NEM
AA
N
O
I
T
C
A
IT
N
N
U
O
I
R
T
O
A
T
T
I
O
T
M
AN
U
Q
L
A
I
T
OTEN
WO
P
HO
RKS
P
Paul
us,
a
h
k
r
. Ba
E
ssor
rofe
iate P
Assoc
AMERICAN ASSOCIATION OF NEUROMUSCULAR &
ELECTRODIAGNOSTIC MEDICINE
2621 Superior Drive NW
Rochester, MN 55901
(507) 288-0100
[email protected]
www.aanem.org
MD
AMERICAN ASSOCIATION OF NEUROMUSCULAR &
ELECTRODIAGNOSTIC MEDICINE
Workshop handouts are prepared as background didactic material to complement a hands-on workshop
session. This workshop handout was originally prepared in September 1994 and revised in October 2001.
The idea and opinions in this publication are solely those of the author(s) and do not necessarily represent
those of the AANEM.
Copyright © October 2001
AMERICAN ASSOCIATION OF NEUROMUSCULAR &
ELECTRODIAGNOSTIC MEDICINE
2621 Superior Drive NW
Rochester, MN 55901
Motor Unit Action Potential Quantitation
Paul E. Barkhaus, MD
Department of Neurology
Medical College of Wisconsin
“Since the measuring device has been constructed by the observer...We have to remember that
what we observe is not nature itself but nature exposed to our method of questioning.”
Werner Karl Heisenberg
(Physics and Philosophy, 1958)
INTRODUCTION
This workshop will focus on motor unit action potential
(MUAP) quantitation. It is assumed that the participant is familiar with the basic, routine needle electromyography (EMG)
examination. The goal of this workshop is to enhance his/her
ability to maximize the data they acquire in routine needle
EMG. This will be achieved in 3 general steps: review of basic
MUAP features (refer to references 24, 25, 31 and 35 for detailed reviews), demonstration of acquisition techniques, and
analysis of the MUAPs that are acquired.
Even basic contemporary electromyographs offer many options
in instrumentation, yet many electrodiagnostic medicine (EDX)
consultants have an aversion to touching the machine during
needle EMG signal acquisition except to turn the preamplifier
on and off. The general trend in what might be termed ‘subjective’ needle EMG is a passive process of watching the screen
with minimal to no interference from the operator. MUAP assessments are made on general impression (particularly amplitude)2 and the only movement is when the needle electrode is
displaced to a different recording site. At the other extreme is
computer-assisted MUAP quantitation where a great deal of operator input is required. This approach requires training, takes
time, and is not indicated in most studies.25,31,36
There is a middle ground, however, which we have called ‘objective-interactive’ needle EMG.8 This disciplines the EDX consultant to truly monitor the signals being acquired on the screen
and to make ongoing semi-quantitative or ‘objective’ assessments of amplitude, duration, complexity, area, and stability.
But these assessments are usually not achieved with a single instrumentation setting (i.e., a single sensitivity, sweep speed, and
fixed filters). Hence the operator must constantly ‘interact’ with
his/her electromyograph and manipulate settings as needed to
acquire the information desired. Combining these two concepts
yields the descriptive term, ‘objective-interactive’ needle EMG.
There are numerous computer-assisted algorithms available at
present on commercial electromyographs.15,23,36 It is not possible
to review all of them, nor is that the intention of this workshop.
Instead, a more ‘back to basics’ approach will be made that is
universal in understanding any of them, i.e., the ‘Buchthalian’
basic technique along with trigger/delay averaging.9,36 Both
require operator interaction with the electromyograph as described above.
In conjunction with the above steps toward achieving our goal,
three themes deserve mention at the outset. First, there is no
substitute for signal quality. This is something the operator controls at every level (i.e., patient cooperation, instrumentation,
analysis, and interpretation). Second, the EDX consultant must
always bear in mind that the recorded MUAP is a result of the
motor unit’s anatomy and physiology as modified by the recording device (particularly the physical characteristics of the needle
electrode). Third, despite their impressive ability to ‘quantitate’,
computers are ‘idiot savants’ and are only as good a servant as
the EDX consultant is a master.
INSTRUMENTATION
Few topics are guaranteed to alienate readers faster than instrumentation. By this we mean how the electromyograph may influence acquisition of the signal. For detailed reviews on filters,
averaging, sensitivity, and sampling rate, the participant should
2
Motor Unit Action Potential Quantitation
see references 19, 26, 30, and 31. The effects of commonly manipulated instrumentation settings on the EMG signal are summarized in Table 1.
Table 1 Effects of Instrument Settings on EMG Signals
Instrument Setting
Effect on EMG signals
Increasing highpass frequency
(low filter) from 3 Hz - 20 Hz.
Usually, the MUAP duration
decreases. The baseline becomes
stable. An artifact in the form of a
negative phase may result at the
end of the MUAP giving longer
durations.
Setting highpass frequency
MUAP amplitude and duration
(low filter) to 500 Hz.
decrease significantly. EMG may
appear ‘myopathic’.
Decreasing lowpass frequency
MUAP amplitude decreases and
(high filter) from 10 kHz to 1 kHz. the rise time increases.
Increasing the numerical value
Amplitude of large MUAPs can
of display sensitivity (i.e., from
be measured. The duration of
100 V/div to 500 V/div).
MUAPs by visual assessment may
be reduced.
Increasing the duration
Late components can be
of the sweep.
detected, but the waveform
details of the main spike may
be obscured.
Averaging waveforms.
Gives a better baseline for
duration measurements. In
unstable complex MUAPs,
the averaged waveform may lose
turns and phases that are seen in
individual discharges
of the MUAP.
Decreasing the sampling rate
Waveform amplitude may appear
below 10 kHz.
variable even in a normal MUAP
giving a false impression of
instability. Amplitude will be
underestimated.
Decreasing the number
For large amplitude MUAPs, the
of bits of the
least bit resolution may be
analog-to-digital converter.
inadequate for quantitative
analysis.
AANEM Workshop
MU Territory: “The area in a muscle over which the MFs belonging to an individual MU are distributed.”27
Motor Unit Action Potential (MUAP): “The action potential reflecting the electric activity of a single anatomic MU. It is the
compound action potential of those MFs within the recording
range of an electrode.”27
See Figure 1 for concentric needle (CN) electrode recordings.
The CN has an elliptical recording surface of 150 x 580µ
(=0.07mm2) ground to an angle of 15 degrees.18,24,25 The cannula
acts as the reference electrode. The various features of the CN
MUAP (e.g., amplitude, duration: see below) are dependent on
the location of the MFs within a given MU territory relative to
the position of the CN (see Figure 2). The so-called ‘facial’ CN
has a smaller recording surface and thinner diameter cannula
than the standard CN electrode. Therefore, reference values for
standard and facial are not interchangeable.4 Monopolar needle
(MN) electrodes differ from CN electrodes and will be considered separately below.21
Figure 1 Schematic (A) longitudinal and (B) cross sectional
view of a CN electrode recording a MUAP from a MU. The
sum of the action potentials generated by the individual MFs
(C) constitutes the MUAP (D). [Nandedkar SD. Introduction
to quantitative analysis in needle electromyography. In Johnson
EW, Pease WS (editors): Practical Electromyography, 3rd
edition. Baltimore: Williams & Wilkins; 1997. (with
permission)]
TERMINOLOGY
ACQUISITION OF MOTOR UNIT ACTION POTENTIALS
To lay the groundwork for the discussion that follows, it may be
useful to review definitions. The following are excerpted from
the AAEE Glossary of Terms in Clinical Electromyography.27
Motor Unit (MU): “The anatomic unit of an anterior horn cell, its
axon, the neuromuscular junctions, and all of the muscle fibers
(MFs) innervated by the axon.”27
The aim of the needle electrode study is to assess a sufficiently
large sample of MUAPs from a muscle. When mean values of
MUAP features (e.g., amplitude, duration, etc.) are being quantitated, 20 MUAPs are considered sufficient. The following
steps may be helpful in recording good quality signals and representing as many MUs as possible in a time-efficient manner:
AANEM Workshop
Motor Unit Action Potential Quantitation
3
In most cases, 3 corridors of sampling from one skin insertion are adequate. If another skin insertion is needed; it
must be made lateral or medial to the first insertion along
the long axis of the MFs. If one inserts proximal or distal to
the first insertion site along the long axis of the MFs, there
will be a high likelihood of sampling the same MUAPs.6,31
B. Insert the needle electrode perpendicular to the muscle with
the bevel of the CN perpendicular to the direction of the
MFs. A good strategy is to avoid the end-plate area (usually
painful to the patient) as well as the very distal portions of
the muscle at its tendinous insertions/origins.
Figure 2 Scanning EMG study through a normal biceps
brachii muscle. With a stable trigger, a CN electrode is slowly
withdrawn through a MU’s territory. A MUAP is recorded at
each “step” or stop-point along the corridor. (Courtesy of
Professor Erik V. Stålberg).
A. A helpful strategy is to move the needle electrode slowly
along ‘corridors’ while the patient minimally activates the
muscle.6 A corridor is the path along which the recording
tip of the electrode passes into the muscle, typically from
superficial to deep. The best positions for activation will
vary with each muscle, but may need to be modified if significant weakness is present.6
When the electrode reaches the end of the corridor (limited
by either the length of the electrode or the size of the
muscle), the needle electrode is withdrawn to the subcutaneous border of the muscle and the tip redirected about 45
degrees lateral to each side of the first corridor, but perpendicular to the long axis of the MFs. This gives 2 additional corridors.6 When using a CN electrode, the choice of
recording site may influence the MUAPs recorded in a
muscle.17 MUAPs recorded from superficial sites may have
smaller amplitudes and shorter durations due to the influence of the cannula.17 MUAPs may also vary depending on
where they are recorded between the end-plate zone and
tendon.17 The latter also applies to CN electrodes, but likely
MN electrodes as well, though this is yet to be demonstrated. In general, this is most significant in larger muscles
(e.g., biceps brachii). In recording MUAPs, this emphasizes
the need for sampling from different depths when using a
CN electrode. Some authors recommend the use of standardized recording sites.17
C. MU territory will vary between muscles (e.g., approximately
10 mm in the biceps brachii) which should be used as an aid
to know how far to move the electrode to exit the previous
MU’s presumed territory. Recruitment threshold may also
reveal whether the same MUAP is being recorded, but
from a different position in the MU. Depending on where
in the MU territory it is recorded, the MUAP may change
considerably in shape. This is readily demonstrated on scanning EMG studies (see Figure 2).24,25
D. Patient comfort and cooperation is critical in acquiring high
quality signals. In planning an expedient study, decide on
what muscles need first priority to best address the clinical
question. If possible, try to defer typically uncomfortable
muscles until last. Needle electrodes will dull with prolonged use as in lengthy studies. For this reason, there may
be merit in some instances to studying sensitive muscles
(e.g., abductor pollicis brevis) earlier when the electrode is
sharpest.
Figure 3 MUAP Features. Dots represent phases and
arrows indicate turns. Shaded portion represents area.
[Nandedkar SD. Introduction to quantitative analysis in needle
electromyography. In Johnson EW, Pease WS (editors):
Practical Electromyography, 3rd edition. Baltimore: Williams &
Wilkins; 1997. (with permission)]
4
AANEM Workshop
Motor Unit Action Potential Quantitation
Table 2 Anatomic Correlates of MUAP MUAP Features
MUAP Feature
Anatomic Correlate
Amplitude
Size and number of muscle fibers that are
within 0.5 mm of the active recording
surface.
Size and distance of the nearest muscle
fiber of the motor unit.
Area
Size and number of muscle fibers that are
within 2 mm of the active recording surface.
Amplitude
Figure 4 Topographic distribution of MFs contributing to
Phases
Temporal dispersion of action
the various MUAP features in 2 schematic motor units, ‘A’ and
‘B’, using a CN and MN, respectively. These are based on
computer simulations. In ‘A’, the radius from the CN electrode tip to the perimeters demarcating amplitude and
phases/turns are 500 , and 1000 , respectively and
conform to the semicircles at the recording surface of the CN
(to the left of the vertical dotted line). In contrast, the uptake
area for duration is circular: MFs as far away as 2500 contribute to duration. The progressively enlarging concentric
circles in ‘B’ show the relative uptake areas for MN MUAP features: amplitude, phases/turns, and duration, respectively.
[Nandedkar SD, Barkhaus PE. Quantitative EMG analysis. In:
Katirji B, Kaminski HJ, Preston DC, Ruff RL, Shapiro BE
(editors): Neuromuscular disorders in clinical practice. Boston:
Butterworth-Heinemann; 2001. (with permission)]
and
potentials of muscle fibers that are
Turns
within 1 mm of the active recording
surface, i.e., action potential propagation
velocity in terminal axon branches and
muscle fibers, action potential propagation
velocity in muscle fibers, and the width of
the end-plate zone.
Duration
Number and size of muscle fibers that
are within 2.5 mm of the active
recording surface.
Waveform Stability
Efficacy of neuromuscular transmission at
the motor end-plates of muscle fibers that
are within 1 mm of the active recording
surface.
MOTOR UNIT ACTION POTENTIAL FEATURES
The MUAP is described in terms of various features (e.g., amplitude, area, etc.). These features have different implications as
to how many MFs may be contributing to them, e.g., amplitude
versus duration.22,24,25 These are summarized in Table 2. This
section applies only to the CN electrode (Figures 3 and
4).8,11,12,18,24,25 The differences in MN electrode features will be discussed later.
Amplitude: This is measured from the maximum negative to the
maximum positive peak (see Figure 3). The smaller facial CN
electrode records higher amplitudes than the standard CN electrode.4
Duration: Duration is measured from the initial deviation of the
signal from the baseline to the return to the baseline and is composed of the action potentials of those MFs lying within a 2.5
mm radius of the recording surface of the CN. Temporal dispersion of the single MF action potentials may be increased in
pathological processes, such as myopathy, 1,5,7,8,25 resulting in increased MUAP durations without a commensurate increase in
the number of MFs in the MU under consideration. Thus a
typical finding is one of long- duration, complex MUAPs in my-
opathy. This is to be separated from long-duration, complex
MUAPs as seen in neurogenic processes which are due to an increased number of MFs within a MU (i.e., reinnervation).
Therefore, in order to increase diagnostic sensitivity, polyphasic
MUAPs are conventionally excluded from duration measurements.25
Area: This is measured as the area under the rectified waveform
within a defined duration.24,25
Phase: A phase is the portion of the MUAP waveform between
the departure from, and return to, the baseline.27 A simple
method to measure phases is to count the number of crossings
of the baseline and add one. Simple CN-EMG MUAPs have
fewer than 5 phases. In most normal muscle, fewer than 12% of
MUAPs may be polyphasic with slightly higher percentages
allowed in proximal muscles.3,8,24,25
Turn: A turn is essentially a positive or negative peak within the
MUAP duration. Successive turns are separated by a threshold
amplitude (25-100 (V) to exclude peaks generated by noise. The
use of the term in this context is not synonymous with the use
of ‘turn’ as in interference pattern analysis. Increased turns
AANEM Workshop
Motor Unit Action Potential Quantitation
5
(more than 5 turns) were found to be a sensitive but nonspecific
feature in MUAPs from patients with neuromuscular diseases.23,24,35 In practical application, slight movement of the
recording electrode may change a phase into a turn and vice
versa.
Complexity: When turns and/or phases are increased or late components are present, the MUAP is considered ‘complex.’ An increase in phases, turns or late components (i.e., satellite
components) may result from an increase in the number of MFs
in a MU. This occurs in reinnervation. When there is an increase
in variability in the diameter of MFs of a MU, or increase in the
velocity recovery function of the MFs. These mechanisms are
typical for myopathic process. Therefore, an increase in complexity per se is a sensitive, but nonspecific indicator of abnormality. A small percent of MUAPs in normal muscle may be
complex, especially in proximal muscles.23,36
Rise Time: For a MUAP, rise time represents the time interval
between the maximum amplitude to the preceding maximum
positive deflection or the onset of the main spike. Traditional
teaching suggests that this be less than 0.5 ms. This was based
upon earlier guidelines which were recently reconsidered to be
too restrictive.2,31 One generally believes that the electrode
recording tip is ‘closest’ to the MU when the amplitude is
maximal, rather than solely relying on rise time.2 It must also be
noted that rise time may, by necessity, be longer, as in neurogenic disease, when amplitude and area are increased, thereby
resulting in enlargement of the main spike. Even in normal
muscle, a rise time of greater than 0.5 ms may be acceptable for
MUAPs.2,31
Stability: This is a very useful, but often overlooked, feature. It is
synonymous with variability, which is defined as the “change in
the MUAP shape between consecutive discharges.”8,25,27 Unless
the MUAP is complex or amplitude variations are marked, this
may be difficult to perceive at standard instrumentation settings.
An alternative and more revealing approach is to ‘focus’ the
recording tip to only those MFs quite close to it (Figure 5),
thereby increasing the ‘selectivity’ of the recording electrode.8,25,28
This can be done with either a CN or MN electrode.
To assess stability, changes in instrumentation are necessary.
The operator activates the trigger/delay, increases the low-frequency filter to 500 Hz (this ought to be familiar to those who
have tried single fiber EMG), increases the sweep speed to at
least 1 or 0.5 ms/div, and if possible, superimposes the consecutively acquired traces. If ‘instability’ is present, it will be seen as
increased variation between consecutive interpotential intervals
(i.e., increased jitter), or in some cases, actual ‘blocking’ (i.e., loss
of one or more of the presumed single MF action potentials observed). Large degrees of instability may be seen at slower sweep
speeds (Figure 6).8,25 Stålberg and colleagues have also intro-
Figure 5 MUAP stability. Observation: Using the 500 Hz
low filter, as in Single Fiber EMG, enhances the action potentials of the closest muscle fibers. It also reduces the MUAP
amplitude from 800 V to 500 V and reduced the duration
from 7 ms to 4 ms. Conclusion: This is a stable MUAP.
duced the concept of vertical instability of the MUAP, or ‘jiggle’,
which will not be considered here.34
Instability is rare in normals, but common in neuromuscular diseases, particularly neurogenic ones.8,25 In the latter, denervation
is typically a fairly short-term process (i.e., days) whereas reinnervation may take weeks to months. In chronic neurogenic
processes we infer that the instability most likely represents
ongoing reinnervation changes. The MUAPs in myopathic disorders are generally stable.1,5,7,8,24,25 Instability is also the electrophysiologic hallmark of disorders of neuromuscular
transmission. If one observes marked instability in a weak
patient with otherwise normal appearing MUAPs, then a disease
such as myasthenia gravis should be considered. An important
caveat is to be certain that the low-frequency filter is set back to
2 Hz. This can be a common oversight rendering the subsequent MUAPs ‘myopathic’ in appearance!8,25
Area/Amplitude Ratio: This feature quantifies the thickness of the
MUAP waveform.20 The ‘thin’ waveforms seen in myopathic
processes have a commensurately reduced numerical value of
this feature.
Effect of Electrode Position on MUAP Features: Except for the
MUAP duration and the area/amplitude ratio, all MUAP
feature values may change significantly with a slight change in
the position of the recording electrode within the MU territory.20
6
Motor Unit Action Potential Quantitation
AANEM Workshop
being studied, the distribution of the endplates within the
muscle, and etceteras.
In contrast to the schematic representation in Figure 4 of the
recording territories of each electrode being similar in size, the
actual overall recording area of the MN is larger than the CN.
In the author’s experience, however, MUAP durations for these
electrodes are similar. This is probably explained by the minimal
contributions to the MUAP by the more distant MFs and the
greater inherent ‘noise’ associated with the MN electrode.
Duration may also be affected (i.e., reduced) by recording
MUAPs at higher low-frequency settings, e.g., 10-20 Hz versus
2-3 Hz.
Figure 6 MUAP stability. Observations: Increased jitter and
block. Conclusion: This is an unstable MUAP. Is there abnormal neuromuscular transmission? This will depend on further
observations of features and stability in other MUAPs, in addition to other EDX studies.
As a result, one may observe several different MUAP waveforms for the same MU with minimal displacements of the
needle electrode (Figures 2, 15).20,32
MONOPOLAR NEEDLE (MN) ELECTRODE
Over the years there has been ongoing debate as to the superiority of needle recording electrodes in EMG, mainly CN versus
MN electrodes. For routine or quantitative EMG, either is acceptable, as long as the electrode is of good quality and the operator appreciates the differences in their physical characteristics
and how this affects the MUAPs they record. The reference
values for a muscle using CN cannot be applied to MN electrode recordings, although MUAP durations tend to be
similar.4,21
The features that tend to show the most difference are MUAP
amplitude and phases, both being higher when using the MN
electrode.21 This is due to the differences in the physical characteristics of the MN electrode, which in turn alter the pattern by
which the MFs within a MU’s territory are recorded (Figure 4).
The MN electrode’s conical tip has a larger recording surface
than the standard CN. Therefore, the MN electrode records
from more MFs immediately adjacent to the tip whose action
potentials contribute to the amplitude feature. Also, because
these MFs are often not discharging completely in synchrony,
there tends to be some phase interaction and therefore polyphasia in the MUAP recorded from the MN. This is a general statement and is of variable importance depending on the muscle
QUANTITATIVE EMG TECHNIQUES
Evolution of Techniques
Given the number of techniques for quantitating MUAPs, it
may be useful to briefly review, in chronological order of development, the 3 main categories: manual,9,31 trigger/delay,31,36 and
decomposition.15,23,25
In the manual method, MUAPs are identified from strip chart
recordings of the EMG signal or by direct observation of the
signal on the screen (see Figures 9 and 10). In the trigger-delay
method, successive discharges of a single MUAP are averaged
to extract the waveform (see Figure 11). Decomposition is a
computer-based method that mimics the manual method and
may be more successful at high activation levels at which the
manual method may not be able to resolve individual MUAP
waveforms.
One can make the same observation of these styles as Dale
Berra did of his father, Yogi: “Our similarities are different.”
Although the techniques differ, MUAP duration remains a very
robust measurement and really shows no differences between
techniques in normal controls.23 This is not surprising since
MUAP duration shows minimal change when manipulating the
CN electrode.20
MUAP amplitude values are largest in the trigger/delay technique. This might be anticipated since the CN electrode is intentionally manipulated to maximize amplitude. Amplitude
values are lowest in the manual technique where CN electrode
position is not manipulated. Values for MUAP amplitude using
the decomposition technique fall between the trigger/delay and
manual techniques.23
The percent of complex MUAPs is lowest in the manual technique where the needle is least manipulated and highest in the
AANEM Workshop
Motor Unit Action Potential Quantitation
7
Selection of MUAP Quantitation Technique
There are a number of computer-assisted techniques available
to quantitate MUAPs. Discussion of the details of the advanced,
computer-assisted programs such as those utilizing decomposition are beyond the scope of this handout. Commercially available electromyographs typically offer a MUAP analysis program
using one or more of the above techniques. There is no ‘best’
technique. The technique that will be most suitable will depend
Figure 8 Trigger/delay technique of measurement using 3
high quality superimposed sweeps which can easily be averaged with little loss of amplitude or turns (i.e., serration less
preferred term). Observations: A triphasic, serrated MUAP
having an amplitude of 450 V and 9.2 ms duration.
Conclusion: This is a normal MUAP.
on the operator, the equipment available, and the purpose of the
study. If choosing a technique is an option, then it is recommended that any available literature or studies utilizing the technique be reviewed prior to use.
Outlier Approach
Figure 7 Free run MUAPs at 10 ms/div (total 100 ms each
sweep; sensitivity 200 V/ division vertical in A and B and 100
V /division vertical in C). Normal MUAPs (A and B) with myopathy (C). On free run mode, a waveform should occur at
least 3 times before being considered a bona fide MUAP.
Therefore in ‘A,’ note the probable ‘pseudo-MUAP’ on trace
4, representing the phase interaction or simultaneous occurrence of the other two MUAP waveforms which are each
seen more than 3 times. Exercise: How many MUAPs are in
each series (A, B, and C) and how fast are they firing?
Discussion is at the end of this handout.
trigger/delay technique for the opposite reason. The decomposition technique shows values similar to the manual technique.
Even the most efficient recording techniques and algorithms
consume time and it can be argued that ‘quantitative EMG’ is
only needed in equivocal cases or for research. How can these
seemingly opposite views be reconciled since every EDX consultant needs to have the ability to discriminate ‘normal’ from
‘abnormal’?
One method is the outlier approach, which is similar to that utilized in single fiber EMG. Instead of collecting a cohort of
MUAPs from a muscle and calculating their mean values to determine normalcy, one can identify individual MUAPs with features lying clearly outside of the range of reference values.33 If a
8
Motor Unit Action Potential Quantitation
AANEM Workshop
muscle obviously and rapidly shows consistent abnormality (i.e.,
similar changes in MUAP features) in more than 3 to 4 MUAPs,
then the study of that muscle is concluded and the examiner
proceeds to the next muscle. Only if the pattern of abnormality
is inconsistent (e.g., both reduced and increased MUAP duration) would further evaluation be necessary. The outlier approach is somewhat similar to routine ‘subjective’ needle EMG
in that individual MUAPs are being assessed. The ‘outlier’ approach in ‘objective-interactive’ EMG simply disciplines the operator to observe the MUAPs critically in terms of their features,
for a consistent pattern of abnormality that lies outside reference values for individual MUAPs. Therefore in lieu of the subjective descriptor ‘large MUAP’, one ascertains that the MUAP
is captured on the screen, measured, and quickly identified as
being in or outside of reference limits for amplitude or duration
with additional comment on area, complexity, and stability. The
EDX consultant must constantly run MUAPs through this
‘checklist’ of MUAP features in assessing each MUAP.
Although seemingly arduous and time consuming, with practice
it can be performed quite rapidly.8,25
comparison to other reference values. Therefore the EDX consultant must ensure that there is no variation in needle electrode
or instrumentation (i.e., no variation in filter or sensitivity settings in the amplifier). It must also be stressed that most reference values published are mean values29 and that the concept of
ranges of individual values is a more recent consideration in
MUAP measurement.33
Description of MUAPs
Recruitment pattern and firing rate can easily be measured on
free run mode, especially when rastered sweeps are reviewed
(see Figures 9 and 10). If a MUAP appears to ‘shift left’ on successive traces, then the interdischarge interval (1/ firing rate in
Hertz) is less than the sweep duration on the screen and may be
quickly inferred to be firing at more than 10 Hz. A ‘shift right’
indicates the opposite, i.e., discharging at less than 10 Hz.
In reporting findings, the EDX consultant must remember that
there is a difference between electrophysiologic descriptions of
MUAPs versus making diagnostic implications.14 The concept
of ‘myopathic’ versus ‘neurogenic’ MUAPs is just that, only a
general gestalt that borders on jargon. It is preferable to describe
the MUAPs in terms of their features. Using the objective-interactive approach describe above, one can offer a number of
observations on the signals recorded, and in turn, make some
correlations to neuromuscular pathology. This is shown by
example, as much as possible, in the Figure legends (e.g., Figures
5-10 and 12-14).
Recruitment/Interference Pattern
While the size and configuration of the MUAP may give clues
to abnormality, the rate of firing of a particular MUAP and
overall number of MUAPs firing at a given level of effort may
also contribute information. Interference pattern (IP) is defined
as “the electrical activity recorded from a muscle with a needle
electrode at maximal voluntary effort.”27 Recruitment is defined
as the “orderly activation of the same and new MUs with increasing strength of voluntary contraction.”27 Attempts to
analyze IP have been ongoing for many years pari passu with
MUAP analysis.24,25
Recruitment of a new MUAP can be easily recognized on a
rastered display (i.e., the display of consecutive sweeps as in
Based on all of the electrophysiologic data, the EDX consultant
may then offer some interpretation, but not with the intention
of misleading the referring clinician away from other possible diagnoses. For instance, a general summary such as “rapidly recruited, stable MUAPs of normal to reduced amplitude, reduced
duration, occasionally complex with thin appearing main
spikes” followed by an interpretation of ‘myopathic process’
informs the referring clinician that there is cogent electrophysiologic evidence to support the presence of a myopathy.
Reference Values
This term is preferred to that of ‘normal’ values. Based on the
above, it should be obvious that reference values can be, in part,
derived from the literature3,23,29,36 and indeed each EMG laboratory should utilize published values as a means of validating
their own results. However, it cannot be emphasized enough,
that in doing so, the technique used in one’s lab must be absolutely symmetric to that used in the literature so as to permit
Figure 9 Visual assessment of recruitment: 2 MUAPs.
Observations: (1) The initial MUAP seen beginning at the top
trace is triphasic with 500 V amplitude and 6-8 ms duration
firing at 11-12 Hz (note it is ‘shifting left’ on successive traces.
(2) Recruitment of a new MUAP is observed on the 4th trace
(amplitude = 200 V, duration = 9 ms). The recruitment interval is 87 ms and corresponds to a recruitment frequency of
11 Hz. (3) The newly recruited MUAP is discharging at slightly
more than 5 Hz. Conclusion: Normal recruitment of normal
appearing MUAPs, but based on ‘objective’ observation.
AANEM Workshop
Motor Unit Action Potential Quantitation
9
The various features may be abnormal in isolation or in combination. The variation in size and configuration of MUAPs is
Figure 11 Schematic cross-section of a normal MU (left) and
that of a MU in muscular dystrophy (center). There are 3 positions of the needle electrode shown along a single corridor
through the MU territory. The MUAPs at these positions are
shown on the right (a,b,c) to emphasize how variable the
waveforms may be within the same MU. [From Nandedkar
SD. Quantitative electromyography in clinical electrodiagnosis.
In: Goodgold J (editor): Rehabilitation Medicine. St. Louis: CV
Mosby; 1988, p 97. (with permission)]
Figure 10 Visual assessment of recruitment using a 10 s
epoch. Horizontal marker (‘H’ bar demarcated by arrows) in
the top trace (A) shows 1s from the beginning of the trace that
is displayed immediately below. Observation: A single MUAP is
firing at 8 Hz. In the lower trace from the same epoch (B), the
marker is shifted to the right so as to display a segment where
more recruitment activity is occurring. Observation: (1) The
same MUAP as seen above is again appreciated, but now discharging at 12 Hz; (2) second, larger amplitude MUAP is now
recruited with an initial discharge frequency of 8 Hz.
Conclusion: This is normal recruitment.
Figures 7, 9, 13) of the EMG signal. The firing rate immediately
before the recruitment of a new MUAP is the recruitment frequency (Figure 9).25 Recruitment frequency is typically increased
in neurogenic processes and is subjectively recognized as fastfiring single MUs on the screen.
Neuromuscular Diseases
The above discussion has been aimed toward MUAPs in
normal controls. A comprehensive discussion of findings in
neuromuscular disorders is beyond the scope of this
handout.5,7,8,10,13,24,25,36 Here, as described above, how to define the
various MUAP features is stressed, and in a basic way, how to
recognize when these may be ‘outside’ a given set of reference
values (i.e., outliers).33
Table 3 Patterns of MUAP Abnormalities in Discriminating
between Neurogenic and Myopathic Disorders
(Specific abnormalities are indicated by the ◆.)
MUAP Feature
Amplitude
Myopathic Process
Neurogenic Process
Usually normal or reduced◆ Increased and may
Occasionally mild increase
be 5 to 10 times
normal◆
Area
Normal or reduced◆
Increased◆
Duration
Reduced for simple MUAPs◆ Increased◆
Reduced or increased for
complex MUAPs
Area/Amplitude Reduced◆
Normal or
Ratio
increased◆
Phases and
Increased
Increased
Turns
Stability
Usually stable
Unstable during
disease activity and
progression
best appreciated when groups of MUAPs from normal and neuromuscular diseases are compared (see Figure 15). This emphasizes why several MUAPs from a single muscle may need to be
recorded in mild or questionable situations as alterations in MU
architecture in neuromuscular disease are not homogeneous
(see Figure 11).
10
Motor Unit Action Potential Quantitation
Typical patterns of MUAP abnormality in neuromuscular disorders are summarized in Table 3 (see also Figures 12, 13, 14). The
relationships of individual MUAP features in pathology are
summarized in Table 4. This is useful to interpret the MUAP
findings when the patterns observed in a particular patient do
not appear to conform to expected abnormalities such as those
described in Table 3. This is more likely to occur in severely
weakened or mild (i.e., subclinical) cases.7 Finding consistency in
AANEM Workshop
pattern of abnormality in the MUAPs of a muscle, and in turn
other affected muscles, greatly strengthens the EDX consultant’s ability to reach an EDX impression of the underlying
disease process, based on all electrophysiologic data. This
handout, focuses only on MUAPs. Obviously nerve conduction
data, the presence of rogue, ‘sub-MUAP’ signals (e.g., complex
repetitive or myotonic discharges), and fibrillation or fasciculation potentials may greatly influence the overall interpretation of
the study.
CONCLUSION
The advent of computers has greatly facilitated quantitation of
the EMG signal. It is important, however, to remember that
quantitative analysis as a technique is relatively old, beginning
with the work of Buchthal and colleagues in the late 1940’s. It is
a tribute to the meticulous quality of their efforts that their ref-
Figure 12 Measurement of a MUAP: Note the increased
sweep speed. Observations: The MUAP shows 3 fairly well
superimposed sweeps of a biphasic MUAP of 400 V amplitude and 1.74 ms duration. Conclusion: This pattern is compatible with a myopathic process. Although amplitude is
normal, duration is reduced, a specific finding in myopathy.
Figure 14 The objective approach in measurement of a
MUAP. Observations: (1) 2 superimposed sweeps of this
MUAP show that it is triphasic with an amplitude of 3 mV. (2)
The duration is at least 19.8 ms. Conclusion: These observations are indicative of a neurogenic process.
Figure 13 Observations: (1) There are at least 4 simple and
thin MUAPs (best appreciated on the second trace) with 200800 V amplitude (normal amplitude values). (2) The durations are reduced, perhaps less than 5 ms. (3) The firing rates
approach 20 Hz. Since all are firing during these sweeps, recruitment interval cannot be measured. (4) The initial negative
deflection in all of the MUAPs indicates that the recording
electrode is near the end-plate zone. Conclusion: These
MUAPs are compatible with a myopathic process.
erence mean values for MUAP durations are still widely utilized.29 This in turn should reinforce to the EDX consultant the
importance of consistency in technique. Each study must be approached with a high degree of organization and method. One
may then extract much more meaningful information from each
needle EMG study with the above ‘objective-interactive’ approach, without really expending more effort. To quote Yogi
Berra, “You can see a lot just by observing.”
The author acknowledges Sanjeev D. Nandedkar for reviewing
this manuscript.
AANEM Workshop
11
Motor Unit Action Potential Quantitation
Table 4 Pathologic Correlations of MUAP features
Neurogenic Process
➾
Hypertrophy of muscle
fibers
Reinnervation
Question selective
loss of smaller
motor units
➾
Atrophy of muscle fibers
Atrophy of muscle
fibers
➾
Increased variability of
fiber diameter
Reinnervation
Question selective
loss of smaller
motor units
➾
Loss of muscle fibers
Question fractionation of reinnervated
motor units
➾
Increased variability of
fiber diameter
Increased width of
end-plate zone.
Slow conduction in
atrophic muscle
fibers
Slow conduction in
newly formed
terminal axon
branches
Question reinnervation
Reinnervation
Amplitude
Amplitude
Duration
Duration
Polyphasia
Figure 15 CN electrode recorded MUAPs in myopathic
process (A), normal (B), and neurogenic process (C).
Observations: (1) There are a wide range of MUAP waveforms within each group. (2) Note how some MUAPs from
each ‘abnormal’ category could easily be interchanged with
those in the ‘normal’ group, emphasizing the need for adequate sampling in mild disorders (Courtesy of Professor Erik V.
Stålberg).
REFERENCES
11. Barkhaus PE, Nandedkar SD, Sanders DB. Quantitative EMG in
inflammatory myopathy. Muscle Nerve 1990;3:247-253.
12. Barkhaus PE, Nandedkar SD. On the selection of concentric needle
EMG motor unit action potentials for analysis: is the rise time criterion too restrictive? Muscle Nerve 1995;19:1554-1560.
13. Barkhaus PE, Periquet MI, Nandedkar SD. Quantitative MUAP
analysis in paraspinal muscles. Muscle Nerve 1997;20:373-375.
14. Barkhaus PE, Roberts MM, Nandedkar SD. “Facial” and concentric
needles are not interchangeable. Muscle Nerve 1998;21:1602.
15. Barkhaus PE. The electrodiagnostic evaluation of the patient with
suspected myopathy. In Kincaid JC (editor): 1998 Course E: The
young, the old, and the weak: a basic course in electrodiagnostic medicine. Rochester, MN: American Association of Electrodiagnostic
Medicine, 1998, pp 23-39.
16. Barkhaus PE, Nandedkar SD. Electronic myoanatomic atlas for clinical electromyography (on CD-ROM). Hopewell Junction: CASA
Engineering; 1998.
17. Barkhaus PE, Periquet MI, Nandedkar SD. Quantitative electromyographic studies in sporadic inclusion body myositis. Muscle
Nerve 1999;22:480-487.
Pathologic Correlate
Myopathic Process
➾
MUAP Abnormality
Instability
Increased velocity
recovery function in
muscle fibers
18. Barkhaus PE, Nandedkar SD. Electronic atlas of waveforms in clinical electromyography (on CD-ROM). Hopewell Junction: CASA
Engineering; 1999.
19. Buchthal F. AAEM Minimonograph #15: An introduction of electromyography. Rochester, MN: American Association of
Electrodiagnostic Medicine; 1981.
10. Buchthal F. Electrophysiological signs of myopathy as related with
muscle biopsy. Acta Neurologica (Napoli) 1977;32:1-29.
11. Buchthal F, Guld C, Rosenfalck P. Action potential parameters in
normal human muscle and their dependence on physical variables.
Acta Physiol Scandan 1954;32:200-219,1954.
12. Buchthal F, Pinelli P, Rosenfalck P. Action potential parameters in
normal human muscle and their physiological determinants. Acta
Physiol Scandan 1954;32:219-229.
13. Buchthal F, Kamieniecka Z. The diagnostic yield of quantified electromyography and quantified muscle biopsy in neuromuscular disorders. Muscle Nerve 1982;5:265-280.
14. Daube JR. The description of motor unit potentials in electromyography. Neurology 1978;28:623-625.
15. Dorfman LJ and McGill KC. AAEM Minimonograph #29:
Automatic quantitative electromyography. Rochester, MN:
American Association of Electrodiagnostic Medicine; 1988.
16. Dumitru D, King JC, Nandedkar SD. Comparison of single-fiber
and macro electrode recordings: relationship to motor unit action
potential duration. Muscle Nerve 1997;20:1381-1388.
12
Motor Unit Action Potential Quantitation
17. Falck B, Stålberg EV, Bischoff C. Influence of recording site within
the muscle on motor unit potentials. Muscle Nerve 1995;18:13851389.
18. Gydikov A, Gerilovsky L, Kostov K, Gatev P. Influence of some
features of the muscle structure on the potentials of motor units by
means of different types of needle electrodes. Electromyogr Clin
Neurophysiol 1980;20:299-321.
19. Maccabee PJ and Nasser NF. AAEM Minimonograph #39: Digital
filtering: basic concepts and application to evoked potentials.
Rochester, MN: American Association of Electrodiagnostic
Medicine; 1992.
20. Nandedkar SD, Barkhaus PE, Sanders DB, Stålberg EV. Analysis of
amplitude and area of concentric needle EMG motor unit action potentials. Electroenceph Clin Neurophysiol 1988;69:561-567.
21. Nandedkar SD, Sanders DB. Recording characteristics of monopolar EMG electrodes. Muscle Nerve 1991;14:108-112.
22. Nandedkar SD, Sanders DB, Stålberg EV, Andreassen S. Simulation
of the concentric needle EMG motor unit action potential. Muscle
Nerve 1988;11:151-159.
23. Nandedkar SD, Barkhaus PE, Charles A. Multi-motor unit action
potential analysis (MMA). Muscle Nerve 1995;18:1155-1166.
24. Nandedkar SD. Introduction to quantitative analysis in needle electromyography. In Johnson EW, Pease WS (editors): Practical electromyography, 3rd edition. Baltimore: Williams & Wilkins; 1997.
25. Nandedkar SD, Barkhaus PE. Quantitative EMG analysis. In: Katirji
B, Kaminski HJ, Preston DC, Ruff RL, Shapiro BE (editors):
Neuromuscular disorders in clinical practice. Boston: ButterworthHeinemann; 2001.
26. Nilsson J, Panizza M, Hallett M. Principles of digital sampling of a
physiologic signal. Electroenceph Clin Neurophysiol 1993;89:349358.
27. Nomenclature Committee. AAEE glossary of terms in clinical electromyography. Muscle Nerve 1987;10(8S).
28. Payan J. The blanket principle: a technical note. Muscle Nerve
1978;1:423-426.
29. Rosenfalck P, Rosenfalck A. Electromyography and sensory/motor
conductions: findings in normal subjects. Copenhagen: Laboratory
of Clinical Neurophysiology, Rigshospitalet; 1975.
30. Spehlmann R. Evoked potential primer. Boston: Butterworth;1985,
21-51.
31. Stålberg EV, Andreassen S, Falck B, Lang H, Rosenfalck A,
Trojaborg W. Quantitative analysis of individual motor unit action
potentials: A proposition for standardized terminology and criteria
for measurement. J Clin Neurophysiol 1986;3(4):313-348.
32. Stålberg EV, Dioszeghy P. Scanning EMG in normal muscle and in
neuromuscular disorders. Electroenceph Clin Neurophysiol
1991;81: 403-416.
AANEM Workshop
33. Stålberg EV, Bischoff C, Falck B. Outliers, a way to detect abnormality in quantitative EMG. Muscle Nerve 1994;17:392-399.
34. Stålberg EV, Sonoo M. Assessment of variability in the shape of the
motor unit action potential, the “jiggle” at consecutive discharges.
Muscle Nerve 1994;17:1135-1144.
35. Stålberg EV, Nandedkar SD, Sanders DB, Falck B. Quantitative
motor unit action potential analysis. J Clin Neurophysiol
1996;13:401-422.
36. Stewart C, Nandedkar SD, Massey JM, Gilchrist JM, Sanders DB.
Evaluation of an automatic method of measuring features of motor
unit action potentials. Muscle Nerve 1989;12:141-148.
Answers to Exercise (Figure 7)
(A) There are 2 MUAPs, both seen on the first trace. The larger, first
MUAP, is discharging about 8 Hz: it is shifting right on subsequent
discharges and therefore firing at less than 10 Hz. The second,
smaller MUAP, is discharging slightly faster at 9-10 Hz throughout
the traces.
(B) Three distinct waveforms are seen. The initial, smallest one, is discharging at about 12 Hz and 3 similar waveforms are noted on traces
1, 2, and 4. Three discharges of the second MUAP are also seen
(traces 2, 3, and 4) shifting right with an estimated discharge frequency of 9 Hz. The third MUAP (second MUAP on trace 2) is seen
only twice: therefore, since it is seen less than 3 times in these sweeps,
we will not count it.
(C) This is tricky! The first, really discernible MUAP is the 280 V amplitude waveform on trace 1 at the third division marker. It may have
recurred again at the end of trace 1 but it is cut off. It recurs near the
eighth division marker on trace 2, but on trace 3 at the fifth division
marker it appears different: the smaller 80 V component to the
right of the main spike is missing. Is this a different MUAP or is this
smaller component blocking? Note the time interval between these
components changes between its first and second discharge, suggesting instability and therefore favoring the latter possibility. Use of
trigger/ delay with the low pass filter set up to 500 Hz could resolve
this, or by carefully measuring more discharges manually. Therefore,
this would not be called a single unique MUAP until more information is available.
The second 200 V amplitude, complex MUAP (trace 1, just before
the seventh division marker) appears consistent and shifting left,
therefore discharging at a frequency of greater than 10 Hz. The
smaller MUAPs along the baseline are difficult to assess and it is important to make certain they are not satellites or linked components
to larger waveforms. Remember that MUAPs of less than 50 V amplitude would not be included in analysis.
NEM
AA
N
O
I
T
C
A
IT
N
N
U
O
I
R
T
O
A
T
T
I
O
T
M
AN
U
Q
L
A
I
T
OTEN
WO
P
HO
RKS
P
Paul
us,
a
h
k
r
. Ba
E
ssor
rofe
iate P
Assoc
AMERICAN ASSOCIATION OF NEUROMUSCULAR &
ELECTRODIAGNOSTIC MEDICINE
2621 Superior Drive NW
Rochester, MN 55901
(507) 288-0100
[email protected]
www.aanem.org
MD
AMERICAN ASSOCIATION OF NEUROMUSCULAR &
ELECTRODIAGNOSTIC MEDICINE