatic nerve dis-
placement during flexion, which was excellent. These
results are matched with Figure 5.
Table 2 gives means and standard deviations for all
primary outcome measures for both exercise conditions
plus P values obtained from the linear mixed model. No
significant differences in sciatic nerve–hamstring muscle
interface shear strain and sciatic nerve displacement
were seen between the active and passive exercise condi-
tions (Table 2 and Figure 6).
A repeated analysis of variance was performed to
examine the EMG data, for the lateral and medial ham-
string muscles, across the 4 phases for both active and
passive exercise conditions (Figure 7). There was no sig-
nificant difference in the level of EMG activation for any
of the exercise conditions across the 4 EMG phases.
This finding indicates that the EMG activities for both
the active and passive exercises, for both the medial and
lateral hamstring muscles, were similar throughout the
knee flexion and extension movements.
Discussion
The influence that adjacent tissues have on peripheral
nerve mechanics may play an important role in under-
standing clinical pathologic conditions, such as
entrapment neuropathies, in which peripheral nerve
biomechanics are believed to be adversely affected.
For clinical conditions such as deep gluteal syndrome,
irritation of the sciatic nerve against the interfacing
tissues of the gluteal or hamstring muscles may be an
etiologic factor.5,6 To our knowledge, this work was
the first study to specifically examine the relationship
between the sciatic nerve and its interface with the
hamstring muscles by assessing interface shear strain
using US elastographic techniques.
Another feature of this study was the use of both
active and passive exercise conditions to induce periph-
eral nerve excursion. Most previously reported studies
that used US imaging to assess peripheral nerve mechan-
ics used passive movements. The ability to assess the
influence of both active and passive exercises holds clini-
cal value. For example, many therapeutic techniques,
such as neural mobilization exercises, are performed
either passively or actively. For patients to use neural
mobilization exercises outside the clinical setting, the
exercises are often performed actively. Furthermore,
7
Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain

active exercises afford patients more control, which better for the active compared to the passive exercise
allows them greater control over potential provocative conditions.
movements. The findings of this study showed that The intrarater within-session reliability of using US
the reliability of measuring the primary outcomes was elastography to assess the 2 primary outcome measures

Figure 5. Primary outcome measures.

8 J Ultrasound Med 2018; 00:00–00

 Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain

was excellent to fair. These levels of reliability are similar
to those from other studies that used US elastography to
examine mechanical interface shear strain16 and longitu-
dinal sciatic nerve displacement.38,39,48
With regard to sciatic nerve–hamstring muscle inter-
face shear strain, there was no significant difference seen
between the active versus passive exercise conditions.
Likewise, there were no significant differences seen in
longitudinal sciatic nerve displacement between the
active versus passive exercise. Shear strain was gener-
ally greater during the knee flexion (73% active and
81% passive) compared to knee extension (60% active
and 63% passive). Unsurprisingly, a similar trend was
seen for longitudinal sciatic nerve displacement, with

Table 1. Intraclass Correlation Coefficient Values of All Primary Outcomes With 95% Confidence Interval for Each Testing Condition
Sciatic Nerve Sciatic Nerve
Shear Strain: Shear Strain: Displacement: Displacement:
Condition Flexion Extension Flexion Extension

Passive knee movement 0.70 (0.14, 0.92) 0.52 (20.52, 0.88) 0.83 (0.5, 0.96) 0.64 (20.14, 0.9)
Active knee movement 0.84 (0.48, 0.96) 0.81 (0.32, 0.96) 0.79 (0.25,0.96) 0.89 (0.66, 0.98)

Figure 6. Means of outcome measures with 95% confidence intervals (CI) by knee movements (active and passive).

J Ultrasound Med 2018; 00:00–00 9

Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain

greater movement during knee flexion (5.38 mm active
and 4.68 mm passive) compared to knee extension
(4.52 mm active and 4.34 mm passive).
The use of hamstring EMG was deliberate to assess
the relevant muscle activity during the active and passive
exercise conditions. As the hamstring muscles do not
perform knee extension, EMG activity during this part
of the active test condition was expected to be quiet.
Furthermore, as gravity is the primary force in lowering
the lower leg and foot as the knee flexes after extension
(in sitting), the EMG activity of the hamstrings was also
expected to be quiet. This expectation was the case, as
seen with no differences in EMG signals between the
active and passive exercise conditions compared to the

Table 2. Values for Sciatic Nerve–Hamstring Muscle Interface Shear Strain and Sciatic Nerve Displacement
Outcome Active Knee Movement Passive Knee Movement P

Shear strain: flexion, % 73.06 6 41.45 80.92 6 43.37 .13

Shear strain: extension, % 59.84 6 32.24 63.26 6 40.29 .70
Sciatic nerve displacement: flexion, mm 5.38 6 2.20 4.68 6 3.14 .07
Sciatic nerve displacement: extension, mm 4.52 6 2.63 4.34 6 2.81 .69

Figure 7. Lateral (L) and medial (M) hamstring muscle EMG data.

10 J Ultrasound Med 2018; 00:00–00

 Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
respective baseline period. This finding aligned with the
fact that no difference in sciatic nerve–hamstring muscle
interface shear strain was seen between the active and
passive exercise conditions, indicating that the active
hamstring contraction was not a factor and did not influ-
ence the primary outcomes.
The participants in this study were healthy individu-
als. It has been previously reported that hamstring EMG
activity increased during a straight-leg raise in people
with lumbar-related leg pain compared to healthy con-
trol participants.20 A future extension of this study would
be to replicate the study design, using US elastography
combined with EMG, to ascertain whether the nerve-
muscle interface interaction is altered in a subset of
patients with lumbar spine–related leg pain and to deter-
mine the clinical relevance of this potential interaction.
Limitations
This study had a small sample size of 12 participants,
with 2 being excluded for poor US image quality. The
findings with regard to reliability and the primary out-
come measures need to be taken into context with this
small cohort number. The cohort included healthy peo-
ple; therefore, the findings cannot be generalized to clini-
cal populations. However, research such as this,
performed in healthy cohorts, is important to provide
foundation results that can contribute to a better under-
standing of clinical populations once they are examined.
Furthermore, it must be noted that this study looked at
the hamstring muscle interface with the sciatic nerve and
not the gluteal muscle interface. In future research,
which may look at clinical conditions such as deep glu-
teal syndrome, both muscular interfaces (gluteal and
hamstring) would be recommended for assessment, as
both could be areas of sciatic nerve irritation.
Conclusions
Although US elastography was seen to be a reliable
method for calculating sciatic nerve–hamstring muscle
interface shear strain and sciatic nerve displacement,
some caution is required for this interpretation, given
the small sample size. Both active and passive knee
extension resulted in substantial shear strain (ranging
from 60% to 81%) at the nerve-muscle interface. Dis-
placement and shear strain measurement was more reli-
able for the flexion exercise condition. In healthy
participants, there is very little hamstring activity seen
during seated exercises using knee extension and flexion
to induce sciatic nerve movement. There was no
J Ultrasound Med 2018; 00:00–00
difference between active and passive knee extension-
flexion exercises, performed while sitting, with regard to
sciatic nerve–hamstring muscle interface shear strain and
sciatic nerve displacement. An examination of these bio-
mechanical features in clinical populations may provide
a deeper understanding of etiologic factors in conditions
such as entrapment neuropathies and, furthermore, for
the design of therapeutic exercises to influence nerve
mechanics.
References
1. Hill JC, Konstantinou K, Egbewale BE, Dunn KM, Lewis M, Van
Der Windt D. Clinical outcomes among low back pain consulters
with referred leg pain in primary care. Spine (Phila Pa 1976) 2011;
36:2168–2175.
2. Konstantinou K, Dunn KM. Sciatica: review of epidemiological stud-
ies and prevalence estimates. Spine (Phila Pa 1976) 2008; 33:2464–
2472.
3. Masters S, Lind R. Musculoskeletal pain: presentations to general
practice. Aust Fam Physician 2010; 39:425–428.
4. Koes BW, Van Tulder MW, Peul WC. Diagnosis and treatment of
sciatica. Br Med J 2007; 334:1313–1317.
5. Hernando MF, Cerezal L, Perez-Carro L, Abascal F, Canga A. Deep
gluteal syndrome: anatomy, imaging, and management of sciatic nerve
entrapments in the subgluteal space. Skeletal Radiol 2015; 44:919–934.
6. Martin HD, Shears SA, Johnson JC, Smathers AM, Palmer IJ. The
endoscopic treatment of sciatic nerve entrapment/deep gluteal syn-
drome. Arthroscopy 2011; 27:172–181.
7. Shacklock MO. Clinical Neurodynamics: A New System of Neuromuscu-
loskeletal Treatment. Oxford, England: Butterworth Heinemann; 2005.
8. Slater H, Butler DS, Shacklock MO. The dynamic central nervous sys-
tem: examination and assessment using tension tests. In: Boyling J, Pala-
stanga N (eds). Grieve’s Modern Manual Therapy: The Vertebral Column.
2nd ed. Edinburgh, Scotland: Churchill Livingstone; 1994:587–606.
9. Coppieters MW, Alshami AM, Babri AS, Souvlis T, Kippers V,
Hodges PW. Strain and excursion of the sciatic, tibial, and plantar
nerves during a modified straight leg raising test. J Orthop Res 2006;
24:1883–1889.
10. Topp KS, Boyd BS. Peripheral nerve: from the microscopic functional
unit of the axon to the biomechanically loaded macroscopic structure.
J Hand Ther 2012; 25:142–152.
11. Filius A, Scheltens M, Bosch HG, et al. Multidimensional ultrasound
imaging of the wrist: changes of shape and displacement of the
median nerve and tendons in carpal tunnel syndrome. J Orthop Res
2015; 33:1332–1340.
12. Hough AD, Moore AP, Jones MP. Reduced longitudinal excursion of
the median nerve in carpal tunnel syndrome. Arch Phys Med Rehabil
2007; 88:569–576.
11
Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
13. Ellis RF, Blyth R, Arnold N, Miner-Williams W. Is there a rela-
tionship between impaired median nerve excursion and carpal
tunnel syndrome? A systematic review. J Hand Ther 2017; 30:
3–12.
14. Efstathiou MA, Stefanakis M, Savva C, Giakas G. Effectiveness of neu-
ral mobilization inpatients with spinal radiculopathy: a critical review.
J Bodyw Mov Ther 2015; 19:205–212.
15. Kobayashi S, Shizu N, Suzuki Y, Asai T, Yoshizawa H. Changes in
nerve root motion and intraradicular blood flow during an intraopera-
tive straight-leg-raising test. Spine (Phila Pa 1976) 2003; 28:1427–
1434.
16. Liao YY, Lee WN, Lee MR, et al. Carpal tunnel syndrome: US strain
imaging for diagnosis. Radiology 2015; 275:205–214.
17. Abe Y, Doi K, Kawai S. An experimental model of peripheral nerve
adhesion in rabbits. Br J Plast Surg 2005; 58:533–540.
18. Boyd BS, Puttlitz C, Gan J, Topp KS. Strain and excursion in the rat
sciatic nerve during a modified straight leg raise are altered after trau-
matic nerve injury. J Orthop Res 2005; 23:764–770.
19. Coppieters MW, Stappaerts KH, Wouters LL, Janssens K. Aberrant
protective force generation during neural provocation testing and the
effect of treatment in patients with neurogenic cervicobrachial pain.
J Manipulative Physiol Ther 2003; 26:99–106.
20. Hall T, Zusman M, Elvey RL. Adverse mechanical tension in the
nervous system? Analysis of straight leg raise. Man Ther 1998; 3:140–
146.
21. Kerr JM, Vujnovich AL, Bradnam L. Changes in alpha-motorneuron
excitability with positions that tension neural tissue. Electromyogr Clin
Neurophysiol 2002; 42:459–472.
22. Liong K, Lahiri A, Lee S, Chia D, Biswas A, Lee HP. Predominant
patterns of median nerve displacement and deformation during indi-
vidual finger motion in early carpal tunnel syndrome. Ultrasound Med
Biol 2014; 40:1810–1818.
23. Ridehalgh C, Moore A, Hough A. Sciatic nerve excursion during a
modified passive straight leg raise test in asymptomatic participants
and participants with spinally referred leg pain. Man Ther 2015; 20:
564–569.
24. Silva A, Manso A, Andrade R, Domingues V, Brand~ao MP, Silva AG.
Quantitative in vivo longitudinal nerve excursion and strain in
response to joint movement: a systematic literature review. Clin Bio-
mech 2014; 29:839–847.
25. Boyd BS, Gray AT, Dilley A, Wanek L, Topp KS. The pattern of tibial
nerve excursion with active ankle dorsiflexion is different in older peo-
ple with diabetes mellitus. Clin Biomech 2012; 27:967–971.
26. Coppieters MW, Hough AD, Dilley A. Different nerve-gliding exer-
cises induce different magnitudes of median nerve longitudinal excur-
sion: an in vivo study using dynamic ultrasound imaging. J Orthop
Sports Phys Ther 2009; 39:164–171.
27. Kasehagen B, Ellis RF, Mawston G, Allen S, Hing WA. Assessing the
reliability of using ultrasound imaging to examine radial nerve excur-
sion. Ultrasound Med Biol 2016; 42:1651–1659.
12
28. Carroll M, Yau J, Rome K, Hing W. Measurement of tibial
nerve excursion during ankle joint dorsiflexion in a weight-
bearing position with ultrasound imaging. J Foot Ankle Res 2012;
5:5.
29. Shum GL, Attenborough AS, Marsden JF, Hough AD. Tibial
nerve excursion during lumbar spine and hip flexion measured
with diagnostic ultrasound. Ultrasound Med Biol 2013; 39:784–
790.
30. Greening J, Dilley A. Posture-induced changes in peripheral nerve
stiffness measured by ultrasound shear-wave elastography. Muscle
Nerve 2017; 55:213–222.
31. Konofagou E, Ophir J. A new elastographic method for estimation
and imaging of lateral displacements, lateral strains, corrected axial
strains and Poisson’s ratios in tissues. Ultrasound Med Biol 1998; 24:
1183–1199.
32. Ophir J, Cespedes I, Ponnekanti H, Yazdi Y, Li X. Elastography: a
quantitative method for imaging the elasticity of biological tissues.
Ultrason Imaging 1991; 13:111–134.
33. Langevin HM, Fox JR, Koptiuch C, et al. Reduced thoracolumbar fas-
cia shear strain in human chronic low back pain. BMC Musculoskelet
Disord 2011; 12:203.
34. Andrade RJ, Nordez A, Hug F, et al. Non-invasive assessment of sci-
atic nerve stiffness during human ankle motion using ultrasound shear
wave elastography. J Biomech 2016; 49:326–331.
35. Kantarci F, Ustabasioglu FE, Delil S, et al. Median nerve stiffness mea-
surement by shear wave elastography: a potential sonographic
method in the diagnosis of carpal tunnel syndrome. Eur Radiol 2014;
24:434–440.
36. Basson A, Olivier B, Ellis R, Coppieters M, Stewart A, Mudzi W.
The effectiveness of neural mobilisation for neuro-musculoskeletal
conditions: a systematic review and meta-analysis. J Orthop Sports
Phys Ther 2017; 47:593–615.
37. Ellis R, Osborne S, Whitfield J, Parmar P, Hing W. The effect of spinal
position on sciatic nerve excursion during seated neural mobilisation
exercises: an in vivo study using ultrasound imaging. J Man Manipula-
tive Ther 2017; 25:98–105.
38. Ellis RF, Hing WA, Dilley A, McNair PJ. Reliability of measuring sci-
atic and tibial nerve movement with diagnostic ultrasound during a
neural mobilisation technique. Ultrasound Med Biol 2008; 34:1209–
1216.
39. Ellis RF, Hing WA, McNair PJ. Comparison of longitudinal sciatic
nerve movement with different mobilization exercises: an in vivo
study utilizing ultrasound imaging. J Orthop Sports Phys Ther 2012;
42:667–675.
40. Kellis E, Ellinoudis A, Intziegianni K. Reliability of sonographic assess-
ment of biceps femoris distal tendon strain during passive stretching.
Ultrasound Med Biol 2017; 43:1769–1779.
41. Tosovic D, Muirhead JC, Brown JMM, Woodley SJ. Anatomy of the
long head of biceps femoris: an ultrasound study. Clin Anat 2016; 29:
738–745.
J Ultrasound Med 2018; 00:00–00
 Ellis et al—Quantification of Sciatic Nerve Displacement and Shear Strain
42. Fornage BD. Peripheral nerves of the extremities: imaging with US.
Radiology 1988; 167:179–182.
43. Peer S, Kovacs P, Harpf C, Bodner G. High-resolution sonography of
lower extremity peripheral nerves. J Ultrasound Med 2002; 21:315–
322.
44. R Core Team. R: a language and environment for statistical comput-
ing. R Foundation website; 2017. https://www.R-project.org/.
45. Hallgren KA. Computing inter-rater reliability for observational
data: an overview and tutorial. Tutor Quant Methods Psychol 2012;
8:23–24.
J Ultrasound Med 2018; 00:00–00
46. Cicchetti DV. Guidelines, criteria, and rules of thumb for evaluating
normed and standardized assessment instruments in psychology. Psy-
chol Assess 1994; 6:284–290.
47. Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team. nlme: linear
and nonlinear mixed effects models. R package version 3.1-131. R Foun-
dation website; 2017. https://CRAN.R-project.org/package=nlme.
48. Coppieters MW, Andersen LS, Johansen R, et al. Excursion of the sci-
atic nerve during nerve mobilization exercises: an in vivo cross-
sectional study using dynamic ultrasound imaging. J Orthop Sports
Phys Ther 2015; 45:731–737.
13