Comparison of the Amplitudes of the H-reflex of Post-stroke Hemiplegia Patients and Normal Adults during Walking.

[Purpose] The purpose of this study was to compare H-reflex characteristics during gait of hemiplegic stroke patients. [Subjects] Twenty-five patients and age-matched twenty-five volunteers in good health were studied. All the subjects could walk independently. [Methods] An MP150 (BIOPAC Systems, Inc., Goleta, CA, USA) was used to record the electromyography (EMG) data collected with Ag-Ag/Cl measurement electrodes (BIOPAC, diameter of 2 cm).
[Results] The comparison showed significant differences of Hmax/Mmax ratio (%) in all gait cycles between the stroke group and the control group.
[Conclusion] In conclusion, this study furnished basic reference data for gait strategies and functional training programs for hemiplegic stroke patients.

INTRODUCTION

Hemiplegic stroke patients are restricted in their daily life due to lowered exercise ability, weakened muscular strength, and muscle stiffening1). Generally, about 50 to 80% of stroke survivors acquire the ability to maintain balance three weeks after the occurrence of the stroke or after leaving the hospital, and once motility is secured, they become capable of walking2).

However, many patients still have disorders such as a slower walking pace or reduced endurance, and are restricted in their independent mobility at home or in the workplace3). Therefore, the recovery of walking ability becomes an important goal in the rehabilitation of stroke patients4).

The distinctive walking patterns of stroke patients include a slow gait cycle and walking speed, time asymmetry between the leg on the affected side and the leg on the unaffected side, and a relatively short stance phase on the affected side and a long swing phase5). In addition, stroke patients lack the ability to generate voluntary muscle contraction to a normal extent and cannot generate the proper timing and intensity of muscle activities2). The focus of the rehabilitation of stroke patients is to increase their balance ability and generate proper functional movements6). Therefore, in order to evaluate functional losses or treatment results and project prognoses, an objective and quantitative evaluation of hemiplegic patients is required. The quantitative evaluations used in the clinical setting are the Ashworth scale7), the pendulum test8), and an electrodiagnostic method using the H-reflex9).

The spinal cord motor reflex examined by the electrically stimulated H-reflex plays an important role in posture control. The H-reflex controls posture in the performance of different postural tasks, movements requiring high levels of changes, and activities involving changes in postural angle10).

It is essential to find out the accurate causes and body areas of abnormality in patients with pathologic gait, and compare and evaluate the findings through the process of objectifying and quantifying. The abnormal gaits of hemiplegic patients have been a topic of continuous research, and many recent studies have carried out gait analyses based on electromyography (EMG)11, 12). However, EMG-based studies about the H-reflex of hemiplegic stroke patients and normal adults during the gait cycle still remain inadequate.

The purpose of this study was to examine the effects of the H-reflex on the soleus muscle of hemiplegic stroke patients and normal adults during walking and on their gait cycle, to furnish basic data for diagnosis and adjustment of the control of abnormal muscles in hemiplegic stroke patients.

SUBJECTS AND METHODS

Subjects

The subjects of this study were normal adults (n=25), and patients who were diagnosed with hemiplegia resulting from a stroke (n=25). Our experiment was conducted after explaining the purpose of this study to the subjects and gaining their consent. The normal group consisted of 15 women and 10 men with an age range of 58 to 75 who had no orthopedic diseases that could affect their gait or balance. The patient group consisted of patients who had hemiplegia due to cerebral infarction or hemorrhage and were being treated at Hospital B in Ulsan. The patient group consisted of 18 men and seven women with Brunnstrom stages ranging from three to five. The average duration since stroke onset in the patient group was 20.88±8.35 months. Five patients had hemiplegia on the right side, and 20 patients had it on the left side. The averages of NIHSS (National Institutes of Health Stroke Scale) and Brunnstrom stage of the patient group were 9.08±2.57 score and 3.88±0.78 respectively. The patients could walk independently over 10 m indoors without a walking aid and were capable of performing the experiment according to the experimenter’s requests (Table 1).

Table 1.

General characteristics of subjects (M±SD)

	Normal (n=25)	Stroke (n=25)
Age (yrs)	65.52±10.63	67.56±11.25
Height (cm)	165.44±9.36	167.84±9.56
Weight (kg)	65.84±11.02	72.72±9.86
Leg length (cm)	80.64±4.36	87.76±5.21

M±SD: Mean ± standard deviation

Leg length is the length from the anterior-superior spine of the ilium to the medial malleolus.

Methods

To record and process EMG measurements a MP150 (BIOPAC System Inc., Goleta, CA, USA) was used with Ag-Ag/Cl (BIOPAC, diameter 2 cm) electrodes. The EMG signals were collected using a sampling rate of 1000 Hz, and processed by full-wave rectification. While the subjects were walking, data was collected using Acknowledge 4.1 (BIOPAC System Inc., Goleta, CA, USA) software. In order to measure the H-reflex, data were band-pass filtered between 30 to 500 Hz, and for the removal of noise, the signals were processed by notch-filtering at 60 Hz.

In order to measure the H-reflex, a line was drawn between the center of the folds of the popliteal fossa and the medial malleolus of each subject, and a recording electrode was attached to the inner muscle belly of the soleus muscle, which lay at the midpoint of the line. In addition, a standard electrode was attached to the distal part, 15 cm from the achilles tendon. The stimulating electrode used in this experiment was a bipolar electrode. It was attached to the popliteal fossa with its cathode positioned on the proximal part so that it would stimulate the tibial nerve. For the ground electrode, a 1×1 cm self-adhesive disposable electrode (Neuroline Disposable Neurology Electrodes 700 10-k, Medicotest A/S, Olstykke, Denmark) was attached between the stimulating and recording electrodes. The electric stimulation was delivered to the center of the folds of the hamstring muscle to the posterior tibial nerve using the bipolar electrode at the level of submaximal stimulation at intervals of two seconds. The H-reflex was elicited by low voltage with duration of a current about 1 msec and at a rate of once every 2 sec to prevent the direct stimulation of motor neuron axons and the generation of M-waves. In an EMG analysis, the effective blocking of noise is of the utmost importance; there, we paid careful attention to the selection of electrode locations. Moreover, in order to prevent noise from the connecting leads, the leads were firmly fixed with tape.

Each subject was instructed to walk on a treadmill comfortably while wearing EMG pads, and the experiment began when the subject exhibited his/her natural gait. The experiment was carried out while the subject did not feel fatigued by having to pay attention and concentrate. While the subject was walking, stimuli were delivered to elicit the H-reflex in the four gait phases of heel-strike, mid-stance, toe-off, and mid-swing. The stimuli for the H-reflex were manually delivered as a square voltage pulse of 0.1 seconds in the four gait phases, and the timing of the stimulation was displayed on a screen. The amplitude of the H-reflex and the amplitude of the M-wave were measured, and the amplitude was measured as peak-to-peak values. After checking for the presence of the H-wave, the stimulus strength was slowly adjusted to find Hmax and Mmax, and the Hmax/Mmax ratio was calculated as the maximum amplitude ratio of the two action potentials13). The research results were statistically analyzed using SPSS version 12.0 (ICC, Chicago, IL, USA). The physical characteristics of each group, and the measurement data of each variable are presented as the average and standard deviation which were calculated using descriptive statistics. In order to compare the H-reflexes of the patient group and the normal group, the independent t-test was performed. Statistical significance was accepted for values of p<0.05.

RESULTS

The H-reflex amplitudes of the hemiplegic patients were 28.41±7.46% in the swing phase, 28.22±7.75% in the heel-strike phase, 46.15±15.43% in the mid-stance phase, and 36.19±15.20% in the toe-off phase. Those of the normal adults were 8.82±2.49% in the swing phase, 12.68±4.29% in the heel-strike phase, 34.62±13.31% in the mid-stance phase, and 7.52±1.69% in the toe-off phase (Table 2). Additionally, according to the independent t-test, statistically significant differences were present in all the phases, with values of p=0.000 in the swing phase, p=0.00 in the heel-strike phase, p=0.004 in the mid-stance phase, and p=0.00 in the toe-off phase (p<0.01).

Table 2.

The H-reflex amplitudes of the stroke group and the normal group during walking (unit: %)

	Group	N	M±SD
Swing phase	Stroke	25	28.41±7.46	***
	Normal	25	8.82±2.49
Heel strike	Stroke	25	28.22±7.75	***
	Normal	25	12.68±4.29
Mid-stance	Stroke	25	46.15±15.43	**
	Normal	25	34.62±13.31
Toe-off	Stroke	25	36.19±15.20	***
	Normal	25	7.52±1.69

M±SD: Mean ± standard deviation

* p<0.05, **: p<0.01, ***: p<0.001

DISCUSSION

When treating hemiplegic stroke patients, improving walking ability is very important, as walking is an essential element for gaining functional independence in stroke patients’ daily lives14). In order to improve walking ability, the studies of normal gait, which provide gait standards for patients, are necessary. Moreover, analyses of the causes of abnormal gait patterns exhibited by patients and the objective assessments of their treatment results are also important.

The H-reflex is a single synapse reflex conducted by the central nervous system and provides information that is functionally important to peripheral sensory and motor nerves15). The H-reflex is a single synapse reflex that is used to evaluate functional conditions of the spinal reflex loop and has been employed in the evaluation of spinal excitability. Particularly, H-reflex research has focused on its relation with the various disease conditions that lower the excitability of spinal motor neurons16, 17).

The H-reflex amplitude shows wide variations with change in stimulus and is highly affected by the central nervous system. It is also heavily influenced by subjects’ gait, degree of tension release, intensity and duration of electrostimulation, location of electrodes, age, and body temperature18). Therefore, the H-reflex is known to vary widely not only between individuals, but also within the same individual19). Such a phenomenon occurs due to the existence of many factors that can affect the α-motor neuron and changes in the excitability of the synapse in the central nervous system between the Ia afferent nerve and the motor neuron in the muscle spindles19). Therefore, measurement of spinal neuron excitability change is possible by measuring the amplitude of the H-wave, and calculating the Hmax/Mmax ratio is the commonly used method of H-reflex normalization20). The Hmax/Mmax ratio can be said to be an accurate index that shows the frequency and amount of excitatory alpha motor neurons according to the activity of descending tract13). In the present study, we measured the maximum amplitudes of the H-wave and the M-wave to compare Hmax amplitude to Mmax amplitude ratios.

In a study that examined the H-reflex amplitudes during walking at a speed of 1.2 km/h by two age groups of 7 to 14 and 13 to 16, the highest amplitude was reported to have occurred in the mid-stance phase, which follows the heel-strike phase, whereas the toe-off, swing, and heel-strike phases were observed to have similar amplitudes21). A similar result was also reported in a gait analysis of normal adults aged 20 to 40 in another study22).

In contrast, people with upper motor neuron syndrome (UMNS), such as stroke patients, are reported to have a gait pattern of abnormal muscle activation in their lower limb muscles23). The reason for this is that, the dorsiflexion of the ankles is disturbed at the initial activity during the stance phase because their calf muscles are stretched24).

According to the results of this study, the H-reflex amplitude of the hemiplegic stroke patients was at its highest in the mid-stance phase and at its lowest in the heel-strike phase. In the normal adults, the H-reflex amplitude was the highest in the mid-stance phase and the lowest in the toe-off phase. In both the hemiplegic stroke patients and the normal adults, the highest amplitude of the H-reflex was detected in the mid-stance phase, a result which is a similar to those of previous studies. However, unlike in the normal group, the lowest amplitude in the patient group was during the heel-strike phase25). Additionally, in all the phases of the gait cycle, the H-reflex amplitude exhibited a statistically significant difference between the patient and normal groups. We attribute this to the nature of the disease, as hemiplegic stroke patients tend to show a gait pattern of stiff walking due to muscle spasticity, which is accompanied by a tendency toward circumduction or dragging of the feet. Moreover, hemiplegic stroke patients with neurologic lesions have difficulties generating appropriate levels of muscle contraction and control.

This study demonstrated that statistically significant differences exist in the amplitudes of the H-reflex during the gait cycle between hemiplegic stroke patients and normal adults. This study investigated the H-reflexes during walking of hemiplegic stroke patients and normal adults. We hope the data of this report will contribute to the provision of effective gait strategies and functional training programs for hemiplegic stroke patients in the future.