Analysis of motor unit firing characteristics in patients with motor neuron diseases.

This study was designed to evaluate firing rate variability in patients with upper/lower motor neuron disorders. Twenty healthy subjects and 19 patients with motor neuron disorders participated in the study. Consecutive motor unit action potential pairs from extensor digitorum communis (EDC) muscle were recorded from each subject with trigger-delay line mode. Patients with motor neuron disorders (17.7 ± 10.8 ms) showed significantly higher mean time variability of interpotential interval value than healthy volunteers (10.3 ± 0.1 ms) (p < 0.001).

Keywords: ALS; firing rate variability; interpotential intervals; LMN; motor neuron diseases; motor unit firing characteristics

Motor neuron diseases are disorders of the upper or lower motor neurons or both. Needle electromyography reveals ongoing denervation and re-innervation by means of reduced interference, fibrillation potentials, and long duration, high-amplitude motor unit potentials (Sonoo and Stålberg [18]). Changes in motor unit potential configuration are indicators of lower motor neuron loss and increased fiber density by collateral sprouting. Lower motor neuron loss impairs recruitment order and reduces interference at full contraction (Petajan [15]). A reduced recruitment and a faster motor unit firing could be seen in the needle electromyographic examination as a result of loss of anterior horn cells.

In the case of upper motor neuron (UMN) involvement, the firing frequency of the lower motor neuron (LMN) declines without any change in the motor unit potential configurations. In amyotrophic lateral sclerosis (ALS), which is a disorder of both upper and lower motor neurons, a combination of the above findings occurs.

The time interval between consecutive discharges of the same motor unit during sustained voluntary activation is a function of both upper and lower motor neurons. It takes time for motor neurons to trigger and propagate an action potential. Triggering an action potential is more complex than the potential for action propagation over the axons; it involves soma and synaptic connections over the dendrites. The time interval between consecutive discharges of the same motor unit changes from one firing to the next, even under physiological conditions (Petajan and Philip [14]; Person and Kudina [13]). This variability may be an indication of resting potential membrane changes and local depolarizing or hyperpolarizing postsynaptic currents driven by synaptic connections over the soma of the lower and upper motor neurons. Disorders affecting the upper and/or lower motor neurons may increase this variability; this is what the present study proposed to test.

This study was carried out for 1 year. Written informed consent was obtained from each subject following a detailed explanation of the objectives and protocol of the study, which was conducted in accordance with the ethical principles stated in the "Declaration of Helsinki" and approved by the institutional ethics committee of Istanbul University (788-556).

The patients' group consisted of 19 subjects (mean age: 45.7 ± 13.0 years) including 9 patients with ALS (3 females, 6 males; 46.5 ± 13.4 years), 8 patients with LMN diseases (2 females, 6 males; 43.0 ± 14.3 years), and 2 patients with primary lateral sclerosis (PLS) (2 males; 53 ± 0 years), who were followed at the outpatient clinics of neurology departments in Acıbadem University School of Medicine and in Istanbul University, Istanbul Faculty of Medicine. Motor neuron disorders were diagnosed according to clinical and electrophysiological findings. All of the ALS patients were labeled as "definitive ALS" according to the El-Escorial criteria (Brooks et al. [2]). Of the 8 patients with LMN involvement, 2 had prior poliomyelitis and the remaining 6 were classified as spinal muscular atrophy (SMA) type 3. Patients with only UMN involvement underwent a thorough work-up to exclude secondary causes but not pure degeneration. Pringle criteria were used for diagnosing PLS (Pringle et al. [17]). Twenty healthy volunteers (10 females, 10 males; mean age: 38.3 ± 9.8 years) without any neurological diseases agreed to participate in the study. Their physical examinations were normal and they were not taking any medication which might affect electrophysiological assessment.

Needle electromyography (EMG) of the extensor digitorum communis (EDC) muscle was done unilaterally in each subject by using a disposable concentric needle electrode (i.e., 0.46 mm in diameter and 37 mm in length; Medelec Synergy, Oxford Instruments, Surrey, UK, catalogue number: X53156). Pre-installed single fiber EMG-jitter measurement software of a five-channel electromyograph was used with some modifications for acquiring and analyzing data (Medelec Synergy, Oxford Instruments, Surrey, UK). Amplitude triggering was used during recording. A low-cut filter was set to 2 Hz and its sweep speed slowed down to 100 ms/div. Each patient was asked to extend and keep his or her fingers and wrist at an angle of 45° in order to activate the muscle under investigation. The concentric needle electrode was properly placed to record a motor unit action potential (MUAP) of the highest amplitude and shortest rise time. This MUAP was used to trigger the sweep. After reaching a stable firing level, at least 60 consecutive traces harboring triggering and jittering MUAPs were acquired and stored. The interpotential interval (IPI) between triggering and jittering MUAPs was calculated over each trace between the midpoints of the rising edges. The consecutive differences between the IPI were calculated and the mean value of the consecutive differences was accepted as "time variability of IPI" (vIPI). At least seven different MUAPs from each patient were recorded and analyzed.

The mean, median, standard deviation (SD), minimum, and maximum value of the vIPI were calculated in the healthy volunteer and patient groups. A Mann–Whitney U-test compared the normal control and patients' data. A Kruskal–Wallis test was used to compare the vIPI values between the groups (the healthy volunteer and the patients with LMN and ALS). We used Dunn's test for multiple comparison of groups. Patients with PLS were excluded for subgroup analysis due to the small number of participants. A generalized linear mixed effect model was used to compare the mean vIPI values of the three groups. The level of significance was set at p < 0.05.

Table 1 shows descriptive features of the study population and the number of recorded MUAPs from the patients. One patient with ALS was familial; his symptoms had begun when he was 24 years old. Another patient with ALS had experienced asymmetrical weakness when she was 23, although her case was not familial. The oldest ALS patient was 63 years old and he had been suffering from the disease for 1 year. The shortest duration of the disease for ALS patients was 8 months. Of the 8 patients with LMN disorders, 2 had had late poliomyelitis and had contracted the paralytic infection 34 and 44 years ago. In these patients the extremities which were examined with EMG were mildly affected and none of them had shown clinical or electrophysiological signs of post-polio syndrome. In the three other patients with LMN involvement, the disease history of clinical signs was compatible with SMA type 3. The youngest of them was 29 years old and had had the disease for 22 years, longer than anyone else in the group. The remaining 3 patients with LMN disorder were classified as "pure LMN disease" because they had asymmetrical weakness with muscular atrophy, without any accompanying signs of UMN involvement. All of the patients had sufficient muscle power in EDC during EMG recording to maintain sustained contraction against gravity (3/5 or more, according to the Medical Research Council).

Table 1. Descriptive features of the patients with motor neuron diseases.

1 MUAPs: motor unit action potentials; SMA: spinal muscular atrophy; LMN: lower motor neuron; ALS: amyotrophic lateral sclerosis; FALS: familial amyotrophic lateral sclerosis.

Overall, 205 different MUAPs were recorded from the patients' group. Among the healthy volunteers, 240 different MUAPs were analyzed. Table 2 shows descriptive data of the control group and the patients (including and without patients with PLS) and also subgroups (the patients with LMN diseases and ALS). According to the pooled data, the mean value of 205 vIPIs was 17.7 ± 10.8 ms, whereas in the healthy volunteers the mean value of 240 vIPIs was 10.3 ± 0.1 ms. A Mann–Whitney U-test found statistical differences between the mean values of vIPIs of the healthy volunteers and the patients (p < 0.001). Dunn's test was compared to the mean vIPIs between subgroups: (normal and LMN (p = 0.01)), (normal and ALS (p = 0.006)), (LMN and ALS (p = 1.000)), respectively. Figure 1 shows the traces harboring triggering and jittering MUAPs for 1 healthy volunteer, 1 patient with ALS, and 1 patient with LMN disease.

Graph: Figure 1. The superimposed traces harboring triggering and jittering MUAPs for 1 healthy volunteer (sweep speed: 1 ms/div, sensitivity: 1 mV) (A), 1 patient with ALS (sweep speed: 5 ms/div, sensitivity: 10 mV) (B), and 1 patient with LMN disease (sweep speed: 2 ms/div, sensitivity: 5 mV) (C).

Table 2. Descriptive data of the healthy control, the patients, the patients without PLS, and all subgroups (the patients with LMN diseases and ALS).

2 vIPI: variation of interpotential interval; µs: microsecond; n1: number of subjects; n2: number of motor unit action potentials; PLS: primary lateral sclerosis; LMN: lower motor neuron; ALS: amyotrophic lateral sclerosis.

A generalized mixed effect model was used to describe the relationships between variable (group) and independent variables (age and sex). Age and sex were not significant for the group comparisons for vIPI (df: 1; p = 0.146).

The present study showed that the physiological variation of interdischarge interval of each motor unit increased in those with motor neuron diseases. Simply checking the variation of consecutive IPIs of MUAP by using SFEMG software, made it possible to demonstrate the instability of the firing rate.

Dorfman et al. ([5]) published a study in which they showed an increased variability of firing rate in patients with central motor neuron disorders. They had used an automatic decomposition program to capture individual MUAPs and calculated the interspike intervals (ISIs) between consecutive MUAPs. They expressed the ISIs of an individual MUAP recorded over the 10-s epoch as a histogram. As a measure of firing rate variability, the authors calculated the coefficient of ISI variability (CIV) which is defined as the standard deviation of the fundamental peak in the ISI histogram. In patients with motor neuron diseases, firing rates of MUAP increased, but in patients with multiple sclerosis (MS) they found that the mean firing rate was significantly subnormal. In patients with both motor neuron diseases and MS, the authors found significantly increased ISI variability, which they described as increased mean CIV. However, they did not find significant abnormality of CIV in patients with myopathy. The authors commented that their findings were in parallel with the observations of previous researchers (Grimby and Hannerz [10]; Freund et al. [9]; Andreassen and Rosenfalck [1]).

The excitability of the motor cortex in motor neuron diseases has been thoroughly studied by transcranial magnetic stimulation (TMS) techniques. Resting motor threshold (RMT), motor evoked potential (MEP) amplitude, and cortical silent period (CSP) duration are elicited by single-pulse TMS techniques. Short-interval intracortical inhibition (SICI) and intracortical facilitation (ICF) are elicited by pair-pulse TMS techniques. Cortical excitability and disinhibition have been implicated in ALS pathogenesis, with neuronal degeneration mediated by an anterograde glutamate-mediated excitotoxic process (Eisen et al. [6]). Studies have shown that TMS is a sensitive tool for the demonstration of UMN dysfunction in ALS patients and the motor cortex and its surroundings are hyperexcitable in the early periods of disease (Caramia et al. [3]; Mills and Nithi [12]; Eisen et al. [7]; Di Lazzaro et al. [4]; Pouget et al. [16]). A recent study also showed that cortical hyperexcitability precedes LMN dysfunction in ALS, suggesting a cortical origin of the disease (Menon et al. [11]). Menon et al. ([11]) studied cortical and axonal excitability in patients with ALS. They found that significant reduction of SICI, RMT, and CSP duration along with increases in MEP amplitude and ICF recorded over the abductor pollicis brevis (APB) muscle utilizing TMS. The study also showed the anatomical and functional integrity of LMN function recording from APB muscle utilizing qualitative and quantitative EMG techniques along with axonal excitability studies.

Our study demonstrated an increased time variation of interdischarge interval of each motor unit in the patients with MNDs, which were made up mostly of ALS and mixed subdiagnosis of LMN. The mean vIPI values of healthy controls (10.3 ± 0.1 ms) and the patients (17.7 ± 10.8 ms) were significantly different (p < 0.001). Subgroup analysis showed the statistical difference between the healthy controls and the patients with LMN diseases (14.9 ± 0.4 ms; p = 0.01) and also ALS (21.3 ± 14.7 ms; p = 0.006). The mean vIPI values of patients with PLS were not compared to the control group statistically because of an insufficient number of patients. Table 2 shows the mean vIPI values for all patients (18.3 ± 11.3 ms) and without PLS patients (17.7 ± 10.8 ms). In those patients with PLS the mean vIPI value (13.2 ± 0.2 ms) was higher than the normal control, but we could not make any comment because of an insufficient number of patients. The generalized mixed effect model showed that age and sex did not contribute any difference for vIPI values between groups. In those with ALS the variation of consecutive MUAPs is two times higher compared to the healthy control. Moreover, patients with ALS have very high standard deviation levels compared to healthy subjects, which is probably due to MUAPs from recording both normal and degenerative motor neurons. This finding could be interpreted as a reflection of cortical hyperexcitability which has been identified as an important pathogenic mechanism in MND and a synaptic dysfunction (an increase of the time interval due to loss of synaptic input) or excitability changes at the level of the spinal motor neuron. In ALS, both the alpha motor neuron and the corticomotoneuronal synaptic input from the UMN are affected during the disease process, on the other hand, the alpha motor neuron and its synaptic connections were affected in LMDs. The time interval between consecutive discharges of the same motor unit during sustained voluntary activation is a function of both upper and lower motor neurons. This time interval consecutively reflects the generation and propagation of action potential over the motor neurons and their axons. Action potential propagation over the axon is relatively simple and involves saltatory conduction. However, triggering an action potential requires sufficient excitatory postsynaptic potential over the dendrites of a motor neuron, which is a more dynamic and variable procedure depending upon the converging synapses. In other words, the firing rate variability of a particular motor neuron mostly reflects the changes in the ability to generate rather than to propagate an action potential. This speculation is supported by the changes in the observed excitability of upper and lower motor neurons, as previously demonstrated (Caramia et al. [3]; Mills and Nithi [12]; Eisen et al. [7]; Di Lazzaro et al. [4]; Pouget et al. [16]; Eisen [8]).

Although the present small size study showed the variability of the firing rate of motor units in motor neuron diseases, it has quite a few limitations. First, it is essential to get sustained contraction in order to measure the variability of the firing rate properly. In patients with severe weakness and easy fatigability it is sometimes impossible to reach a sustained level of contraction. This increases the variability, as it does in patients with poor cooperation. Second, if the firing rate variability were an indication of excitability changes in both upper and lower motor neurons, then it should be more pronounced early in the disease, but the patients who were recruited for the study had to have progressed sufficiently in the disease to have allowed a definitive diagnosis to be made. Finally, the software used to calculate the firing rate variability was in some ways also sensitive to MUAP shape variability (jiggle).

In conclusion, patients with upper and/or lower motor neuron diseases show increased variability in the firing rate of motor units, which can be demonstrated easily by using a trigger-delay line. Our data is limited, so clinical relevance of this information could not be interpreted or how this could influence diagnosis and/or management of MNDs. However, this electrophysiological examination is not time consuming and if it is additionally done during EMG study, it may help to differentiate ALS from mimic disorders, to prevent a diagnostic delay and to additionally support a diagnosis of MND. Finally, larger studies need to determine the normal values of the vIPI in healthy subjects and also in patients with upper/lower motor neuron diseases. In that case, it may prove useful as a diagnostic investigation for ALS.

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.