Homosynaptic Depression of the Soleus Predicts Post-Stroke Gait Speed

Abstract

Homosynaptic depression (HD) refers to the reduction in the magnitude of the monosynaptic spinal reflex resulting from prior activation of the circuit, often evoked with the H-reflex. Previous literature has reported HD of the soleus H-reflex is reduced post-stroke. However, it remains unclear if HD plays a role in functional impairments. The goal of this study was to characterize HD of the soleus H-reflex in individuals with post-stroke gait impairments and examine the relationship with functional measures of gait. Our results revealed that individuals after stroke experienced reduced depression at longer (8s) interstimulus intervals compared to age-matched neurologically intact individuals. However, we did not observe a difference in the change in HD across interstimulus intervals between groups, contrary to previous reports. This finding could not be explained by age of participants. In addition, we found a strong correlation between faster gait speed and reduced change in depression in individuals after stroke. While the underlying mechanisms linking HD with gait are unclear, this finding represents the first piece of evidence of the potential role of HD in function. Further research is needed to understand the parameters that guide HD and clarify how useful the mechanism is for improving the assessment and treatment of post-stroke impairments.

Keywords

homosynaptic depression spinal reflex modulation H-reflex stroke hyperreflexia spasticity gait

Introduction

Hyperactive spinal reflex excitability, a component of spasticity known specifically as hyperreflexia, is a common symptom of stroke affecting up to 80% of survivors (Menoux et al., 2019). Hyperreflexia has been linked to post-stroke gait impairments such as Stiff-Knee gait, where hyperactive quadriceps reflexes reduce knee flexion during the swing phase of gait (Akbas et al., 2020). While common, hyperreflexia lacks a consistently effective treatment (Tenniglo et al., 2023), highlighting the need to characterize the underlying neurophysiology.

Across literature, several mechanisms found within the spinal cord have been theorized to contribute to hyperreflexia including reciprocal inhibition, heteronymous facilitation, and homosynaptic depression (Aymard et al., 2000; Colard et al., 2024; Dyer et al., 2009; Lamy et al., 2009; Metz et al., 2023; Pierrot-Deseilligny & Burke, 2012). In this manuscript, we focus on homosynaptic depression (HD), alternatively referred to as post-activation depression (Pierrot-Deseilligny & Burke, 2012). HD is defined as the depression in magnitude of the monosynaptic spinal reflex resulting from the prior activation of the circuit (Crone & Nielsen, 1989; Hultborn et al., 1996; Pierrot-Deseilligny & Burke, 2012). While many mechanisms have been associated with post-stroke hyperreflexia, unlike others, the link between HD and function is largely unknown, especially pertaining to impaired function.

HD has been evaluated using a preceding tendon tap, brief agonist or antagonist muscle contraction, or reflex evoked from the homonymous pathway (Crone & Nielsen, 1989; Pierrot-Deseilligny & Burke, 2012). Initially observed in animal models, the exact mechanisms underlying HD are unclear, but are hypothesized to yield a reduced probability of neurotransmitter release at the Ia-α-motoneuron synapse (Curtis & Eccles, 1960; Curtis & Lacey, 1998; Eccles et al., 1961; Hirst et al., 1981; Kuno, 1964; Metz et al., 2023; Pierrot-Deseilligny & Burke, 2012). HD has frequently been used to test if motor activity is reflexive in humans by employing peripheral nerve stimulation to evoke a reflex (defined as the H-reflex) (Aymard et al., 2000; Crone & Nielsen, 1989; Dukkipati et al., 2023; Floeter & Kohn, 1997; Kawaishi et al., 2017; Lamy et al., 2009; Palmieri et al., 2004; Pierrot-Deseilligny & Burke, 2012; Stein et al., 2007). When utilized in neurologically intact human participants, short interstimulus intervals (ISI=1–6s) result in significant depression of H-reflex amplitude, relative to longer intervals that yield minimal (ISI=8s) to no (ISI=12s) depression (Aymard et al., 2000; Crone & Nielsen, 1989).

Recent studies have examined HD in individuals after stroke reporting reduced inhibition of reflex amplitude in clinically impaired upper (flexor carpi radialis) and lower (soleus) limb muscles compared to the contralateral limb and neurologically intact participants (Aymard et al., 2000; Lamy et al., 2009). Lamy et al. (2009) linked the reduction after stroke of reflex depression to hyperreflexia, suggesting that disrupted pre-synaptic regulation increases the efficacy or strength of the Ia afferent synaptic transmission. Along those lines, Aymard et al. (2000) suggested that the pre-synaptic and post-synaptic regulation is less efficient after stroke, resulting in the Ia afferent releasing a greater amount of neurotransmitters relative to neurologically intact individuals (Aymard et al., 2000).

Despite studies finding that depression is reduced in individuals after stroke, to date, the relationship between HD and upper or lower limb function remains poorly understood. Previously, literature has hypothesized that HD encodes long-term use-dependent plasticity by influencing the timing of muscle activation through the modulation of reflex excitability (Crone & Nielsen, 1989; Pierrot-Deseilligny & Burke, 2012). Supporting this idea, several studies have shown that both neurologically intact participants and individuals after stroke can increase inhibition of H-reflex amplitude and therefore increase pre-synaptic efficacy through physical motor training (Meunier et al., 2007; Obata et al., 2022; Trompetto et al., 2013). However, other literature has reported that HD may be disrupted following neurological injury and may be diminished during function altogether. For example, several studies have observed mixed results regarding changes in HD and their relation to functional improvements (Kiser et al., 2005; Phadke et al., 2009; Trimble et al., 1998; Trompetto et al., 2014). Notably, Mahmoud et al. (2023) found that HD did not change despite increases in corticospinal tract integrity within a heterogeneous group of individuals after stroke (Mahmoud et al., 2023). These findings coincide with data showing that HD is reduced during voluntary contraction relative to passive contraction and static conditions, suggesting it does not play any role in function (Colard et al., 2023, 2024; Stein et al., 2007). Taken together, the functional significance of HD after stroke remains unclear and provides the main motivation for the present study.

The primary goal of this study was to characterize HD of the soleus H-reflex in individuals with post-stroke gait impairments and examine the relationship with functional measures of gait. HD at short (2s), long (8s) and control (12s) interstimulus intervals (ISI) was assessed in individuals after stroke with age-matched neurologically intact participants included for comparison. Based on the work of Aymard et al. (2000) and Lamy et al. (2009), we hypothesized inhibition of the soleus H-reflex would be reduced in individuals after stroke at ISI=2s, but not ISI=8s or 12s, relative to neurologically intact participants. To investigate the functional relevance of HD post-stroke, we analyzed the correlation of HD with gait speed and reflex excitability. Building from previous literature, we hypothesized increased inhibition of the H-reflex is associated with increased gait speed and reduced reflex excitability post-stroke. Determining whether HD has functional significance can lead towards its use as a biomarker of impairment and recovery.

Methods

Participants

We recruited volunteers with no known gait-related neuromuscular impairments and individuals with post-stroke gait impairments who exhibited reduced knee flexion during swing. All participants provided written informed consent in accordance with the procedures approved by the MetroHealth Medical System Institutional Review Board. Individuals after stroke were included based on: (1) diagnosed with a stroke at least 90 days prior to enrollment, (2) sustained a gait impairment resulting from the stroke with the ability to stand upright and walk for at least 5 min, (3) presented with reduced knee flexion during swing as defined by a physical therapist, and (4) presented a soleus H-reflex. All individuals were excluded based on: (1) sustained unrelated neurological or musculoskeletal gait-related impairments, (2) ≤12 weeks since previous botulinum toxin dosage for lower limbs, and (3) implanted with an electronic device such as a cardiac pacemaker. All participants completed the 10-meter walk test overground using any assistive devices to collect their fast and comfortable gait speed (Steffen & Seney, 2008). Two repetitions were collected at each speed. Individuals after stroke also completed the Modified Ashworth Scale (MAS) and Manual Muscle Test (MMT) for their plantar-flexors and dorsiflexors (Kendall et al., 2005). MAS scores were collected with participants relaxed in the supine position with a foam roller under their knee (Bohannon & Smith, 1987).

Experimental Setup

Electromyography (EMG) was acquired using a 9mm Ag/AgCl bipolar surface electrode with contacts spaced 22mm apart (Norotrode 20, Myotronics, WA, USA). Prior to donning electrodes, the skin was abraded using electrode gel (Nuprep, Weaver and Company, CO, USA), cleaned with isopropyl alcohol wipes, and dried. EMG was sampled at 10kHz (MA400 EMG amplifier, Motion Labs, LA, USA) using a custom MATLAB application (MATLAB Version 2024b, MathWorks, MA, USA). EMG from the soleus on the clinically impaired (post-stroke) or dominant (neurologically intact) leg was recorded using electrodes positioned 2/3 down the posteromedial aspect of the shank, distal to the medial gastrocnemius (Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles, n.d.; van Melick et al., 2017). A 5 cm square electrode was placed on the patella as a reference electrode. Pre-wrap athletic tape was added over EMG electrodes to provide additional support (Cramer Tape Underwrap, Cramer Sports Medicine, IL, USA). Collected EMG signals were amplified (x350) and filtered using a 2kHz low-pass anti-aliasing filter. Transcutaneous stimulation of the tibial nerve was performed using a constant current electrical stimulator (DS8R, Digitimer, FL, USA). The nerve was found using a bar electrode (MLADDF30, AD Instruments, CO, USA) positioned medially on the posterior aspect of the popliteal fossa with the anode ∼1cm distal and 0.5cm medial to the cathode (Figure 1). Upon finding the optimal stimulation position, a surface electrode was positioned at the previous location of the bar electrode with an elastic band added to provide additional support and apply pressure (Digitimer, FL, USA).

Experimental Procedure

Participants were seated on a treatment table in a semi-recumbent position with their hip flexed 60°, knee flexed 20°, and ankle plantarflexed 20° (relative to anatomical position; Figure 1) to remain consistent with previous literature (Aymard et al., 2000; Crone & Nielsen, 1989; Lamy et al., 2009). Participants were instructed to remain relaxed and limit movement as much as possible. Due to the monotonous nature of the experimental protocol, participants were allowed to listen to music to help them remain attentive. Tibial nerve stimulation was elicited using a 1-ms pulse-width monopolar square pulse of varying amplitude. In some participants, a 0.5-ms pulse-width was utilized to help elicit a consistent H-reflex and M-wave (Lagerquist & Collins, 2008) (Table 1). A soleus H-reflex recruitment curve was performed by varying the stimulation intensity in steps of 1–2 mA while using 3 repetitions, with the purpose of identifying the optimal stimulation intensity and collecting the maximal H-reflex amplitude. Larger steps (5–10 mA) and only 2 repetitions were utilized to collect the maximal M-wave amplitude. The recruitment curve was completed by stimulating with an ISI ≥10s to avoid modulation due to homosynaptic depression (Crone & Nielsen, 1989). The optimal stimulation intensity was chosen as the current amplitude which elicited an H-reflex on the ascending portion of the recruitment curve at ∼50% of the maximal H-reflex amplitude while also producing an M-wave. HD of the soleus was assessed using 15 pairs of evoked reflexes where a conditioned H-reflex (H₂) was preceded by a test H-reflex (H₁) after a short (ISI=2s), long (ISI=8s), or control (ISI=12s) interval (Figure 1). Pairs of reflexes were collected for each ISI in a pseudo-random order with 12s between pairs.

Figure 1.

Homosynaptic depression protocol. Homosynaptic depression of the soleus was investigated using an H-reflex elicited through tibial nerve stimulation. (A) For testing, participants were instructed to remain relaxed while seated on a treatment table with their hip flexed 60°, knee flexed 20°, and ankle plantarflexed 20°. (B) The tibial nerve was found using a bar electrode positioned medially on the posterior aspect of the popliteal fossa with the anode (lower red circle) ∼1cm distal and 0.5cm medial to the cathode (upper black circle). Reflexes were evoked using a current amplitude which elicited an H-reflex on the ascending portion of the recruitment curve at ∼50% of the maximal H-reflex amplitude while also producing an M-wave. Soleus electromyography was recorded by positioning electrodes (grey circles) 2/3 down the posteromedial aspect of the shank, distal to the medial gastrocnemius (Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles, n.d.; van Melick et al., 2017). (C) Homosynaptic depression was examined by conditioning an H-reflex (H₂) with a preceding test H-reflex (H₁) at a short (2s), long (8s), or control (12s) interval. Fifteen conditioned reflexes were collected for each interval in a pseudo-random order. Reflex magnitude was computed using the peak-to-peak amplitude.

Table 1.

Participant Information.

Neurologically Intact Participants				Participants After Stroke
ID	Age (Year) Sex	Comfortable/Fast Gait Speed (m/s)	Simulation Pulse Width (mS)	Age/Sex	Comfortable/Fast Gait Speed (m/s)	Simulation Pulse Width (mS)
1	77M	1.32/1.69	1.0	77M	0.64/0.68	1.0
2	77F	1.5 /1.99	1.0	77F*	0.73/1.17	1.0
3	71F	1.16/1.49	1.0	72M	0.16/0.18	1.0
4	71M	1.7 /2.32	1.0	72M	0.89/1.59	1.0
5	66M	1.31/1.87	1.0	66F	0.27/0.37	1.0
6	64M	1.00/1.68	0.5	63M	0.80 /1.12	1.0
7	63F	1.25/1.61	1.0	62M	0.43/0.64	1.0
8	63M	1.26/2.36	1.0	62M	0.30/0.38	1.0
9	60F	1.15/1.32	1.0	60M	0.99/1.23	1.0
10	57F	1.16/1.85	1.0	59M	0.47/0.53	1.0
11	54M	1.63/2.08	1.0	58M	0.41/0.61	1.0
12	54M*	1.74/2.06	0.5	54M	0.74/0.88	0.5
13	52F	1.62/2.79	1.0	52M	1.10 /1.38	1.0
14	52M	1.39/2.38	1.0	50M	1.13/1.48	1.0
15	44M	1.17/2.09	1.0	48M	0.56/0.64	1.0
16	32M	1.22/1.72	0.5	36M	0.73/1.15	1.0
17				75M	0.40/0.43	1.0
18				69M	0.64/0.92	1.0
19				67M	0.15/0.15	1.0
20				59M	0.60/0.83	1.0
21				49M	0.31/0.36	1.0
	59.8±11.9	1.35±0.22	13–1.0	61.3±10.7	0.59±0.30	20–1.0
	6-F/10-M	1.96±0.38	3–0.5	2-F/19-M	0.80±0.43	1–0.5

The mean ± standard deviation is reported for Age and Gait Speed, while the counts are reported for Sex and Stimulation Pulse Width.

*Identified as outliers and were thus excluded from analysis.

Data Processing

EMG data was processed using custom code developed in MATLAB (MATLAB Version 2024b, MathWorks, MA, USA). EMG signals were filtered using a 3rd-order Butterworth filter from 10–1000Hz. The windows for the start and end of the H-reflex and M-wave were selected through visual inspection. Within each trial, the stimulation artifact corrupting the H-reflex and M-wave was reduced by detrending data within each window using the linear best-fit line (Akbas et al., 2020). The amplitude of each wave was then calculated using the peak-to-peak amplitude. The background signal was also extracted from trials by calculating the root-mean-square of 100ms of EMG data, 1ms prior to stimulation. Trials of HD were excluded if the background signal or M-wave amplitude was outside ± 2 standard deviations with respect to each specific participant.

Two outcomes were derived from the HD data, the normalized reflex modulation and the 2/8s ratio. The normalized reflex modulation was introduced as a new outcome measure for HD that allows for the quantification of reflex modulation relative to a reference response evoked after a baseline interval of 12s. This outcome provides a precise representation of the change in reflex amplitude as it accounts for baseline variability. The 2/8s ratio was also calculated to compare results with previous literature and measure the change in reflex modulation at short intervals relative to longer intervals where minimal depression is expected. Based on Aymard et al. (2000), we utilized the 2/8s ratio to describe the change in pre-synaptic efficacy across ISIs, while the normalized reflex modulation relates to pre-synaptic efficacy relative to baseline. For a given trial with a test reflex (H₁) followed by a conditioned reflex (H₂), the normalized reflex modulation is given as:

H_{n o r m} = \frac{H_{2}}{H_{1}} * 100

(1)

The 2/8s ratio was calculated using the average conditioned reflexes (H₂), at ISI=2s and ISI=8s such that:

2 / 8 s R a t i o = \frac{H_{2 a t I S I = 2 s}}{H_{2 a t I S I = 8 s}} * 100

(2)

Reflex excitability was calculated by normalizing the maximal H-reflex to M-wave amplitudes. Comfortable and fast gait speeds were calculated by averaging the speeds collected across two trials for each condition.

Statistical Analysis

Statistical analyses were conducted in MATLAB (MATLAB Version 2024b, MathWorks, MA, USA). The normality of each outcome within groups was assessed using Shapiro-Wilk tests and by inspecting quantile-quartile plots (Cuadra et al., 2023). Student's t-tests were utilized for normally distributed data, while non-normally distributed data used the Wilcoxon signed rank test for one-sample and paired-sample tests and the Mann-Whitney U test for two-sample tests. Participants were excluded as outliers if any outcome exceeded ± 3 standard deviations from the group average.

A linear mixed-effects model of the normalized reflex modulation was used to account for repeated measures where group and ISI were fixed effects, participant was a random effect, age was a covariate, and sex as a bivariate. Post-hoc tests compared ISIs across groups, groups across ISIs, and the presence of reflex modulation (deviation from unmodulated response (H₁ = 100%). The 2/8s ratio was analyzed to assess the presence of depression of reflex modulation at ISI=2s compared to ISI=8s and for comparison between groups. Pearson's correlation was used to evaluate the relationship between HD outcomes (normalized reflex modulation at ISI=2s and ISI=8s and the 2/8s ratio) and both gait speed and reflex excitability. Statistical significance was set to p<0.05. All statistical values are included in the Supplementary Material. Significant results are reported in the Results section with the corresponding average and mean ± standard deviation (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

Results

Participant Recruitment

We recruited a total of 21 individuals with post-stroke gait impairments and 25 neurologically intact individuals with no known neuromuscular gait impairments. Sixteen neurologically intact individuals (6 female, 10 male) were age-matched to 21 participants after stroke (2 female, 19 male). Participant data is shown in Table 1. Information about stroke-related impairments can be found in Table 2. Table 3 compares participant data of older and younger individuals. There was no significant difference in age between groups while the average difference between matched individuals was approximately 16 months (Supplementary Material Figure S1). Results of the statistical analysis showed data from ISI=8s and 12s was not normally distributed so further analyses utilized non-parametric tests (Supplementary Material Table S2).

Table 2.

Stroke-Related Impairment Information.

ID	Stroke Type	Time Since Stroke (year)	Time Between MAS/MMT and HD (day)	Plantar Flexor MAS	Dorsiflexor MAS	Plantar Flexor MMT	Dorsiflexor MMT	Assistive Devices	Taking Oral Baclofen?
1	I	2.69	0	0	0	2	4	Cane	Yes
2***	I	2.71	1	0	0	2	4	Cane
3	I	0.63	7	0	0	2	3-	Walker	Yes
4	I	2.85	6	0	0	2	4	Cane
5	H	2.32	7	0	1	2	3-	Cane	Yes
5	H	2.32	7	0	1	2	3-	AFO	Yes
6	H	4.51	0	0	0	2	5
7	I	20.40	1	0	0	2	1	AFO
8	I	12.59	3	0	0	1	1	Cane
8	I	12.59	3	0	0	1	1	AFO
9	H	7.73	3	0	0	2	5	Cane
10	I	0.59	3	4*	4*	2	1	Cane**
11	I	0.29	4	0	0	2	3-	Quad-Cane	Yes
12	H	2.39	0	0	1	2	2	Cane
13	I	1.66	6	0	0	5	5
14	I	3.43	38	4*	4*	2	3-
15	H	0.52	0	4*	4*	2	3-	Cane	Yes
16	I	2.78	2	0	0	1	0		-
17	H	10.02	0	4*	4*	2	1	Cane	Yes
18	I	14.12	1	0	0	1	3-	Cane
18	I	14.12	1	0	0	1	3-	AFO
19	I	8.37	4	4*	4*	1	3-	Quad-Cane
20	I	0.50	1	0	0	2	4	Quad-Cane
21	H	2.04	2	4*	4*	2*	3-	Quad-Cane	Yes
	7-H14-I	4.91±5.38	4.24±8.09						7-Yes

The mean ± standard deviation is reported for the Time Since Stroke and Time Between Homosynaptic Depression and collecting Modified Ashworth Scale (MAS) Scores and Manual Muscle Test (MM) Scores. The counts are reported for Type of Stroke, and if participants were taking Oral Baclofen during testing.

*Individuals were noted to have severe ankle tightness with limited range-of-motion that limited application of the MAS. **Individual used a functional electrical stimulation device to assist in dorsiflexion but did not wear it for any testing. ***Outlier that was excluded from analysis.

Table 3.

Older and Younger Neurologically Intact Participant Information.

Older (≥60 years) Neurologically Intact Participants				Younger (≤35 years) Neurologically Intact Participants
ID	Age/Sex	Comfortable/Fast Gait Speed (m/s)	Simulation Pulse Width (mS)	Age/Sex	Comfortable/Fast Gait Speed (m/s)	Simulation Pulse Width (mS)
1	77M	1.32/1.69	1.0	32M	1.22/1.72	0.5
2	77F	1.50/1.99	1.0	30M	1.44/2.3	0.5
3	71F	1.16/1.49	1.0	30F	1.28/1.69	0.5
4	71M	1.70/2.32	1.0	28F	0.92/1.41	0.5
5	66M	1.31/1.87	1.0	28M	1.21/1.63	0.5
6	64M	1.00/1.68	0.5	28M	1.36/1.79	0.5
7	63F	1.25/1.61	1.0	26F	1.43/1.69	1.0
8	63M	1.26/2.36	1.0	26F	1.16/1.85	0.5
9	60F	1.15/1.32	1.0	24F	1.07/1.57	0.5
10				24M	1.14/1.79	0.5
	68.0±6.3	1.29±0.21	8–1.0	27.6±2.6	1.22±0.16	1–1.0
	4-F/5-M	1.81±0.36	1–0.5	5-F/5-M	1.74±0.23	9–0.5

The mean ± standard deviation is reported for Age and Gait Speeds, while the counts are reported for Sex and Stimulation Pulse Width.

Comparing Reflex Modulation in Neurologically Intact Participants to Individuals with Gait Impairments After Stroke

A linear mixed-effects model revealed ISI and its interaction with group as significant factors in normalized reflex modulation while age and sex did not have an influence. Figure 2 illustrates the reflex modulation responses at different ISIs. Our analysis revealed two outliers: one neurologically intact participant and one individual after stroke (see triangles in Figure 2, identified in Table 1). Values are reported without outliers unless their inclusion changed any results. Post-hoc testing showed both neurologically intact participants (75.5±17.6%, p<0.0001) and individuals with post-stroke gait impairments (81.2±16.7%, p<0.001) observed depression at ISI=2s that was not statistically different. At ISI=8s, there was a significant difference between groups (p<0.05) as neurologically intact participants yielded depression (94.9±7.75%, p<0.01) while individuals’ after stroke observed no modulation (99.6±6.10%). At ISI=12s, the neurologically intact group observed facilitation (106±12.2%, p<0.05), while the post-stroke group observed no modulation (103±6.71%). However, when excluding the neurologically intact outlier value, there was no modulation (104±7.42%). The full statistical results can be found in Supplementary Materials Tables S2-4 and visualized in Figure 2.

Figure 2.

Reflex modulation in neurologically intact and post-stroke participants. Normalized reflex modulation at the intervals of 2, 8, and 12s were calculated to assess homosynaptic depression of the soleus H-reflex in neurologically intact participants (blue, left) and individuals with post-stroke gait impairments (orange, right). The soleus reflex was inhibited at 2s in both groups, while there was only inhibition at 8s in neurologically intact participants. In accordance, the only significant difference in normalized reflex modulation between groups was observed at 8s. The triangles in the figure represent outlier values.

The 2/8s ratio was depressed in both the neurologically intact participants (75.7±18.0%, p<0.001) and the participants after stroke (80.1±11.8%, p<0.0001). There was no significant difference between the groups. Full statistical calculations can be found in Supplementary Material Tables S2-3 with data plotted in Figure 3.

Figure 3.

2/8s ratio results. The 2/8s ratio was calculated by normalizing the average amplitude of conditioned reflexes at 2s to 8s. There was no difference between neurologically intact participants and individuals with post-stroke gait impairments contrasting previously reported findings. The approximate results reported in Aymard et al. (2000) (78.29±11.7%* in 10 individuals with middle cerebral artery strokes, 60.1% in 16 neurologically intact individuals) and Lamy et al. (2009) (76.67% in 17 individuals after stroke, 62.01% and 60.24% in the dominant and non-dominant soleus of 26 neurologically intact participants) are also plotted for comparison. Similarly, there was no difference between older and younger neurologically intact participants. * Converted from standard error of mean found in original text to standard deviation.

Comparing Older to Younger Neurologically Intact Participants

A linear mixed-effects model showed age and sex were not significant factors while ISI and the interaction with group played a significant role. Post-hoc testing revealed both the older (78.4±17.0%, p<0.01) and younger (69.8±13.9%, p<0.0001) participants observed depression at ISI=2s despite no difference in response. Similarly, there was no observed difference between groups at ISI=8s although older adults exhibited depression (90.9±9.1%, p<0.01) while there was no modulation in younger adults (97.1±4.2%). At ISI=12s, older adults experienced facilitation (112.2±13.3%, p<0.05) while younger adults did not observe modulation (103.3±10.5%). There were no statistical differences between groups. All statistical values can be found in Supplementary Material Tables S2-4 with data plotted in Figure 4.

Figure 4.

Reflex modulation in older and younger neurologically intact participants. Normalized reflex modulation was compared between older (≥60 years; yellow, left) and younger (≤35 years; purple, right) neurologically intact participants. Both groups exhibited depression at 2s and 8s. The older participants also observed facilitation at 12s. There were no differences between groups.

Both the older (83.5±15.5%, p<0.05) and younger (72.2±13.5%, p<0.001) neurologically intact participants experienced depression in the 2/8s ratio that was not statistically different between groups. Data is shown in Figure 3 while statistical calculations can be found in Supplementary Material Tables S2-3.

Correlating Homosynaptic Depression to Gait Speed

The 2/8s ratio of HD of the soleus was correlated with both comfortable (r=0.645, p<0.01) and fast (r=0.737, p<0.001) gait speed in individuals after stroke (Figure 5). In contrast, there was no significant correlation between gait speed and HD in neurologically intact participants. Statistical values can be found in Supplementary Material Table S5.

Figure 5.

Homosynaptic depression correlation to gait speed. Comfortable and fast gait speeds were collected from the 10-meter walk test performed overground. Pearson correlation's show the 2/8s ratio was significantly related to both gait speeds only in participants with post-stroke gait impairments. These results suggest homosynaptic depression of the soleus may have a functional role in walking post-stroke.

Correlating Homosynaptic Depression to Soleus Reflex Excitability

We did not observe any correlations of HD of the soleus with spinal reflex excitability in either participants after stroke or those who were neurologically intact. Statistical calculations are presented in Supplementary Material Table S6 with data shown in Supplementary Material Figure S3.

Discussion

We investigated HD of the soleus H-reflex in individuals with post-stroke gait impairments, as well as its relationship with functional measures of gait. We hypothesized that HD would be reduced at short intervals and would correlate with both gait speed and reflex excitability in individuals after stroke. Our results indicate that HD is reduced at long intervals in individuals with post-stroke gait impairments relative to age-matched neurologically intact participants. In addition, our data showed that a reduced change in HD was strongly associated with faster gait speeds post-stroke, while no relation was found with neurologically intact individuals. These findings suggest that HD may be linked to lower limb function, although the underlying mechanisms remain unclear.

Pre-Synaptic Efficacy of Soleus Ia–α-Motoneuron Synapse Reduced at Long Intervals After Stroke

We assessed HD of the soleus H-reflex using two outcomes: the normalized reflex modulation to quantify reflex modulation relative to a baseline response, and the 2/8s ratio to quantify the change in reflex modulation across intervals. Based on the former, we found that depression is reduced at 8s intervals after stroke (Figure 2). However, when comparing groups based on the 2/8s ratio, there were no significant differences (Figure 4). These findings suggest that pre-synaptic efficacy is reduced at long intervals after stroke-related gait impairments. This conclusion aligns with previous literature that has theorized that the reduced inhibition of spinal reflex pathways contributes to hyperreflexia and post-stroke impairments (Akbas et al., 2020; Dyer et al., 2009; Lamy et al., 2009; Pierrot-Deseilligny & Burke, 2012). Based on our results, we speculate that reduced inhibition resulted in reduced pre-synaptic efficacy that diminished inhibition of the evoked reflex at 8s intervals. However, given the limited ISIs examined in this study, we were unable to characterize the temporal dynamics of this mechanism. Thus, future work should seek to investigate HD across more intervals between 2–8s.

In contrast with previous literature, our data did not observe a difference in the 2/8s ratio between neurologically intact participants and individuals after stroke (Aymard et al., 2000; Lamy et al., 2009). Both Aymard et al. (2000) and Lamy et al. (2009) reported individuals after stroke yielded a significant reduction in depression compared to neurologically intact participants (Figure 3). One potential reason for the discrepancy may stem from differences in the heterogeneity of participants with post-stroke gait impairments included in each study. Previous work has reported an association between HD and both upper and lower limb spasticity after stroke, indicating impairment severity may have a significant impact on results (Lamy et al., 2009; Yang et al., 2015). Here, we also found a correlation between HD and function through gait speed. Despite observing similar results in individuals after stroke compared to Aymard et al. (2000) and Lamy et al. (2009), it is unknown how the individuals who participated across studies differ in key demographic values. Therefore, it is possible that the heterogeneity of individuals after stroke contributed to the discrepancy in results observed between studies.

Conversely, the discrepancy could arise from differences within the neurologically intact participants across studies. Results in Figure 3 show data from the control groups of Aymard et al. (2000) and Lamy et al. (2009) exhibited increased depression compared to the participants of this study (Aymard et al., 2000; Lamy et al., 2009). One potential source of this difference could be the age of participants. While the average age of neurologically intact participants across studies was different, data shown here and across other reports all suggest age is unlikely to impact HD (Lamy et al., 2009; Trompetto et al., 2014). As a result, we also find it unlikely that the discrepancy stems from the age of the neurologically intact participants, and are thus unsure of its source.

Change in Pre-Synaptic Efficacy Strongly Linked to Gait Speed After Stroke

We observed a previously unreported correlation between the change in HD of the soleus H-reflex with gait speed in individuals with post-stroke gait impairments. In contrast, there were no relations found in neurologically intact participants. As the correlation was observed with the 2/8s ratio, the result suggests that the change in pre-synaptic efficacy across intervals may play a role in lower limb function. However, further research is needed to help define within what populations and what characteristics this holds true.

We observed a correlation between gait speed and the 2/8s ratio post-stroke, while no relation was found with the normalized reflex modulation at ISI=2s or 8s. These outcomes differ as the normalized reflex modulation is calculated in reference to a baseline response, while the 2/8s ratio is calculated in reference to the average response at ISI=8s without accounting for the test reflex (H₁). In the participants with post-stroke gait impairments, the normalized reflex modulation (at ISI=8s) is not statistically different to the reference baseline response (Figure 2). Thus, one might expect a significant correlation at ISI=2s given the similarity of context to the 2/8s ratio outcome. However, we find that reflex modulation at 2s is not correlated to gait speed potentially because of the increased variance found within the outcome. Our data shows there was less variance in the 2/8s ratio compared to the normalized reflex modulation at ISI=2s within individuals after stroke. Thus, we infer the heterogeneity and resulting variance of our outcomes contributed to our significant finding.

Our finding challenges the assumption that the nervous system seeks to minimize excess neurotransmitter release, or increase HD, to optimize motor output (Meunier et al., 2007; Obata et al., 2022; Trompetto et al., 2013). Considering this contradiction, we speculate that HD may serve a compensatory role in lower limb function, enabling individuals to walk faster despite reduced synaptic efficacy. Notably, while the correlation observed was only within individuals after stroke, this may extend to individuals with no known neurological injuries who may walk slower for other reasons.

We find it important to note that this relationship was observed in gait, a gross motor function. Dukkipati et al. (2023) reported the opposite relation between HD and manual dexterity in the upper limbs, a fine motor task requiring precise movements (Dukkipati et al., 2023). Therefore, the role of HD may depend on the precision of the movement or function.

Homosynaptic Depression Has a Limited Role in Reflex Excitability

In this study, we observed no significant correlations between homosynaptic depression and soleus reflex excitability in age-matched neurologically intact participants or individuals with post-stroke gait impairments. This finding is consistent with previous work reporting no association between HD and soleus reflex excitability in neurologically intact participants (Meunier et al., 2007). These results suggest that the role of HD is limited to modulating reflex excitability under specific temporal conditions where the synapse does not have sufficient time to recover.

Limitations

We acknowledge several limitations within the present manuscript. First, our analysis did not control for antagonistic tibialis anterior (TA) activity. While we are unable to definitively exclude the potential influence from TA activation, our signal processing approach, which removed trials with abnormal background soleus activity, may mitigate any heterogeneous muscle effects on our data. Along the same lines, our methodological approach did not utilize static joint angles representative of gait, or examine HD under dynamic or functional conditions. As the goal of this study was to examine whether a connection exists between HD and lower limb function, we utilized an approach consistent with established protocols (Aymard et al., 2000; Crone & Nielsen, 1989; Lamy et al., 2009). Therefore, the inclusion of additional static or dynamic conditions was outside the scope of this manuscript. We acknowledge recent studies that have shown that joint angles and motion can impact HD (Colard et al., 2023, 2024). Given the relative consistency in the change in HD across participants reported in Colard et al. (2024), together with our findings, we hypothesize that the relationship between HD and gait speed will remain robust across static or dynamic contexts. Future work may evaluate this by incorporating these additional conditions.

Second, it should be noted that our participants with post-stroke gait impairments largely identified as male (2 female, 19 male), while our age-matched neurologically intact participants were more evenly distributed (6 female, 10 male). It is unclear if participant sex is a significant factor of HD, as the impact of sex on spinal circuitry has been reported in previous literature (Miller et al., 2010). Lastly, it should be noted that HD is not believed to act in isolation, it is one of many spinal mechanisms that could influence synaptic transmission and gait function (Aymard et al., 2000; Cuadra et al., 2025; Dyer et al., 2009; Lamy et al., 2009; Pierrot-Deseilligny & Burke, 2012). Future investigations could include the additional assessment of other mechanisms such as reciprocal inhibition or heteronymous facilitation.

Conclusion

The goal of this study was to characterize HD of the soleus H-reflex in individuals with post-stroke gait impairments and examine its relationship with functional measures of gait. We found that while both groups observed depression at short intervals, depression was diminished at longer intervals after stroke compared to age-matched neurologically intact participants. Post-hoc analysis comparing older to younger neurologically intact participants suggests that age did not account for these differences. We also found a correlation between faster gait speed and reduced depression in individuals after stroke. Thus, HD may serve a role in compensating for reduced gait speeds. Future research is needed to understand the demographic, impairment-related, and neurophysiological parameters that guide HD to potentially utilize the mechanism in the assessment and treatment of hyperreflexia.

Supplemental Material

sj-docx-1-rnn-10.1177_09226028251413958 - Supplemental material for Homosynaptic Depression of the Soleus Predicts Post-Stroke Gait Speed

Supplemental material, sj-docx-1-rnn-10.1177_09226028251413958 for Homosynaptic Depression of the Soleus Predicts Post-Stroke Gait Speed by J Sebastián Correa, Ricardo Siu, Dana Lorenz, Kristine Hansen, David A Cunningham and James S Sulzer in Restorative Neurology and Neuroscience

Footnotes

Acknowledgements

The authors would like to thank all the participants who took part in the study. The authors would also like to thank Shreya Ramani and William Kozak for supporting data collection.

Ethical Considerations

This study was approved by the MetroHealth System Institutional Review Board (Study 00000128).

Consent to Participate

All participants gave informed written consent prior to taking part in the study.

Consent for Publication

Not Applicable

Author Contributions

JC, RS, DC, and JS designed research. JC, R, DL, and KH performed research. JC, DC, and JS analyzed data. JC wrote the paper. All authors reviewed the manuscript and approved the final version.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Development of the National Institute of Health under Award Number R01HD100416. The content is solely the responsibility of the authors and does not necessarily represent the official view of the National Institute of Health. J. Sebastian Correa was supported by a National Science Foundation Graduate Research Fellowship No. 1937968.

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

Data used in this work is available upon request.

Supplemental Material

Supplemental material for this article is available online.

ORCID iDs

J Sebastián Correa

Ricardo Siu

Dana Lorenz

David A Cunningham

James S Sulzer

References

Akbas

Kim

Doyle

Manella

Lee

Spicer

Knikou

Sulzer

(2020). Rectus femoris hyperreflexia contributes to Stiff-Knee gait after stroke. Journal of NeuroEngineering and Rehabilitation, 17(1), 117. https://doi.org/10.1186/s12984-020-00724-z

Aymard

Katz

Lafitte

Pénicaud

Pradat-Diehl

Raoul

(2000). Presynaptic inhibition and homosynaptic depression: A comparison between lower and upper limbs in normal human subjects and patients with hemiplegia. Brain, 123(8), 1688–1702. https://doi.org/10.1093/brain/123.8.1688

Bohannon

R. W.

Smith

M. B.

(1987). Interrater reliability of a modified Ashworth scale of muscle spasticity. Physical Therapy, 67(2), 206–207. https://doi.org/10.1093/ptj/67.2.206

Colard

Jubeau

Crouzier

Duclay

Cattagni

(2024). Effect of muscle length on the modulation of H-reflex and inhibitory mechanisms of Ia afferent discharges during passive muscle lengthening. Journal of Neurophysiology, 132(3), 890–905. https://doi.org/10.1152/jn.00142.2024

Colard

Jubeau

Duclay

Cattagni

(2023). Regulation of primary afferent depolarization and homosynaptic post-activation depression during passive and active lengthening, shortening and isometric conditions. European Journal of Applied Physiology, 123(6), 1257–1269. https://doi.org/10.1007/s00421-023-05147-x

Crone

Nielsen

(1989). Methodological implications of the post activation depression of the soleus H-reflex in man. Experimental Brain Research, 78(1), 28–32. https://doi.org/10.1007/BF00230683

Cuadra

Wolf

S. L.

Lyle

M. A.

(2023). Differential effect of heteronymous feedback from femoral nerve and quadriceps muscle stimulation onto soleus H-reflex. PLOS ONE, 18(8), e0290078. https://doi.org/10.1371/journal.pone.0290078

Cuadra

Wolf

S. L.

Lyle

M. A.

(2025). Heteronymous feedback from quadriceps onto soleus in stroke survivors. Journal of NeuroEngineering and Rehabilitation, 22(1), 39. https://doi.org/10.1186/s12984-025-01572-5

Curtis

D. R.

Eccles

J. C.

(1960). Synaptic action during and after repetitive stimulation. The Journal of Physiology, 150(2), 374–398. https://doi.org/10.1113/jphysiol.1960.sp006393

10.

Curtis

D. R.

Lacey

(1998). Prolonged GABA(B) receptor-mediated synaptic inhibition in the cat spinal cord: An in vivo study. Experimental Brain Research, 121(3), 319–333. https://doi.org/10.1007/s002210050465

11.

Dukkipati

S. S.

Walker

S. J.

Trevarrow

M. P.

Busboom

M. T.

Kurz

M. J.

(2023). Spinal cord H-reflex post-activation depression is linked with hand motor control in adults with cerebral palsy. Clinical Neurophysiology, 148, 9–16. https://doi.org/10.1016/j.clinph.2023.01.004

12.

Dyer

J.-O.

Maupas

Melo

S. d. A.

Bourbonnais

Fleury

Forget

(2009). Transmission in heteronymous spinal pathways is modified after stroke and related to motor incoordination. PLOS ONE, 4(1), e4123. https://doi.org/10.1371/journal.pone.0004123

13.

Eccles

J. C.

Eccles

R. M.

Magni

(1961). Central inhibitory action attributable to presynaptic depolarization produced by muscle afferent volleys. The Journal of Physiology, 159(1), 147–166. https://doi.org/10.1113/jphysiol.1961.sp006798

14.

Floeter

M. K.

Kohn

A. F.

(1997). H-reflexes of different sizes exhibit differential sensitivity to low frequency depression. Electroencephalography and Clinical Neurophysiology/Electromyography and Motor Control, 105(6), 470–475. https://doi.org/10.1016/S0924-980X(97)00032-5

15.

Hirst

G. D.

Redman

S. J.

Wong

(1981). Post-tetanic potentiation and facilitation of synaptic potentials evoked in cat spinal motoneurones. The Journal of Physiology, 321, 97–109. https://doi.org/10.1113/jphysiol.1981.sp013973

16.

Hultborn

Illert

Nielsen

Paul

Ballegaard

Wiese

(1996). On the mechanism of the post-activation depression of the H-reflex in human subjects. Experimental Brain Research, 108(3), 450–462. https://doi.org/10.1007/BF00227268

17.

Kawaishi

Matsumoto

Nishiwaki

Hirano

(2017). Postactivation depression of soleus H-reflex increase with recovery of lower extremities motor functions in patients with subacute stroke. Journal of Physical Therapy Science, 29(9), 1539–1542. https://doi.org/10.1589/jpts.29.1539

18.

Kendall

M. E.

Geise

P. P.

McIntyre

R. M.

Romani

(2005). Muscles testing and function with posture and pain (Fifth). Lippincott Williams and Wilkins.

19.

Kiser

T. S.

Reese

N. B.

Maresh

Hearn

Yates

Skinner

Pait

T. G.

Garcia-Rill

(2005). Use of a motorized bicycle exercise trainer to normalize frequency-dependent habituation of the H-reflex in spinal cord injury. The Journal of Spinal Cord Medicine, 28(3), 241–245. https://doi.org/10.1080/10790268.2005.11753818

20.

Kuno

(1964). Mechanism of facilitation and depression of the excitatory synaptic potential in spinal motoneurones. The Journal of Physiology, 175(1), 100–112. https://doi.org/10.1113/jphysiol.1964.sp007505

21.

Lagerquist

Collins

(2008). Stimulus pulse-width influences H-reflex recruitment but not Hmax/Mmax ratio. https://onlinelibrary.wiley.com/doi/10.1002/mus.20957

22.

Lamy

J.-C.

Wargon

Mazevet

Ghanim

Pradat-Diehl

Katz

(2009). Impaired efficacy of spinal presynaptic mechanisms in spastic stroke patients. Brain, 132(3), 734–748. https://doi.org/10.1093/brain/awn310

23.

Mahmoud

Hultborn

Zuluaga

Zrenner

Ziemann

Ramos-Murguialday

(2023). Testing spasticity mechanisms in chronic stroke before and after intervention with contralesional motor cortex 1 Hz rTMS and physiotherapy. Journal of NeuroEngineering and Rehabilitation, 20, 150. https://doi.org/10.1186/s12984-023-01275-9

24.

Menoux

Jousse

Quintaine

Tlili

Yelnik

A. P.

(2019). Decrease in post-stroke spasticity and shoulder pain prevalence over the last 15 years. Annals of Physical and Rehabilitation Medicine, 62(6), 403–408. https://doi.org/10.1016/j.rehab.2018.03.003

25.

Metz

Matos

I. C.

Hari

Bseis

Afsharipour

Lin

Singla

Fenrich

K. K.

Bennett

D. J.

Gorassini

M. A.

(2023). Post-activation depression from primary afferent depolarization (PAD) produces extensor H-reflex suppression following flexor afferent conditioning. The Journal of Physiology, 601(10), 1925–1956. https://doi.org/10.1113/JP283706

26.

Meunier

Kwon

Russmann

Ravindran

Mazzocchio

Cohen

(2007). Spinal use-dependent plasticity of synaptic transmission in humans after a single cycling session. The Journal of Physiology, 579(Pt 2), 375–388. https://doi.org/10.1113/jphysiol.2006.122911

27.

Miller

E. M.

Slota

G. P.

Agnew

M. J.

Madigan

M. L.

(2010). Females exhibit shorter paraspinal reflex latencies than males in response to sudden trunk flexion perturbations. Clinical Biomechanics, 25(6), 541–545. https://doi.org/10.1016/j.clinbiomech.2010.02.012

28.

Obata

Kim

Ogawa

Sekiguchi

Nakazawa

(2022). Effect of long-term classical ballet dance training on postactivation depression of the soleus hoffmann-reflex. https://doi.org/10.1123/mc.2021-0079

29.

Palmieri

R. M.

Ingersoll

C. D.

Hoffman

M. A.

(2004). The hoffmann reflex: Methodologic considerations and applications for use in sports medicine and athletic training research. Journal of Athletic Training, 39(3), 268–277. https://pmc.ncbi.nlm.nih.gov/articles/PMC522151/

30.

Phadke

C. P.

Flynn

S. M.

Thompson

F. J.

Behrman

A. L.

Trimble

M. H.

Kukulka

C. G.

(2009). Comparison of single bout effects of bicycle training versus locomotor training on paired reflex depression of the soleus H-reflex after motor incomplete spinal cord injury. Archives of Physical Medicine and Rehabilitation, 90(7), 1218–1228. https://doi.org/10.1016/j.apmr.2009.01.022

31.

Pierrot-Deseilligny

Burke

(2012). The circuitry of the human spinal cord: Spinal and corticospinal mechanisms of movement (Vol. 88). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4345533/

32.

Steffen

Seney

(2008). Test-Retest reliability and minimal detectable change on balance and ambulation tests, the 36-item short-form health survey, and the unified Parkinson disease rating scale in people with parkinsonism. Physical Therapy, 88(6), 733–746. https://doi.org/10.2522/ptj.20070214

33.

Stein

R. B.

Estabrooks

K. L.

McGie

Roth

M. J.

Jones

K. E.

(2007). Quantifying the effects of voluntary contraction and inter-stimulus interval on the human soleus H-reflex. Experimental Brain Research, 182(3), 309–319. https://doi.org/10.1007/s00221-007-0989-x

34.

Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles. (n.d.). [Dataset]. http://www.seniam.org/

35.

Tenniglo

M. J. B.

Nene

A. V.

Rietman

J. S.

Buurke

J. H.

Prinsen

E. C.

(2023). The effect of botulinum toxin type A injection in the rectus femoris in stroke patients walking with a stiff knee gait: A randomized controlled trial. Neurorehabilitation and Neural Repair, 37(9), 640–651. https://doi.org/10.1177/15459683231189712

36.

Trimble

M. H.

Kukulka

C. G.

Behrman

A. L.

(1998). The effect of treadmill gait training on low-frequency depression of the soleus H-reflex: Comparison of a spinal cord injured man to normal subjects. Neuroscience Letters, 246(3), 186–188. https://doi.org/10.1016/S0304-3940(98)00259-6

37.

Trompetto

Marinelli

Mori

Canneva

Colombano

Traverso

Currà

Abbruzzese

(2014). The effect of age on post-activation depression of the upper limb H-reflex. European Journal of Applied Physiology, 114(2), 359–364. https://doi.org/10.1007/s00421-013-2778-5

38.

Trompetto

Marinelli

Mori

Cossu

Zilioli

Simonini

Abbruzzese

Baratto

(2013). Postactivation depression changes after robotic-assisted gait training in hemiplegic stroke patients. Gait & Posture, 38(4), 729–733. https://doi.org/10.1016/j.gaitpost.2013.03.011

39.

van Melick

Meddeler

Hoogeboom

(2017). How to determine leg dominance: The agreement between self-reported and observed performance in healthy adults. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0189876

40.

Yang

Xiao

Song

(2015). Post-activation depression of the lower extremities in stroke patients with spasticity and spastic equinovarus deformity. Arquivos De Neuro-Psiquiatria, 73(6), 493–498. 10.1590/0004-282X20150052.

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.

0.00 MB

0.20 MB