AAEM minimonograph #25: Single-fiber electromyography

Single-fiber electromyography (SFEMG) is a selective recording technique in which a needle electrode with a small recording surface in the side is used to identify action potentials from individual muscle fibers. The SFEMG parameters of greatest clinical use are fiber density (FD) and neuromuscular jitter. FD reflects the local organization of muscle fibers within the motor unit; jitter reflects the safety factor of neuromuscular transmission at individual neuromuscular junctions. SFEMG can be of great value in demonstrating or excluding abnormalities in mild or questionable disease of nerve, muscle, or the neuromuscular junction. The neuromuscular jitter may be measured during nerve stimulation, which is particularly useful in uncooperative patients or when it is desirable to control the firing rate precisely, or during voluntary muscle activation, which is less subject to technical artifact. The SFEMG findings may not be specific to a particular diseases, but they frequently increase understanding of the disease process by demonstrating abnormal neuromuscular transmission or rearrangement of muscle fibers within the motor unit, which complements information from more conventional EMG examinations. © 1996 American Association of Electrodiagnostic Medicine. Published by John Wiley & Sons, Inc.     

Effect of triggering potential on calculations of jitter in single-fiber EMG

Jitter values are calculated in reference to a triggering potential during single-fiber electromyography (SFEMG) performed during voluntary contraction. When there are more than two single-fiber action potentials (SFAP) and the selected triggering potential is from an abnormal end-plate, all the jitters calculated from the same trace will be affected. In this study, the effect of triggering potential on calculated jitter was investigated in myasthenic and healthy volunteers by switching the triggering potential and recalculating the jitter off-line. Selecting a triggering potential from an abnormal end-plate increased the number of abnormal individual jitters as well as the mean jitter. Therefore, if the equipment software has the capacity to change the trigger potential, the triggering potential should not be from an abnormal end-plate if all possible jitter values are to be calculated for traces having three or more single-fiber potentials. Otherwise, only one jitter value should be included from any one trace to prevent false-positive results.     

Stimulated single-fiber electromyography in the rat

We constructed an animal model of stimulated single-fiber electromyography (SFEMG) by testing Wistar rats under anesthesia. Stimuli of 1 Hz were applied to the sciatic nerve through an insulated monopolar needle electrode. Single-fiber action potentials were acquired from the gastrocnemius muscle. Jitter was assessed by the mean consecutive difference (MCD). Eighty-seven fibers were obtained from 12 rats. Their MCDs ranged from 2 to 72 μs (17.7 ± 13.4). Seven of these values were less than or equal to 5 μs, and three exceeded 50 μs. Neuromuscular blocking agents injected into some of the rats induced considerable increases in jitter and blocking. A rat with one fiber with an MCD less than 5 μs also received an injection of curare. The jitter showed the same pattern of increment, evidence that the small jitter was not attributable to direct muscle stimulation. These results show that SFEMG can be used on rats. In addition, jitter reflects the changes in motor end-plate function. The findings also suggest the presence of an extremely high safety factor in rat neuromuscular junctions. © 1998 John Wiley & Sons, Inc. Muscle Nerve 21:482–489, 1998.     

Reference values for jitter recorded by concentric needle electrodes in healthy controls: A multicenter study

Introduction: The aim of this study was to create reference values for jitter measured with concentric needle electrodes. Methods: Operators worldwide contributed recordings from orbicularis oculi (OO), frontalis (FR), and extensor digitorum (ED) muscles in healthy controls. Criteria for acceptable signal quality were agreed upon in advance. Fifteen or 20 recordings of acceptable quality from each muscle were required for voluntary and electrical stimulation recordings, respectively. Results: Recordings from 59 to 92 subjects were obtained for each muscle and activation type. Outlier limits for mean consecutive difference and individual jitter data for voluntary activation were: OO, 31 and 45 µs; FR, 28 and 38 µs; ED, 30 and 43 µs; and for electrical stimulation they were: OO, 27 and 36 µs; FR, 21 and 28 µs; ED, 24 and 35 µs. Conclusion: Reference jitter values from concentric needle electrode recordings were developed from signals of defined quality while seeking to avoid creating supernormal values. Muscle Nerve 53: 351–362, 2016     

Electrophysiological study of neuromuscular junction in congenital myasthenic syndromes,congenital myopathies,and chronic progressive external ophthalmoplegia

This study was designed to analyze the sensitivity, specificity, and accuracy of jitter parameters combined with repetitive nerve stimulation (RNS) in congenital myasthenic syndrome (CMS), chronic progressive external ophthalmoplegia (CPEO), and congenital myopathies (CM). Jitter was obtained with a concentric needle electrode during voluntary activation of the Orbicularis Oculi muscle in CMS (n = 21), CPEO (n = 20), and CM (n = 18) patients and in controls (n = 14). RNS (3 Hz) was performed in six different muscles for all patients (Abductor Digiti Minimi, Tibialis Anterior, upper Trapezius, Deltoideus, Orbicularis Oculi, and Nasalis). RNS was abnormal in 90.5% of CMS patients and in only one CM patient. Jitter was abnormal in 95.2% of CMS, 20% of CPEO, and 11.1% of CM patients. No patient with CPEO or CM presented a mean jitter higher than 53.6 µs or more than 30% abnormal individual jitter (> 45 µs). No patient with CPEO or CM and mild abnormal jitter values presented an abnormal decrement. Jitter and RNS assessment are valuable tools for diagnosing neuromuscular transmission abnormalities in CMS patients. A mean jitter value above 53.6 µs or the presence of more than 30% abnormal individual jitter (> 45 µs) strongly suggests CMS compared with CPEO and CM.     

Axonal stimulation for end-plate jitter studies.

This single fibre EMG study compares the standard method of neuromuscular jitter measurement in voluntarily activated muscle to that by intramuscular electrical stimulation of motor axons in a group of normal subjects. The latter method avoids the interdischarge interval-dependent jitter, as well as a possible failure to recognise split muscle fibres. The mean MCD on axonal stimulation was only 5.2 microseconds less than in the voluntary activation study and was thus 8% more than theoretically expected for single motor end plates. The difference could be due to an axonal jitter and some other factors. Axonal stimulation has proved to be a relatively easy and reliable method for routine estimation of neuromuscular jitter, provided that the resolution of time measurement is better than 2 microseconds, so that low jitter due to occasional direct muscle fibre stimulation is not mistaken for a normal reading. The upper normal limits for the extensor digitorum communis muscle suggested by the present study are 40 microseconds (individual muscle fibres) and 25 microseconds (mean of 30 muscle fibres).     

Simulation of the normal concentric needle electromyogram by using a muscle model.

OBJECTIVES: To study the correlation between anatomical parameters and EMG signals by means of simulations. METHODS: A mathematical model of the electrical activity from muscle fibres and motor units has been developed. The electrical fields around the muscle fibres are simulated using a line source model. The model permits the simulation of single muscle fibre action potentials obtained by SFEMG, concentric and Macro EMG electrodes. By using appropriate anatomical parameters EMG recordings with these electrodes can be simulated. The model is flexible and permits a number of anatomical parameters to be changed such as; number of muscle fibres in a motor unit, fibre diameter distribution, and motor end-plate geometry. Some physiological parameters can be optionally varied; firing rate, threshold for recruitment, jitter. RESULTS: In this study, simulations of CNEMG are performed and the influence of a number of parameters on the CNEMG signal is studied. It is shown that the model produces motor unit potentials reasonably well resembling those from live recordings. More important is however the relative change in MUP parameters when certain conditions are changed; number of muscle fibres in a motor unit, recording position, muscle fibre diameters and some special effects of the recording conditions. CONCLUSIONS: The simulated muscle and corresponding EMG recording can be used both as a research tool and for teaching.     

Concentric needle electrode for neuromuscular jitter analysis

We used a concentric needle electrode (CNE) with 2 kHZ low-cut filter and a single fiber electrode (SFE) in the same subjects for neuromuscular jitter measurement in the extensor digitorum communis (EDC) and orbicularis oculi (OOc) muscles. At the same session, 20 jitter values were obtained from each subject with each electrode. For EDC (during voluntary contraction), mean jitter values with SFE and CNE were 23.4 +/- 8 micros and 23.3 +/- 8 micros in 10 normals; and 56.8 +/- 28 micros and 57.4 +/- 33 micros in 10 myasthenics. For OOc (during electrical stimulation), mean jitter values with SFE and CNE were 17.9 +/- 5 micros and 16.3 +/- 4 micros in 11 normal subjects, and 41.2 +/- 29 micros and 36.7 +/- 27 micros in 10 myasthenics. For both muscles, the numbers of individual abnormal jitter values with SFE and CNE were highly comparable. Both needles labeled the same patients as having "normal" or "abnormal" neuromuscular transmission. CNE may be an alternative to SFE in neuromuscular jitter analysis.     

The effect of continuous voluntary activation on neuromuscular transmission: a SFEMG study of myasthenia gravis and anterior horn cell disorders

In normal subjects the neuromuscular jitter does not increase during continuous voluntary activation. In patients with myasthenia gravis, spinal muscular atrophy and motor neurone disease we have found that the neuromuscular jitter may increase during recordings of several minutes of continuous voluntary activation at steady innervation rates. In some units this led to impulse blocking, and in other units an initially normal jitter increased beyond the normal range. Measurement of jitter during continuous voluntary activation at steady innervation rates provides relevant information in the evaluation of neuromuscular transmission and fatigue in these disorders.     

Response of gait deficits to neuromuscular electrical stimulation for stroke survivors

Persistent gait deficits after stroke can cause falls, elevated energy cost and poor endurance. Coordination impairment is an underlying cause of gait deficits. Few efficacious interventions have been described that have targeted and measured restoration of coordinated gait components. Neuromuscular electrical stimulation can provide the critical gait practice characteristic of close-to-normal movements, by electrically inducing muscle contractions and coordinated movements that are not possible under volitional effort. Two-channel, surface neuromuscular electrical stimulation can be synchronized with phases of gait and can provide faster, more symmetrical neuromuscular electrical stimulation-assisted gait than gait with no neuromuscular electrical stimulation. Difficulties encountered during the use of surface neuromuscular electrical stimulation for gait training led to the development of neuromuscular electrical stimulation with implanted technologies. Implanted electrodes and/or stimulators proved to be feasible for gait training in stroke survivors. Gait training with a multichannel neuromuscular electrical stimulation system with implanted electrodes proved more advantageous than gait training without neuromuscular electrical stimulation, according to measures of volitional coordinated gait components (neuromuscular electrical stimulation deactivated).     

Characterization of the relationship between motor end-plate jitter and the safety factor

A mathematical model developed for a personal computer was used to simulate the jitter phenomenon in neuromuscular transmission in order to characterize the relationship between jitter and the safety factor. Four models of normal and abnormal neuromuscular transmission were investigated. In a human experiment, the mean consecutive difference (MCD) values for stimulated single-fiber electromyography at firing rates of 1, 2, 5, and 10 Hz were measured in 137 motor end-plates of myasthenic patients. Results of the computer simulations show that the relationship between jitter and the safety factor is exponential like. Variations in jitter are most prominent in end-plates with low safety factors. This relationship agrees with results of human end-plate studies. Changes in the MCD values caused by presynaptic depression or facilitation are linearly correlated to the initial jitter, whereas the logarithmic values of MCD are not. It is very important to keep in mind this nonlinear relationship when relating single-fiber jitter to the safety factor. © 1998 John Wiley & Sons, Inc. Muscle Nerve 21:628–636, 1998.     

Safety factor in single human motor end-plates studied in vivo with single fibre electromyography.

In a single fibre EMG investigation on healthy subjects the relationship between the neuromuscular jitter and the safety factor of the motor end-plate relative to curare was studied. A positive correlation was found between the jitter increase after a standard dose of D-tubocurarine and the initial jitter. The conclusion was drawn that the neuromuscular jitter can be used as an indicator of the safety factor. There are a number of nerve-muscle disorders in which the jitter is increased, probably indicating a reduced safety factor which may be of clinical importance when these patients are subjected to diseases or drugs which would otherwise not produce neuromuscular symptomatology.     

Jitter analysis with concentric needle electrode in the masseter muscle for the diagnosis of generalised myasthenia gravis

Objectives

The purpose of our study was to show neuromuscular transmission abnormality in the masseter muscle of generalised myasthenia gravis (MG) patients and to compare motor end-plate failure of the masseter with the extensor digitorum communis (EDC) and periocular muscles.

Methods

Motor end-plate function was evaluated during voluntary contraction of the masseter muscle of 20 generalised MG patients aged between 16 and 63 years, as well as 20 age-matched healthy volunteers. The mean jitter value was calculated for each group and compared. The upper limit of normal jitter was also calculated and the number of jitters exceeding this cut-off value was counted for each group for comparison. In MG patients, jitter analysis was also performed in periocular and EDC muscles along with the masseter and the number of single fibre-like potentials with abnormal jitter was counted for each muscle. All tests were performed during the same session with a concentric needle electrode (CNE).

Results

For the masseter muscle, the mean jitter of all potential pairs was significantly higher in the patient group (24.7 ± 9.6 μs in healthy volunteers, 71.9 ± 41 μs in patients). The calculated mean jitter for the 18th highest value in healthy volunteers was 33.8 ± 5.9 μs (upper 95% confidence limit was 45.6 μs). The number of abnormal jitters (?46 μs) was significantly higher in the patient group (276 out of 402 jitters) compared to healthy volunteers (10 out of 400 jitters). In the patient group, the number of single fibre-like potentials with abnormal jitter was found to be similar for the masseter, periocular and EDC muscles.

Conclusion

The masseter muscle has diagnostic importance in generalised MG. The ratio of high jitters to all of the calculated jitters in a particular muscle was similar for masseter, periocular and EDC muscles.

Significance

Jitter analysis of the masseter muscle during voluntary contraction is easy to perform and it was found as informative as other muscles in patients with generalised MG.     

Single‐fiber EMG and clinical correlation in Lambert‐Eaton myasthenic syndrome

Objectives: We analyzed 82 single‐fiber EMG (SFEMG) tests in the extensor digitorum communis muscle in 30 Lambert‐Eaton myasthenic syndrome (LEMS) patients to study the relationship between electrodiagnostic findings and clinical severity. Methods: The repetitive nerve stimulation test was performed in the abductor digiti quinti and flexor carpi ulnaris muscles. SFEMG was performed in the extensor digitorum communis muscle using the conventional method. Results: Fiber density was normal in all patients. Jitter was abnormal in all patients at the first evaluation regardless of clinical severity. The jitter was increasingly abnormal with worsening disease severity. Mean MCD correlated well with clinical and electrophysiological severity. In 5 potential pairs in 3 patients, MCD analysis in relation to firing rate showed improvement with increasing firing rates, which is consistent with presynaptic neuromuscular transmission disorders. Conclusions: In all LEMS studies, SFEMG was abnormal on the first evaluation. The mean MCD correlated well with clinical and electrophysiological disease severity on the repetitive nerve stimulation test. Muscle Nerve 47: 664–667, 2013     

Neuromuscular activation and motor-unit firing characteristics in cerebral palsy

Muscle strength, neuromuscular activation, and motor-unit firing characteristics (firing rate, recruitment, and short-term synchronization) were assessed during voluntary contractions of the medial gastrocnemius (GAS) and tibialis anterior (TA) muscles of 10 participants with spastic diplegic or hemiplegic cerebral palsy (CP). The participants (six females, four males; age range 6 to 37y) walked with equinus gait at Gross Motor Function Classification System levels II to III. These were compared with 10 age-matched controls (five females; age range 7 to 35y). Neuromuscular activation was estimated by the ratio of surface electromyogram amplitude to M-wave amplitude elicited by supramaximal electrical nerve stimulation. Participants with CP produced significantly less torque (normalized by leg length) compared with controls (TA: mean 2.3, SD 1.6 vs mean 8.9, SD 3.4Nm/m; GAS mean 13.7, SD 7.1 vs mean 28.6, SD 5.1Nm/m, p < 0.001). Neuromuscular activation during maximum voluntary contraction was significantly reduced in the participants with CP compared with controls (mean 2.4, SD 1.5 vs mean 9.7, SD 2.7Nm/m for TA; mean 1.04, SD 0.41 vs mean 3.1, SD 1.2Nm/m for GAS, p < 0.001). When compared at the same submaximal level of neuromuscular activation, motor-unit recruitment and firing rates were not different between the groups, although short-term synchronization in TA was reduced in the participants with CP. These data indicate that weakness, known to be an important component of the motor deficit in CP, has a strong central component. Although the relation between recruitment and firing rate remained substantially intact at the low and moderate force contractions tested, results suggest that the participants with CP were unable to recruit higher threshold motor units or to drive lower threshold motor units to higher firing rates.     

Single-fiber electromyography in experimental autoimmune myasthenia gravis

The sensitivity of stimulated single-fiber electromyography in the detection of early abnormalities in neuromuscular transmission in experimental autoimmune myasthenia gravis (EAMG) was tested. Increased jitter and blocking were seen up to 3 weeks before clinical illness or decrement developed. Stimulation at 10 Hz appeared more sensitive in detection of abnormalities than stimulation at 3 or 5 Hz. Jitter values did not correlate with anti-Torpedo acetylcholine receptor (AChR), nor with anti-rat AChR antibody titer. No correlation was found between jitter and AChR loss or AChR-antibody complexes in muscle. It is concluded that, in addition to AChR loss and the presence of AChR-antibody complexes, other factors must determine the neuromuscular dysfunction in EAMG and possibly myasthenia gravis.     

Some electrical properties of motor units and their effects on the methods of estimating motor unit numbers.

Long train (50) electrical pulses were applied to the ulnar, median, and deep peroneal nerves. The probability of firing, or firing index, of the first motor units (MU) recorded by surface and needle electrodes in the first dorsal interosseous, thenar, and extensor digitorum brevis muscles varied from 0-100% as the stimulus was increased from a minimum threshold voltage (V) to V + delta V (delta V = 2-4 volts). The voltage interval, delta V (firing level range), of even the first few MU recruited greatly overlapped in normal subjects, but may overlap less in some neuromuscular disorders such as amyotrophic lateral sclerosis. The mean size of the single MU isolated by (1) isometric voluntary contraction method, (2) F recurrent discharge method, and (3) stimulation at multiple locations along the nerve, were more than twice as large as the mean size of the incremental steps evoked by graded electrical stimulation. These results suggest that methods of estimating the number of MU in a muscle should incorporate a correction for the fluctuations in excitability and overlap in firing levels of MU, and in addition should include larger MU in the estimation of the mean MU potential.     

Jitter correction: a computer algorithm for reduction of the velocity recovery function artifact

The measurement of neuromuscular jitter in single fiber electromyography may be artifactually raised by a component of interdischarge interval (IDI)-dependent jitter caused by the velocity recovery function (VRF) in muscle fibers. We have developed a computer algorithm for on-line mathematical correction for this artifact, thus improving the reliability of neuromuscular jitter estimates. The method, based on a modeling technique, was validated using intramuscular stimulation in order to either exclude an IDI-dependent component (using regular stimulation) or to include an IDI-dependent component (using pseudorandom stimulation). In 10 normal subjects the distribution of 106 corrected jitter values obtained using voluntary activity showed no difference from the measured values. This finding implies that previously published values for normal jitter are not likely to have been influenced by the VRF effect.     

Stimulated single fiber electromyography in the mouse: techniques and normative data.

As the number of new transgenic mouse models of human neuromuscular disease continues to increase, the development of sophisticated electrophysiologic techniques for assessing the peripheral nervous system in these models has become important. Neuromuscular junction (NMJ) dysfunction, in particular, is often not detectable by morphologic or other techniques. To enable sensitive testing of murine NMJ function, we developed and tested a method for stimulated single fiber electromyography (S-SFEMG) in the gastrocnemius muscles of anesthetized mice. Jitter was assessed by measuring the mean consecutive latency difference (MCD) of single fiber responses to sciatic nerve stimulation at 2 HZ. Mean MCD values in normothermic mice were in the range of 6-8 micros for different strains, with no MCD values exceeding 25 micros. Reduced core temperature (to 29 degrees--30 degrees C) resulted in increased jitter, whereas intubation and mechanical ventilation of mice did not alter these values. Intraperitoneal and intravenous injection of vecuronium, however, resulted in progressively increased jitter followed by blocking in continuously monitored fibers. These observations validate the utility of S-SFEMG in mice as an index of NMJ function under a variety of physiologic conditions, and suggest that a high safety factor for neuromuscular transmission exists at mouse NMJs.     

Detecting single fiber contributions to motor unit action potentials

The ability to detect muscle fiber action potential (MFAP) contributions to motor unit action potentials (MUAPs) measured using single fiber (SF) and concentric needle (CN) electrodes was studied using simulated MUAPs. Various MFAP-acceleration thresholds were used to define significant fiber contributions. Attempts to detect the significant MFAP contributions, by locating peaks in filtered MUAPs or MUAP accelerations using various MUAP-based thresholds, were then made. Considering filtered MUAPs and a significant contribution threshold of 7.5 kV/s2, and using fiber-density peak-detection criteria, at best 46% and 50% of significant MFAP contributions were detected for the SF and CN MUAPs, respectively. Considering MUAP accelerations and a significant contribution threshold of 7.5 kV/s2, 80% and 84% of significant MFAP contributions could be detected, respectively. Most significant contributions were created from fibers located within approximately 350 microm of the electrode. The results suggest that significant peaks, defined using MUAP-based thresholds, within the acceleration of CN MUAPs can strongly correspond to individual fiber activity and may be useful for measuring fiber density and neuromuscular jitter.