Effect of Stroke and Aging on Plantarflexor Strength and Voluntary Activation

Abstract

Plantarflexor strength and voluntary activation are key metrics for characterizing lower extremity function in stroke survivors. However, the extent to which stroke and/or aging affects these neuromuscular properties remains unclear. This study addressed this gap by testing plantarflexor strength and voluntary activation in fifty-two (stroke: 19, older: 15, young: 18) participants. Testing was done bilaterally in stroke survivors and on the dominant leg of the control participants using the central activation ratio (CAR) and interpolated twitch technique (ITT) with triplets. Stroke survivors demonstrated significantly reduced raw and mass-normalized plantarflexor strength on both legs compared with controls, and on their more affected leg compared with the less affected leg (all p's ≤ 0.02). Regardless of technique, voluntary activation was significantly lower only in the more affected leg compared with the less affected leg and control leg (all p's ≤ 0.02). Older adults also demonstrated lower plantarflexor strength (p ≤ 0.01), but not voluntary activation, compared with young adults. These findings indicate that both stroke and aging affect plantarflexor strength; however, voluntary activation is only affected by stroke. Additionally, quantification technique influenced voluntary activation estimates, with CAR consistently demonstrating higher activation relative to ITT. Collectively, these findings highlight the need for targeting plantarflexor strength and voluntary activation during post-stroke rehabilitation.

Keywords

twitch interpolation inhibition age-related hemiparesis impairment elderly

Introduction

Ankle plantarflexor strength is critical for maintaining optimal gait speed, balance and mobility in older adults and individuals with stroke (Aguirre-Villanueva et al., 2025; Dorsch et al., 2021; Liang et al., 2025). Voluntary activation failure, which is the inability to completely contract the muscle during a maximal contraction (Kent-Braun & Le Blanc, 1996; Todd et al., 2004), can contribute to plantarflexor weakness (Klein et al., 2010) and negatively affect gait and mobility. Voluntary activation failure can happen due to incomplete recruitment of all motor units and/or suboptimal firing of the recruited motor units. While voluntary activation is typically >90% in neurologically intact adults (Dalton et al., 2009; Newham & Hsiao, 2001; Vandervoort & McComas, 1986), voluntary activation failure may develop due to factors such as aging (Akagi et al., 2018; Morse et al., 2004; Rozand et al., 2020) and stroke (Klein et al., 2010; Klein et al., 2013; Knorr et al., 2011). Thus, characterizing aging and stroke effects on plantarflexor strength and voluntary activation is critical for developing optimal post-stroke gait and mobility interventions.

Despite ongoing research investigating the effects of aging and stroke, the role of aging on plantarflexor strength and voluntary activation is conflicting (Akagi et al., 2018; Barber et al., 2013; Morse et al., 2004; Simoneau et al., 2005). For example, some studies reported reduced strength and voluntary activation (Akagi et al., 2018; Morse et al., 2004), while others reported reduced strength without reductions in voluntary activation (Barber et al., 2013; Simoneau et al., 2005) in older adults compared with young adults. Moreover, the effects of aging and stroke have been investigated separately (Klein et al., 2010, 2013; Knorr et al., 2011; Rozand et al., 2020), making it difficult to distinguish their individual and cumulative effects on plantarflexor strength and voluntary activation. In addition, the few studies that have evaluated plantarflexor voluntary activation in stroke survivors typically only compare between limbs without comparing with an age-matched control group (Klein et al., 2010). Thus, concurrent evaluation of both aging and stroke would provide comprehensive information on how these factors affect plantarflexor strength and voluntary activation.

A potential factor that could contribute to conflicting findings in aging-related voluntary activation deficits is the quantification method used for computing voluntary activation. Voluntary activation is commonly computed using the percent activation derived from the central activation ratio (CAR) or the interpolated twitch technique (ITT). The CAR is calculated by dividing the maximal voluntary torque produced prior to delivery of the electrical stimulus relative to the total torque following the superimposed stimulus (Kent-Braun & Le Blanc, 1996; Rozand et al., 2020). In the ITT method, voluntary activation is calculated as the electrically evoked torque during the maximal contraction normalized to the electrically evoked torque at rest (i.e., resting twitch torque) (Garcia et al., 2022). An advantage of ITT is that it accounts for the inability of the electrical stimulus to produce true maximal muscle torque and ensures the evoked torque during maximal voluntary isometric contraction (MVIC) is normalized to the torque from the stimulated muscles (Krishnan & Williams, 2010; Shield & Zhou, 2004; Zarkou et al., 2017). In contrast, CAR has consistently demonstrated higher voluntary activation, with these differences becoming more pronounced with larger voluntary activation deficits (Grindstaff & Threlkeld, 2014; Krishnan & Williams, 2010; Zarkou et al., 2017). As a result, CAR may be less sensitive in its ability to detect voluntary activation deficits, particularly in populations with substantial voluntary activation deficits. While researchers have established quantification methods can influence estimates of voluntary activation of other lower extremity muscles (Garcia et al., 2022; Krishnan & Williams, 2010; Olsen et al., 2021), the effect of these quantification techniques have yet to be compared in the plantarflexor muscles of stroke survivors and neurologically intact adults.

Therefore, the primary purpose of this study was to evaluate the effect of stroke and aging on plantarflexor strength and voluntary activation. We hypothesized that (1) stroke survivors would demonstrate lower plantarflexor strength and voluntary activation bilaterally compared with the age-matched neurologically intact control participants and (2) older adults would demonstrate lower plantarflexor strength and voluntary activation compared with young adults. The secondary purpose of this study was to evaluate the effect of quantification method (CAR vs. ITT) on estimates of plantarflexor voluntary activation. We hypothesized that CAR would underestimate voluntary activation deficits, with higher voluntary activation estimates determined for CAR compared with ITT.

Methods

Participants

A total of 52 participants (19 stroke survivors [12 males, 7 females], 15 older adults [10 males, 5 females]; 18 young adults [10 males, 8 females]) participated in this study. The control participant pool consisted of 19 age- and sex-matched participants that served as a control group for the stroke group. Older adults were defined as ≥60 years, and young adults were defined as 18–35 years, consistent with prior voluntary activation studies that used these age ranges to distinguish age-related differences in voluntary activation (Rozand et al., 2020). Participant demographics are reported in Table 1, and they were part of a larger ongoing study to investigate the functional implications of stroke and botulinum neurotoxin on ankle mechanical properties. Stroke survivors were included in this study if they were: (1) over the age of 18, (2) were at least 6 months from the onset of stroke, and (3) had no documented major deficits of sensation or proprioception. Participants were excluded if they: (1) were unable to provide written informed consent, (2) had a Mini-Mental State Exam score < 22, (3) had uncontrolled diabetes or hypertension, (4) had a history of botulinum neurotoxin injection for spasticity management, (5) had a recent history of lower limb fractures or surgery, (6) had severe joint contractures, (7) had severe aphasia, (8) were incapable of standing independently, with or without the use of external support or orthopedic devices, (9) had a cerebellar stroke, and (10) had any other medical conditions that will significantly impact the study results. All participants read and signed a written informed consent document approved by the University of Michigan Medical School Institutional Review Board (IRBMED HUM00164858).

Table 1.

Demographics of Participants.

Group	Sex	Age (year ± SD)	Mass (kg ± SD)	Height (cm ± SD)
Stroke (n = 19)	12M/7F	64.84 ± 13.05	79.43 ± 23.30	172.72 ± 8.51
Age-matched Control (n = 19)	12M/7F	61.47 ± 12.15	74.84 ± 12.69	172.79 ± 7.67
Older Control (n = 15)	10M/5F	66.13 ± 4.17	74.51 ± 13.08	171.80 ± 7.19
Younger Control (n = 18)	10M/8F	21.72 ± 3.74	71.46 ± 15.25	175.88 ± 9.86

Abbreviations: cm, centimeters; F, females; kg, kilogram; M, males; SD, standard deviation

Experimental Set-up and Protocol

Participants were positioned and secured on an isokinetic dynamometer (Biodex Medical Systems, Inc., Shirley, NY, USA) with the trunk set to 70° of flexion and the knee set to 20–30° of flexion, such that shank was parallel to the ground (Tuominen et al., 2023). In addition, the ankle of the tested leg was set to about 15° of dorsiflexion. The tested foot was strapped to the ankle plantarflexion/dorsiflexion footplate, with the lateral malleolus aligned with the dynamometer's axis of rotation according to the manufacturer's guidelines (Figure 1).

Figure 1.

The top panel depicts the participant set-up in the isokinetic dynamometer for plantarflexor strength and voluntary activation testing, and the bottom panel provides a schematic for quantification techniques used to estimate voluntary activation ratios. The central activation ratio (CAR) and interpolated twitch technique (ITT) were used to estimate voluntary activation. In the respective equations: a, represents the total torque evoked by the electrical stimulus, b, represents the torque at the time of electrical stimulus during maximal voluntary isometric contraction, and c, represents the evoked torque at rest.

Plantarflexor strength and voluntary activation were evaluated using the electrical superimposition method during maximum voluntary isometric contraction (MVIC), which is the greatest force a muscle can generate without joint movement. (Krishnan & Williams, 2011). Testing was performed bilaterally for stroke survivors and on the dominant leg of control participants. Leg dominance was determined by the participant's self-reported preferred leg for kicking a ball (Krishnan & Williams, 2014; van Melick et al., 2017). After cleaning the skin over the electrode stimulation site with alcohol wipes, two 5 cm square Dura-Stick Plus self-adhesive electrodes were placed on the medial and lateral gastrocnemius muscles prior to testing. A high-voltage, constant-current electrical stimulator (DS7AH; Digitimer North America, LLC) was connected to the stimulation electrodes to electrically stimulate the plantarflexors. The plantarflexors were stimulated with pulse trains (3 pulses, 100 Hz, 200-μs pulse duration, 400 V) at rest in increasing intervals of 100 mA until the electrically evoked torque from the stimulated muscle contractions no longer increased. The current intensity was then decreased by 50 mA and a final stimulation was delivered. The current intensity that elicited the largest evoked ankle plantarflexion torque at rest was used for testing (Garcia et al., 2022). Participants performed a brief warm-up consisting of a series of submaximal isometric ankle plantarflexion contractions, performing two repetitions at 25%, 50%, and 75% of their perceived maximum effort, as well as one repetition at 100% effort. After a brief rest, participants performed three 5-s MVIC trials with 2 min of rest between trials. During each MVIC, participants received visual feedback of their torque output and strong verbal encouragement was given to facilitate maximum effort. A validated, automated torque-based triggered approach (Krishnan et al., 2009) was used to electrically stimulate the plantarflexors and assess the percent voluntary activation of the plantarflexor muscles. The electrical stimulation delivered during MVIC was set to the optimal current intensity determined as previously explained. Plantarflexor strength was determined as the voluntary peak torque prior to the electrical stimulation, and was normalized to body mass (N•m/kg). Estimates of voluntary activation were computed for the CAR and ITT techniques using Equation 1 and Equation 2, respectively.

C A R (% V o l u n t a r y A c t i v a t i o n) = \frac{b}{a} \times 100

(1)

I T T (% V o l u n t a r y A c t i v a t i o n) = (1 - \frac{a - b}{c}) \times 100

(2)

$a = total torque evoked by the electrical stimulus$

b = torque at the time of electrical stimulus during MVIC

c = evoked torque at rest

Voluntary activation estimates <90% for ITT and <95% for CAR were indicative of voluntary activation deficits, as these criteria are commonly used as a threshold for meaningful deficits and given that the measurement techniques have some level of measurement error due to methodological factors as well as participant characteristics (Coventry et al., 2025; Garcia et al., 2022; Kooistra et al., 2007; Lynch et al., 2012; Shield & Zhou, 2004; Stevens et al., 2003).

Data Management and Statistical Analysis

Data collection and analysis were performed using custom-written LabView programs. Torque signals and Digitimer synchronization pulses were sampled at 1000 Hz. Torque data were filtered using a digital bandpass filter (zero-lag Butterworth filter, 4^th order, 10 Hz cut-off). Sync data were used to identify stimulation onset and the maximum amplitude occurring prior to stimulation for plantarflexor strength. The peak plantarflexor strength and voluntary activation estimates determined from the three MVIC trials were used for analyses.

All statistical analyses were performed using JASP (Version 0.19.3.0, JASP Team, 2025). Descriptive statistics were computed for all variables. Mixed design repeated measures ANOVA models were used to evaluate the effects of group and quantification technique on voluntary activation. Group was treated as a between-subject factor (stroke vs. age-matched controls; older vs. younger adults), and quantification technique (ITT vs. CAR) as a within-subject factor. In addition, a two-way repeated measures ANOVA with quantification technique (two levels: ITT and CAR) and leg (two levels: more affected and less affected) as within-subject factors was performed to evaluate the effect of quantification technique and leg on voluntary activation in stroke survivors. For all analyses, a significant main effect or interaction was followed by post-hoc comparisons using Holm-Bonferroni method to adjust for multiple comparisons. Welch's independent and paired samples t-tests were used (due to violations of homogeneity of variances) to compare raw and mass-normalized plantarflexor strength values between groups and legs, respectively. The extent of association between the voluntary activation estimates of ITT and CAR was also computed using Pearson product-moment correlation. A significance level of α = 0.05 was established for all statistical analyses. Estimates of effect size were reported using partial η2 for ANOVA and Cohen's D for t-tests.

Results

Effect of Stroke on Strength and Voluntary Activation

Plantarflexor Strength

The more affected leg of stroke survivors demonstrated significantly lower raw and mass-normalized plantarflexor strength compared with the less affected leg [Raw MVIC: t(18.00) = 2.79, p = 0.01, Cohen's d = 0.64; Normalized MVIC: t(18.00) = 2.51, p = 0.02, Cohen's d = 0.58, Table 2]. Additionally, stroke survivors demonstrated significantly lower raw and mass-normalized plantarflexor strength in the more affected leg [Raw MVIC: Welch's t(34.01) = 4.00, p < 0.01, Cohen's d = 1.30; Normalized MVIC: Welch's t(34.47) = 4.66, p < 0.01, Cohen's d = 1.51, Table 2] and the less affected leg [Raw MVIC: Welch's t(35.72) = 2.50, p = 0.02, Cohen's d = 0.81; Normalized MVIC: Welch's t(35.54) = 2.64, p = 0.01, Cohen's d = 0.86, Table 2] compared with age-matched controls.

Table 2.

Group Averaged Values for the Peak raw and Mass-Normalized Plantarflexor Strength and Voluntary Activation Estimates for the More Affected and Less Affected leg of Stroke Survivors, age-Matched Control leg, Older Control leg, and Younger Control leg.

	Stroke More Affected Leg (Mean ± SD)	Stroke Less Affected Leg (Mean ± SD)	Age-Matched Control Leg (Mean ± SD)	Older Control Leg (Mean ± SD)	Younger Control Leg (Mean ± SD)
Peak ITT (%)	66.83 ± 34.92	92.28 ± 10.97	95.75 ± 6.88	94.88 ± 7.50	96.26 ± 9.21
Peak CAR (%)	81.29 ± 24.29	97.22 ± 4.41	99.00 ± 1.76	98.79 ± 1.92	99.03 ± 2.48
Peak MVIC (Nm)	48.57 ± 33.45	67.21 ± 23.92	87.52 ± 26.14	81.25 ± 24.17	107.07 ± 32.21
Normalized Peak MVIC (Nm/kg)	0.63 ± 0.40	0.89 ± 0.36	1.18 ± 0.32	1.10 ± 0.29	1.51 ± 0.35

Abbreviations: ITT, interpolated twitch technique; CAR, central activation ratio; kg, kilograms; MVIC, maximum voluntary isometric contraction; N•m, newton•meters; %, percentage.

Voluntary Activation

When comparing the more affected leg and less affected leg of stroke survivors, there was a significant leg-by-quantification technique interaction [F(1,18) = 5.86, partial η2 = 0.25, p = 0.03, Supplementary Table 1]. Post-hoc pairwise comparisons revealed that voluntary activation was significantly lower for the more affected leg compared with the less affected leg for both ITT (p = 0.02, Figure 2A) and CAR (p = 0.02, Figure 2B), indicating that both techniques were sensitive to detect differences between legs. In addition, ITT voluntary activation estimates were significantly lower compared with CAR estimates for the more affected leg (p < 0.01, ITT: 66.83 ± 34.92, CAR: 81.29 ± 24.29) and less affected leg (p = 0.02, ITT: 92.28 ± 10.97, CAR: 97.22 ± 4.41).

Figure 2.

Raincloud plots depicting comparisons of voluntary activation between the more and less affected legs of stroke survivors using A. ITT, and B. CAR. Individual data points are shown with lines connecting paired values from the same participant. Box plots display the mean (center line), interquartile range (box) and full range (whiskers). Rainclouds and box plots show the distribution of voluntary activation values across participants. Abbreviations: ITT, interpolated twitch technique; CAR, central activation ratio.

When comparing the more affected leg of stroke survivors and the age-matched control leg, there was a significant group-by-quantification technique interaction effect on voluntary activation [F(1,36) = 7.89, partial η2 = 0.18, p < 0.01, Supplementary Table 2]. Post-hoc comparisons revealed that voluntary activation was significantly lower in the more affected leg of stroke survivors compared with control participants when using ITT (p < 0.01, More affected: 66.83 ± 34.92, Control: 95.75 ± 6.88, Figure 3A) and CAR (p < 0.01, More affected: 81.29 ± 24.29, Control: 99.00 ± 1.76, Figure 3B), indicating that both techniques were sensitive to detect differences between groups. However, voluntary activation estimates were significantly affected by the quantification technique in stroke survivors but not in the control group. Specifically, ITT voluntary activation estimates were significantly lower than CAR estimates in stroke survivors (p < 0.01) but not in controls (p = 0.26).

Figure 3.

Raincloud plots depicting comparisons of voluntary activation between A. control leg and more affected leg using ITT, B. control leg and more affected leg using CAR, C. control leg and less affected leg using ITT, D. control leg and less affected leg using CAR. Individual data points represent an individual participant. Box plots display the mean (center line), interquartile range (box) and full range (whiskers). Rainclouds and box plots show the distribution of voluntary activation values across participants. Abbreviations: ITT, interpolated twitch technique; CAR, central activation ratio.

When comparing the less affected leg and control leg, there was no significant main effect of group [F(1,36) = 1.71, partial η2 = 0.05, p = 0.20, Supplementary Table 3, Figures 3C & 3D]. However, there was a significant main effect of quantification technique [F(1,36) = 17.44, partial η2 = 0.33, p < 0.01, Table 2]. Post-hoc tests revealed CAR estimates were significantly higher than ITT estimates (p < 0.01), indicating CAR underestimated voluntary activation regardless of group. There was no significant group-by-quantification technique interaction [F(1,36) = 0.74, partial η2 = 0.02, p = 0.39, Table 2].

Effect of Aging on Strength and Voluntary Activation

Older adults demonstrated significantly lower raw and mass-normalized plantarflexor strength compared with young adults [Raw MVIC: Welch's t(31.00) = 2.63, p = 0.01, Cohen's d = 0.91; Normalized MVIC: Welch's t(31.00) = 3.68, p < 0.01, Cohen's d = 1.28]. However, there was no significant main effect of group [F(1,31) = 9.21, partial η2 = 0.01, p = 0.67, Supplementary Table 4, Figures 4A & 4B] or group-by-quantification technique interaction effect on voluntary activation [F(1,31) = 0.27, partial η2 = 0.01, p = 0.61]. When evaluating the main effect of quantification technique, CAR yielded higher voluntary activation estimates than ITT across both age groups [F(1,31) = 9.21, partial η2 = 0.23, p < 0.01] (Table 2).

Figure 4.

Raincloud plots depicting comparisons of voluntary activation between legs of older and younger control participants using A. ITT, and B. CAR. Individual data points represent an individual participant. Box plots display the mean (center line), interquartile range (box) and full range (whiskers). Rainclouds and box plots show the distribution of voluntary activation values across participants. Abbreviations: ITT, interpolated twitch technique; CAR, central activation ratio.

Relationship Between ITT and CAR

There were significant correlations between voluntary activation estimates obtained from ITT and CAR (r = 0.92, p < 0.001). The mapping equation between this linear relationship is the following: $I T T = 1.44 C A R - 47.74$ , which is roughly : $I T T = 1.5 C A R - 50$ .

Discussion

The purpose of this study was to comprehensively evaluate the effect of stroke and aging on plantarflexor strength and voluntary activation and investigate how quantification technique affects voluntary activation estimates in both stroke and neurologically intact adults. Several key findings emerged from this study. First, as hypothesized, the stroke survivors’ more affected leg demonstrated lower plantarflexor strength and voluntary activation compared with both their less affected leg and the age-matched control leg, Second, although stroke survivors exhibited bilateral deficits in plantarflexor strength (both raw and normalized) when compared with the age-matched control group, contrary to our hypothesis, voluntary activation deficits were not observed in the less affected leg of the stroke survivors. Third, as hypothesized, older adults demonstrated lower plantarflexor strength when compared with young adults. Fourth, contrary to our hypothesis, older adults did not exhibit significant voluntary activation deficits when compared with young adults. Finally, as hypothesized, CAR demonstrated higher plantarflexor voluntary activation estimates compared with ITT. These findings, provide a comprehensive understanding of the stroke and aging related alterations in the neuromuscular function of the plantarflexors.

A key finding of this study was that the more affected leg demonstrated lower plantarflexor strength and voluntary activation compared with the less affected leg and age-matched control leg. These findings extend the results of prior studies by demonstrating that stroke survivors exhibit bilateral plantarflexor strength deficits compared with age-matched controls (Fimland et al., 2011; Klein et al., 2010). Further, voluntary activation deficits were present only in the more affected leg, which was surprising considering that bilateral strength deficits were observed, and voluntary activation deficits have been shown to be the primary source of post-stroke plantarflexor weakness (Klein et al., 2010). The voluntary activation deficits in the more affected leg are likely explained by some combination of diminished motor unit recruitment and/or firing rates (Frontera et al., 1997; Jakobsson et al., 1991). However, the observation of bilateral weakness with unilateral activation deficits suggests other factors (e.g., muscle atrophy) likely contribute to the plantarflexor weakness in the less affected leg of stroke survivors (Jorgensen & Jacobsen, 2001; Klein et al., 2010). Although evidence is conflicting on whether significant muscle atrophy occurs in the plantarflexors of the less affected leg compared with controls (D’Souza et al., 2020; Hunnicutt & Gregory, 2017), decreased muscle volume could contribute to reduced force generation during maximal contractions. Regardless of the potential mechanisms contributing to reduced plantarflexor strength in the less affected leg, these changes persist long after stroke, highlighting the need to address plantarflexor dysfunction bilaterally during rehabilitation.

Another notable finding of this study was that aging contributed to lower plantarflexor strength, but not voluntary activation. These aging-related reductions in plantarflexor strength are consistent with previous literature (Dalton et al., 2014; Scaglioni et al., 2016) and confirms previous evidence demonstrating the lack of aging-related changes in voluntary activation (Dalton et al., 2014; Scaglioni et al., 2016). Although as a group, older adults did not exhibit reductions in voluntary activation (i.e., <90%), it is to be noted that a small minority of participants in both groups demonstrated reductions in voluntary activation. This finding is consistent with prior studies indicating that aging does not influence plantarflexor voluntary activation (Barber et al., 2013; Simoneau et al., 2005); however, is in conflict with some studies that have reported decreased voluntary activation in older adults (Akagi et al., 2018; Morse et al., 2004). One potential explanation for this discrepancy in study results could be related to the differences in the quantification technique that was used to estimate voluntary activation. However, we did not observe aging-related differences in voluntary activation irrespective of whether CAR or ITT was used, indicating that quantification technique may not explain this discrepancy. Another potential factor that could explain the conflicting results in the literature could be related to fatigue during testing. For instance, studies reporting no aging-related differences in voluntary activation, including this study, provided at least two minutes of rest between MVIC trials (Dalton et al., 2014; Scaglioni et al., 2016), while those reporting differences provided inadequate rest (≤1 min) (Akagi et al., 2018; Morse et al., 2004). Thus, it is likely that studies reporting voluntary activation differences between older and young adults could have been confounded by fatigue, as inadequate rest would diminish the ability to produce true maximal contractions, thereby reducing voluntary activation estimates. Nonetheless, given that aging does not appear to affect the ability to drive the muscles maximally during a contraction (i.e., voluntary activation), the reduction in plantarflexor strength is most likely stemming from age-related reductions in muscle size (i.e., sarcopenia) (Abellan van Kan et al., 2011). Therefore, interventions designed to address plantarflexor strength deficits, rather than voluntary activation deficits, should be considered to improve gait and balance in older adults.

An interesting finding of the current study is that CAR is equally sensitive to detect differences in voluntary activation, even though it generally demonstrated higher plantarflexor voluntary activation estimates compared with ITT. This finding extends previous results in neurologically intact populations (Garcia et al., 2022; Krishnan & Williams, 2010) to stroke survivors. The CAR estimates were generally much higher than ITT in the more affected leg (∼21% increase) when reductions in voluntary activation were larger, whereas when reductions in voluntary activation were minimal (e.g., in the less affected leg and control leg), estimates were only slightly higher (≤5%). This observation is consistent with prior findings where it has been shown that the differences in voluntary activation estimates between CAR and ITT are more pronounced only when there are larger reductions in voluntary activation (Krishnan & Williams, 2010; Zarkou et al., 2017). The observation of higher voluntary activation estimates with CAR than ITT is likely due to the assumption that the combination of the electrically-induced torque and the MVIC torque can evoke the true maximal torque, which cannot be achieved in practice even with maximal stimulation (Shield & Zhou, 2004). Additionally, it is also possible that the evoked torque at rest in ITT could have been underestimated due to mechanical factors such as tendon slack, axis misalignment, etc. Regardless, it is important to recognize that our findings confirmed previous work in healthy control and knee injured populations demonstrating that CAR and ITT are highly correlated measures and both are sensitive to detect reductions in voluntary activation (Garcia et al., 2022; Krishnan & Williams, 2010). Specifically, we found a strong correlation between CAR and ITT (r = 0.92), supporting the use of both methods to estimate voluntary activation. Further, CAR generally is more comfortable for participants, as it does not require an electrical stimuli at rest, making it more appropriate for populations with low pain tolerance (Garcia et al., 2022; Grindstaff & Threlkeld, 2014). In contrast, the evoked torque at rest provides important information regarding peripheral changes in the muscle, making ITT more suitable for experiments focusing on muscle physiology.

Notably, the current study demonstrates that there are cumulative changes in plantarflexor strength due to both stroke and aging, while only stroke influenced voluntary activation. Contrary to our findings, a recent meta-analysis reported that age-related effects on voluntary activation do occur in stroke survivors (Coventry et al., 2025). However, these age-related effects demonstrated very low certainty and included studies with predominantly middle-age and older participants. Thus, the conflicting findings are likely due to methodological differences such as stimulation technique (tibial nerve vs. neuromuscular stimulation) and knee and ankle positioning, as well as differences in participants (e.g., age, stroke severity, time since stroke onset, etc.). In addition, this study is the first to report cumulative effects of stroke and aging on plantarflexor strength. These cumulative plantarflexor strength deficits have strong clinical implications, as these muscles generate the majority of propulsive force during gait (Zelik & Adamczyk, 2016), which is critical for gait function. In fact, prior studies have shown that plantarflexor weakness negatively affects gait speed in both older adults and stroke survivors (Kanayama et al., 2022; Kim & Eng, 2003; Nadeau et al., 1999). Hence, these cumulative effects underscore the need to restore plantarflexor strength, particularly in older stroke survivors with greater weakness. Additionally, given that voluntary activation deficits have been shown to contribute to a greater extent than atrophy to plantarflexor weakness post-stroke (Klein et al., 2010), restoring voluntary activation appears to be critical to improving plantarflexor strength. Thus, interventions that are capable of improving both muscle strength and voluntary activation, such as high-intensity neuromuscular electrical stimulation, light or heavy resistance power training, and eccentric exercise (Lattouf et al., 2021; Lee et al., 2016; Reid et al., 2015; Snyder-Mackler et al., 1994), may be particularly effective for addressing concurrent deficits in plantarflexor strength and voluntary activation post-stroke.

There are some limitations to this study. First, this study only evaluated plantarflexor strength and voluntary activation in chronic (>6 months) stroke survivors, with most participants several years post-stroke. While previous research has reported plantarflexor deficits in stroke survivors at least two years post-stroke (Klein et al., 2010), we cannot generalize the results from the current study to stroke survivors in the acute or subacute stages. Based on existing literature on the quadriceps, we anticipate that chronic stroke survivors would demonstrate higher strength and voluntary activation, as they may have regained muscle function through spontaneous recovery and/or rehabilitation (Newham & Hsiao, 2001). However, it is also possible that chronic inactivity may contribute to lower voluntary activation compared with acute or sub-acute phases, particularly after the completion of rehabilitation. Therefore, future research is required to fully understand the longitudinal changes in plantarflexor strength and voluntary activation after stroke. In addition, it is possible that the muscle-tendon unit could have a small amount of slack in the testing position used, which could reduce the evoked torque at rest and thereby affect voluntary activation estimates (Herbert et al., 2011; Landin et al., 2015; Lieber & Friden, 2000). However, a dorsiflexed position aligned with ankle joint center was used to minimize potential slack in the ankle joint and to ensure maximal measurements of plantarflexor strength and evoked torque at rest (Maganaris et al., 1998; Sale et al., 1982; Simoneau et al., 2007). Further, CAR estimates, which do not require evoked torque at rest, also showed significant voluntary activation deficits, indicating that slack in muscle-tendon unit did not confound the study results. Finally, we used triplets (i.e., 3 pulses) instead of 5 or 10 pulses to quantify voluntary activation deficits to minimize discomfort. While voluntary activation estimates from ITT are not typically affected by the number of pulses used during testing, it is likely that the CAR estimates could have been higher than expected due to the number of pulses used (Garcia et al., 2022; Grindstaff & Threlkeld, 2014).

Conclusion

The findings of this study provide a comprehensive understanding of the effect of stroke and aging on plantarflexor strength and voluntary activation, as well as the impact of quantification technique on voluntary activation estimates. Specifically, stroke survivors exhibited bilateral plantarflexor strength deficits; however, voluntary activation deficits were only observed in the more affected leg. Additionally, older adults exhibited reduced plantarflexor strength, but not voluntary activation, compared with young adults. Finally, voluntary activation estimates obtained via CAR were consistently higher relative to ITT; however, CAR was equally sensitive to detect voluntary activation failure when compared with ITT. Collectively, these findings advance our understanding of neuromuscular changes associated with stroke and aging. The findings also highlight the need to target plantarflexor strength and voluntary activation during post-stroke rehabilitation, while carefully considering the quantification technique used to estimate voluntary activation.

Supplemental Material

sj-docx-1-rnn-10.1177_09226028261425424 - Supplemental material for Effect of Stroke and Aging on Plantarflexor Strength and Voluntary Activation

Supplemental material, sj-docx-1-rnn-10.1177_09226028261425424 for Effect of Stroke and Aging on Plantarflexor Strength and Voluntary Activation by Danny Shin, Kazandra M Rodriguez, Edward S Claflin and Chandramouli Krishnan in Restorative Neurology and Neuroscience

Footnotes

Acknowledgments

The author(s) have no other disclosures or acknowledgements to report.

Ethical Considerations and Consent to Participate

All participants read and signed a written informed consent document approved by the University of Michigan Medical School Institutional Review Board (IRBMED HUM00164858).

Consent for Publication

Not applicable.

Author Contributions

C.K. contributed to study conception and design. C.K., D.S., K.M.R., and E.S.C. contributed to the acquisition, analysis, and/or interpretation of data. C.K., D.S., and K.M.R., contributed to drafting the manuscript. C.K., D.S., K.M.R., and E.S.C. contributed to critically reviewing and editing the manuscript. C.K., D.S., K.M.R., and E.S.C. read and approved the final manuscript.

Funding

This work was supported by the National Institutes of Health Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01-HD111567) and the NIH StrokeNet – University of Michigan Regional Coordinating Stroke Center (5U24NS107214). Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the funding sources.

National Institutes of Health- University of Michigan Regional Coordinating Stroke Center, National Institutes of Health, (grant number 5U24NS107214, R01-HD111567).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability

The data supporting the findings of this study are available on request from the corresponding author.

Supplemental Material

Supplemental material for this article is available online.

ORCID iDs

Danny Shin

Kazandra M Rodriguez

Edward S Claflin

Chandramouli Krishnan

References

Abellan van Kan

Cderbaum

J. M.

Cesari

Dahinden

Fariello

R. G.

Fielding

R. A.

Goodpaster

B. H.

Hettwer

Isaac

Laurent

Morley

J. E.

Pahor

Rooks

Roubenoff

Rutkove

S. B.

Shaheen

Vamvakas

Vrijbloed

J. W.

Vellas

(2011). Sarcopenia: Biomarkers and imaging (international conference on sarcopenia research). The Journal of Nutrition, Health and Aging, 15(10), 834–846. https://doi.org/10.1007/s12603-011-0365-1

Aguirre-Villanueva

E. A.

Cano-de-la-Cuerda

Laguarta-Val

(2025). Isokinetic muscle strengthening of the lower limbs in people with stroke: A systematic review of randomized controlled trials. Journal of Clinical Medicine, 14(7), 2215. https://doi.org/10.3390/jcm14072215

Akagi

Suzuki

Kawaguchi

Miyamoto

Yamada

Ema

(2018). Muscle size-strength relationship including ultrasonographic echo intensity and voluntary activation level of a muscle group. Archives of Gerontology and Geriatrics, 75, 185–190. https://doi.org/10.1016/j.archger.2017.12.012

Barber

L. A.

Barrett

R. S.

Gillett

J. G.

Cresswell

A. G.

Lichtwark

G. A.

(2013). Neuromechanical properties of the triceps surae in young and older adults. Experimental Gerontology, 48(11), 1147–1155. https://doi.org/10.1016/j.exger.2013.07.007

Coventry

M. E.

Latella

Green

Mosler

A. B.

Peak

Perez Armendariz

M. L.

Rio

E. K.

Murphy

M. C.

(2025). Voluntary activation of the ankle plantar flexors: A systematic review and meta-analysis. JSAMS Plus, 6, 100117. https://doi.org/10.1016/j.jsampl.2025.100117

Dalton

B. H.

Allen

M. D.

Power

G. A.

Vandervoort

A. A.

Rice

C. L.

(2014). The effect of knee joint angle on plantar flexor power in young and old men. Experimental gerontology, 52, 70–76. https://doi.org/10.1016/j.exger.2014.01.011

Dalton

B. H.

Harwood

Davidson

A. W.

Rice

C. L.

(2009). Triceps surae contractile properties and firing rates in the soleus of young and old men. Journal of Applied Physiology, 107(6), 1781–1788. https://doi.org/10.1152/japplphysiol.00464.2009

Dorsch

Ada

Sorial

Fanayan

(2021). The relationship between strength of the affected leg and walking speed after stroke varies according to the level of walking disability: A systematic review. Physical Therapy, 101(12): pzab233. https://doi.org/10.1093/ptj/pzab233

D'Souza

Bolsterlee

Herbert

R. D.

(2020). Architecture of the medial gastrocnemius muscle in people who have had a stroke: A diffusion tensor imaging investigation. Clinical Biomechanics, 74, 27–33. https://doi.org/10.1016/j.clinbiomech.2020.02.004

10.

Fimland

M. S.

Moen

P. M.

Hill

Gjellesvik

T. I.

Torhaug

Helgerud

Hoff

(2011). Neuromuscular performance of paretic versus non-paretic plantar flexors after stroke. European Journal of Applied Physiology, 111(12), 3041–3049. https://doi.org/10.1007/s00421-011-1934-z

11.

Frontera

W. R.

Grimby

Larsson

(1997). Firing rate of the lower motoneuron and contractile properties of its muscle fibers after upper motoneuron lesion in man. Muscle & Nerve, 20(8), 938–947. https://doi.org/10.1002/(SICI)1097-4598(199708)20:8<938::AID-MUS2>3.0.CO;2-7

12.

Garcia

S. A.

Rodriguez

K. M.

Brown

S. R.

Palmieri-Smith

R. M.

Krishnan

(2022). Estimates of voluntary activation in individuals with anterior cruciate ligament reconstruction: Effects of type of stimulator, number of stimuli, and quantification technique. Journal of Sport and Health Science, 11(1), 85–93. https://doi.org/10.1016/j.jshs.2019.12.001

13.

Grindstaff

T. L.

Threlkeld

A. J.

(2014). Optimal stimulation parameters to detect deficits in quadriceps voluntary activation. Journal of Strength and Conditioning Research, 28(2), 381–389. https://doi.org/10.1519/JSC.0b013e3182986d5f

14.

Herbert

R. D.

Clarke

Kwah

L. K.

Diong

Martin

Clarke

E. C.

Bilston

L. E.

Gandevia

S. C.

(2011). In vivo passive mechanical behaviour of muscle fascicles and tendons in human gastrocnemius muscle-tendon units. The Journal of Physiology, 589(Pt 21), 5257–5267. https://doi.org/10.1113/jphysiol.2011.212175

15.

Hunnicutt

J. L.

Gregory

C. M.

(2017). Skeletal muscle changes following stroke: A systematic review and comparison to healthy individuals. Topics in Stroke Rehabilitation, 24(6), 463–471. https://doi.org/10.1080/10749357.2017.1292720

16.

Jakobsson

Edstrom

Grimby

Thornell

L. E.

(1991). Disuse of anterior tibial muscle during locomotion and increased proportion of type II fibres in hemiplegia. Journal of the Neurological Sciences, 105(1), 49–56. https://doi.org/10.1016/0022-510X(91)90117-P

17.

Jorgensen

Jacobsen

B. K.

(2001). Changes in muscle mass, fat mass, and bone mineral content in the legs after stroke: A 1 year prospective study. Bone, 28(6), 655–659. https://doi.org/10.1016/S8756-3282(01)00434-3

18.

Kanayama

Minami

Yamamoto

Ohmine

Fujiwara

Murakami

Okuno

Ueba

Iwata

(2022). Examination of the impact of strength and velocity of the knee and ankle on gait speed in community-dwelling older adults. Healthcare (Basel), 10(10), 2093. https://doi.org/10.3390/healthcare10102093

19.

Kent-Braun

J. A.

Le Blanc

(1996). Quantitation of central activation failure during maximal voluntary contractions in humans. Muscle & Nerve, 19(7), 861–869. https://doi.org/10.1002/(SICI)1097-4598(199607)19:7<861::AID-MUS8>3.0.CO;2-7

20.

Kim

C. M.

Eng

J. J.

(2003). The relationship of lower-extremity muscle torque to locomotor performance in people with stroke. Physical Therapy, 83(1), 49–57. https://doi.org/10.1093/ptj/83.1.49

21.

Klein

C. S.

Brooks

Richardson

McIlroy

W. E.

Bayley

M. T.

(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. https://doi.org/10.1152/japplphysiol.00804.2009

22.

Klein

C. S.

Power

G. A.

Brooks

Rice

C. L.

(2013). Neural and muscular determinants of dorsiflexor weakness in chronic stroke survivors. Motor Control, 17(3), 283–297. https://doi.org/10.1123/mcj.17.3.283

23.

Knorr

Ivanova

T. D.

Doherty

T. J.

Campbell

J. A.

Garland

S. J.

(2011). The origins of neuromuscular fatigue post-stroke. Experimental Brain Research, 214(2), 303–315. https://doi.org/10.1007/s00221-011-2826-5

24.

Kooistra

R. D.

de Ruiter

C. J.

de Haan

(2007). Conventionally assessed voluntary activation does not represent relative voluntary torque production. European Journal of Applied Physiology, 100(3), 309–320. https://doi.org/10.1007/s00421-007-0425-8

25.

Krishnan

Allen

E. J.

Williams

G. N.

(2009). Torque-based triggering improves stimulus timing precision in activation tests. Muscle & Nerve, 40(1), 130–133. https://doi.org/10.1002/mus.21279

26.

Krishnan

Williams

G. N.

(2010). Quantification method affects estimates of voluntary quadriceps activation. Muscle & Nerve, 41(6), 868–874. https://doi.org/10.1002/mus.21613

27.

Krishnan

Williams

G. N.

(2011). Factors explaining chronic knee extensor strength deficits after ACL reconstruction. Journal of Orthopaedic Research, 29(5), 633–640. https://doi.org/10.1002/jor.21316

28.

Krishnan

Williams

G. N.

(2014). Effect of knee joint angle on side-to-side strength ratios. Journal of Strength and Conditioning Research, 28(10), 2981–2987. https://doi.org/10.1519/JSC.0000000000000476

29.

Landin

Thompson

Reid

(2015). Knee and ankle joint angles influence the plantarflexion torque of the gastrocnemius. Journal of Clinical Medicine Research, 7(8), 602–606. https://doi.org/10.14740/jocmr2107w

30.

Lattouf

N. A.

Tomb

Assi

Maynard

Mesure

(2021). Eccentric training effects for patients with post-stroke hemiparesis on strength and speed gait: A randomized controlled trial. NeuroRehabilitation, 48(4), 513–522. https://doi.org/10.3233/NRE-201601

31.

Lee

Jeong

(2016). Mirror therapy with neuromuscular electrical stimulation for improving motor function of stroke survivors: A pilot randomized clinical study. Technology and Health Care, 24(4), 503–511. https://doi.org/10.3233/THC-161144

32.

Liang

S. G.

Chow

J. C. M.

Leung

N. M.

Y. N.

T. M. H.

Woo

C. L. C.

Lam

F. M. H.

(2025). The effects of ankle and foot exercises on ankle strength, balance, and falls in older people: A systematic review and meta-analysis. Physical Therapy, 105(1), pzae157. https://doi.org/10.1093/ptj/pzae157

33.

Lieber

R. L.

Friden

(2000). Functional and clinical significance of skeletal muscle architecture. Muscle & Nerve, 23(11), 1647–1666. https://doi.org/10.1002/1097-4598(200011)23:11%3C1647::AID-MUS1%3E3.0.CO;2-M

34.

Lynch

A. D.

Logerstedt

D. S.

Axe

M. J.

Snyder-Mackler

(2012). Quadriceps activation failure after anterior cruciate ligament rupture is not mediated by knee joint effusion. Journal of Orthopaedic & Sports Physical Therapy, 42(6), 502–510. https://doi.org/10.2519/jospt.2012.3793

35.

Maganaris

C. N.

Baltzopoulos

Sargeant

A. J.

(1998). Differences in human antagonistic ankle dorsiflexor coactivation between legs; can they explain the moment deficit in the weaker plantarflexor leg? Experimental Physiology, 83(6), 843–855. https://doi.org/10.1113/expphysiol.1998.sp004164

36.

Morse

C. I.

Thom

J. M.

Davis

M. G.

Fox

K. R.

Birch

K. M.

Narici

M. V.

(2004). Reduced plantarflexor specific torque in the elderly is associated with a lower activation capacity. European Journal of Applied Physiology, 92(1-2), 219–226. https://doi.org/10.1007/s00421-004-1056-y

37.

Nadeau

Gravel

Arsenault

A. B.

Bourbonnais

(1999). Plantarflexor weakness as a limiting factor of gait speed in stroke subjects and the compensating role of hip flexors. Clinical Biomechanics, 14(2), 125–135. https://doi.org/10.1016/S0268-0033(98)00062-X

38.

Newham

D. J.

Hsiao

S. F.

(2001). Knee muscle isometric strength, voluntary activation and antagonist co-contraction in the first six months after stroke. Disability and Rehabilitation, 23(9), 379–386. https://doi.org/10.1080/0963828001006656

39.

Olsen

Signal

Niazi

I. K.

Alder

Rashid

Nedergaard

R. B.

Taylor

(2021). Reliability of tibialis anterior muscle voluntary activation using the interpolated twitch technique and the central activation ratio in people with stroke. Brain Sciences, 11(2), 176. https://doi.org/10.3390/brainsci11020176

40.

Reid

K. F.

Martin

K. I.

Doros

Clark

D. J.

Hau

Patten

Phillips

E. M.

Frontera

W. R.

Fielding

R. A.

(2015). Comparative effects of light or heavy resistance power training for improving lower extremity power and physical performance in mobility-limited older adults. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 70(3), 374–380. https://doi.org/10.1093/gerona/glu156

41.

Rozand

Sundberg

C. W.

Hunter

S. K.

Smith

A. E.

(2020). Age-related deficits in voluntary activation: A systematic review and meta-analysis. Medicine & Science in Sports & Exercise, 52(3), 549–560. https://doi.org/10.1249/MSS.0000000000002179

42.

Sale

Quinlan

Marsh

McComas

A. J.

Belanger

A. Y.

(1982). Influence of joint position on ankle plantarflexion in humans. The Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology, 52(6), 1636–1642. https://doi.org/10.1152/jappl.1982.52.6.1636

43.

Scaglioni

Narici

M. V.

Martin

(2016). Neural activation during submaximal contractions seems more reflective of neuromuscular ageing than maximal voluntary activation. Frontiers in Aging Neuroscience, 8, 19. https://doi.org/10.3389/fnagi.2016.00019

44.

Shield

Zhou

(2004). Assessing voluntary muscle activation with the twitch interpolation technique. Sports Medicine, 34(4), 253–267. https://doi.org/10.2165/00007256-200434040-00005

45.

Simoneau

Martin

Van Hoecke

(2005). Muscular performances at the ankle joint in young and elderly men. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 60(4), 439–447. https://doi.org/10.1093/gerona/60.4.439

46.

Simoneau

Martin

Van Hoecke

(2007). Effects of joint angle and age on ankle dorsi- and plantar-flexor strength. Journal of Electromyography and Kinesiology, 17(3), 307–316. https://doi.org/10.1016/j.jelekin.2006.04.005

47.

Snyder-Mackler

Delitto

Stralka

S. W.

Bailey

S. L.

(1994). Use of electrical stimulation to enhance recovery of quadriceps femoris muscle force production in patients following anterior cruciate ligament reconstruction. Physical Therapy, 74(10), 901–907. https://doi.org/10.1093/ptj/74.10.901

48.

Stevens

J. E.

Stackhouse

S. K.

Binder-Macleod

S. A.

Snyder-Mackler

(2003). Are voluntary muscle activation deficits in older adults meaningful? Muscle & Nerve, 27(1), 99–101. https://doi.org/10.1002/mus.10279

49.

Todd

Gorman

R. B.

Gandevia

S. C.

(2004). Measurement and reproducibility of strength and voluntary activation of lower-limb muscles. Muscle & Nerve, 29(6), 834–842. https://doi.org/10.1002/mus.20027

50.

Tuominen

Leppanen

Jarske

Pasanen

Vasankari

Parkkari

(2023). Test-retest reliability of isokinetic ankle, knee and hip strength in physically active adults using biodex system 4 pro. Methods and Protocols, 6(2), 26. https://doi.org/10.3390/mps6020026

51.

Vandervoort

A. A.

McComas

A. J.

(1986). Contractile changes in opposing muscles of the human ankle joint with aging. Journal of Applied Physiology, 61(1), 361–367. https://doi.org/10.1152/jappl.1986.61.1.361

52.

van Melick

Meddeler

B. M.

Hoogeboom

T. J.

Nijhuis-van der Sanden

M. W. G.

van Cingel

R. E. H.

(2017). How to determine leg dominance: The agreement between self-reported and observed performance in healthy adults. PLoS One, 12(12), e0189876. https://doi.org/10.1371/journal.pone.0189876

53.

Zarkou

Stackhouse

Binder-Macleod

S. A.

Lee

S. C. K.

(2017). Comparison of techniques to determine human skeletal muscle voluntary activation. Journal of Electromyography and Kinesiology, 36, 8–15. https://doi.org/10.1016/j.jelekin.2017.05.011

54.

Zelik

K. E.

Adamczyk

P. G.

(2016). A unified perspective on ankle push-off in human walking. Journal of Experimental Biology, 219(Pt 23), 3676–3683. https://doi.org/10.1242/jeb.140376

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.03 MB

0.00 MB