Long-Term Endurance and Power Training May Facilitate Motor Unit Size Expansion to Compensate for Declining Motor Unit Numbers in Older Age.

The evidence concerning the effects of exercise in older age on motor unit (MU) numbers, muscle fiber denervation and reinnervation cycles is inconclusive and it remains unknown whether any effects are dependent on the type of exercise undertaken or are localized to highly used muscles. MU characteristics of the vastus lateralis (VL) were assessed using surface and intramuscular electromyography in eighty-five participants, divided into sub groups based on age (young, old) and athletic discipline (control, endurance, power). In a separate study of the biceps brachii (BB), the same characteristics were compared in the favored and non-favored arms in eleven masters tennis players. Muscle size was assessed using MRI and ultrasound. In the VL, the CSA was greater in young compared to old, and power athletes had the largest CSA within their age groups. Motor unit potential (MUP) size was larger in all old compared to young (p < 0.001), with interaction contrasts showing this age-related difference was greater for endurance and power athletes than controls, and MUP size was greater in old athletes compared to old controls. In the BB, thickness did not differ between favored and non-favored arms (p = 0.575), but MUP size was larger in the favored arm (p < 0.001). Long-term athletic training does not prevent age-related loss of muscle size in the VL or BB, regardless of athletic discipline, but may facilitate more successful axonal sprouting and reinnervation of denervated fibers. These effects may be localized to muscles most involved in the exercise.

Failure to expand the motor unit size to compensate for declining motor unit numbers distinguishes sarcopenic from non-sarcopenic older men.

KEY POINTS: The age-related loss of muscle mass is related to the loss of innervating motor neurons and denervation of muscle fibres. Not all denervated muscle fibres are degraded; some may be reinnervated by an adjacent surviving neuron, which expands the innervating motor unit proportional to the numbers of fibres rescued. Enlarged motor units have larger motor unit potentials when measured using electrophysiological techniques. We recorded much larger motor unit potentials in relatively healthy older men compared to young men, but the older men with the smallest muscles (sarcopenia) had smaller motor unit potentials than healthy older men. These findings suggest that healthy older men reinnervate large numbers of muscle fibres to compensate for declining motor neuron numbers, but a failure to do so contributes to muscle loss in sarcopenic men. ABSTRACT: Sarcopenia results from the progressive loss of skeletal muscle mass and reduced function in older age. It is likely to be associated with the well-documented reduction of motor unit numbers innervating limb muscles and the increase in size of surviving motor units via reinnervation of denervated fibres. However, no evidence exists to confirm the extent of motor unit remodelling in sarcopenic individuals. The aim of the present study was to compare motor unit size and number between young (n = 48), non-sarcopenic old (n = 13), pre-sarcopenic (n = 53) and sarcopenic (n = 29) men. Motor unit potentials (MUPs) were isolated from intramuscular and surface EMG recordings. The motor unit numbers were reduced in all groups of old compared with young men (all P < 0.001). MUPs were higher in non-sarcopenic and pre-sarcopenic men compared with young men (P = 0.039 and 0.001 respectively), but not in the vastus lateralis of sarcopenic old (P = 0.485). The results suggest that extensive motor unit remodelling occurs relatively early during ageing, exceeds the loss of muscle mass and precedes sarcopenia. Reinnervation of denervated muscle fibres probably expands the motor unit size in the non-sarcopenic and pre-sarcopenic old, but not in the sarcopenic old. These findings suggest that a failure to expand the motor unit size distinguishes sarcopenic from pre-sarcopenic muscles.

The reliability of methods to estimate the number and size of human motor units and their use with large limb muscles.

PURPOSE: Current methods for estimating muscle motor unit (MU) number provide values which are remarkably similar for muscles of widely differing size, probably because surface electrodes sample from similar and relatively small volumes in each muscle. We have evaluated an alternative means of estimating MU number that takes into account differences in muscle size. METHODS: Intramuscular motor unit potentials (MUPs) were recorded and muscle cross-sectional area (CSA) was measured using MRI to provide a motor unit number estimate (iMUNE). This was compared to the traditional MUNE method, using compound muscle action potentials (CMAP) and surface motor unit potentials (sMUPs) recorded using surface electrodes. Data were collected from proximal and distal regions of the vastus lateralis (VL) in young and old men while test-retest reliability was evaluated with VL, tibialis anterior and biceps brachii. RESULTS: MUPs, sMUPs and CMAPs were highly reliable (r = 0.84-0.91). The traditional MUNE, based on surface recordings, did not differ between proximal and distal sites of the VL despite the proximal CSA being twice the distal CSA. iMUNE, however, gave values that differed between young and old and were proportional to the muscle size. CONCLUSION: When evaluating the contribution that MU loss makes to muscle atrophy, such as in disease or ageing, it is important to have a method such as iMUNE, which takes into account any differences in total muscle size.

Decomposition of intramuscular EMG signals using a knowledge -based certainty classifier algorithm.

An automated system for resolving an intramuscular electromyographic (EMG) signal into its constituent motor unit potential trains (MUPTs) is presented. The system is intended mainly for clinical applications where several physiological parameters for each motor unit (MU), such as the motor unit potential (MUP) template and mean firing rate, are required. The system decomposes an EMG signal off-line by filtering the signal, detecting MUPs, and then grouping the detected MUPs using a clustering and a supervised classification algorithm. Both the clustering and supervised classification algorithms use MUP shape and MU firing pattern information to group MUPs into several MUPTs. Clustering is partially based on the K-means clustering algorithm. Supervised classification is implemented using a certainty-based classifier technique that employs a knowledge-based system to merge trains, detect and correct invalid trains, as well as adjust the assignment threshold for each train. The accuracy (93.2%±5.5%), assignment rate (93.9%±2.6%), and error in estimating the number of MUPTs (0.3±0.5) achieved for 10 simulated EMG signals comprised of 3-11 MUPTs are encouraging for using the system for decomposing various EMG signals.

Probabilistic muscle characterization using QEMG: application to neuropathic muscle.

Clinicians who use electromyographic (EMG) signals to help determine the presence or absence of abnormality in a muscle often, with varying degrees of success, evaluate sets of motor unit potentials (MUPs) qualitatively and/or quantitatively to characterize the muscle in a clinically meaningful way. The resulting muscle characterization can be improved using automated analysis. As such, the intent of this study was to evaluate the performance of automated, conventional Means/Outlier and Probabilistic methods in converting MUP statistics into a concise, and clinically relevant, muscle characterization. Probabilistic methods combine the set of MUP characterizations, derived using Pattern Discovery (PD), of all MUPs detected from a muscle into a characterization measure that indicates normality or abnormality. Using MUP data from healthy control subjects and patients with known neuropathic disorders, a Probabilistic method that used Bayes' rule to combine MUP characterizations into a Bayesian muscle characterization (BMC) achieved a categorization accuracy of 79.7% compared to 76.4% using the Mean method (P > 0.1) for biceps muscles and 94.6% accuracy for the BMC method compared to 85.8% using the Mean method (P < 0.01) for first dorsal interosseous muscles. The BMC method can facilitate the determination of "possible," "probable," or "definite" levels for a given muscle categorization (e.g., neuropathic) whereas the conventional Means and Outlier methods support only a dichotomous "normal" or "abnormal" decision. This work demonstrates that the BMC method can provide information that may be more useful in supporting clinical decisions than that provided by the conventional Means or Outlier methods.

Decision support for QEMG.

For clinicians to use quantitative electromyography (QEMG) to help determine the presence or absence of neuromuscular disease, they must manually interpret an exhaustive set of motor unit potential (MUP) or interference pattern statistics to formulate a clinically useful muscle characterization. A new method is presented for automatically categorizing a set of quantitative electromyographic (EMG) data as characteristic of data acquired from a muscle affected by a myopathic, normal or neuropathic disease process, based on discovering patterns of MUP feature values. From their numbers of occurrence in a set of training data, representative of each muscle category, discovered patterns of MUP feature values are expressed as conditional probabilities of detecting such MUPs in each category of muscle. The conditional probabilities of each MUP in a set of MUPs acquired from an examined muscle are combined using Bayes' rule to estimate conditional probabilities of the examined muscle being of each category type. Using simulated and clinical data, the ability of a "pattern discovery" based Bayesian (PD-based Bayesian) method to correctly categorize sets of test MUP data was compared to conventional methods which use data means and outliers. The simulated data were created by modeling the effects of myopathic and neuropathic diseases using a physiologically based EMG signal simulator. The clinical data was from controls and patients with known neuropathic disorders. PD-based Bayesian muscle characterization had an accuracy of 84.4% compared to 51.9% for the means and outlier based method when using all MUP features considered. PD-based Bayesian methods can accurately characterize a muscle. PD-based Bayesian muscle characterization automatically maximizes both sensitivity and specificity and provides transparent rationalizations for its characterizations. This leads to the expectation that clinicians using PD-based Bayesian muscle characterization will be provided with improved decision support compared to that provided by the status quo means and outlier based methods.

Bayesian characterization of external anal sphincter muscles using quantitative electromyography.

OBJECTIVES: Based on the analysis of electromyographic (EMG) data muscles are often characterized as normal or affected by a neuromuscular disease process. The objective of this work was to compare the accuracy of Bayesian muscle characterization to conventional means and outlier analysis of motor unit potential (MUP) feature values. METHODS: Quantitative MUP data from the external anal sphincter muscles of control subjects and patients were used to compare the sensitivity, specificity, and accuracy of the methods under examination. RESULTS: The results demonstrated that Bayesian muscle characterization achieved similar accuracy to combined means and outlier analysis. Thickness and number of turns were the most discriminative MUP features for characterizing the external anal sphincter (EAS) muscles studied in this work. CONCLUSIONS: Although, Bayesian muscle characterization achieved similar accuracy to combined means and outlier analysis, Bayesian muscle characterization can facilitate the determination of "possible", "probable", or "definite" levels of pathology, whereas the conventional means and outlier methods can only provide a dichotomous "normal" or "abnormal" decision. Therefore, Bayesian muscle characterization can be directly used to support clinical decisions related to initial diagnosis as well as treatment and management over time. Decisions are based on facts and not impressions giving electromyography a more reliable role in the diagnosis, management, and treatment of neuromuscular disorders. SIGNIFICANCE: Bayesian muscle characterization can help make electrophysiological examinations more accurate and objective.

Motor unit potential characterization using "pattern discovery".

Typically in clinical practice, electromyographers use qualitative auditory and visual analysis of electromyographic (EMG) signals to help infer if a neuromuscular disorder is present and if it is neuropathic or myopathic. Quantitative EMG methods exist that can more accurately measure feature values but require qualitative interpretation of a large number of statistics. Electrophysiological characterization of a neuromuscular system can be improved through the quantitative interpretation of EMG statistics. The aim of the present study was to compare the accuracy of pattern discovery (PD) characterization of motor unit potentials (MUPs) to other classifiers commonly used in the medical field. In addition, a demonstration of PD's transparency is provided. The transparency of PD characterization is a result of observing statistically significant events known as patterns. Using clinical MUP data from normal subjects and patients with known neuropathic disorders, PD achieved an error rate of 30.3% versus 29.8% for a Naïve Bayes classifier, 30.1% for a Decision Tree and 29% for discriminant analysis. Similar results were found for simulated EMG data. PD characterization succeeded in interpreting the information extracted from MUPs and transforming it into knowledge that is consistent with the literature and that can be valuable for the capture and transparent expression of clinically useful knowledge.

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.

Decomposition and quantitative analysis of clinical electromyographic signals.

Procedures for the quantitative analysis of clinical electromyographic (EMG) signals detected simultaneously using selective or micro and non-selective or macro electrodes are presented. The procedures first involve the decomposition of the micro signals and then the quantitative analysis of the resulting motor unit action potential trains (MUAPTs) in conjunction with the associated macro signal. The decomposition procedures consist of a series of algorithms that are successively and iteratively applied to resolve a composite micro EMG signal into its constituent MUAPTs. The algorithms involve the detection of motor unit action potentials (MUAPs), MUAP clustering and supervised classification and they use shape and firing pattern information along with data dependent assignment criteria to obtain robust performance across a variety of EMG signals. The accuracy, extent and speed with which a set of 10 representative 20-30 s, concentric needle detected, micro signals could be decomposed are reported and discussed. The decomposition algorithms had a maximum and average error rate of 2.5% and 0.7%, respectively, on average assigned 88.7% of the detected MUAPs and took between 4 to 8 s. Quantitative analysis techniques involving average micro and macro MUAP shapes, the variability of micro MUAPs shapes and motor unit firing patterns are described and results obtained from analysis of the data set used to evaluate the decomposition algorithms are summarized and discussed.

A longitudinal study of the pathophysiological changes in single human thenar motor units in amyotrophic lateral sclerosis.

The sequence of pathophysiological changes in amyotrophic lateral sclerosis (ALS) at the single motor unit (MU) level is not well understood. Using a recently described technique, a comprehensive range of physiological properties in two thenar MUs in ALS were intensively studied. In the first MU, despite a marked decline in the ability of the subject to voluntarily recruit the MU, the physiological properties of this MU remained remarkably stable over a 2-year period. In contrast, the physiological properties of the other MU declined rapidly over 5 months despite the fact that this MU could be recruited with ease throughout the study period. These differences between the progressively dysfunctional changes in these two MUs illustrates the value of such longitudinal studies of specific MUs in improving our understanding of the evolution of changes in single motoneurons in ALS. The broader application of longitudinally tracking the pathophysiological changes of the surviving MUs may prove to be a sensitive measure of disease progression and in evaluating the effectiveness of treatments.

Methods for estimating the numbers of motor units in human muscles.

After the introduction by A. J. McComas of the original method for estimating the number of motor units based on manual incremental stimulation of a motor nerve, several new techniques have been developed, designed to correct for some of the errors inherent in the original technique. These methods incorporate algorithms to adjust for alternation and, to a greater or lesser extent, automate the methods, rendering the techniques less subject to operator bias and various physiological and technical errors. This review explores the advantages and drawbacks in the multiple-point stimulation (MPS), spike-triggered averaging (STA), and decomposition-enhanced STA techniques, illustrates some of the current applications of the techniques, and explores some tantalizing prospects for new studies of motor-unit physiology in the future.

Physiological properties of single thenar motor units in the F-response of younger and older adults.

The purpose of this study was to characterize the properties of single thenar motor units in the F-response of healthy younger (n = 15; age 33 +/- 11 years) and older subjects (n = 15; age 68 +/- 3 years). Trains of 300 stimuli at intensities evoking M-potentials 10%, 20%, and 30% of the peak-to-peak amplitude of the maximum M-potential, were delivered to the median nerve. In the young, observed firing probabilities of surface-detected motor unit action potentials (S-MUAPs) extracted from the F-response ranged from less than 1-10%, the S-MUAPs varied in size from 0.015% to 5.3% of the maximum M-potential negative peak area, and they were similar in size to the population of S-MUAPs collected by multiple point stimulation of the median nerve. The percentage difference between the slowest and fastest conducting fibers for individual subjects ranged from 8% to 20%, which translated to conduction velocities (CVs) of 48-68 m.s-1 (mean 59 +/- 4). The preceding were all independent of stimulus intensity. The S-MUAP sizes were significantly larger in older subjects (39%), and the range and distribution of motor unit CVs (38-61 m.s-1; mean 52 +/- 3) were markedly shifted to reflect a slower population of motor fibers. These findings suggest that age-related axonal slowing may uniformly affect all median motor fibers.

Motor unit number estimates based on the automated analysis of F-responses.

An automated technique for estimating the number of motor units based on single motor unit action potentials in the F-response is described. The average surface detected motor unit action potential (S-MUAP) was calculated from the datapoint-by-datapoint average of a sample of S-MUAPs automatically selected from a population of F-responses. The technique was applied to the thenar muscles of young (n = 18, aged 31 +/- 11 years) and older (n = 15, aged 68 +/- 3) subjects. Motor unit number estimates based on the automated selection of S-MUAPs from the F-responses compared well with those derived using a computer-assisted manual method for selecting S-MUAPs from the F-response (automated 245 +/- 105 vs. manual 241 +/- 100, r = 0.93) and were similar to estimates obtained using multiple point stimulation (219 +/- 77). The advantages of the automated technique for collecting S-MUAPs from the F-response include the ready tolerance of the technique by subjects, the minimal amount of operator interaction required, and the additional information relating to the conduction velocities and latencies of single motor axons.

Determinants of mean motor unit size: impact on estimates of motor unit number.

The purpose of this study was to compare two fundamentally different methods of deriving the average surface-detected motor unit action potential (S-MUAP) size from which to calculate a motor unit number estimate (MUNE), namely: (1) the simple arithmetic average of S-MUAP parameter values; and (2) a computer-derived datapoint by datapoint average waveform which takes account of differences in S-MUAP shapes and durations. Multiple point stimulation was used to collect representative samples of between 11 and 20 S-MUAPs (mean 15 +/- 2 SD) from the median-innervated thenar muscles of 20 healthy control subjects between 20 and 76 years of age (mean 48 +/- 19 SD). The average S-MUAP size based on peak-to-peak amplitude, negative peak amplitude, and negative peak area measurements was calculated using the two different methods. The mean S-MUAP sizes based on the average waveform were significantly lower in all cases than those based on the simple average of S-MUAP parameter values. Differences tended to be greatest for MUNEs based on peak-to-peak amplitude (35%), less for negative peak amplitude (20%), and least for negative peak area (16%).

Simulation of electromyographic signals.

A technique for simulating electromyographic (EMG) signals is reported. The four step method is physiologically based and begins with the modelling of a cross-section of a muscle. Within this cross-section motor unit territories of various sizes are randomly distributed and within a detection area at the centre of the cross-section individual muscle fibres are modelled and randomly assigned to appropriate motor units. For the motor units with fibres in the detection area, recruitment and firing time behaviours, as a function of an assumed level of contraction, are then simulated. For each active motor unit with fibres in the detection area, motor unit action potentials (MUAPs) are created using a line source volume conductor model. MUAPs can be created for various types of detecting electrodes including concentric and monopolar needle electrodes. Finally, the individual motor unit firing time behaviours and MUAPs are combined to create a complete EMG signal. The routines are interfaced through a series of user-friendly menus and forms, are implemented in C and can be run on any IBM compatible machine. Exemplary data are presented along with results from attempts to evaluate the model. Suggested uses of the simulation package, especially with respect to EMG signal decomposition, are discussed.

Probabilistic inference-based classification applied to myoelectric signal decomposition.

A new probabilistic inference based technique (IBC) for the classification of motor unit action potentials (MUAP's) is presented. This new technique discovers statistically significant relationships in the data and uses these relationships to generate classification rules. The technique was applied to the classification of MUAP's extracted from simulated myoelectric signals. Its performance was compared to that of classical template matching algorithms (TBC) applied to the same data. Using 32 time samples as features to represent the MUAP's it was found that the IBC based technique performed significantly better (p less than 0.005) than the TBC algorithms (83.0 +/- 2.6% versus 78.1 +/- 2.8% peak correct classification performance). As the size of the training set was reduced or as increasing numbers of random classification errors were introduced into the training data, the performance of the IBC and TBC techniques declined similarly. IBC performance remained superior until very small training sets (less than 30 MUAP's per motor unit) or training sets with large numbers of errors (greater than 50%) were used. Because the probabilistic inference technique can utilize nominal data it has the potential to use declarative problem domain knowledge which conceivably could improve its performance.