Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 May 1.
Published in final edited form as: Clin Biomech (Bristol). 2020 Apr 19;75:105007. doi: 10.1016/j.clinbiomech.2020.105007

Limited fascicle shortening and fascicle rotation may be associated with impaired voluntary force-generating capacity in pennate muscles of chronic stroke survivors

Jongsang Son 1,2, William Z Rymer 1,2, Sabrina S M Lee 3
PMCID: PMC7234905  NIHMSID: NIHMS1587803  PMID: 32339945

Abstract

Background:

Muscle weakness is one of the most common motor impairments after stroke. A variety of progressive muscular changes are reported in chronic stroke survivors, and it is now feasible to consider these changes as an added source of weakness. However, the net contributions of such muscular changes towards muscle weakness have not been fully quantified.

Methods:

Accordingly, this study aims: (1) to compare muscle architecture of the human medial gastrocnemius between paretic and non-paretic sides in seven chronic hemispheric stroke survivors under passive conditions; (2) to characterize fascicle behavior (i.e., fascicle shortening and fascicle rotation) of the muscle during voluntary isometric contractions; and (3) to assess potential associations between muscle architectural parameters and muscle weakness. Muscle architecture of the medial gastrocnemius (including fascicle length, fascicle pennation angle, and muscle thickness) was characterized using B-mode ultrasonography, and fascicle behavior was then quantified as a function of isometric plantarflexion torque normalized to body mass.

Findings:

Our experimental results showed that under passive conditions, there was a significant difference in fascicle length and muscle thickness between paretic and non-paretic muscles, but no difference in resting fascicle pennation angle. However, during isometric contraction, both fascicle shortening and fascicle rotation on the paretic side were significantly decreased, compared to the non-paretic side. Moreover, the relative (i.e., paretic / non-paretic) fascicle rotation-shortening ratio (i.e., fascicle rotation per fascicle shortening) was strongly correlated with the relative maximum voluntary isometric plantarflexion torque.

Interpretation:

This association implies that such fascicle changes could impair the force-generating capacity of the muscle in chronic stroke survivors.

Keywords: Fascicle shortening, Fascicle rotation, Stroke, Muscle weakness, Muscle mechanics

Introduction

Muscle weakness is one of the most common motor impairments in chronic stroke, and it routinely affects activities of daily living in an adverse manner. There remains a significant need for better understanding of the potential mechanisms inducing muscle weakness, in order to help us develop or refine novel interventions that might help to maximize functional recovery.

Although neural changes after stroke seem to be the most important factor contributing to muscle weakness, such altered neural commands are also associated with progressive muscular changes such as a reduction in both fiber size and overall muscle mass (Gray et al., 2012). Reduction in muscle size may lead to additional deficit of force-generating capacity, but investigation into the association between muscle weakness and muscular factors such as muscle architecture remains limited.

Earlier studies using ultrasonography have shown differences in muscle architecture in stroke-impaired muscle compared to those of individuals without neurological impairments. These include shorter fascicle lengths in brachialis (Li et al., 2007), soleus (Zhao et al., 2015), and medial gastrocnemius (MG) (Gao and Zhang, 2008; Gao et al., 2009; Zhao et al., 2015; Dias et al., 2017). There are also consistent trends towards smaller pennation angles in plantarflexors after stroke (Gao et al., 2009; Ramsay et al., 2014; Dias et al., 2017) and towards decreased overall muscle thickness in plantarflexors (Gao et al., 2009; Zhao et al., 2015; Dias et al., 2017). One important qualification is that stroke-impaired muscle architecture values reported in the majority of the previous studies were recorded solely under passive conditions. However, recently, Dias and colleagues reported on muscle architectural parameters in the MG muscle of stroke survivors at rest and during maximum voluntary contraction (MVC) and found decreased fascicle excursion between rest and MVC on the paretic side as compared to non-paretic side (Dias et al., 2017). These observations could potentially account for some aspects of muscle weakness after stroke in terms of muscle mechanics such as force-length-velocity relations (Gray et al., 2012). However, considering that activities of daily living are typically done at submaximal levels (Scaglioni et al., 2016), it is also important to understand how muscle fascicles behave during submaximal voluntary isometric muscle contraction.

Accordingly, the purpose of this study is to characterize muscle fascicle behavior during submaximal voluntary isometric muscle contractions of the MG muscles in chronic stroke survivors, using ultrasonography. As a secondary goal of this study, we propose to explore the associations between fascicle behavior and torque-generating capacity, in order to determine whether altered fascicle behavior is associated with reduced torque-generating capacity in chronic stroke survivors. Our hypothesis is that fascicle behavior is different between the paretic and non-paretic sides and that the altered fascicle behavior is associated with a reduction in maximum torque-generating capacity.

Methods

Participants

Seven chronic stroke survivors participated in this study (Table 1). All participants did not have any equinus deformity or injuries to their lower limbs within the previous 12 months. They were independently ambulatory, and were not currently receiving physical therapy. They had no history of botulinum toxin treatments for at least twelve months before testing. Written informed consent was obtained from all participants prior to testing and Northwestern University’s Institutional Review Board approved all procedures.

Table 1.

Subject characteristics (P: paretic; NP: non-paretic)

Subject ID Sex Paretic side Age (years) Time post-stroke (years) Height (m) Body mass (kg) Fugl-Meyera,b Maximum knee extension (deg)b Knee joint range of motion (deg)b Maximum dorsiflexion (deg)b Ankle joint range of motion (deg)b
Passive (P / NP) Active (P / NP) Passive (P / NP) Active (P / NP) Passive (P / NP) Active (P / NP) Passive (P / NP) Active (P / NP)
1 M L 68 5.9 1.73 90.7 28 16 / 5 35 / 10 99 / 120 50 / 95 0 / 10 2 / 8 47 / 55 37 / 48
2 F R 67 11.8 1.60 61.0 19 14 / 8 15 / 17 47 / 52 10 / 57 9 / 13 −15 / 5 46 / 56 25 / 50
3 M R 60 11.7 1.78 93.9 26 8 / 0 16 / 0 58 / 104 42 / 87 12 / 17 −3 / 9 41 / 55 22 / 33
4 M R 59 4.6 1.73 82.6 21 5 / 2 17 / 14 112 / 133 62 / 91 9 / 10 1 / 9 59 / 60 46 / 59
5 M L 46 11.6 1.78 64.9 25 9 / 5 21 / 17 116 / 122 61 / 83 5 / 11 −13 / 3 47 / 51 24 / 45
6 F R 54 5.2 1.63 69.2 - 13 / 3 18 / 6 107 / 122 32 / 80 7 / 12 2 / 6 44 / 64 35 / 52
7 M L 45 4.9 1.70 97.5 - - - - - - - - -
Mean (SD) 2/5 (F/M) 3/4 (L/R) 57.0 (9.2) 8.0 (3.5) 1.71 (0.07) 80.0 (14.9) 23.8 (3.7) 10.8 (4.2)* / 3.8 (2.8) 20.3 (7.5) / 10.7 (6.7) 89.8 (29.7)* / 108.8 (29.4) 42.8 (19.7)* / 82.2 (13.5) 7.0 (4.1)* / 12.2 (2.6) −4.3 (7.7)* / 6.7 (2.4) 47.3 (6.2)* / 56.8 (4.5) 31.5 (9.4)* / 47.8 (8.7)
Open in a new tab
a

Score assessed for lower extremity (Max score: 34)

b

Data not provided were excluded.

*

Significant difference between paretic (P) and non-paretic (NP) sides (p < 0.05)

Experimental Setup

Participants were seated upright in a fully-adjustable chair with the trunk and thigh firmly strapped to the chair. The foot was secured to the footplate of the dynamometer (Biodex Systems 3 Pro, Biodex Medical System, Inc., Shirley, NY, USA) while the knee joint was extended as much as possible to a comfortable position, and the ankle joint was fixed at neutral (0° defined as perpendicular between the shank and the foot). A slight adjustment in the subject’s posture was then allowed to make sure the subject was comfortable (in most cases, the knee was flexed ~10°), and the knee and ankle joint configurations were similar between paretic and non-paretic sides.

B-mode ultrasound images were captured using an Aixplorer ultrasound system (Supersonic Imagine, Aix-en-Provence, France) with a linear transducer array (4–15 MHz, SuperLinear SL15–4, Vermon, Tours, France). The optimal location to observe muscle architecture of the MG was first determined in order to avoid interference between the ultrasound probe and the electromyogram (EMG) electrode. The ultrasound transducer was positioned at the mid-belly region of the MG muscle, oriented parallel to the fascicle plane, and secured to the shank using a custom neoprene sleeve, in order to minimize translation and pressure induced by the transducer.

A single active differential EMG electrode (Bagnoli, Delsys Inc., Boston, MA, USA) was then placed over the MG muscle belly, next to the ultrasound transducer, to record muscle electrical activity. A ground electrode was attached to the patella. The area for the electrodes was cleaned with alcohol pads before positioning.

Before voluntary contraction trials, a baseline trial was collected to allow estimation of torque or EMG signal offsets; these offsets were then subtracted before data analysis. Subjects performed three plantarflexor MVC trials for five seconds each, with a minute break between each MVC trial to minimize fatigue. The average value of three maximum isometric plantarflexion torques for each MVC trial was used to calculate the level of plantarflexion contraction intensity for visual feedback. Subjects were given real-time visual feedback of their %MVC torque to help them match the desired %MVC torque target set by the experimenter. Three isometric plantarflexion contractions were recorded for each designated contraction level (0, 10, 20, 40, and 60%MVC). Contraction levels were randomized. For each trial, subjects were instructed to remain relaxed, and then to perform isometric plantarflexion contractions for five seconds with a minimum of a 30-s break between each trial.

During each experiment, isometric plantarflexion torque, surface EMG signals, and B-mode ultrasound images were collected. The torque and EMG signals were recorded at a sampling rate of 2 kHz and synchronized through a data acquisition (DAQ) system (National Instrument, Austin, TX, USA). B-mode ultrasound images were also recorded, time-synchronized with corresponding torque and EMG signals using an external trigger pulse sent from the ultrasound machine to the DAQ system.

Data analysis

Muscle architectural parameters such as fascicle length, pennation angle, and muscle thickness were manually estimated from the B-mode ultrasound images for all trials (Figure 1) (Lee and Piazza, 2009; Kim et al., 2013; Randhawa et al., 2013). Two points on each designated fascicle were digitized and fit with a straight line. Fascicle length was then calculated as the distance between the insertions of the digitized fascicles to the superficial and to the deep aponeuroses. Representative fascicle length (lf) was taken as the mean of three to five digitized fascicles per image. Pennation angle (α) was calculated as the mean of the angles that the digitized fascicles intersected with the superficial (α1) and deep aponeuroses (α2) and was averaged across the digitized fascicles for each trial. Muscle thickness (W) was the shortest distance from the superficial and to the deep aponeuroses through the center of the digitized fascicles (w = lf sin α).

Figure 1.

Figure 1.

Open in a new tab

Representative ultrasound image with superficial and deep aponeuroses, muscle fascicle, and muscle belly thickness highlighted by yellow lines. Note that w is the muscle thickness, lf the fascicle length, α1 the angle between the fascicle and superficial aponeurosis, and α2 the angle between the fascicle and deep aponeurosis.

In order to further characterize fascicle behavior related to force output, the fascicle rotation per millimeter of fascicle shortening (Azizi et al., 2008), named the fascicle rotation-shortening ratio, was calculated (as described in Statistical analysis section). The fascicle rotation was computed as the absolute change in pennation angle from resting pennation angle and the fascicle shortening as the change from resting fascicle length at each contraction level, respectively. Earlier studies suggested that the magnitude of fiber rotation characterized by fascicle rotation-shortening ratio depends on dynamic changes in muscle shape during a contraction which can significantly alter the force and velocity output of a pennate muscle (Azizi et al., 2008; Eng et al., 2018).

Torque signals were processed by applying a zero-phase second-order Butterworth low pass filter with a cut-off frequency of 6 Hz, followed by root mean square (RMS) envelope with moving window of 500 ms. The maximum isometric plantarflexion torque was determined by averaging peak RMS envelope values corresponding to three MVC trials, and all torque values were normalized by body mass (N m/kg) for further analyses. Similarly, EMG signals recorded from the MG muscle were processed by applying a zero-phase second-order Butterworth bandpass filter (bandwidth: 20–450 Hz), followed by the RMS envelope with the same moving window. Both processed torque and EMG values at the time of image capture were used for further analyses.

All signal processing procedures described above were performed using custom-written programs in MATLAB (Mathworks, Natick, USA).

Statistical analysis

Linear mixed-effects models were used to test whether fascicle behavior during isometric contraction remained intact following stroke (Figure 2A–C). Three separate dependent factors were considered: fascicle length, pennation angle and muscle thickness. Fixed effects were the intercept (i.e., fascicle length, pennation angle, and muscle thickness under passive conditions), side (non-paretic or paretic), torque, and interaction between side and torque. Separate analyses were conducted for each of the dependent factors. In all analyses, subjects were treated as a random effect to take into account the variability of our participants. Moreover, as all dependent factors are repeated measures variables for each side within subjects, the model included a random intercept and random slope (i.e., torque) for each side within subjects.

Figure 2.

Figure 2.

Open in a new tab

Fascicle length (A), pennation angle (B), and muscle thickness (C) as a function of joint torque normalized by body mass during voluntary isometric contraction for paretic (in red) and non-paretic (in blue) sides. With increasing the normalized torque, the fascicle length decreases and the pennation angle increases, but muscle thickness does not change. Each marker indicates each trial across individuals, and thick lines indicate the mean regression line averaged over the complete sample of each side across individuals. These lines indicate the regression line for each side of each individual. The confidential interval of 95% is represented by the shaded area.

The fascicle rotation per millimeter of fascicle shortening, named fascicle rotation-shortening ratio, was also calculated using a linear mixed-effects model for repeated measures data, with fascicle rotation as a dependent factor, and fascicle shortening and interaction between side and fascicle shortening as fixed effects (Figure 2D). The model included a random slope (i.e., fascicle shortening) for each side within subjects.

Finally, in order to test whether any of the muscle architectural parameters (i.e. fascicle length, pennation angle and muscle thickness under passive conditions, and fascicle rotation-shortening ratio) was modified with the maximum isometric plantarflexion torque, Spearman correlation coefficients were calculated with relative parameters (paretic / non-paretic).

All statistical analyses were done using MATLAB (Mathworks, Natick, USA), and the significance level (α) of 0.05 was used. A post-hoc power calculation was then conducted to understand actual statistical power (1 – β, a type II error). The R package SIMR based on Monte Carlo simulations was used to calculate statistical power for the linear mixed-effects models (Green and Macleod, 2016). Statistical power for the Spearman correlation analysis was determined using a one-tailed bivariate normal model (Faul et al., 2009).

Results

As expected, maximum voluntary isometric plantarflexion torque at the ankle on the paretic side was measurably reduced, in comparison with the non-paretic side. The torque on the paretic side (0.44 (SD 0.21) N·m/kg) was decreased by 37.2% compared to the non-paretic side (0.68 (SD 0.17) N·m/kg). Wilcoxon signed-rank test showed that the maximum torque on the paretic side was significantly smaller than on the non-paretic side (p = 0.016).

Fascicle length under passive conditions (i.e., at rest in neutral position) was also significantly smaller by approximately 5% on the paretic side than on the non-paretic side (F(1, 206) = 4.79, p = 0.030, power = 0.83) (Table 2). Fascicle shortening increased on both sides with increasing the torque (F(1, 206) = 133.16, p < 0.0001, power = 1.00) (Figure 2A), but smaller magnitudes of fascicle shortening were found on the paretic side, as indicated by a significant interaction between side and torque (F(1, 206) = 15.79, p < 0.0001, power = 1.00) (Figure 3A).

Table 2.

Muscle architecture under passive condition of each participant.

Fascicle length (mm) Pennation angle (deg) Muscle thickness (mm)
Non-paretic Paretic Non-paretic Paretic Non-paretic Paretic
S1 41.9 36.8 18.5 20.2 13.9 12.8
S2 47.6 47.9 22.6 15.2 18.1 12.6
S3 57.4 52.4 18.0 17.2 17.9 15.3
S4 48.7 48.4 22.5 21.8 18.6 18.1
S5 56.2 55.9 15.0 14.9 14.7 14.5
S6 53.5 50.0 19.8 12.7 18.4 10.9
S7 48.9 46.0 18.9 16.6 15.7 12.9
Mean (SD) 50.6 (5.4) 48.2 (6.0)* 19.3 (2.6) 17.0 (3.2) 16.7 (2.0) 13.9 (2.3)*
Open in a new tab
*

Significant difference between paretic and non-paretic sides.

Figure 3.

Figure 3.

Open in a new tab

Comparison of fascicle shortening (A), fascicle rotation (B), and muscle thickness change (C) per joint torque normalized by body mass during voluntary isometric contraction between paretic and non-paretic sides. Positive values indicate greater fascicle shortening or rotation or increase in muscle thickness with increasing contraction intensity. Note that significantly less fascicle shortening and fascicle rotation are shown on the paretic side compared to the non-paretic side, but changes in muscle thickness do not change on both sides. (D) Fascicle rotation-shortening ratio (i.e., fascicle rotation per fascicle shortening) is also significantly smaller on the paretic side than on the non-paretic side. Each marker indicates an individual and gray lines indicate paired data. Asterisk (*) indicates a significant difference between sides (p < 0.05).

There was no significant difference in pennation angle under passive conditions between the paretic and non-paretic sides (F(1, 206) = 2.99, p = 0.085, power = 0.59) (Table 2). However, as the torque increased, the pennation angle on both sides significantly increased (F(1, 206) = 46.84, p < 0.0001, power = 1.00) (Figure 2B). There was a significant interaction between side and torque (F(1, 206) = 5.84, p = 0.017, power = 0.88) (Figure 3B), indicating that less fascicle rotation was found on the paretic side compared to the non-paretic side.

Under passive conditions, muscle thickness on the paretic side was smaller by approximately 16% than on the non-paretic side (F(1, 206) = 7.72, p < 0.01, power = 0.85) (Table 2). However, both sides did not show a significant change in the muscle thickness with increasing torque (F(1, 206) = 0.04, p = 0.846, power = 0.07) (Figure 2C) as well as no interaction between side and torque (F(1, 206) = 1.90, p = 0.170, power = 1.00) (Figure 3C).

Fascicle rotation increased with increasing fascicle shortening on both sides (F(1, 208) = 128.72, p < 0.0001, power = 1.00) (Figure 2D). However, there was a significant interaction between side and fascicle shortening (F(1, 208) = 6.48, p = 0.012, power = 1.00), indicating that the magnitude of fascicle rotation-shortening ratio (i.e., fascicle rotation per fascicle shortening) was significantly smaller on the paretic side compared to the non-paretic side (Figure 3D).

A correlation matrix is summarized in Table 3. The relative fascicle rotation-shortening ratio showed a strong positive correlation with the relative maximum isometric plantarflexion torque (r = 0.821, p = 0.034, power = 0.83; Figure 4A). However, there was no correlation between the relative maximum isometric torque and relative muscle architectural parameters under passive conditions: fascicle length (p = 0.783, power = 0.02), pennation angle (p = 0.354, power = 0.26), and muscle thickness (p = 0.354, power = 0.26). The relative pennation angle was significantly correlated with muscle thickness (r = 0.893, p = 0.012, power = 0.94; Figure 4B).

Table 3.

Spearman correlation matrix between relative (paretic / non-paretic) parameters. Data are presented with correlation coefficient (r), significance (p) and actual statistical power (1 – β). r (p, power).

Relative parameters Fascicle length Pennation angle Muscle thickness Fascicle rotation-shortening ratio
Pennation angle −0.250 (0.595, 0.135)
Muscle thickness 0.071 (0.906, 0.067) 0.893 (0.012, 0.939)*
Fascicle rotation-shortening ratio −0.214 (0.662, 0.118) 0.429 (0.354, 0.257) 0.571 (0.200, 0.413)
Maximum torque −0.143 (0.783, 0.090) 0.429 (0.354, 0.257) 0.429 (0.354, 0.257) 0.821 (0.034, 0.829)*
Open in a new tab
*

Significant relationship (p < 0.05).

Figure 4.

Figure 4.

Open in a new tab

Relationship of relative joint torque vs. relative fascicle rotation-shortening ratio (A) and relative pennation angle vs. relative muscle thickness (B). The ratio of the parameters was calculated by dividing parameter values on the paretic side (P) by the corresponding values on the non-paretic side (NP). Each marker indicates an individual. The confidence interval of 95% is represented by the shaded area.

Discussion

To understand how muscle fascicles behave during submaximal voluntary isometric muscle contractions in stroke survivors, we characterized the fascicle shortening, the fascicle rotation, and the fascicle rotation-shortening ratio in paretic and non-paretic MG muscles, using ultrasonography. We also investigated the relationship between fascicle behavioral characteristics and torque-generating capacity. Our results showed that the fascicle shortening and fascicle rotation during voluntary isometric submaximal contractions are significantly reduced on the paretic side, as compared to the non-paretic side. In addition, the relative fascicle rotation-shortening ratio was positively correlated with the relative joint torque generation capability.

Architecture of paretic muscle after stroke under passive conditions

Our study showed that fascicle lengths on the paretic side were approximately 5% shorter than on the non-paretic side, which is comparable to previous results reported at similar knee and ankle positions. For example, the average MG fascicle length, taken from the previously reported values measured at rest (Gao and Zhang, 2008; Gao et al., 2009; Ramsay et al., 2014), was 47 mm (range: 38–51 mm) and 57 mm (range: 51–65 mm) on the paretic and non-paretic side of stroke survivors, respectively. A shortened fascicle length can affect force-generating capacity (Gray et al., 2012) such that the shape and location of the force-length curve are affected. After a hemispheric stroke, muscle fascicles often sit in a shortened position for a prolonged time. This may be related to contractures which tend to make muscles adapt to the new shortened length. Considering that the number of sarcomeres can also decrease after a period of immobilization in a shortened position (Williams and Goldspink, 1978), the shorter fascicle length on the paretic side in our study may be associated with a reduction in the number of sarcomeres, potentially leading to a loss of force-generating capacity (i.e., 37.2% reduction in maximum isometric plantarflexion torque in our study), as was also suggested by a previous study (Gao and Zhang, 2008).

Our study further showed that pennation angles under passive conditions were 2.3° smaller on the paretic side than on the non-paretic side, although this was not a significant difference. Smaller pennation angles in stroke-impaired muscle than contralateral muscle have again been reported earlier (Gao et al., 2009; Ramsay et al., 2014; Dias et al., 2017). The range of resting pennation angles reported in the literature was 18.2 to 20.6° for the non-paretic side, but 15.7 to 17.5° for the paretic side. Since fascicle forces are transmitted to the tendon, and since that force is proportional to the cosine of pennation angle, a decrease in pennation angle on the paretic side itself may not be a major contributor to a reduced force-generating capacity; since as pennation angle is smaller, the cosine of the angle is closer to 1 (i.e., the amount of force transmission is greater at smaller angle). Considering that there was a strong linear correlation between relative changes in muscle thickness with respect to the non-paretic side and the corresponding relative changes in pennation angle (Figure 3B), the smaller pennation angle alone after stroke may potentially be a result of muscle atrophy, that has been reported consistently after stroke (Ramsay et al., 2011; Knarr et al., 2013), and as was also suggested by a previous study (Ramsay et al., 2014).

Altered fascicle behavior during submaximal voluntary isometric contraction

We found that an increase in muscle contraction intensity was associated with a decrease in fascicle length of the MG muscle and an increase in fascicle rotation, in keeping with previous findings in both healthy individuals (Narici et al., 1996; Kawakami et al., 1998; Maganaris et al., 1998; Arampatzis et al., 2007; Dias et al., 2017) and stroke survivors (Dias et al., 2017), showing comparable magnitude of fascicle shortening and fascicle rotation on both sides. In agreement with an earlier study (Dias et al., 2017), the magnitude of fascicle shortening or fascicle rotation was significantly smaller on the paretic side than on the non-paretic side.

There was no change in the MG muscle thickness with increasing muscle contraction in both healthy individuals (Narici et al., 1996; Kawakami et al., 1998; Maganaris et al., 1998; Dias et al., 2017) and stroke survivors (Dias et al., 2017). This behavior may be explained classically in that a constant muscle thickness is compensated by a combination of fascicle shortening and fascicle rotation during isometric contraction (Gans and Gaunt, 1991; Narici et al., 1996).

Potential implication of decreased fascicle rotation-shortening ratio

Fascicle rotation-shortening ratio depends on dynamic changes in muscle shape during a contraction (Azizi et al., 2008; Eng et al., 2018). The muscle shape change during a contraction is thought to be related jointly to the changes in fascicle length and to the changes in pennation angle (Eng et al., 2018). Especially in a pennate muscle like MG, the direction of shape change (i.e., thickness, width, or both) may modulate the force output (Azizi et al., 2008).

In our current study, muscle thickness did not change during voluntary isometric contraction. Therefore, we deduce that changes in muscle shape during the contraction were presumably in the direction of muscle width. Arellano et al. (2016) reported that in wild turkeys, the strain ratio of aponeurosis width to fascicle length decreases as the MG muscle force increases during isotonic contraction, suggesting that changes in the aponeurosis width depend on the interplay between the longitudinal and transverse forces generated during a muscle contraction. We can then speculate that the decreased fascicle rotation-shortening ratio on the paretic side (i.e., smaller changes in pennation angle at given muscle fascicle shortening on the paretic side) may be associated with less strain in the direction of muscle width potentially due to an altered inter-/intramuscular mechanics (e.g., altered interplay between the longitudinal and transverse forces), potentially altering muscle performance.

It is currently unclear which factors could most affect inter-/intramuscular mechanics in stroke-impaired muscle. One potential explanation is muscle fibrosis that is associated with an abnormal accumulation of extracellular matrix (ECM). The increased concentration of collagen, a major component of the ECM, has been observed in spastic muscle in individuals with cerebral palsy (Lieber and Ward, 2013; Mathewson and Lieber, 2015) and in aging skeletal muscle (Alnaqeeb et al., 1984). Although the majority of findings regarding the changes in the ECM is from work in individuals with cerebral palsy or in aging muscle, it is conceivable that similar changes in the ECM occur in a stroke-impaired muscle. Our evidence here comes from indirect muscle stiffness estimates. For example, shear waves travel faster in paretic MG muscle than in non-paretic muscle of chronic stroke survivors (Lee et al., 2015; Jakubowski et al., 2017). Moreover, the efficacy of shock wave therapy for spasticity treatment in stroke survivors has been thought to be associated with reducing connective tissue stiffness caused by fibrosis of chronic hypertonic muscles (Manganotti and Amelio, 2005; Li et al., 2016; Xiang et al., 2018). Collectively, these altered material properties could change muscle transverse isotropy, potentially leading to a reduction in muscle performance.

Reduction in voluntary isometric torque-generating capacity

The mean maximum plantarflexion torque at the neutral position was 0.44 N m/kg on the paretic side and 0.68 N m/kg on the non-paretic side, showing a 37.2% deficit on the paretic side compared to the non-paretic side. The magnitude of these torque values are smaller than the values previously reported at similar knee and ankle positions (Klein et al., 2010; Knarr et al., 2013; Dias et al., 2017; Freire et al., 2017). For example, the average of maximum plantarflexion torque in the previous studies was 0.7 N m/kg (range: 0.6–0.8 N m/kg) and 1.4 N m/kg (range: 1.1–1.9 N m/kg) on the paretic and non-paretic side or stroke survivors, respectively. However, the relative deficit of the maximum plantarflexion torque on the paretic side to the non-paretic side is within the reported values (range: 33–62%) (Klein et al., 2010; Knarr et al., 2013; Dias et al., 2017; Freire et al., 2017).

Plantar flexors provide the forward and upward propulsive forces during walking (Hamner and Delp, 2013), and the plantarflexion torque explained 72% of the variance in the gait speed of chronic stroke survivors (Kim and Eng, 2003). During walking, peak ankle plantarflexion moment has been reported to be approximately 0.7–0.9 N m/kg on the paretic side and 1.2–1.5 N m/kg on the non-paretic side (Olney and Richards, 1996; Lamontagne et al., 2002) which are greater than the maximum torque values in this study. Although this discrepancy may appear to imply that the plantar flexor strength in our participants might not be enough for push-off during walking, direct comparison is not advised since the knee and ankle joint angles when peak plantarflexion moment occurs are different in walking and the experimental set-up of this study. Since the gastrocnemius muscle is biarticular, differences in the muscle length have a great effect on the force-generating capacity of the muscles.

Considering that chronic stroke survivors who are unable to walk independently may experience progressive changes in skeletal muscle on both the paretic and non-paretic sides (Akazawa et al., 2018), altered muscle architecture might influence the maximum torque-generating capacity and in turn, walking performance. However, this study was limited to a small sample size to assess this hypothesis (Table 3); thus, further studies are required to understand how muscular changes could adversely affect muscle function as well as motor performance.

Potential mechanisms of muscle weakness resulting from altered contractile properties: A muscle modeling approach

As our protocol was not designed to investigate potential changes in contractile properties, we developed a detailed simulation, based on a modified Hill-type model (Thelen, 2003), in order to explore potential effects of altered contractile properties (Gao and Zhang, 2008) on fascicle behavior and fascicle force production. All model parameters required to define the MG muscle were obtained from the generic model ‘3DGaitModel2392’ in OpenSim v3.3 (Delp et al., 2007). The details of the model are described elsewhere (Son et al., 2018).

Model parameters were adjusted to reflect the previous findings including shortened optimal fascicle length and narrower active force-length curve width (Figure 5A). Our simulation showed that altered contractile properties can significantly limit fascicle shortening (Figure 5B), fascicle rotation (Figure 5C), fascicle rotation-shortening ratio (Figure 5D) and fascicle force production (Figure 5E), suggesting that shortened optimal fascicle length may be the more dominant factor related to those limitations rather than narrower width of active force-length curve. Moreover, our modeling approach indicates that a significant reduction in fascicle shortening, fascicle rotation and fascicle rotation-shortening ratio found in the present study may contribute to the decrease in the maximum plantarflexion torque on the paretic side, compared to the non-paretic side, potentially explaining a strong correlation between the relative fascicle rotation-shortening ratio and the relative maximum plantarflexion torque.

Figure 5.

Figure 5.

Open in a new tab

(A) Hypothetical active fascicle force-length curves with no adjustment of active force-length curve (blue solid line), with narrower width of active force-length curve (red solid line), with shortened optimal fiber length (red dashed line), and with both narrower width and shortened optimal fiber length (red dotted line). Note that altered contractile properties can significantly affect fascicle shortening (B), fascicle rotation (C), fascicle rotation-shortening ratio (D), and force output (E). Moreover, reduced volitional activation reduces force output dominantly, but force output is more affected with altered force-length curves, indicating muscular changes may also lead to additional impairment in force production (F).

Further studies are required to better understand potential contributions of muscular changes to altered force-generating capacity after stroke, with a more precise muscle model including other observations on pennation angle (Gao et al., 2009; Ramsay et al., 2014; Dias et al., 2017), muscle thickness (Gao et al., 2009; Ramsay et al., 2014; Dias et al., 2017), physiological cross-sectional area (Ramsay et al., 2011; Knarr et al., 2013), fascicle stiffness (Gao et al., 2009; Zhao et al., 2015; Jakubowski et al., 2017), and tendon stiffness (Svantesson et al., 2000; Zhao et al., 2009; Zhao et al., 2015); although our hypothetical approaches could account for some aspects of the potential effects of altered contractile properties on fascicle behavior and force production without considering those parameters.

Potential impact of muscular changes on muscle weakness

Using the burst superimposition test, Knarr et al. (2013) reported that the ratio of paretic to non-paretic maximum plantarflexion torques had a strong positive correlation with the central activation ratio (i.e., the ratio of the voluntary force to the total force including any force increment (Stackhouse et al., 2001); CAR), suggesting that the reduced volitional activation on the paretic side likely leads to the decreased MVC on the paretic side. Interestingly, Knarr et al. (2013) also showed that the average of the maximum force generating ability (MFGA) in the paretic calf muscles was 65% of the non-paretic MFGA. Considering that the burst superimposition test is used to estimate MFGA of a muscle by providing additional signals large enough to maximally activate the muscle, it seems plausible that the reduction in the paretic MFGA by 35% may come from muscular changes. Although it is yet possible to assign a contribution to each potential muscular factor, the impact of muscular changes on muscle weakness would be significant.

Based on the previous finding that the average CAR of the paretic limb was 61% of the non-paretic limb (Knarr et al., 2013), we performed similar simulations as above but with 60% muscle activation to reflect the reduced volitional activation on the paretic side. Our simulation results indicate that the reduced volitional activation is a predominant factor contributing to muscle weakness. However, when the reduced volitional activation and the altered active force-length curve were jointly assumed to contribute, the predicted force was more affected (Figure 5F), supporting the idea that muscular changes may lead to additional impairments in force production. Future studies are guaranteed to investigate underlying mechanisms of how muscular changes could contribute to additional deficit of voluntary muscle strength.

Limitations

Our findings, and thus, conclusions, are potentially limited by relatively low statistical power, mainly due to the limited sample size. Under passive conditions, the mean pennation angle on the non-paretic side was greater than on the paretic side, but the pennation angle was not significantly different between sides (Table 2). Moreover, the significant relationship between the relative joint torque and the relative fascicle rotation-shortening ratio might be also driven by a specific data point (e.g., the subject at (1, 1) position in Figure 4A). Collectively, special care should be taken to generalize our findings.

In the current study, MG muscle architectural parameters at neutral ankle position were compared between the paretic and non-paretic sides of stroke survivors. Given that the ankle and knee joint angles were comparable between the sides in our participants, it is reasonable to assume that fascicles on the paretic side are more stretched and are under higher tension, compared to the non-paretic side. This could be a reason for less fascicle shortening and less fascicle rotation in the paretic MG muscles. In addition, it is also worthwhile to note that our findings obtained from voluntary isometric contraction at the neutral position may not directly address fascicle behavior in other situations including different muscle-tendon arrangements or dynamic movements. Indeed, since the MG muscle is a bi-articular muscle crossing both knee and ankle joints, the fascicle behavior of the MG muscle is different for changes in both ankle and knee positions during either concentric or eccentric contraction, due to the force-length-velocity relations, slackness of tendinous tissues, and/or tendon compliance (Wakahara et al., 2007, 2009; Zhao et al., 2009; Zhao et al., 2015; Dias et al., 2017; Fukutani et al., 2017). Future studies are needed to investigate the fascicle behavior of stroke-impaired muscle in various situations, allowing us to better understand the potential impact of altered muscle architecture on mechanical properties.

In our study, muscle architectural parameters were determined from the MG muscle belly of both paretic and non-paretic sides in our participants. In MG muscles of intact individuals, there was no significant difference in fascicle length and in pennation angle between different regions (i.e., proximal, central, and distal) along with different sections (i.e., lateral, mid-sagittal and medial) of the muscle (Maganaris et al., 1998). A similar finding was also reported that the effect of the locations (i.e., distal, central, and proximal) on fascicle length and curvature was not significant (Muramatsu et al., 2002). However, there may be a possibility of heterogeneity in muscle morphology and function after stroke, potentially due to fiber atrophy (Scelsi et al., 1984; Hachisuka et al., 1997), connective tissue infiltration (Ryan et al., 2002; Ramsay et al., 2011), shifts in fiber type (Scelsi et al., 1984; Dattola et al., 1993; Hachisuka et al., 1997; Lukács et al., 2008), and fiber re-innervation (Dattola et al., 1993; Lukács, 2005; Lukács et al., 2009). These changes in muscle morphology potentially lead to a spatial difference in fascicle behavior over the whole muscle, although it is unclear whether the motor units in the human MG are spatially localized (Vieira et al., 2011) or uniformly distributed (Héroux et al., 2015). Investigating fascicle behavior of stroke-impaired muscle at different measurement sites will improve our knowledge of heterogeneity in the human MG muscle after stroke.

It is important to consider clinical treatment history. For example, some of the subjects wore ankle-foot orthosis during times outside of their homes. Ankle orthoses or other bracing devices are typically prescribed to prevent a severe foot drop during swing phase of walking in individuals with stroke as well as to improve walking ability (Leung and Moseley, 2003). Although the potential effects of long-term use of such clinical devices restricting joint motion have not been investigated systematically (Singer et al., 2014), it is plausible that muscular changes might be involved over a prolonged time. In addition, it is evident that muscle architecture and material properties can change due to repeat injections of botulinum toxin (Fortuna et al., 2011; Tok et al., 2011). The subjects in this study did not have any botulinum toxin injections in the previous year, though a few did have injections previously. Thus, long-term clinical treatment history could potentially contribute to an altered muscle architecture and fascicle behavior during voluntary isometric contraction.

We used net ankle plantarflexion torque to determine contraction levels. As other muscles may have contributed to this torque, it is important to note that the actual contribution of the MG muscles to the plantarflexion torque could be different across trials and/or participants. As the magnitude of co-contraction of dorsiflexors becomes greater, the actual plantarflexion torque becomes greater for a given net plantarflexion torque. Considering that the paretic muscles may have a greater co-contraction ratio than the non-paretic muscles (Banks et al., 2017), our results may not be changed by the magnitude of co-contraction of dorsiflexors. In addition, the relative contribution of other plantar flexors to the net plantarflexion torque may not be consistent across trials and/or subjects. For future studies, recording sEMG signals from the other muscles contributing to the net joint torque is necessary to confirm the potential impact of co-contraction and/or load-sharing.

Conclusions

This study characterized and compared fascicle behaviors in paretic and non-paretic MG muscles during voluntary isometric contraction. The main findings were: (1) paretic muscles have altered muscle architecture under passive conditions (i.e., shorter fascicle length and reduced muscle thickness) and; (2) paretic muscle fascicles behave differently during voluntary isometric contractions (i.e., reduced fascicle shortening and rotation, and small fascicle rotation-shortening ratio); (3) the relative fascicle rotation-shortening ratio is significantly associated with the relative maximum isometric plantarflexion torque. This implies that altered muscle architecture could affect fascicle behavior associated with isometric torque-generating capacity.

Highlights.

  • Paretic MG muscles show altered muscle architecture under passive conditions.

  • Paretic muscles show altered fascicle behavior during voluntary isometric contractions.

  • Relative fascicle rotation-shortening ratio is associated with relative maximum torque.

Acknowledgments

We thank all the participants in this study. We also thank James Pisano for help collecting the data and Masha Kocherginsky for statistical consulting. This study was supported by grants from the National Institute on Disability, Independent Living, and Rehabilitation Research (90SFGE0005), the Davee Foundation Stroke Research Seed Grant initiative and the Northwestern University Department of Neurology, Division of Stroke and Neurocritical Care, and the National Institute of Health (R01HD089952 and K12HD073945).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Akazawa N, Harada K, Okawa N, Tamura K, Hayase A, Moriyama H, 2018. Relationships between muscle mass, intramuscular adipose and fibrous tissues of the quadriceps, and gait independence in chronic stroke survivors: A cross-sectional study. Physiotherapy 104 (4), 438–445. [DOI] [PubMed] [Google Scholar]
  2. Alnaqeeb MA, Al Zaid NS, Goldspink G, 1984. Connective tissue changes and physical properties of developing and ageing skeletal muscle. Journal of Anatomy 139 (Pt 4), 677–689. [PMC free article] [PubMed] [Google Scholar]
  3. Arampatzis A, Mademli L, De Monte G, Walsh M, 2007. Changes in fascicle length from rest to maximal voluntary contraction affect the assessment of voluntary activation. Journal of Biomechanics 40 (14), 3193–3200. [DOI] [PubMed] [Google Scholar]
  4. Arellano CJ, Gidmark NJ, Konow N, Azizi E, Roberts TJ, 2016. Determinants of aponeurosis shape change during muscle contraction. Journal of Biomechanics 49 (9), 1812–1817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Azizi E, Brainerd EL, Roberts TJ, 2008. Variable gearing in pennate muscles. Proceedings of the National Academy of Sciences 105 (5), 1745–1750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Banks CL, Huang HJ, Little VL, Patten C, 2017. Electromyography exposes heterogeneity in muscle co-contraction following stroke. Frontiers in Neurology 8 (699). [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dattola R, Girlanda P, Vita G, Santoro M, Roberto ML, Toscano A et al. , 1993. Muscle rearrangement in patients with hemiparesis after stroke: An electrophysiological and morphological study. European Neurology 33 (2), 109–114. [DOI] [PubMed] [Google Scholar]
  8. Delp SL, Anderson FC, Arnold AS, Loan P, Habib A, John CT et al. , 2007. Opensim: Open-source software to create and analyze dynamic simulations of movement. IEEE Transactions on Biomedical Engineering 54 (11), 1940–1950. [DOI] [PubMed] [Google Scholar]
  9. Dias CP, Freire B, Goulart NBA, Onzi ES, Becker J, Gomes I et al. , 2017. Muscle architecture and torque production in stroke survivors: An observational study. Topics in Stroke Rehabilitation 24 (3), 206–213. [DOI] [PubMed] [Google Scholar]
  10. Eng CM, Azizi E, Roberts TJ, 2018. Structural determinants of muscle gearing during dynamic contractions. Integrative and Comparative Biology 58 (2), 207–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Faul F, Erdfelder E, Buchner A, Lang A-G, 2009. Statistical power analyses using g*power 3.1: Tests for correlation and regression analyses. Behavior Research Methods 41 (4), 1149–1160. [DOI] [PubMed] [Google Scholar]
  12. Fortuna R, Aurélio Vaz M, Rehan Youssef A, Longino D, Herzog W, 2011. Changes in contractile properties of muscles receiving repeat injections of botulinum toxin (botox). Journal of Biomechanics 44 (1), 39–44. [DOI] [PubMed] [Google Scholar]
  13. Freire B, Dias CP, Goulart NBA, de Castro CD, Becker J, Gomes I et al. , 2017. Achilles tendon morphology, plantar flexors torque and passive ankle stiffness in spastic hemiparetic stroke survivors. Clinical Biomechanics 41, 72–76. [DOI] [PubMed] [Google Scholar]
  14. Fukutani A, Misaki J, Isaka T, 2017. Relationship between joint torque and muscle fascicle shortening at various joint angles and intensities in the plantar flexors. Scientific Reports 7 (1), 290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gans C, Gaunt AS, 1991. Muscle architecture in relation to function. Journal of Biomechanics 24, Supplement 1 (0), 53–65. [DOI] [PubMed] [Google Scholar]
  16. Gao F, Grant TH, Roth EJ, Zhang L-Q, 2009. Changes in passive mechanical properties of the gastrocnemius muscle at the muscle fascicle and joint levels in stroke survivors. Archives of Physical Medicine and Rehabilitation 90 (5), 819–826. [DOI] [PubMed] [Google Scholar]
  17. Gao F, Zhang L-Q, 2008. Altered contractile properties of the gastrocnemius muscle poststroke. Journal of Applied Physiology 105 (6), 1802–1808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gray V, Rice CL, Garland SJ, 2012. Factors that influence muscle weakness following stroke and their clinical implications: A critical review. Physiotherapy Canada 64 (4), 415–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Green P, Macleod CJ, 2016. SIMR: An R package for power analysis of generalized linear mixed models by simulation. Methods in Ecology and Evolution 7 (4), 493–498. [Google Scholar]
  20. Hachisuka K, Umezu Y, Ogata H, 1997. Disuse muscle atrophy of lower limbs in hemiplegic patients. Archives of Physical Medicine and Rehabilitation 78 (1), 13–18. [DOI] [PubMed] [Google Scholar]
  21. Hamner SR, Delp SL, 2013. Muscle contributions to fore-aft and vertical body mass center accelerations over a range of running speeds. Journal of Biomechanics 46 (4), 780–787. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Héroux ME, Brown HJ, Inglis JT, Siegmund GP, Blouin JS, 2015. Motor units in the human medial gastrocnemius muscle are not spatially localized or functionally grouped. Journal of Physiology 593 (16), 3711–3726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jakubowski KL, Terman A, Santana RVC, Lee SSM, 2017. Passive material properties of stroke-impaired plantarflexor and dorsiflexor muscles. Clinical Biomechanics 49 (Supplement C), 48–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kawakami Y, Ichinose Y, Fukunaga T, 1998. Architectural and functional features of human triceps surae muscles during contraction. Journal of Applied Physiology 85 (2), 398–404. [DOI] [PubMed] [Google Scholar]
  25. Kim CM, Eng JJ, 2003. The relationship of lower-extremity muscle torque to locomotor performance in people with stroke. Physical Therapy. [PubMed] [Google Scholar]
  26. Kim S, Son J, Yi C, Kim D, Kim Y, 2013. Pennation angles of ankle dorsiflexor and plantarflexors depending on muscle contraction intensity. International Journal of Precision Engineering and Manufacturing 14 (5), 855–858. [Google Scholar]
  27. Klein CS, Brooks D, Richardson D, McIlroy WE, Bayley MT, 2010. Voluntary activation failure contributes more to plantar flexor weakness than antagonist coactivation and muscle atrophy in chronic stroke survivors. Journal of Applied Physiology 109 (5), 1337–1346. [DOI] [PubMed] [Google Scholar]
  28. Knarr BA, Ramsay JW, Buchanan TS, Higginson JS, Binder-Macleod SA, 2013. Muscle volume as a predictor of maximum force generating ability in the plantar flexors post-stroke. Muscle & Nerve 48 (6), 971–976. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lamontagne A, Malouin F, Richards CL, Dumas F, 2002. Mechanisms of disturbed motor control in ankle weakness during gait after stroke. Gait & Posture 15 (3), 244–255. [DOI] [PubMed] [Google Scholar]
  30. Lee SSM, Piazza SJ, 2009. Built for speed: Musculoskeletal structure and sprinting ability. Journal of Experimental Biology 212 (22), 3700–3707. [DOI] [PubMed] [Google Scholar]
  31. Lee SSM, Spear S, Rymer WZ, 2015. Quantifying changes in material properties of stroke-impaired muscle. Clinical Biomechanics 30 (3), 269–275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Leung J, Moseley A, 2003. Impact of ankle-foot orthoses on gait and leg muscle activity in adults with hemiplegia. Physiotherapy 89 (1), 39–55. [Google Scholar]
  33. Li L, Tong KY, Hu X, 2007. The effect of poststroke impairments on brachialis muscle architecture as measured by ultrasound. Archives of Physical Medicine and Rehabilitation 88 (2), 243–250. [DOI] [PubMed] [Google Scholar]
  34. Li T-Y, Chang C-Y, Chou Y-C, Chen L-C, Chu H-Y, Chiang S-L et al. , 2016. Effect of radial shock wave therapy on spasticity of the upper limb in patients with chronic stroke. Medicine 95 (18), e3544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lieber RL, Ward SR, 2013. Cellular mechanisms of tissue fibrosis. 4. Structural and functional consequences of skeletal muscle fibrosis. American Journal of Physiology. Cell Physiology 305 (3), C241–C252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lukács M, 2005. Electrophysiological signs of changes in motor units after ischaemic stroke. Clinical Neurophysiology 116 (7), 1566–1570. [DOI] [PubMed] [Google Scholar]
  37. Lukács M, Vécsei L, Beniczky S, 2008. Large motor units are selectively affected following a stroke. Clinical Neurophysiology 119 (11), 2555–2558. [DOI] [PubMed] [Google Scholar]
  38. Lukács M, Vécsei L, Beniczky S, 2009. Changes in muscle fiber density following a stroke. Clinical Neurophysiology 120 (8), 1539–1542. [DOI] [PubMed] [Google Scholar]
  39. Maganaris CN, Baltzopoulos V, Sargeant AJ, 1998. In vivo measurements of the triceps surae complex architecture in man: Implications for muscle function. Journal of Physiology 512 (2), 603–614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Manganotti P, Amelio E, 2005. Long-term effect of shock wave therapy on upper limb hypertonia in patients affected by stroke. Stroke 36 (9), 1967–1971. [DOI] [PubMed] [Google Scholar]
  41. Mathewson MA, Lieber RL, 2015. Pathophysiology of muscle contractures in cerebral palsy. Physical Medicine & Rehabilitation Clinics of North America 26 (1), 57–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Muramatsu T, Muraoka T, Kawakami Y, Shibayama A, Fukunaga T, 2002. In vivo determination of fascicle curvature in contracting human skeletal muscles. Journal of Applied Physiology 92 (1), 129–134. [DOI] [PubMed] [Google Scholar]
  43. Narici MV, Binzoni T, Hiltbrand E, Fasel J, Terrier F, Cerretelli P, 1996. In vivo human gastrocnemius architecture with changing joint angle at rest and during graded isometric contraction. Journal of Physiology 496 (1), 287–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Olney SJ, Richards C, 1996. Hemiparetic gait following stroke. Part i: Characteristics. Gait & Posture 4 (2), 136–148. [Google Scholar]
  45. Ramsay JW, Barrance PJ, Buchanan TS, Higginson JS, 2011. Paretic muscle atrophy and non-contractile tissue content in individual muscles of the post-stroke lower extremity. Journal of Biomechanics 44 (16), 2741–2746. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Ramsay JW, Buchanan TS, Higginson JS, 2014. Differences in plantar flexor fascicle length and pennation angle between healthy and poststroke individuals and implications for poststroke plantar flexor force contributions. Stroke Research and Treatment 2014, 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Randhawa A, Jackman ME, Wakeling JM, 2013. Muscle gearing during isotonic and isokinetic movements in the ankle plantarflexors. European Journal of Applied Physiology 113 (2), 437–447. [DOI] [PubMed] [Google Scholar]
  48. Ryan AS, Dobrovolny CL, Smith GV, Silver KH, Macko RF, 2002. Hemiparetic muscle atrophy and increased intramuscular fat in stroke patients. Archives of Physical Medicine and Rehabilitation 83 (12), 1703–1707. [DOI] [PubMed] [Google Scholar]
  49. Scaglioni G, Narici MV, Martin A, 2016. Neural activation during submaximal contractions seems more reflective of neuromuscular ageing than maximal voluntary activation. Frontiers in Aging Neuroscience 8 (19). [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Scelsi R, Lotta S, Lommi G, Poggi P, Marchetti C, 1984. Hemiplegic atrophy: Morphological findings in the anterior tibial muscle of patients with cerebral vascular accidents. Acta Neuropathologica 62 (4), 324–331. [DOI] [PubMed] [Google Scholar]
  51. Singer ML, Kobayashi T, Lincoln LS, Orendurff MS, Foreman KB, 2014. The effect of ankle–foot orthosis plantarflexion stiffness on ankle and knee joint kinematics and kinetics during first and second rockers of gait in individuals with stroke. Clinical Biomechanics 29 (9), 1077–1080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Son J, Indresano A, Sheppard K, Ward SR, Lieber RL, 2018. Intraoperative and biomechanical studies of human vastus lateralis and vastus medialis sarcomere length operating range. Journal of Biomechanics 67, 91–97. [DOI] [PubMed] [Google Scholar]
  53. Stackhouse SK, Stevens JE, Lee SC, Pearce KM, Snyder-Mackler L, Binder-Macleod SA, 2001. Maximum voluntary activation in nonfatigued and fatigued muscle of young and elderly individuals. Physical Therapy 81 (5), 1102–1109. [PubMed] [Google Scholar]
  54. Svantesson U, Takahashi H, Carlsson U, Danielsson A, Stibrant Sunnerhagen K, 2000. Muscle and tendon stiffness in patients with upper motor neuron lesion following a stroke. European Journal of Applied Physiology 82 (4), 275–279. [DOI] [PubMed] [Google Scholar]
  55. Thelen DG, 2003. Adjustment of muscle mechanics model parameters to simulate dynamic contractions in older adults. Journal of Biomechanical Engineering 125 (1), 70–77. [DOI] [PubMed] [Google Scholar]
  56. Tok F, Ozcakar L, Safaz I, Alaca R, 2011. Effects of botulinum toxin-a on the muscle architecture of stroke patients: The first ultrasonographic study. J Rehabil Med 43 (11), 1016–1019. [DOI] [PubMed] [Google Scholar]
  57. Vieira TMM, Loram ID, Muceli S, Merletti R, Farina D, 2011. Postural activation of the human medial gastrocnemius muscle: Are the muscle units spatially localised? Journal of Physiology 589 (2), 431–443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Wakahara T, Kanehisa H, Kawakami Y, Fukunaga T, 2007. Fascicle behavior of medial gastrocnemius muscle in extended and flexed knee positions. Journal of Biomechanics 40 (10), 2291–2298. [DOI] [PubMed] [Google Scholar]
  59. Wakahara T, Kanehisa H, Kawakami Y, Fukunaga T, 2009. Effects of knee joint angle on the fascicle behavior of the gastrocnemius muscle during eccentric plantar flexions. Journal of Electromyography and Kinesiology 19 (5), 980–987. [DOI] [PubMed] [Google Scholar]
  60. Williams PE, Goldspink G, 1978. Changes in sarcomere length and physiological properties in immobilized muscle. Journal of Anatomy 127 (Pt 3), 459–468. [PMC free article] [PubMed] [Google Scholar]
  61. Xiang J, Wang W, Jiang W, Qian Q, 2018. Effects of extracorporeal shock wave therapy on spasticity in post-stroke patients: A systematic review and meta-analysis of randomized controlled trials. J Rehabil Med 50 (10), 852–859. [DOI] [PubMed] [Google Scholar]
  62. Zhao H, Ren Y, Roth EJ, Harvey RL, Zhang L-Q, 2015. Concurrent deficits of soleus and gastrocnemius muscle fascicles and Achilles tendon post stroke. Journal of Applied Physiology 118 (7), 863–871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Zhao H, Ren Y, Wu Y-N, Liu SQ, Zhang L-Q, 2009. Ultrasonic evaluations of Achilles tendon mechanical properties poststroke. Journal of Applied Physiology 106 (3), 843–849. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES