M responses and twitch contractions were evoked in single motor units (MUs) of the first dorsal interosseus muscle by intramuscular microstimulation of motor axons. Two-hundred nine MUs were studied in 21 subjects. Thirty-five MUs (17%) showed F waves in addition to M responses. Twitch force was used to provide an indirect measure of MU size; additionally, twitch contraction time was measured. There was no select group of MUs generating F waves with regard to the above contraction parameters. However, four of five MUs with very high twitch forces, above 70 mN, generated F waves. We conclude that MUs of all sizes produce F waves with similar probability. Only few MUs with very strong twitch forces, i.e., very large MUs, may be more subject to F-wave production and may be involved in the generation of the so-called repeater F waves. 0 1992 John Wiley & Sons, Inc.
Key words: F wave
motor unit twitch force contraction time MUSCLE & NERVE 15~1138-1142 1992
F WAVES AND MOTOR UNIT SIZE REINHARD DENGLER, MD, ANDON KOSSEV, PhD, KAI WOHLFAHRT, MD, MARGOT SCHUBERT, MD, JOSEF ELEK, MD, and WERNER WOLF, Drlng
T h e muscle compound action potential elicited by a supramaximal shock to a motor nerve is frequently followed by a small, variable response called the F wave.I ,5.8- 1 l . I h . 2 0 I t is caused by backfiring of a fraction of motoneurons (about 1% of the pool) in response to aritidromic excitation. ' ' - I 4 This fact, and the finding that identical waveforms probably produced by the same motor units may occur in consecutive trials, led to the question of involvement of a select group of motoneurons. Kimura et al." typed motoneurons according to functional axon properties, i.e., excitability and conduction velocity, and could not find a select group being more subject to F-wave production. T h e recurrence of identical waveforms, the so-called repeater F waves,"' remained unexplained. They concluded that niotoneuron properties not tested in their study might play a role in
From the Department of Neurology, University of Bonn, Bonn, Germany (Drs. Dengler, Kossev, Wohlfahrt. Schubert. and Elek), and the Department of Computer Sciences of Bundeswehr University Munich, Neubiberg. Germany (Dr Wolf) Acknowledgments. This study was supported by the Deutsche Forschungsgemeinschaft and by the Alexander-von-Humboldt Foundation (Dr Kossev) Dr Kossev's present address is Acad G Bontchev Str BI. 21, Central Laboratory of Biophysics, Bulgarian Academy of Sciences, 113 Sofia, Bulgaria. Address reprint requests to Dr Reinhard Dengler, Neurologische Klinik der Universitat Bonn, Sigmund-Freud-Str 25, D-5300 Bonn 1, Germany. Accepted for publication March 25, 1992
CCC 0148-639X/92/101138-05 0 1992 John Wiley & Sons, Inc
1138
F Waves and Motor Unit Size
F-wave generation. l'herefore, we investigated F waves in individual MUs which w e could type according to contractile properties. We used motor unit (MU) twitch force, a size-dependent parameter, and contraction time, which is fairly independent of MU size in man.15,18319 MATERIALS AND METHODS
Twenty-one normal subjects, aged from 23 to 54 years (13 men a n d 8 women), were investigated. T h e first dorsal interosseus muscle of the hand was studied in all cases. MU action potentials and twitch contractions were evoked by intramuscular microstimulation of motor axons by a modification" of the method described by 'Taylor and Stevens.'H Single axons wert stimulated by means of a bipolar needle electrode inserted into the muscle proximal to the midbelly. T h e evoked M U action potentials were monitored using surface electrodes in a belly tendon arranagement (bandpass 20 to 3000 Hz). l ' h e associated MU twitches were recorded by a virtually isometric force transducer positioned at the radial side of the proximal phalanx of the index finger (filters 0 to 500 Hz). T h e stimulation electrode was moved in minute steps to find intramuscular nerve fibers. Stimuli were applied at a rate of 1 Hz and an intensity of less than 3 mA (duration 0.05 to 0.1 ms). A single M U response was assumed if an identical action potential occurred in an all-or-none fashion when changing the stimulus intensity several times."," T h e responses of 20 to 40 stimuli were
MUSCLE & NERVE
October 1992
on-line digitized (sampling rate for EMG 10,000 Hz and for force 1000 Hz), averaged, and stored on hard disc. Measurements were repeated once or twice, and each identified MU was stimulated at least 50 times. M U action potentials were monitored to ensure that the same MU was stimulated throughout. The twitches were analyzed with respect to their peak forces [twitch force, (TF)] and contraction times (CT). An original record showing the evoked M U action potential and the associated twitch contraction is illustrated in Figure 1. Generally, the stimulus evoked only one MU action potential with short latency corresponding to an M response. In some MUs, the M response was occasionally followed by a later one, which revealed an identical waveform and a latency compatible with that of an F wave. Such trials were documented on paper, but excluded from averaging to avoid errors in MU twitch parameter assessment. T h e latencies of the F waves were measured on a display. RESULTS
A total of 209 MUs were collected in the 21 subjects. N o F waves were found in the 82 MUs of 6 subjects. The remaining 15 subjects revealed F waves in 27% of their MUs (35 of 127 MUs), which is 17% in relation to the total number of
A
T 1 0 . 1 mV
MUs. The rate of occurrence of F waves in a given MU was not systematically analyzed, but was generally low and similar in all MUs. Usually, one F wave was found in every 10 to 20 consecutive trials as is seen in the record illustrated in Figure 2. Three F waves in 10 consecutive trials, observed in 1 subject, were an extreme and a clear exception. The TFs of all 209 MUs are shown in Figure 3A. An exponential distribution with a clear preponderance of weaker, i.e., small MUs, became apparent. The distribution of the TFs of the 82 MUs of the subjects without F waves is plotted in Figure 3C and that of all 127 MU of the subjects with F waves in Figure 3B. The TFs of only the 35 MUs with F waves are plotted in Figure 3D. Comparison of the four distributions revealed a close similarity, indicating that there was no difference between these groups of Mus with respect of their TFs. The fact that the exponential shape of- the distribution in Figure 3D was somewhat distorted seemed rather an effect of the small absolute number of MUs with F waves than a true biological phenomenon. Obviously, MUs with weak and strong TFs, i.e., small and large motoneurons, generated F waves with similar probabilities. The MUs with weaker TFs produced F waves in a greater number than their stronger counterparts simply because they were more numerous. There may, however, be an indication of a special role of a few MUs with very strong TFs. Four of the five MUs with twitch forces above 70 mN produced F waves (see Fig. 3D).
Twitch Forces (TFs).
B I
1’ I
50 ms FIGURE 1. Motor unit action potential (M response) (A) and twitch contraction (6)evoked by intramuscular microstimulation of a motor axon. The stimulus artifact is seen in front of the M response. Twenty-five trials were averaged.
F Waves and Motor Unit Size
-H 5 ms FIGURE 2. Ten consecutive trials with intramuscular microstimulation of a motor axon. The M responses are followed in one trace by an F wave with an identical waveform and a latency of 34 ms.
MUSCLE & NERVE
October 1992
1139
I
07 1
0.1
O0 0 2
O3
1
A
A
[ n=209 ]
[ n=209 ]
h
O 2
B
7
[ n=127]
h
5
r: cc 0
e,
L
9
BE
5
[ n=82 ]
-
c)
cd
2
[ n=82]
I)
> .-I
-2
0
.->
C
C
0.3 02 01
0.1
I
0.0
0.0 0.2
0.7
-
-
11111 0
20
1L 1
1
[ n=35 ]
[ n=35 ]
40
60
80
100
120
140
0
FIGURE 3. Distribution of motor units in relative numbers (n = absolute number) according to their twitch forces (bin width 4 mN). (A) All motor units studied. (B)All motor units of the subjects (15) with F waves. (C) All motor units of the subjects (6) without F waves. (D) All motor units with F waves. The four distributions are fairly similar. However, note that 4 of 5 motor units above 70 mN generated F waves.
T h e C T values were more symmetrically distributed and revealed no correlation with the TFs (r = -0.002) in accordance with findings of other authors. 15,18,19 T h e CTs of all MUs and of the other three groups of MUs as defined for the TFs above (see Fig. SA-D) are shown in Figure 4A-D. As with TFs, the four distributions were similar and did not reveal a select group of MUs producing F waves. Contraction Times (CTs).
T h e latencies of the F waves ranged from 30 to 35 ms. As expected, they were longer than those for F waves evoked from the ulnar nerve at the wrist in routine diagnosis. Since the
Latencies.
F Waves and Motor Unit Size
20
40
60
80
100
120
140
Contraction time [msl
Twitch force [mN]
1140
D
FIGURE 4. Distribution of motor units in relative numbers (n = absolute number) according to their contraction times. For (A) to (D) see legend to Figure 3. The four distributions are fairly similar.
site of the intramuscular axon stimulation could not be exactly determined and compared among the different MUs, we did not use the F-wave latencies for typing the MUs. DlSCUSSlON
Our study shows that an MU can generate F waves even if stimulated selectively. Obviously, it is not essential that a large number or all motoneurons of a pool are activated by an antidromic volley, as is assumed by one a ~ t h o r . ~Although ” various factors determining the state of excitability of an alpha-motoneuron may influence the generation of F waves, a putative interaction of the antidromic impulses on the spinal level seems to be of low importance.
MUSCLE & NERVE
October 1992
F waves are produced by a very small fraction of MUs of a muscle.8-13 The question whether a specific subpopulation of motoneurons is more than others subject to F-wave production is old. A series of structural and functional properties of the motoneuron and of the entire MU are determined by motoneuron Theoretically, the larger motoneurons should be more subject to F-wave production, since their initial segments can repolarize more quickly after antidromic depolarization and will more likely be re-excitable for recurrent impulses than the initial segments of the smaller m o t o n e u r o n ~ The .~~~ fact ~ ~that ~~ F waves are produced by a very small fraction of motoneurons could also point to large motoneurons as F-wave generators because of their small number. Kimura et al.,' taking axonal excitability and conduction velocity as measures of motoneuron size, could not find such a relationship. However, the authors discussed some sources of possible error associated with assessment of motor axon excitability from the surface, and with analyzing F-wave latencies without knowledge of the length of the fine axon terminals. We used twitch force and contraction time to characterize the MUs generating F waves. T F is cclosely related to the size of the motoneuron as is axon conduction velocity and excitability used by Kimurd et al.' This is not the case for CT, at least in man,1.5,18,19 This approach allows reliable typing of MUs, although the number of Mus obtained in individual subjects is limited. However, the distribution of the total of T F values in our study resembles that obtained by other techdniques. 15,1'3 It seems that the 209 MUs studied are a fairly representative sample for the first dorsal interosseous muscle and that selection bias is not a critical factor. The results concerning T F affirm the findings of Kimura et al.' We did not find a relationship between the T F of a MU and its capability of F-wave generation. Obviously, motoneurons of' all sizes can produce F waves with a similar probability. Since small motoneurons are more numerous, they generate the majority of F waves. There may, however, be an exception concerning the small number of Mus with very strong TFs, i.e., the very large motoneurons. They may be more susceptible of producing F waves, as indicated by the finding that four of five Mus with twitch forces above 70 mN revealed such responses. One could speculate that these large motoneurons, which are regularly antidromically stimulated by supramaximal nerve
F Waves and Motor Unit Size
stimulation, may be involved in the generation of the repeater F waves. Furthermore, typing of MUs according to their CTs did not result in selection of a subgroup being more subject to F-wave generation. Apparently, the factors which influence CTs of human MUs and which are mainly size-independent15,.'8,19do not play a relevant role in F-wave generation either. The rate of occurrence of F waves within an individual MU proved to be low and similar among the different MUs, in line with findings of a sin le fiber EMG study by Schiller and Stalberg.'7 There was no relationship to the parameters used for typing. Our study provides no information on the conditions which enable only some stimuli to evoke F waves. Both the low rate of occurrence of F waves in an individual MU and the limited number of MUs capable of generating F waves explain why the F wave in response to a supramaximal nerve stimulus is produced by a very small fraction of the motoneurons of the pool. The F-wave latencies recorded in distal hand muscles in routine studies vary within a range of several milliseconds. Since the conduction velocity of a motor axon depends on the size of the mot o n e u r ~ n ~this ' ~ *latency ~ scatter is explained by the finding that motoneurons of all sizes are involved in F-wave generation. Provided that a sufficient number of trials is carried out, the smallest and largest motoneurons may generate F waves and the range of latencies may be fairly representative for the range of motor axon conduction velocities in the nerve under study. In conclusion, motoneurons of all sizes produce F waves with similar probability. Other, sizeindependant factors, e.g., spontaneous or synaptically induced oscillations of the membrane potential and, thus, of the state of the excitability of a motoneuron at the time when the antidromic volley arrives, must account for the fact that the F waves are generated by a fairly small fraction of MUs. Only a few very large motoneurons may be more subjects to F-wave production and may be the generators of the so-called repeater F waves.
REFERENCES 1. Conrad B, Aschoff JC, Fischler M: Der diagnostische Wert d e r F-Wellen-Latenzen.J Neurol 1975;210:151- 159. 2 . Dengler R, Stein RB, Thomas CK: Axonal conduction velocity and force of single human motor units. Musrle Nrrvr 1988;11:136- 145. 3. Elek JM, Dengler R, Konstdnzer A, Hesse S, Wolf W: Mechanical implications of paired motor unit discharges in
MUSCLE & NERVE
October 1992
1141
pathological and voluntary tremor. Electroencephalogr Clin Neurophysiol 1991;8 1 :279- 283. 4. Fisher MA: F waves: comments on the central control of recurrent discharges. Muscle Nerve 1979;2:406. 5. Fisher MA: F-response latency-duration correlations: an argument for the orderly antidromic activation of motoneurons. Muscle Nerve 1980;3:437-438. 6. Henneman E, Mendell LM: Functional organization of motoneuron pool and its inputs, in Brooks VB (ed): Handbook of Physiology. Section 1: The Nervous System. Bethesda, MD, American Physiological Society, 1981, vol 2, pp 423507. 7. Kernell D: Input resistance, electrical excitability, and size of ventral horn cells in cat spinal cord. Science 1966; 152~1637-1640. 8. Kimura J: Clinical value and limitations of F-wave determination. A comment. Muscle Nerve 1978; 1:250-252. 9. Kimura J, Yanagisawa H, Yamada T , Mitsudome A, Sasaki H, Kimura A: Is the F-wave elicited in a selected group of motoneurons? Muscle Nerve 1984;7:392-399. 10. Macleod WN: Repeater F waves: A comparison of sensitivity with sensory antidromic wrist-to-palm latency and distal motor latency in the diagnosis of carpal tunnel syndrome. Neurology 1987;37:773-778. 11. Magladery JW, McDougas DB Jr: Electrophysiological studies of nerve and reflex activity in normal man. 1. Identification of certain reflexes in the electromyogram and the conduction velocity of peripheral nerve fibres. Bull Johns Hopkins Hosp 1950;86:265-290.
1142
F Waves and Motor Unit Size
12. Mayer RF, Feldman RG: Observations on the nature of the F wave in man. Neurology 1967;17:147-156. 13. McLeod JG, Wray SH: An experimental study of the F wave in the baboon. J Neurol Neurosurg Psychzatry 1966; 29:196-200. 14. Miglietta OE: T h e F response after transverse myelotomy, in Desmedt J E (ed): New Developmenls in Electromyography and Clinical Neurophysiology. Basel, Karger, 1973, vol 3, pp 323-327. 15. Milner-Brown HS, Stein RB, Yemm R: The contracile properties of human motor units during voluntary isometric contractions. J Physiol 1973a;228:285-306. 16. Peioglou-Harmoussi S, Fawcett PRW, Howel D, Barwick DD: F-responses: a study of frequency, shape and amplitude characteristics in healthy control subjects. J Neurol Neurosurg Psychiatry 1985;48: 1159- 1164. 17. Schiller HH, Stalberg E: F responses studied with single fibre EMG in normal subjects and spastic patients. J Neurol Neurosurg Psychiatry 1978;4 1:45- 53. 18. Taylor A, Stephans JA: Study of human motor unit contractions by controlled intramuscular microstimulation. Brain Res 1976;117:331-335. 19. Thomas CK, Johansson RS, Westling G , Bigland-Ritchie B: Twitch properties of human thenar motor units measured in response to intraneural motor-axon stimulation. j Neurophysiol 1990;64:1339- 1346. 20. Young RR, Shahani BT: Clinical value and limitations of F wave determination. Muscle Nerve 1978; 1 :248-250.
MUSCLE & NERVE
October 1992
Key words: F wave
motor unit twitch force contraction time MUSCLE & NERVE 15~1138-1142 1992
F WAVES AND MOTOR UNIT SIZE REINHARD DENGLER, MD, ANDON KOSSEV, PhD, KAI WOHLFAHRT, MD, MARGOT SCHUBERT, MD, JOSEF ELEK, MD, and WERNER WOLF, Drlng
T h e muscle compound action potential elicited by a supramaximal shock to a motor nerve is frequently followed by a small, variable response called the F wave.I ,5.8- 1 l . I h . 2 0 I t is caused by backfiring of a fraction of motoneurons (about 1% of the pool) in response to aritidromic excitation. ' ' - I 4 This fact, and the finding that identical waveforms probably produced by the same motor units may occur in consecutive trials, led to the question of involvement of a select group of motoneurons. Kimura et al." typed motoneurons according to functional axon properties, i.e., excitability and conduction velocity, and could not find a select group being more subject to F-wave production. T h e recurrence of identical waveforms, the so-called repeater F waves,"' remained unexplained. They concluded that niotoneuron properties not tested in their study might play a role in
From the Department of Neurology, University of Bonn, Bonn, Germany (Drs. Dengler, Kossev, Wohlfahrt. Schubert. and Elek), and the Department of Computer Sciences of Bundeswehr University Munich, Neubiberg. Germany (Dr Wolf) Acknowledgments. This study was supported by the Deutsche Forschungsgemeinschaft and by the Alexander-von-Humboldt Foundation (Dr Kossev) Dr Kossev's present address is Acad G Bontchev Str BI. 21, Central Laboratory of Biophysics, Bulgarian Academy of Sciences, 113 Sofia, Bulgaria. Address reprint requests to Dr Reinhard Dengler, Neurologische Klinik der Universitat Bonn, Sigmund-Freud-Str 25, D-5300 Bonn 1, Germany. Accepted for publication March 25, 1992
CCC 0148-639X/92/101138-05 0 1992 John Wiley & Sons, Inc
1138
F Waves and Motor Unit Size
F-wave generation. l'herefore, we investigated F waves in individual MUs which w e could type according to contractile properties. We used motor unit (MU) twitch force, a size-dependent parameter, and contraction time, which is fairly independent of MU size in man.15,18319 MATERIALS AND METHODS
Twenty-one normal subjects, aged from 23 to 54 years (13 men a n d 8 women), were investigated. T h e first dorsal interosseus muscle of the hand was studied in all cases. MU action potentials and twitch contractions were evoked by intramuscular microstimulation of motor axons by a modification" of the method described by 'Taylor and Stevens.'H Single axons wert stimulated by means of a bipolar needle electrode inserted into the muscle proximal to the midbelly. T h e evoked M U action potentials were monitored using surface electrodes in a belly tendon arranagement (bandpass 20 to 3000 Hz). l ' h e associated MU twitches were recorded by a virtually isometric force transducer positioned at the radial side of the proximal phalanx of the index finger (filters 0 to 500 Hz). T h e stimulation electrode was moved in minute steps to find intramuscular nerve fibers. Stimuli were applied at a rate of 1 Hz and an intensity of less than 3 mA (duration 0.05 to 0.1 ms). A single M U response was assumed if an identical action potential occurred in an all-or-none fashion when changing the stimulus intensity several times."," T h e responses of 20 to 40 stimuli were
MUSCLE & NERVE
October 1992
on-line digitized (sampling rate for EMG 10,000 Hz and for force 1000 Hz), averaged, and stored on hard disc. Measurements were repeated once or twice, and each identified MU was stimulated at least 50 times. M U action potentials were monitored to ensure that the same MU was stimulated throughout. The twitches were analyzed with respect to their peak forces [twitch force, (TF)] and contraction times (CT). An original record showing the evoked M U action potential and the associated twitch contraction is illustrated in Figure 1. Generally, the stimulus evoked only one MU action potential with short latency corresponding to an M response. In some MUs, the M response was occasionally followed by a later one, which revealed an identical waveform and a latency compatible with that of an F wave. Such trials were documented on paper, but excluded from averaging to avoid errors in MU twitch parameter assessment. T h e latencies of the F waves were measured on a display. RESULTS
A total of 209 MUs were collected in the 21 subjects. N o F waves were found in the 82 MUs of 6 subjects. The remaining 15 subjects revealed F waves in 27% of their MUs (35 of 127 MUs), which is 17% in relation to the total number of
A
T 1 0 . 1 mV
MUs. The rate of occurrence of F waves in a given MU was not systematically analyzed, but was generally low and similar in all MUs. Usually, one F wave was found in every 10 to 20 consecutive trials as is seen in the record illustrated in Figure 2. Three F waves in 10 consecutive trials, observed in 1 subject, were an extreme and a clear exception. The TFs of all 209 MUs are shown in Figure 3A. An exponential distribution with a clear preponderance of weaker, i.e., small MUs, became apparent. The distribution of the TFs of the 82 MUs of the subjects without F waves is plotted in Figure 3C and that of all 127 MU of the subjects with F waves in Figure 3B. The TFs of only the 35 MUs with F waves are plotted in Figure 3D. Comparison of the four distributions revealed a close similarity, indicating that there was no difference between these groups of Mus with respect of their TFs. The fact that the exponential shape of- the distribution in Figure 3D was somewhat distorted seemed rather an effect of the small absolute number of MUs with F waves than a true biological phenomenon. Obviously, MUs with weak and strong TFs, i.e., small and large motoneurons, generated F waves with similar probabilities. The MUs with weaker TFs produced F waves in a greater number than their stronger counterparts simply because they were more numerous. There may, however, be an indication of a special role of a few MUs with very strong TFs. Four of the five MUs with twitch forces above 70 mN produced F waves (see Fig. 3D).
Twitch Forces (TFs).
B I
1’ I
50 ms FIGURE 1. Motor unit action potential (M response) (A) and twitch contraction (6)evoked by intramuscular microstimulation of a motor axon. The stimulus artifact is seen in front of the M response. Twenty-five trials were averaged.
F Waves and Motor Unit Size
-H 5 ms FIGURE 2. Ten consecutive trials with intramuscular microstimulation of a motor axon. The M responses are followed in one trace by an F wave with an identical waveform and a latency of 34 ms.
MUSCLE & NERVE
October 1992
1139
I
07 1
0.1
O0 0 2
O3
1
A
A
[ n=209 ]
[ n=209 ]
h
O 2
B
7
[ n=127]
h
5
r: cc 0
e,
L
9
BE
5
[ n=82 ]
-
c)
cd
2
[ n=82]
I)
> .-I
-2
0
.->
C
C
0.3 02 01
0.1
I
0.0
0.0 0.2
0.7
-
-
11111 0
20
1L 1
1
[ n=35 ]
[ n=35 ]
40
60
80
100
120
140
0
FIGURE 3. Distribution of motor units in relative numbers (n = absolute number) according to their twitch forces (bin width 4 mN). (A) All motor units studied. (B)All motor units of the subjects (15) with F waves. (C) All motor units of the subjects (6) without F waves. (D) All motor units with F waves. The four distributions are fairly similar. However, note that 4 of 5 motor units above 70 mN generated F waves.
T h e C T values were more symmetrically distributed and revealed no correlation with the TFs (r = -0.002) in accordance with findings of other authors. 15,18,19 T h e CTs of all MUs and of the other three groups of MUs as defined for the TFs above (see Fig. SA-D) are shown in Figure 4A-D. As with TFs, the four distributions were similar and did not reveal a select group of MUs producing F waves. Contraction Times (CTs).
T h e latencies of the F waves ranged from 30 to 35 ms. As expected, they were longer than those for F waves evoked from the ulnar nerve at the wrist in routine diagnosis. Since the
Latencies.
F Waves and Motor Unit Size
20
40
60
80
100
120
140
Contraction time [msl
Twitch force [mN]
1140
D
FIGURE 4. Distribution of motor units in relative numbers (n = absolute number) according to their contraction times. For (A) to (D) see legend to Figure 3. The four distributions are fairly similar.
site of the intramuscular axon stimulation could not be exactly determined and compared among the different MUs, we did not use the F-wave latencies for typing the MUs. DlSCUSSlON
Our study shows that an MU can generate F waves even if stimulated selectively. Obviously, it is not essential that a large number or all motoneurons of a pool are activated by an antidromic volley, as is assumed by one a ~ t h o r . ~Although ” various factors determining the state of excitability of an alpha-motoneuron may influence the generation of F waves, a putative interaction of the antidromic impulses on the spinal level seems to be of low importance.
MUSCLE & NERVE
October 1992
F waves are produced by a very small fraction of MUs of a muscle.8-13 The question whether a specific subpopulation of motoneurons is more than others subject to F-wave production is old. A series of structural and functional properties of the motoneuron and of the entire MU are determined by motoneuron Theoretically, the larger motoneurons should be more subject to F-wave production, since their initial segments can repolarize more quickly after antidromic depolarization and will more likely be re-excitable for recurrent impulses than the initial segments of the smaller m o t o n e u r o n ~ The .~~~ fact ~ ~that ~~ F waves are produced by a very small fraction of motoneurons could also point to large motoneurons as F-wave generators because of their small number. Kimura et al.,' taking axonal excitability and conduction velocity as measures of motoneuron size, could not find such a relationship. However, the authors discussed some sources of possible error associated with assessment of motor axon excitability from the surface, and with analyzing F-wave latencies without knowledge of the length of the fine axon terminals. We used twitch force and contraction time to characterize the MUs generating F waves. T F is cclosely related to the size of the motoneuron as is axon conduction velocity and excitability used by Kimurd et al.' This is not the case for CT, at least in man,1.5,18,19 This approach allows reliable typing of MUs, although the number of Mus obtained in individual subjects is limited. However, the distribution of the total of T F values in our study resembles that obtained by other techdniques. 15,1'3 It seems that the 209 MUs studied are a fairly representative sample for the first dorsal interosseous muscle and that selection bias is not a critical factor. The results concerning T F affirm the findings of Kimura et al.' We did not find a relationship between the T F of a MU and its capability of F-wave generation. Obviously, motoneurons of' all sizes can produce F waves with a similar probability. Since small motoneurons are more numerous, they generate the majority of F waves. There may, however, be an exception concerning the small number of Mus with very strong TFs, i.e., the very large motoneurons. They may be more susceptible of producing F waves, as indicated by the finding that four of five Mus with twitch forces above 70 mN revealed such responses. One could speculate that these large motoneurons, which are regularly antidromically stimulated by supramaximal nerve
F Waves and Motor Unit Size
stimulation, may be involved in the generation of the repeater F waves. Furthermore, typing of MUs according to their CTs did not result in selection of a subgroup being more subject to F-wave generation. Apparently, the factors which influence CTs of human MUs and which are mainly size-independent15,.'8,19do not play a relevant role in F-wave generation either. The rate of occurrence of F waves within an individual MU proved to be low and similar among the different MUs, in line with findings of a sin le fiber EMG study by Schiller and Stalberg.'7 There was no relationship to the parameters used for typing. Our study provides no information on the conditions which enable only some stimuli to evoke F waves. Both the low rate of occurrence of F waves in an individual MU and the limited number of MUs capable of generating F waves explain why the F wave in response to a supramaximal nerve stimulus is produced by a very small fraction of the motoneurons of the pool. The F-wave latencies recorded in distal hand muscles in routine studies vary within a range of several milliseconds. Since the conduction velocity of a motor axon depends on the size of the mot o n e u r ~ n ~this ' ~ *latency ~ scatter is explained by the finding that motoneurons of all sizes are involved in F-wave generation. Provided that a sufficient number of trials is carried out, the smallest and largest motoneurons may generate F waves and the range of latencies may be fairly representative for the range of motor axon conduction velocities in the nerve under study. In conclusion, motoneurons of all sizes produce F waves with similar probability. Other, sizeindependant factors, e.g., spontaneous or synaptically induced oscillations of the membrane potential and, thus, of the state of the excitability of a motoneuron at the time when the antidromic volley arrives, must account for the fact that the F waves are generated by a fairly small fraction of MUs. Only a few very large motoneurons may be more subjects to F-wave production and may be the generators of the so-called repeater F waves.
REFERENCES 1. Conrad B, Aschoff JC, Fischler M: Der diagnostische Wert d e r F-Wellen-Latenzen.J Neurol 1975;210:151- 159. 2 . Dengler R, Stein RB, Thomas CK: Axonal conduction velocity and force of single human motor units. Musrle Nrrvr 1988;11:136- 145. 3. Elek JM, Dengler R, Konstdnzer A, Hesse S, Wolf W: Mechanical implications of paired motor unit discharges in
MUSCLE & NERVE
October 1992
1141
pathological and voluntary tremor. Electroencephalogr Clin Neurophysiol 1991;8 1 :279- 283. 4. Fisher MA: F waves: comments on the central control of recurrent discharges. Muscle Nerve 1979;2:406. 5. Fisher MA: F-response latency-duration correlations: an argument for the orderly antidromic activation of motoneurons. Muscle Nerve 1980;3:437-438. 6. Henneman E, Mendell LM: Functional organization of motoneuron pool and its inputs, in Brooks VB (ed): Handbook of Physiology. Section 1: The Nervous System. Bethesda, MD, American Physiological Society, 1981, vol 2, pp 423507. 7. Kernell D: Input resistance, electrical excitability, and size of ventral horn cells in cat spinal cord. Science 1966; 152~1637-1640. 8. Kimura J: Clinical value and limitations of F-wave determination. A comment. Muscle Nerve 1978; 1:250-252. 9. Kimura J, Yanagisawa H, Yamada T , Mitsudome A, Sasaki H, Kimura A: Is the F-wave elicited in a selected group of motoneurons? Muscle Nerve 1984;7:392-399. 10. Macleod WN: Repeater F waves: A comparison of sensitivity with sensory antidromic wrist-to-palm latency and distal motor latency in the diagnosis of carpal tunnel syndrome. Neurology 1987;37:773-778. 11. Magladery JW, McDougas DB Jr: Electrophysiological studies of nerve and reflex activity in normal man. 1. Identification of certain reflexes in the electromyogram and the conduction velocity of peripheral nerve fibres. Bull Johns Hopkins Hosp 1950;86:265-290.
1142
F Waves and Motor Unit Size
12. Mayer RF, Feldman RG: Observations on the nature of the F wave in man. Neurology 1967;17:147-156. 13. McLeod JG, Wray SH: An experimental study of the F wave in the baboon. J Neurol Neurosurg Psychzatry 1966; 29:196-200. 14. Miglietta OE: T h e F response after transverse myelotomy, in Desmedt J E (ed): New Developmenls in Electromyography and Clinical Neurophysiology. Basel, Karger, 1973, vol 3, pp 323-327. 15. Milner-Brown HS, Stein RB, Yemm R: The contracile properties of human motor units during voluntary isometric contractions. J Physiol 1973a;228:285-306. 16. Peioglou-Harmoussi S, Fawcett PRW, Howel D, Barwick DD: F-responses: a study of frequency, shape and amplitude characteristics in healthy control subjects. J Neurol Neurosurg Psychiatry 1985;48: 1159- 1164. 17. Schiller HH, Stalberg E: F responses studied with single fibre EMG in normal subjects and spastic patients. J Neurol Neurosurg Psychiatry 1978;4 1:45- 53. 18. Taylor A, Stephans JA: Study of human motor unit contractions by controlled intramuscular microstimulation. Brain Res 1976;117:331-335. 19. Thomas CK, Johansson RS, Westling G , Bigland-Ritchie B: Twitch properties of human thenar motor units measured in response to intraneural motor-axon stimulation. j Neurophysiol 1990;64:1339- 1346. 20. Young RR, Shahani BT: Clinical value and limitations of F wave determination. Muscle Nerve 1978; 1 :248-250.
MUSCLE & NERVE
October 1992
