819
ORIGINAL ARTICLE
Changes in Passive Mechanical Properties of the
Gastrocnemius Muscle at the Muscle Fascicle and Joint Levels
in Stroke Survivors
Fan Gao, PhD, Thomas H. Grant, MD, Elliot J. Roth, MD, Li-Qun Zhang, PhD
ABSTRACT. Gao F, Grant TH, Roth EJ, Zhang L-Q.
Changes in passive mechanical properties of the gastrocnemius
muscle at the muscle fascicle and joint levels in stroke survi-
vors. Arch Phys Med Rehabil 2009;90:819-26.
Objectives: To investigate the ankle joint–level and muscle
fascicle–level changes and their correlations in stroke survivors
with spasticity, contracture, and/or muscle weakness at the
ankle.
Design: To investigate the fascicular changes of the medial
gastrocnemius muscle using ultrasonography and the biome-
chanical changes at the ankle joint across 0°, 30°, 60°, and 90°
knee flexion in a case-control manner.
Setting: Research laboratory in a rehabilitation hospital.
Participants: Stroke survivors (n⫽10) with ankle spasticity/
contracture and healthy control subjects (n⫽10).
Interventions: Not applicable.
Main Outcome Measurements: At the muscle fascicle
level, medial gastrocnemius muscle architecture including the
fascicular length, pennation angle, and thickness were evalu-
ated in vivo with the knee and ankle flexion changed system-
atically. At the joint level, the ankle range of motion (ROM)
and stiffness were determined across the range of 0° to 90°
knee flexion.
Results: At comparable joint positions, stroke survivors
showed reduced muscle fascicle length, especially in ankle
dorsiflexion (Pⱕ.048) and smaller pennation angle, especially
for more extended knee positions (Pⱕ.049) than those of
healthy control subjects. At comparable passive gastrocnemius
force, stroke survivors showed higher fascicular stiffness
(Pⱕ.044) and shorter fascicle length (Pⱕ.025) than controls.
The fascicle-level changes of decreased muscle fascicle length
and pennation angle and increased medial gastrocnemius fas-
cicle stiffness in stroke were correlated with the joint level
changes of increased joint stiffness and decreased ROM
(P⬍.05).
Conclusions: This study evaluated specific muscle fascicu-
lar changes as mechanisms underlying spasticity, contracture,
and joint-level impairments, which may help improve stroke
rehabilitation and outcome evaluation.
From the Rehabilitation Institute of Chicago, Chicago, IL (Gao, Roth, Zhang);
Departments of Physical Medicine and Rehabilitation (Gao, Roth, Zhang), Orthopae-
dic Surgery (Zhang), Biomedical Engineering (Zhang), and Radiology (Grant), North-
western University, Chicago, IL; Department of Health and Kinesiology, University
of Texas at Tyler, Tyler, TX (Gao).
Supported by the National Institutes of Health (grant nos. HD044295 and
HD043664).
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit on the authors or on any organi-
zation with which the authors are associated.
Reprint requests to Li-Qun Zhang, PhD, Rehabilitation Institute of Chicago, Ste
1406, 345 E Superior St, Chicago, IL 60611, e-mail: l-zhang@northwestern.edu.
0003-9993/09/9005-00495$36.00/0
doi:10.1016/j.apmr.2008.11.004
Key Words: Contracture; Muscle spasticity; Rehabilitation;
Stroke.
© 2009 by the American Congress of Rehabilitation
Medicine
PASTICITY, CONTRACTURE, and muscle weakness are
S commonly observed after stroke and are major sources of
disabilities poststroke. Clinically, the phenomenon of footdrop
is associated with an increase in the tone of the calf muscles.
Muscle tone may result from both reflex and nonreflex changes
and is usually accompanied with increase of passive joint/
muscle stiffness.1 In the spastic lower limb, it is not clear how
hypertonia at the ankle joint is related to changes in the
biomechanical properties of the plantar flexor muscles.2-7 A
better understanding of changes in muscle architecture and its
association with joint biomechanical properties could help us
gain insight into mechanisms underlying spasticity/contracture
and provide guidance to the rehabilitation of patients post-
stroke. The changes in the mechanical properties may be as-
sociated with changes in skeletal muscle architecture, such as
muscle fascicle length, pennation angle, and muscle thickness.
Muscle architecture plays a significant role in normal muscle
function and is closely related to the mechanical properties of
the joint.8 Recent studies based on a biomechanical model
suggested that an increase in ankle joint stiffness could be
attributed to shortened calf muscles.2 However, there has been
little experimental evidence evaluating muscle architecture and
joint stiffness in the same patients with ankle spasticity/con-
tracture.
Ultrasonography has been used in studying muscle and ten-
don function in healthy populations in vivo and noninva-
sively.9-11 However, only a few ultrasonic studies have been
conducted to examine hypertonic muscles in patients with
neurologic disorders, with mixed results reported and with a
lack of study on lower limb muscle architecture poststroke.
Shortland et al12 reported no difference in muscle fascicle
length between children with spastic diplegia and healthy chil-
dren, suggesting muscle architecture changes do not contribute
to contracture in the patients, while Cheatwood et al13 reported
significantly shorter muscle fascicle length in the group of
children with spastic cerebral palsy. Li et al14 found that the
fascicle length of spastic brachialis muscle on the affected side
was significantly shorter than that on the unimpaired side. The
differences in gastrocnemius muscle architecture were also
studied in children with cerebral palsy, and fascicle length of
spastic muscle was significantly shorter than that in nonparetic
muscle of healthy children.15-17
List of Abbreviations
ACSA anatomical cross-sectional area
ROM range of motion
Arch Phys Med Rehabil Vol 90, May 2009
820 FASCICLE- AND JOINT-LEVEL BIOMECHANICAL CHANGES IN STROKE, Gao

The purpose of the this study was to investigate, in vivo and
noninvasively, the biomechanical changes of the medial gas-
trocnemius muscle at both the joint level (characterized by the
ROM, stiffness, resistance torque at controlled position) and
the muscle fascicle level (characterized by the muscle fascicle
length, pennation angle, muscle thickness) at ankles in both
stroke survivors with spasticity/contracture and healthy control
subjects. We hypothesized that there are significant differences
in these biomechanical properties at the fascicle and joint levels
between the 2 groups, and changes at the joint level are
correlated to those at the fascicle level.
METHODS
Participant Selection
A convenience sample of 10 chronic stroke survivors (at
least 1 year poststroke; age, 54.7⫾11y; weight, 84.5⫾15.5kg;
height, 176⫾5.3cm; shank length, 39.4⫾1.0cm) with ankle
spasticity/contracture were recruited. The Modified Ashworth
Score18 was measured at the ankle (2.57⫾0.58). In addition,
the following criteria were used: the subjects were not involved
in any other studies that could potentially affect the test results,
and subjects could walk independently without walking aid and
sit on a chair for 2 hours. Ten age-matched and sex-matched
healthy subjects (age, 56.6⫾20.7y; weight, 87.1⫾17.6kg;
height, 177.3⫾3.7cm; shank length, 38.6⫾1.2cm) without any
neurologic or muscular disorders served as controls. All sub-
jects gave informed consent approved by the institutional re-
view board.
Experimental Setup
A custom knee-ankle joint test device was used to investi-
gate the biomechanic properties of the biarticular medial gas-
trocnemius muscle. The motor at the knee was fixed rigidly to
a frame anchored to the ground, and a leg linkage was mounted
to the knee motor through a 6-axis JR3 force sensor.a The ankle
motor was mounted at the distal end of the leg linkage, and a
footplate was mounted to the ankle motor through another
6-axis force sensor. The ankle motor and footplate could be
adjusted along the leg linkage so that the ankle and knee motors
were aligned with the ankle and knee flexion axes, respectively
(fig 1).
Experimental Protocol
A brief medical history including the date of stroke, ambu-
latory status, use of ankle-foot orthosis, use of antispasticity
drugs, and current therapy was documented for each stroke
survivor. The leg length (from the lateral femoral epicondyle to
the lateral malleolus19) and foot height (from the bottom of foot
to the lateral malleolus) were measured to align the knee and
ankle with the experimental device.
Subjects were seated upright with the thigh and trunk se-
cured using Velcro straps. The leg and foot were attached to the
leg linkage and footplate, respectively (see fig 1). Four knee
positions, starting from full extension with an increment of 30°
of flexion, were tested. At each knee position, the ankle flexion
angle was systematically varied between 20° dorsiflexion and
45° plantar flexion, with increments of 10° in dorsiflexion and
15° in plantar flexion relative to 0° of ankle flexion. The knee
and ankle motors were locked at each of the target positions. At
each knee position, the subject was asked to relax with the
ankle at the resting position. The corresponding ankle resting
angle and torque were recorded. In addition, the resistance
torque at 0° dorsiflexion was measured. At each of the knee and
ankle positions, the subject was asked to relax, and the knee
and ankle torques and angles were recorded for 2 seconds.

Fig 1. (A) Experimental setup. The knee-ankle evaluation device consists of 2 motors and a linkage between. The JR3 force/torque sensors
were mounted on the motor shaft at both joints to measure the joint torques/forces. With the knee flexion axis aligned with the knee motor,
the ankle motor can be adjusted along the leg linkage to align it with the ankle flexion axis. (B) Longitudinal ultrasonic images of the medial
gastrocnemius muscle at rest. The skin is on the top of the image, and the left side corresponds to proximal. The muscle tendon junction
represented the musculo-tendon (muscle aponeurosis) junction. ␣ and ␤ are the posterior and anterior pennation angles, respectively. The
medial gastrocnemius muscle tendon junction was taken as the distal reference point.

Arch Phys Med Rehabil Vol 90, May 2009

 FASCICLE- AND JOINT-LEVEL BIOMECHANICAL CHANGES IN STROKE, Gao
Because some stroke survivors had reduced ROM, the exper-
iment was conducted within the comfort limits of each subject.
Ultrasonic images of the medial gastrocnemius muscle were
collected using a 14-MHz high-resolution matrix probe.b
LOGIQView, a technique of extended field of view, was used
to overcome the limited field of view and register the muscle
images covering the full fascicle lengths. A previous study has
shown the reliability of ultrasound technique for measuring
muscle architecture,20 and the accuracy of extended field of
view has been reported to be better than 5%.21 The probe was
placed perpendicular to the skin and moved smoothly along the
middle line of the medial gastrocnemius throughout its length
(see fig 1). All scans were conducted by an experimenter with
experience in ultrasonic measurements. The scan was repeated
3 times, and the averaged values of these measurements across
scans were used in further analysis.
Data Analysis
The torque signal at each joint was low-pass–filtered with
fourth-order Butterworth filter (5Hz cutoff frequency) and av-
eraged across the 2-second period. Ankle joint stiffness was
calculated as the change of ankle joint resistance torque over
the change of ankle joint angle (K⫽⌬␶/⌬␪) across the ROM. In
this study, the ankle joint stiffness was quantified between
different ankle positions reaching steady state instead of during
continuous dynamic movement. The muscle fascicle length,22
as indicated by the line between the aponeuroses, was mea-
sured at 5cm proximal to the muscle-tendon junction (see fig
1B). In addition to the absolute fascicle length, its normaliza-
tion to the lower-leg length was also calculated for the indi-
vidual subjects. The pennation angle was defined as the angle
between the fascicle and the aponeurosis both posteriorly and
anteriorly (see fig 1B). For simplicity, we focused on the
anterior pennation angle. Muscle thickness was measured at
5cm proximal from the muscle-tendon junction (see fig 1B).
Measurements including joint resistance torque, joint stiffness,
muscle fascicle length, and pennation angle were interpolated
across the ankle flexion using the shape-preserving piecewise
cubic method (by using interp1 function in MATLAB with
option pchipc).
The gastrocnemius contribution to the ankle passive resis-
tance torque was estimated as the difference of the passive
ankle resistance torque between full knee extension and 90°
knee flexion. The calculation was done in the range of 20°
plantar flexion to 15° dorsiflexion. With the gastrocnemius
moment arm for ankle plantar flexion obtained from SIMM,d
the passive resistance force of the gastrocnemius was deter-
mined and related to the measured medial gastrocnemius fas-
cicle length. The moment arm was not normalized for individ-
ual subjects because there were no significant differences in
body height for either stroke survivors (P⫽.15, t test) or
healthy controls (P⫽.74, t test) compared with the body height
used in SIMM (175cm, height of an average man). Further-
more, assuming the passive force sharing between the lateral
and medial heads of gastrocnemius to be proportional to the
ACSA with the medial and lateral gastrocnemius ACSAs at
10.20 and 6.53cm2, respectively,23 the passive medial gastroc-
nemius force was determined as 61% of the total passive
gastrocnemius force. It was assumed that the passive tension of
the gastrocnemius muscle was negligible at 90° knee flexion
because it had been reported that the passive tension was close
to 0 with 10° plantar flexion and greater than 50° knee flex-
ion.24 The corresponding medial gastrocnemius fascicular
force was scaled by 1/cos(␪pennation), with ␪pennation the penna-
tion of medial gastrocnemius fascicles. The medial gastrocne-
821
mius fascicular stiffness was determined as the slope of the
medial gastrocnemius fascicular force and fascicular length
relationship.
Statistical Analysis
Repeated-measures analysis of variance was used to analyze
the response variables (fascicle length, pennation angles, mus-
cle thickness, joint stiffness, ROM) with respect to each factor
(subject population, ankle position, knee position). The signif-
icance level was set at .05, and adjustments were made if a
violation of sphericity was found (Huynh-Feldt adjustment if
the sphericity estimate ⬎0.75, Greenhouse-Geisser otherwise).
The Student t test was used for comparison of the variables
between groups. The Pearson correlation coefficient was used
to quantify associations among the variables, and the Spearman
rank correlation was used to test the monotonic trend (eg, if one
variable is increased, the other follows). Correlations between
the muscle architecture measures (medial gastrocnemius fas-
cicular stiffness, fascicle length, pennation) and the joint-level
variables (passive resistance torque, joint stiffness, ankle dor-
siflexion ROM) were evaluated.
RESULTS
Biomechanic Changes at the Joint Level
Decreased passive ankle range of motion in stroke survi-
vors. Under comparable joint torques (5Nm in dorsiflexion
and 3Nm in plantar flexion), stroke survivors showed reduced
ROM compared with healthy controls. With knee flexion of
30°, the ankle positions ranged from –34.1⫾4.8° to –3.1⫾4.1°
and from –37.7⫾4.8° to 13.5⫾6.9° for stroke survivors and
healthy control groups, respectively. The ROM difference be-
came significant as the ankle was dorsiflexed. The ankle posi-
tions ranged from –31.2⫾7.9° to – 6.4⫾5.1° and from
–35.3⫾6.5° to 6.5⫾5.3° with the knee fully extended for
stroke survivors and healthy control groups, respectively.
Across the different knee flexion angles, patients poststroke
showed a larger ankle resting angle (more into plantar
flexion) than the healthy controls (P⫽.006). At full knee
extension, the ankle resting positions of the stroke and healthy
groups were –18.2⫾6.6° and –13.6⫾3.9°, respectively. With
the knee flexed from 30° to 90°, the resting positions changed
from 19.5⫾5.2° to 17.5⫾4.6° and from 12.7⫾3.5° to
13.1⫾5.2° for the stroke and control groups, respectively.
Although the medial gastrocnemius spans both knee and ankle,
the ankle resting position was not significantly affected by the
knee position (P⫽.841).
Increased ankle stiffness in stroke survivors. Ankle stiff-
ness changed with both ankle (P⬍.001) and knee flexion
(P⬍.001). Stroke survivors exhibited higher ankle stiffness
than healthy controls, especially in dorsiflexion (fig 2; P⬍.05).
At 0° dorsiflexion and full knee extension, for example, the
stiffness was .24⫾.08 and .72⫾.28Nm/° for the control and
stroke groups, respectively. Ankle stiffness ranged from .076 to
.75Nm/° and from .17 to 1.46Nm/° for the control and stroke
groups, respectively. Ankle stiffness was highest at extreme
dorsiflexion and lowest around the ankle resting position (see
fig 2).
Biomechanic Changes at the Muscle Fascicle Level
Muscle fascicle length. As a biarticular muscle, the medial
gastrocnemius muscle fascicle length varied with both ankle
(P⫽.001) and knee flexion (fig 3; P⫽.001, both groups com-
bined), and a significant interaction between the knee and ankle
flexions was also observed (Pⱕ.001). The fascicle length in-
Arch Phys Med Rehabil Vol 90, May 2009
822 FASCICLE- AND JOINT-LEVEL BIOMECHANICAL CHANGES IN STROKE, Gao
creased monotonically as the ankle dorsiflexed but decreased as
the knee flexed (Pⱕ.001; see fig 3). For instance, for stroke
survivors, the muscle fascicle length increased from 32.0⫾
8.1mm to 58.8⫾10.7mm as the ankle moved from 45° plantar
flexion to 15° dorsiflexion with the knee at full extension, while
it decreased from 51.4⫾8.4mm to 34.2⫾7.0mm as the knee
flexed from full extension to 90° flexion with the ankle at the
neutral position.
Although the 2 groups showed similar trends, stroke survi-
vors had significantly shorter muscle fascicles compared with
healthy subjects across the ankle ROM, especially with the
ankle dorsiflexed (see fig 3; P⬍.05). Medial gastrocnemius
fascicle length depended on both knee and ankle positions, and
significant interaction between the 2 joints was observed
(P⬍.05). With the knee fully extended, for example, the fas-
cicle length of stroke survivors was shorter than that of healthy
controls across the range of 45° plantar flexion to 15° dorsi-
flexion. However, the difference diminished as the knee was
flexed. Similar results were observed for normalized muscle
fascicle lengths. Stroke survivors showed significantly shorter
normalized muscle fascicle length than healthy controls, espe-
cially with the ankle dorsiflexed (P⬍.05).
Pennation angle. The pennation angle decreased mono-
tonically as the ankle moved from plantar to dorsiflexion and
increased as the knee flexed from full extension to 90° flexion
(P⫽.001). For instance, for control subjects, the pennation
Arch Phys Med Rehabil Vol 90, May 2009
Fig 2. Passive stiffness at the
ankle joint for both healthy
subjects (dark line) and pa-
tients poststroke (gray line) at
4 knee positions: (A) full knee
extension, (B) 30° knee flex-
ion, (C) 60° knee flexion, and
(D) 90° knee flexion. *Signifi-
cant differences between the 2
populations with P<.05 (t test).
Abbreviations: DF, dorsiflexion;
PF, plantarflexion.
angle increased from 19.5⫾2.6° to 27.4⫾5.1° as the knee
flexed from 0° to 90° flexion with the ankle at 0° dorsiflexion
(fig 4). The pennation angle ranged from 17.1⫾2.4° to
41.4⫾7.8° and from 14.2⫾4.0° to 37.1⫾6.4° for the healthy
and stroke groups, respectively.
Stroke survivors showed smaller pennation angles than
healthy controls, especially at more extended knee positions
(Pⱕ.049; fig 5). For instance, the pennation angles with the
ankle at 0° dorsiflexion and at 30° of knee flexion were
22.1⫾3.1° and 17.5⫾3.9° for the controls and stroke survivors,
respectively. The differences decreased as the ankle ap-
proached extreme dorsiflexion.
Muscle thickness. With the knee extended and/or the ankle
dorsiflexed, the medial gastrocnemius became tighter, and
muscle thickness decreased. As the knee was flexed, the change
in muscle thickness with ankle dorsiflexion also increased. For
instance, for healthy control subjects, as the ankle moved from 15°
dorsiflexion to 45° plantar flexion, the muscle thicknesses in-
creased from 10.0⫾3.7 to 15.0⫾3.1mm, 9.8⫾4.0 to 14.7⫾
2.8mm, 10.7⫾3.2 to 16.2⫾2.5mm, and 10.9⫾2.6 to 17.6⫾
3.5mm at 0°, 30°, 60°, and 90° knee flexion, respectively.
The muscle thickness for stroke survivors was slightly
smaller than that of healthy controls. However, the differ-
ences were significant only at extreme ankle dorsiflexion
and knee extension.
 FASCICLE- AND JOINT-LEVEL BIOMECHANICAL CHANGES IN STROKE, Gao
Fig 3. Muscle fiber lengths of
both healthy subjects (dark
line) and stroke survivors
(gray line) at 4 knee positions:
(A) full knee extension, (B) 30°
knee flexion, (C) 60° knee flex-
ion, and (D) 90° knee flexion.
*
Significant differences be-
tween the 2 populations with
P<.05 (t test). Abbreviations:
DF, dorsiflexion; PF, plantar-
flexion.
Passive force and fascicle length relationship. At compa-
rable levels of medial gastrocnemius fascicular force, ranging
from 20 to 80N, the medial gastrocnemius fascicular stiffness
(slope of the medial gastrocnemius fascicular force and medial
gastrocnemius fascicular length relationship) was significantly
higher for stroke survivors than controls (Pⱕ.044; see fig 4).
For example, at 50N medial gastrocnemius fascicular force, the
slopes were 12.4⫾4.7 and 6.5⫾3.6N/mm for the stroke and
control groups, respectively. Similarly, the gastrocnemius force
and medial gastrocnemius fascicle length curve of stroke sur-
vivors had significantly steeper slope than that of controls at
comparable levels of gastrocnemius force (Pⱕ.044). In addi-
tion, at comparable gastrocnemius passive tension levels rang-
ing from 30 to 130N, stroke survivors had significantly shorter
medial gastrocnemius fascicle length than controls (Pⱕ.025).
For instance, the medial gastrocnemius muscle fascicle lengths
were 48.1⫾13.4 and 65⫾15mm for the stroke and control
groups, respectively, at 80N gastrocnemius passive tension
(Pⱕ.001; see fig 4).
Relationship Between Changes at the Joint and Fascicle
Levels
The joint and fascicle level variables were correlated. The
ankle ROM showed a moderate negative correlation with
the muscle fascicle length (r⫽–.286; P⬍.01), pennation
823
angle (r⫽.304; P⬍.01) and medial gastrocnemius muscle fas-
cicle stiffness (r⫽–.151; P⬍.05) at full knee extension. Joint
stiffness showed a negative correlation with muscle fascicle
length (r⫽–.20; P⬍.01) and a positive correlation with medial
gastrocnemius fascicular stiffness (r⫽.51; Pⱕ.001). Signifi-
cant correlations were observed between the variables evalu-
ated within the fascicle or joint level. At the fascicle level,
muscle fascicle length was correlated with pennation angle
(r⫽–.46; Pⱕ.001). At the joint level, the resistance torque and
joint stiffness were correlated (r⫽.66; Pⱕ.001), and the 2
variables were both correlated with the ankle ROM negatively
(r⫽–.69, Pⱕ.001; and r⫽–.38, Pⱕ.001, respectively).
DISCUSSION
Biomechanic changes at the ankle joint are associated with
spasticity/contracture in stroke survivors, including reduced
ROM and increased resistance and stiffness at the ankle. How-
ever, the corresponding changes at the muscle fascicle level
and the correlations between changes at the 2 levels have not
been investigated together in the stroke population. In this
study, we use ultrasonography combined with biomechanical
measurements to evaluate in vivo muscle fascicle as well as
joint properties and correlate the pathologic changes at the joint
level with the underlying changes at the muscle fascicle level.
Arch Phys Med Rehabil Vol 90, May 2009
824 FASCICLE- AND JOINT-LEVEL BIOMECHANICAL CHANGES IN STROKE, Gao
Fig 4. The relationship between the medial gastrocnemius fascicle
length and the passive gastrocnemius force (thick line) and the
medial gastrocnemius fascicle force (gastrocnemius force ⴛ 0.610 /
cos(␪pennation); thin line) for the stroke and control groups (the group
average with 1-sided SE was shown for both fascicle length and
muscle/fascicle force). The medial gastrocnemius fascicle length
was determined at full knee extension with ankle between 20°
plantar flexion and 15° dorsiflexion.
The current study examined biomechanic changes at the
joint level with more complete knee-ankle positions than pre-
vious studies on healthy subjects.2,4,7 In the current study, 4
knee positions ranging from full extension to 90° flexion were
evaluated, while previous studies usually involved only 1 knee
position,4,7 and 1 study measured changes in gastrocnemius
fascicle length as a function of knee and ankle position without
reporting the absolute fascicle length.25 Our results agreed well
with previously reported results at comparable leg posi-
tions.2,4,7 For instance, the ankle joint stiffness measured at
10° dorsiflexion with 90° knee flexion for stroke survivors
was 0.6⫾0.4Nm/° in the current study and 0.5⫾0.4Nm/°,7
0.44Nm/°,2 and 0.5⫾0.4Nm/°4 in previous studies.
In this study, muscle architecture was compared between
stroke survivors and healthy subjects quantitatively. In general,
measurements of muscle architecture in healthy subjects agreed
well with previously published data.10,20,25-28 In this study, the
medial gastrocnemius fascicle length of healthy control
changed from 28 to 78mm across the ankle and knee positions.
The results on medial gastrocnemius fascicle length normalized
to the lower-leg length also showed similar results. Arampatzis
et al26 reported a range from 37 to 67mm across 6 ankle and
knee angle combinations with the ankle position between 20°
plantar flexion and 10° dorsiflexion. The larger range of fasci-
cle length in our study may arise from the larger range of ankle
and knee positions (ankle position between 45° plantar flexion
and 15° dorsiflexion). Because of the lack of published data,
direct comparisons could not be made for the results on stroke
survivors.
Muscle force is transmitted to the tendon at a pennation
angle with muscle force scaled according to the cosine of the
pennation angle.9,23,29 It has been reported that hypertrophy of
muscles involves an increase in pennation angle.9 In contrast,
Arch Phys Med Rehabil Vol 90, May 2009
the smaller pennation angles in stroke survivors observed in
this study indicated muscle atrophy poststroke. Furthermore,
considering the muscle fascicle length may be reduced post-
stroke, the decreased pennation angle suggests higher fascicu-
lar tension under the passive condition, which may be associ-
ated with the increased tone of the calf muscles in stroke.
Functional impairments at the joint level are closely related
to the changes in muscle architecture properties. The higher
medial gastrocnemius fascicle stiffness, joint stiffness in dor-
siflexion, and reduced ROM at controlled resistance torques in
stroke survivors were associated with shortening of muscle
fascicles as demonstrated by the significant correlations be-
tween the 2 sets of variables. It has been shown that the number
of sarcomeres in series in a muscle is highly adaptable to
changes in muscle length.30-32 Patients poststroke often de-
velop footdrop, which may be a result of shortened plantar
flexors fascicles. As shown in figure 4, stroke survivors had
significantly higher gastrocnemius muscle and medial gastroc-
nemius fascicular stiffnesses, indicating an underlying mecha-
nism for the increased joint stiffness. This was consistent with
an in vitro study by Friden and Lieber,33 who showed that the
spastic muscle cell is stiffer and shorter than the normal con-
trol. Similarly, through modeling analysis, Svantesson et al34
showed that muscle stiffness was significantly higher in the
affected leg than the nonaffected leg of stroke survivors. In
addition, in this study, a significant correlation (r⫽.855;
P⬍.001) was observed between the Modified Ashworth Score
and ankle joint stiffness.
Study Limitations
There were limitations with the present study. First, the
fascicle length and pennation were determined 5cm from the
muscle tendon junction. Although it has been shown that there
is marked uniformity in fascicle length throughout a mus-
cle,35,36 some studies reported heterogeneity of pennation an-
gles37 and fascicle lengths38 along the length of muscle. Thus,
the relationships between joint angles and fascicle arrangement
might differ in different portions of a muscle. In this study, we
chose the location of 5cm from the muscle tendon junction to
obtain a representative measurement and kept it consistent
across the subjects. Second, the measured medial gastrocne-
mius pennation angle was assumed to be the same for the
lateral gastrocnemius. However, the lateral head may have
smaller pennation.10 To distribute the tension between the
lateral and medial heads, we assumed that the tension was
proportional to the ACSA, which represented an approximation
for the 2 populations. It would be interesting to see whether the
ACSA and passive tension ratio of medial head to lateral head
is altered because of stroke. Last, the sample size in the current
study was relatively small, although significant differences in
biomechanical properties were still observed.
CONCLUSIONS
The present study found that compared with healthy sub-
jects, stroke survivors showed simultaneous changes at the
fascicle and joint levels, including decreased length and re-
duced pennation angle of the muscle fascicles, and decreased
ROM and increased stiffness in dorsiflexion at the joint. Bio-
mechanic changes at the fascicle and joint levels were corre-
lated to each other, and the changes at the joint may originate
from the muscle fascicles. Clinical symptoms of spasticity/
contracture may be closely related to changes in muscle archi-
tecture, including shortening of muscle fascicles and reduction
of pennation angle. In vivo evaluation of both muscle and joint
 FASCICLE- AND JOINT-LEVEL BIOMECHANICAL CHANGES IN STROKE, Gao
Fig 5. Anterior pennation an-
gles of both healthy subjects
(dark line) and stroke survi-
vors (gray line) at 4 knee posi-
tions: (A) full knee extension,
(B) 30° knee flexion, (C) 60°
knee flexion, and (D) 90° knee
flexion. * Significant differ-
ences between the 2 popula-
tions. P<.05 (t test). Abbrevia-
tions: DF, dorsiflexion; PF,
plantarflexion.
properties may serve as a quantitative tool in stroke rehabili-
tation and outcome evaluation.
References
1. Singer B, Dunne J, Allison G. Reflex and non-reflex elements of
hypertonia in triceps surae muscles following acquired brain
injury: implications for rehabilitation. Disabil Rehabil 2001;23:
749-57.
2. Harlaar J, Becher JG, Snijders CJ, Lankhorst GJ. Passive stiffness
characteristics of ankle plantar flexors in hemiplegia. Clin Bio-
mech (Bristol, Avon) 2000;15:261-70.
3. Katz RT, Rymer WZ. Spastic hypertonia: mechanisms and mea-
surement. Arch Phys Med Rehabil 1989;70:144-55.
4. Singer B, Dunne J, Singer KP, Allison G. Evaluation of triceps
surae muscle length and resistance to passive lengthening in
patients with acquired brain injury. Clin Biomech (Bristol, Avon)
2002;17:152-61.
5. Vattanasilp W, Ada L, Crosbie J. Contribution of thixotropy,
spasticity, and contracture to ankle stiffness after stroke. J Neurol
Neurosurg Psychiatry 2000;69:34-9.
6. Zhang LQ, Wang G, Nishida T, Xu D, Sliwa JA, Rymer WZ.
Hyperactive tendon reflexes in spastic multiple sclerosis: measures
and mechanisms of action. Arch Phys Med Rehabil 2000;81:901-9.
7. Chung SG, van Rey EM, Bai Z, Roth EJ, Zhang L-Q. Biome-
chanic changes in passive properties of hemiplegic ankles with
spastic hypertonia. Arch Phys Med Rehabil 2004;85:1638-46.
825
8. Lieber RL, Bodine-Fowler SC. Skeletal muscle mechanics: impli-
cations for rehabilitation. Phys Ther 1993;73:844-56.
9. Kawakami Y, Abe T, Fukunaga T. Muscle-fiber pennation angles
are greater in hypertrophied than in normal muscles. J Appl
Physiol 1993;74:2740-4.
10. Maganaris CN, Baltzopoulos V, Sargeant AJ. In vivo mea-
surements of the triceps surae complex architecture in man:
implications for muscle function. J Physiol (Lond) 1998;512:
603-14.
11. Magnusson SP, Aagaard P, Dyhre-Poulsen P, Kjaer M. Load-
displacement properties of the human triceps surae aponeurosis in
vivo. J Physiol 2001;531:277-88.
12. Shortland AP, Harris CA, Gough M, Robinson RO. Architecture
of the medial gastrocnemius in children with spastic diplegia. Dev
Med Child Neurol 2001;43:796-801.
13. Cheatwood AP, Rethlefsen SA, Kay RM, Wren TA. Changes in
medial gastrocnemius architecture with spasticity and contracture.
In: Proceedings of the 52nd Annual Meeting of the Orthopaedic
Research Society; Chicago; 2006.
14. Li L, Tong KY, Hu X. The effect of poststroke impairments on
brachialis muscle architecture as measured by ultrasound. Arch
Phys Med Rehabil 2007;88:243-50.
15. Mohagheghi AA, Khan T, Meadows TH, Giannikas K, Baltzo-
poulos V, Maganaris CN. Differences in gastrocnemius muscle
architecture between the paretic and non-paretic legs in children
Arch Phys Med Rehabil Vol 90, May 2009
826 FASCICLE- AND JOINT-LEVEL BIOMECHANICAL CHANGES IN STROKE, Gao
with hemiplegic cerebral palsy. Clin Biomech (Bristol, Avon)
2007;22:718-24.
16. Mohagheghi AA, Khan T, Meadows TH, Giannikas K, Baltzo-
poulos V, Maganaris CN. In vivo gastrocnemius muscle fascicle
length in children with and without diplegic cerebral palsy. Dev
Med Child Neurol 2008;50:44-50.
17. Fukashiro S, Hay DC, Yoshioka S, Nagano A. Simulation of
muscle-tendon complex during human movements. Int J Sport
Health Sci 2005;3:152-60.
18. Bohannon RW, Smith MB. Interrater reliability of a modified
Ashworth scale of muscle spasticity. Phys Therapy 1987;67:
206-7.
19. Winter DA. Biomechanics and motor control of human move-
ment. 3rd ed. New York: John Wiley & Sons; 2004.
20. Narici MV, Binzoni T, Hiltbrand E, Fasel J, Terrier F, Cerretelli
P. In vivo human gastrocnemius architecture with changing joint
angle at rest and during graded isometric contraction. J Physiol
1996;496:287-97.
21. Weng L, Tirumalai AP, Lowery CM, et al. US extended-field-of-
view imaging technology. Radiology 1997;203:877-80.
22. Kawakami Y, Kumagai K, Huijing PA, Hijakata T, Fukunaga T.
The length-force characteristics of human gastrocnemius and so-
leus muscles in vivo. In: Herzog W, editor. Skeletal muscle
mechanics: from mechanisms to function. Chichester: John Wiley
& Sons; 2000. p 327-41.
23. Fukunaga T, Roy RR, Shellock FG, et al. Physiological cross-
sectional area of human leg muscles based on magnetic resonance
imaging. J Orthop Res 1992;10:928-34.
24. Muraoka T, Chino K, Muramatsu T, Fukunaga T, Kanehisa H. In
vivo passive mechanical properties of the human gastrocnemius
muscle belly. J Biomech 2005;38:1213-9.
25. Herbert RD, Moseley AM, Butler JE, Gandevia SC. Change in
length of relaxed muscle fascicles and tendons with knee and
ankle movement in humans. J Physiol 2002;539:637-45.
26. Arampatzis A, Karamanidis K, Stafilidis S, Morey-Klapsing G,
DeMonte G, Bruggemann GP. Effect of different ankle- and
knee-joint positions on gastrocnemius medialis fascicle length and
EMG activity during isometric plantar flexion. J Biomech 2006;
39:1891-902.
Arch Phys Med Rehabil Vol 90, May 2009
27. Karamanidis K, Arampatzis A. Mechanical and morphological
properties of different muscle-tendon units in the lower extremity
and running mechanics: effect of aging and physical activity. J
Exp Biol 2005;208:3907-23.
28. Maganaris CN. Force-length characteristics of the in vivo human
gastrocnemius muscle. Clin Anat 2003;16:215-23.
29. Gans C, de Vree F. Functional bases of fiber length and angulation
in muscle. J Morphol 1987;192:63-85.
30. Herring SW, Grimm AF, Grimm BR. Regulation of sarcomere
number in skeletal muscle: a comparison of hypotheses. Muscle
Nerve 1984;7:161-73.
31. Tabary JC, Tardieu C, Tardieu G, Tabary C, Gagnard L. Func-
tional adaptation of sarcomere number of normal cat muscle.
J Physiol (Paris) 1976;72:277-91.
32. Williams PE, Goldspink G. Changes in sarcomere length and
physiological properties in immobilized muscle. J Anat 1978;127:
459-68.
33. Friden J, Lieber RL. Spastic muscle cells are shorter and stiffer
than normal cells. Muscle Nerve 2003;27:157-64.
34. Svantesson U, Takahashi H, Carlsson U, Danielsson A, Sunner-
hagen KS. Muscle and tendon stiffness in patients with upper
motor neuron lesion following a stroke. Eur J Appl Physiol
2000;82:275-9.
35. Friederich JA, Brand RA. Muscle fiber architecture in the human
lower limb. J Biomech 1990;23:91-5.
36. Wickiewicz TL, Roy RR, Powell PL, Edgerton VR. Muscle ar-
chitecture of the human lower limb. Clin Orthop Relat Res 1983;
Oct(179):275-83.
37. Scott SH, Brown IE, Loeb GE. Mechanics of feline soleus, I:
effect of fascicle length and velocity on force output. J Muscle Res
Cell Motil 1996;17:207-19.
38. Huijing PA. Architecture of the human gastrocnemius muscle and
some functional consequences. Acta Anat (Basel) 1985;123:
101-7.
Suppliers
a. JR3 Inc, 22 Harter Ave, Woodland, CA 95776.
b. GE LOGIQ-9, PO Box 414, Milwaukee, WI 53201.
c. MathWorks Inc, 3 Apple Hill Dr, Natick, MA 01730.
d. MusculoGraphics Inc, 3617 Westwind Blvd, Santa Rosa, CA
95403.