- Email: [email protected]
Enhanced amplitude reduction of somatosensory evoked potentials by voluntary movement in the elderly
Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
Enhanced amplitude reduction of somatosensory evoked potentials by voluntary movement in the elderly T. Touge a ,*, H. Takeuchi a, I. Sasaki a, K. Deguchi a, N. Ichihara b a
Third Department of Internal Medicine, Kagawa Medical University, 1750-1, Ikenobe, Miki-cho, Kita-gun, Kagawa, 761-07, Japan b Department of Neurology, Takamatsu National Hospital, Kagawa, Japan Accepted for publication: 15 January 1997
Abstract We studied the effects of aging on modification of the median nerve somatosensory evoked potentials (SEPs) by voluntary movement in 17 aged (66.5 ± 8.9 years, mean ± SD) and 12 young normal humans (27.5 ± 5.0 years). The amplitudes of cortical SEP components were generally larger in the aged group than in the young group. Following isometric contraction of the thenar muscle, the aged group showed significant attenuation of the prerolandic P22-N28-P45 and the postrolandic P24-N30-P45, while the young group only demonstrated significant reduction of the prerolandic P22-N28 amplitude. In the prerolandic N28-P45 and the postrolandic P24-N30 and N30-P45, amplitudes reduced by voluntary movement (gated amplitude) significantly correlated with amplitudes at rest (resting amplitude) and with the age of subjects. The effects of stimulus intensity and frequency on gating supported the correlative changes between gated and resting amplitudes. These results suggest that the magnitude of gating depends on SEP amplitudes at rest, and that augmented gating in the aged group is a result of enlarged SEPs. Since the cervical and Erb’s potentials were not changed by movement, and passive movement did not significantly affect the SEPs, a centrifugal mechanism is probably responsible for gating in this study. 1997 Elsevier Science Ireland Ltd. Keywords: Somatosensory evoked potentials (SEPs); Gating; Aging; Voluntary movement; Stimulus intensity; Centrifugal mechanism; Stimulus frequency
1. Introduction Modification, usually attenuation, of somatosensory evoked potentials (SEPs) during or even before the initiation of voluntary movement is known as ‘gating’. Previous studies in animals revealed that gating can occur in the somatosensory ascending pathway at the dorsal column nuclei (Towe and Jabbur, 1961), medial lemniscus (Ghez and Pisa, 1972; Coulter, 1974), thalamus (Tsumoto et al., 1975), and cortical sensory areas (Chapin and Woodward, 1981). Possible mechanisms of gating include influences of sensory feedback from peripheral receptors such as the muscle spindle, joint, or skin and modification of cortical electrophysiological events before and during movement (Jones et al., 1989). For example, activation of cortical motor neu-
* Corresponding author. Tel.: +81 878 985111; fax: +81 878 910115.
0168-5597/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0921-884X(97)9613 6-1
rons affects SEPs in humans and animals (Jiang et al., 1990; Kujirai et al., 1993). Furthermore, Angel et al. (1984) have reported that inhibition of SEPs correlated to decreased discrimination of muscle load sensation during voluntary movement. This result suggests that gating may regulate sensitivity to external sensory stimuli during movement and may be important in the execution of precise movements. No prior study has determined the effects of aging on this gating mechanism. Here, we examined SEP changes during isometric voluntary contraction of the thenar muscles in aged and young normal humans.
2. Methods and materials Seventeen aged (66.5 ± 8.9 years, mean ± SD) and 12 young normal volunteers (27.5 ± 5.0 years) participated in this study. No subjects had any neurologic abnormalities.
EEP 96136
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
Fig. 1. SEPs induced by the stronger (A) or weaker stimulus intensity (B) in a representative elderly subject. Each component of the SEP is labeled. Thick lines show SEPs during rest and thin lines show those during movement.
All subjects gave informed consent to participate in this study. The experimental protocol was approved by the ethics committee of our institution. Subjects were placed in a supine position on a couch in a dark room and were instructed to close their eyes and relax their jaw muscles during recordings. To elicit SEPs, electric rectangular pulses of 200 ms duration were delivered percutaneously to the right wrist at a rate of 1/s through Ag/AgCl surface electrodes 3 cm apart along the median nerve; the cathode was positioned proximally. A stimulus intensity (SI) between 1 and 5 mA was employed to produce the minimum muscle twitch. SEPs were recorded at the pre- and postrolandic areas, 2 cm anterior (C3′) or posterior (P3′) to the vertex and 7 cm lateral to the midsagittal line; at the second vertebral position (C2), just inferior to the processus spinosus of the axis; and at Erb’s point through Ag/AgCl surface electrodes using the linked earlobe as a reference. The raw data were amplified with a bandpass between 10 Hz and 2 kHz and averaged
109
248 times by a Medelec Mystro MS25. Contaminating artifacts and overranged data were automatically or manually rejected from averaging. In the first test trial, SEPs were recorded at rest. During the second recording trial, subjects isometrically pushed a saucer-like pressure transducer of 2 cm diameter which was fixed at their little finger by flexing their ipsilateral thumb. To keep the isometric contraction at 20–30% of the maximum force of the thenar muscle, the pressure transducer was simultaneously monitored on an oscilloscope during SEP recording. When the force was improper, correcting instructions were given orally to subjects by operators. Furthermore, in 10 subjects, SEPs were recorded during passive contraction of the thumb. The thumb was fixed by surgical tape at the same posture as during voluntary contraction. Special care was taken to maintain the wrist position to avoid displacing the stimulus electrodes. Additionally, effects of stimulus intensity on gating were studied in 7 aged normals (66.7 ± 11.8 years, mean ± SD). Subjects in which the maximum amplitudes of the prerolandic P22-N28 or N28-P45 or the postrolandic P24-N30 or N30-P45 were larger than 5 mV were chosen because gating is difficult to detect with small amplitude SEPs. Two SIs were employed. Stronger SI (7.3 ± 3.3 mA, n = 7), which produced minimum muscle twitch of the thenar muscle, evoked the prerolandic P22-N28 or N28-P45 or the postrolandic P24-N30 or N30-P45 beyond 5 mV, while weaker SI (3.7 ± 1.4 mA), which elicited no muscle response, produced responses beyond 3 mV. Under those stimulus conditions, the SEPs were recorded during rest as well as during voluntary contraction of the thenar muscle as described above. Furthermore, SEPs during rest and voluntary contraction of the thenar muscle were recorded using stimulus frequencies (SF) of 0.2 and 1 Hz with SI producing minimum muscle twitch of the thenar muscle in 5 young normals (33.3 ± 7.8 years, mean ± SD). 2.1. Data analysis The SEP peaks were defined according to the polaritylatency convention. The P14, N17, P22, N28, P45, and N60 in the prerolandic leads and P14, N17, P24, N30, P45, and N60 in the postrolandic leads were identified (Fig. 1A); the neighboring peak-to-peak amplitudes and the peak latencies were measured. We used the term N28 which is equivalent to N30 in previous reports to differentiate between this potential and the parietal N30. The potentials at C2, P10N13, and Erb’s point, P8-N10, were also analyzed. Amplitudes and latencies of each SEP component were compared statistically using the Mann-Whitney U test for unpaired data and Wilcoxon signed-ranks test for paired data. Differences of effects of muscle contraction on the SEP components between groups were assessed by the repeated measures analysis of variance (ANOVA). Relationships between SEP parameters or age and gating
110
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
Table 1 Comparisons of SEPs during rest between the aged and young groups Prerolandic lead Amplitudes of cortical SEPs (mV) P14-N17 N17-P22 P22-N28 N28-P45 P45-N60 Latencies of cortical SEPs (ms) P14 N17 P22 N28 p45 N60
Aged group
Young group
Postrolandic lead
2.27 4.71 4.97 6.73 8.81
± ± ± ± ±
l.43 2.86 2.90 3.89 2.91
2.03 2.33 2.67 3.21 4.24
± ± ± ± ±
0.97 1.68* 1.16** 1.59** 2.01†
P14-N17 N17-P24 P24-N30 N30-P45 P45-N60
14.02 18.18 22.66 28.85 44.14 66.17
± ± ± ± ± ±
0.83 l.42 l.42 2.19 6.13 5.21
14.05 17.43 21.75 28.23 41.51 55.95
± ± ± ± ± ±
l.26 l.29 2.01* 3.84 l.99 4.58††
P14 N17 P24 N30 P45 N60
Aged group
Young group
3.77 7.48 5.32 5.73 9.73
± ± ± ± ±
l.31 4.23 3.76 3.00 4.24
2.85 3.43 1.71 2.46 3.91
± ± ± ± ±
1.18* 1.92** 0.52† l.33† l.92††
14.19 19.04 24.27 31.15 43.46 66.74
± ± ± ± ± ±
0.82 l.16 l.43 l.76 5.96 6.24
14.03 18.23 22.54 28.32 40.83 55.27
± ± ± ± ± ±
l.25 l.22 2.05* 4.87 3.06 4.02††
Values are shown as the mean ± SD (n = 17 in the aged group and n = 12 in the young group). *P , 0.05; **P , 0.01; †P , 0.001; ††P , 0.0001, Mann-Whitney U test.
were analyzed using Pearson’s correlation coefficient or simple linear regression. Data are expressed as the mean ± SD.
3. Results 3.1. Comparisons between groups of SEP amplitudes and latencies during rest The amplitude of every cortical SEP component except for the prerolandic P14-N17 were significantly larger in the aged group than in the young group (Table 1). The P10-N13 amplitude at C2 was significantly larger (2.98 ± 0.88 mV) in the young group than in the aged group (1.76 ± 0.57 mV, P , 0.01, n = 29). There was no significant difference of P8-N10 amplitudes between the aged and young groups. The latencies of the prerolandic P22, the postrolandic P24 and the pre- and postrolandic N60 were significantly larger in the aged group compared to the young group (Table 1). The latencies of the P8, N10, P10, and N13 were similar between the groups. 3.2. Effects of voluntary or passive contraction of the thumb on SEPs In the young group, isometric voluntary contraction of the thenar muscles significantly attenuated the prerolandic P22N28 amplitude (Fig. 2). The aged group had significantly smaller amplitudes of the prerolandic P22-N28, N28-P45 and the postrolandic P24-N30 and N30-P45 during voluntary muscle contraction than during rest (Fig. 2). The amplitudes of cervical and Erb’s potentials did not significantly change following movement. Movement did not significantly affect SEP latencies in either group. We compared ratios of the difference between SEP amplitude during rest and movement (gated amplitude)/
SEP amplitude during rest (resting amplitude) in the two groups. Ratios of gated/resting amplitudes in the postrolandic N30-P45 were significantly larger in the aged group (0.13 ± 0.33) as compared to the young group (−0.27 ± 0.65, P , 0.02), and those in the prerolandic P45-N60 were significantly smaller in the young group (−0.14 ± 0.22) than in the aged group (0.03 ± 0.34, P , 0.05). In addition, we evaluated the effects of voluntary muscle contraction on SEP components in both groups. SEP amplitude changes following voluntary muscle contraction differed significantly between the groups in the prerolandic N28-P45 (df = 1,27, F = 4.945, P , 0.05), the postrolandic P24-N30 (F = 7.029, P , 0.02) and N30-P45 amplitudes (F = 7.062, P , 0.02) (Fig. 2). Passive contraction of the thumb did not have any significant effect on amplitudes and latencies of any SEP component in the aged group. 3.3. The effect of age or the relationships between SEP resting and gated amplitudes We studied the relationship between resting and gated amplitudes in components which showed significant gating. In the aged group, the gated and resting amplitudes significantly correlated in the prerolandic N28-P45 (P , 0.002, r = 0.68, n = 19) and the postrolandic P24-N30 (P , 0.0005, r = 0.756) (Fig. 3) and N30-P45 (P , 0.00l, r = 0.720). In the prerolandic P22-N28, ratios of gated/resting amplitudes correlated significantly with resting amplitude in the aged group (P , 0.05, r = −0.537). There were no significant correlations between gated and resting amplitudes or between ratios of gated/resting amplitudes and resting amplitude in the young group. The age of subjects significantly correlated to the gated amplitude of the prerolandic N28-P45 (P , 0.05, r = 0.372, n = 29) and the postrolandic P24-N30 (P , 0.0l, r = 0.475) and N30-P45 (P , 0.05, r = 0.385).
111
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
ger SI condition and the prerolandic P22-N28 amplitude in the weaker SI condition (Fig. 1, Table 2). The prerolandic N28-P45 and the postrolandic N30-P45 had significantly larger gated amplitudes in the stronger SI condition compared to the weaker SI (Table 2). We evaluated the effects of voluntary movement on each SEP component during the two conditions. Significant differences between the two SIs were observed only in the postrolandic N30-P45 (df = 1,8, F = 6.669, P , 0.05). 3.5. Effects of stimulus frequency on the gating in young normals
Fig. 2. Changes of SEP amplitudes (mean ± SEM) during rest and movement. SEP amplitudes during rest and voluntary movement are shown as thick lines in the aged group and thin lines in the young group. Dotted lines refer to SEP amplitudes during rest and passive movement in the aged group. Significant differences of SEP amplitudes between rest and voluntary movement (oblique bars) or effects of voluntary movement on SEP amplitudes between the groups (vertical bars) are shown as *P , 0.05 and **P , 0.01.
3.4. Effect of stimulus intensity on gating in aged normals A stronger SI produced amplitudes of 7.78 ± 2.12 mV (n = 20) during rest and 6.42 ± 1.83 mV during movement in the prerolandic P22-N28, N28-P45 and the postrolandic P24-N30, N30-P45. A weaker SI induced those SEP components with amplitudes of 4.81 ± 1.18 mV during rest and 4.37 ± 1.15 mV during movement. Movement significantly attenuated the prerolandic P22-N28 and N28-P45 amplitudes and the postrolandic N30-P45 amplitude in the stron-
The prerolandic P22-N28 and N28-P45 amplitudes tended to be smaller with SF of 1 Hz compared to SF of 0.2 Hz, and the postrolandic N30-P45 and P45-N60 amplitudes were significantly smaller with SF of 1 Hz than SF of 0.2 Hz (Table 3). The amplitudes of the other SEP components and the latencies of all SEP components did not vary significantly between the SF conditions. With SF of 0.2 Hz, voluntary movement significantly attenuated the prerolandic P22-N28 and N28-P45 and the postrolandic N30-P45, and the prerolandic P22-N28 amplitude was significantly smaller during movement than during rest with SF of 1 Hz (Fig. 4, Table 3). Gated amplitudes in the prerolandic N28-P45 and the postrolandic N30-P45 tended to be larger with SF of 0.2 Hz than SF of 1 Hz, but the differences between the SF conditions were not significant. There were no significant differences of effects of voluntary movement on SEP components between the SF conditions. None of the SEP components showed significant changes in latency by movement under each SF condition.
4. Discussion In this study, our purpose was to determine the effects of aging on SEP modification by voluntary movement. We demonstrated enlarged and prolonged cortical SEPs and
Table 2 Effects of stimulus intensity (SI) on gating in aged normals Prerolandic lead: P22-N28 amplitude
Stronger SI Weaker SI
Prerolandic lead: N28-P45 amplitude
Rest
Movement
Gated amplitude
Rest
Movement
6.91 ± 2.14 4.39 ± 1.10
5.83 ± 2.31* 3.75 ± 0.95*
1.08 ± 0.67 0.64 ± 0.38
9.30 ± 2.35 5.48 ± 1.59
6.84 ± 1.79* 4.36 ± 1.53
Postrolandic lead: P24-N30 amplitude
Stronger SI Weaker SI
Gated amplitude 2.46 ± 1.23** 1.12 ± 0.86
Postrolandic lead: N30-P45 amplitude
Rest
Movement
Gated amplitude
Rest
Movement
Gated amplitude
7.64 ± 2.46 4.54 ± 1.07
7.02 ± 2.39 4.34 ± 0.91
0.62 ± 0.77 0.19 ± 0.66
7.28 ± 0.97 4.82 ± 0.91
6.00 ± 0.91* 5.04 ± 1.05
1.27 ± 1.05** −0.22 ± 0.76
Values are shown as the mean ± SD in mV (n = 5). Amplitudes of SEP components between rest and movement conditions or gated amplitudes between stronger and weaker SI conditions were statistically compared by Wilcoxon signed-ranks test. *P , 0.05, rest vs. movement. **P , 0.05, stronger vs. weaker SI.
112
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
Table 3 Effects of stimulus frequency (SF) on gating in young normals SF (Hz)
Prerolandic lead: P22-N28 amplitude Rest
Movement
0.2 1
2.80 ± 0.96 2.52 ± 1.19
2.50 ± 1.03* 1.85 ± 0.52*
SF (Hz)
Postrolandic lead: N30-P45 amplitude
0.2 1
Prerolandic lead: N28-P45 amplitude Gated amplitude 0.30 ± 0.15 0.67 ± 0.73
Rest
Movement
4.10 ± 0.58 3.09 ± 0.79
2.72 ± 0.50* 2.55 ± 0.83
Gated amplitude 1.38 ± 0.83 0.54 ± 1.31
Postrolandic lead: P45-N60 amplitude
Rest
Movement
Gated amplitude
Rest
Movement
Gated amplitude
3.84 ± 0.73** 2.57 ± 1.03
2.97 ± 0.62* 2.62 ± 1.03
0.87 ± 0.38 -0.05 ± 0.77
4.88 ± 2.09** 3.27 ± 1.47
4.37 ± 2.12 3.79 ± 1.68
0.51 ± 0.52 −0.52 ± 0.85
Values are shown as the mean ± SD in mV (n = 5). Amplitudes of SEP components between rest and movement conditions or amplitudes of SEP components during rest between the two SF conditions were statistically compared by Wilcoxon signed-ranks test. *P , 0.05, rest vs. movement. **P , 0.05 between the two SF conditions.
diminished cervical potentials in the aged group compared with the young group. Those SEP changes in aged normals are consistent with results from previous studies (Lu¨ders, 1970; Desmedt and Cheron, 1980). Our study revealed that the aged group had significantly larger amplitude reductions of the prerolandic P22-N28 and N28-P45 amplitudes and the postrolandic P24-N30 and N30-P45 amplitudes following movement. On the other hand, the young group showed significant attenuation of the prerolandic P22-N28 by movement. In a preliminary study in aged subjects, the reduction of the prerolandic N28-P45 and the postrolandic N30-P45 amplitudes by movement disappeared with decreasing amplitude of the potentials achieved by reducing SI, whereas attenuation of the prerolandic P22-N28 remained. In young subjects, increasing SF from 0.2 Hz to 1 Hz caused attenuation of the pre- and postrolandic potentials as previously reported (Abbruzzese et al., 1990). According to the attenuation of SEPs, the reduction of the prerolandic N28P45 and the postrolandic N30-P45 amplitudes by movement disappeared despite retaining attenuation of the prerolandic P22-N28. Cohen and Starr (1987) have reported reductions of the prerolandic P22, N30, P45, and the postrolandic P27 amplitudes in median nerve SEPs following ballistic flexion of the thumb. Cheron and Borenstein (1987) have described attenuations induced by rapid finger movement of the parietal P27, frontal P22 and N30. In a study by Abbruzzese et al. (1981), isometric or isotonic movement of the thumb significantly attenuated the parietal N20-P25. Rushton et al. (1981) also have described reduction of the parietal N20-P30 and P45-N55 amplitudes during tracking movements of the thumb or finger. Although our experimental conditions differed from these prior studies, our results for gating correlated well with the findings described above. In this study, ANOVA analysis demonstrated differences in gating between the two groups. The attenuation of the prerolandic N28-P45 and the postrolandic P24-N30 and N30-P45 induced by movement was larger in the aged
group than in the young group. The aged group had larger SEP amplitudes and more marked cortical SEP gating effects compared to the young group. These findings suggest some relationship between the magnitude of gating and SEP amplitudes. In the correlation analysis, the gated amplitudes of SEPs significantly correlated with the resting amplitudes of the prerolandic N28-P45 and the postrolandic P24-N30 and N30-P45 in the aged group. We also observed correlative changes between the SEP amplitudes and the magnitude of gating in a preliminary study of the effects of stimulus intensity and frequency on gating. These results suggest that gated amplitudes of cortical SEPs depend on SEP amplitudes at rest. This conjecture does not contradict the view that the gating plays a role in screening out irrelevant sensory signals during movement (Coquery, 1978), or that the cortex filters out irrelevant information (Chapin and Woodward, 1981). Excess neural activities in the cortex responding to somatosensory afferent volleys may be cut off in relation to voluntary movement. Additionally, we found significant correlations between the age of subjects and the gated amplitudes. This increased magnitude of gating may be an effect associated with enlarged cortical SEPs in the elderly. Furthermore, the prerolandic P22-N28 was significantly attenuated by movement in the aged and young groups, but the gated amplitude and the resting amplitude of this potential did not correlate statistically. This finding suggests that central gating primarily occurs in the prerolandic P22-N28 and there is a functional difference among the SEP components which had significant gating. In prior studies, surface and depth recordings of SEPs revealed that the precentral and parietal SEPs originated in the cortical somatosensory area including area 3b and area 1 (Allison et al., 1991). In studies using magnetoencephalography, the somatosensory evoked magnetic fields (SEFs) were localized in the primary somatosensory cortex 20 and 35 ms following electrical stimulation of the median nerve (Forss et al., 1994; Kakigi, 1994; Kawamura et al., 1996). Those authors concluded that SEFs of 20 ms correspond to the N20 in the scalp SEPs and
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
113
The mechanism of the augmented gating effect in aged subjects is not clear. In our study, holding the 2 cm round pressure transducer with a stable muscle strength requires sensory feedback from muscle spindle, joint, and cutaneous receptors. However, the amplitudes of cervical and Erb’s potentials did not significantly change following movement, and passive movement did not have any significant effect upon any SEP parameters. In addition, SEPs attenuated with increased movement speed, but not with increased muscle loading (Ghez and Pisa, 1972; Rushton et al., 1981). Anesthesia of skin or joint related to the movement also does not affect gating (Rushton et al., 1981). Jones et al. (1989) have concluded that sensory feedback from muscle spindle, joint and cutaneous receptors distal to the stimulation did not play a significant role in gating because attenuation of SEPs was at a maximum when the stimuli were delivered at the onset of movement. Therefore, we think that the gating in our study was predominantly due to the centrifugal mechanism as inhibitory discharges from the cortical motor system attenuated incoming afferent flow to
Fig. 3. Relationships between the prerolandic N28-P45 amplitude (A) or the postrolandic P24-N30 amplitude (B) during rest and amplitudes of those components reduced by voluntary movement (gated amplitude). Circles represent the aged group and darkened squares are for the young group. Regression lines in the aged group also are shown in scatter diagrams (A: y = 4.511 + 1.457x, P , 0.005; B: y = 3.005 + 2.453x, P , 0.0005).
those of 35 ms correspond to the N35 or N30. However, clinical studies have demonstrated different features of the prerolandic P22 and N30 as compared to the postrolandic potentials (Mauguie`re et al., 1983; Slimp et al., 1986; Dinner et al., 1987; Mauguie`re and Desmedt, 1991). Desmedt and Bourguet (1985) have postulated that P22 originated in area 4 and the origin of N30 (N28) was in the supplementary motor area. Kawamura et al. (1996) have reported that SEFs that occur approximately 28 ms following electrical stimulation of the median nerve are located in the precentral cortex, area 4. Since we hypothesized that different potential origins can explain these functional differences, our results support the view that the prerolandic P22-N28 originates in the cortical motor areas.
Fig. 4. SEPs induced by the stimulus frequencies of 0.2 Hz (A) or 1 Hz (B) in a representative young subject. Thick lines show SEPs during rest and thin lines show those during movement.
114
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
the primary sensory cortex. The primary motor area has direct interconnections with areas 1, 2, and the anterior part of area 5, and indirectly with areas 3a and 3b. Furthermore, the supplementary motor area is also interconnected with the parietal sensory areas (Jones, 1986). Through such interneuronal connections, the inhibitory or excitatory influences from the cortical motor areas may be able to regulate neuronal activities in the cortical sensory areas receiving somatosensory afferent volleys (Kakigi et al., 1995). We demonstrated enhanced gating associated with enlargement of SEPs in the elderly. When we evaluate the gating of SEPs in humans, it is necessary to consider the effects of aging.
References Abbruzzese, G., Ratto, S., Favale, E. and Abbruzzese, M. Proprioceptive modulation of somatosensory evoked potentials during active or passive finger movements in man. J. Neurol. Neurosurg. Psychiatry, 1981, 44: 942–949. Abbruzzese, G., Dall’Agata, D., Morena, M., Reni, L., Trivelli, G. and Favale, E. Selective effects of repetition rate on frontal and parietal somatosensory evoked potentials (SEPs). In: P.M. Rossini and F. Mauguie`re (Eds.), New Trends and Advanced Techniques in Clinical Neurophysiology (EEG Suppl. 41). Elsevier, Amsterdam, 1990, pp. 145– 148. Allison, T., McCarthy, G., Wood, C.C. and Jones, S.J. Potentials evoked in human and monkey cerebral cortex by stimulation of the median nerve. Brain, 1991, 114: 2465–2503. Angel, R.W., Boylls, C.C. and Weinrich, M. Cerebral evoked potentials and somatosensory perception. Neurology, 1984, 34: 123–126. Chapin, J.K. and Woodward, D.J. Modulation of sensory responsiveness of single somatosensory cortical cells during movement and arousal behaviors. Exp. Neurol., 1981, 72: 164–178. Cheron, G. and Borenstein, S. Specific gating of the early somatosensory evoked potentials during active movement. Electroenceph. clin. Neurophysiol., 1987, 67: 537–548. Cohen, L.G. and Starr, A. Localization, timing and specificity of gating of somatosensory evoked potentials during active movement in man. Brain, 1987, 110: 451–467. Coquery, J.-M. Role of active movement in control of afferent input from skin in cat and man. In: G. Gordon (Ed.), Active Touch. Pergamon Press, New York, 1978, pp. 161–169. Coulter, J.D. Sensory transmission through lemniscal pathway during voluntary movement in the cat. J. Neurophysiol., 1974, 37: 831–845. Desmedt, J.E. and Bourguet, M. Color imaging of parietal and frontal somatosensory potential fields evoked by stimulation of median or posterior tibial nerve in man. Electroenceph. clin. Neurophysiol., 1985, 62: 1–17. Desmedt, J.E. and Cheron, G. Somatosensory evoked potentials to finger stimulation in healthy octogenarians and in young adults: wave forms,
scalp topography and transit times of parietal and frontal components. Electroenceph. clin. Neurophysiol., 1980, 50: 404–425. Dinner, D.S., Lu¨ders, H., Lesser, R.P. and Morris, H.H. Cortical generators of somatosensory evoked potentials to median nerve stimulation. Neurology, 1987, 37: 1141–1145. Forss, N., Hari, R., Salmelin, R., Ahonen, A., Ha¨ma¨la¨inen, M., Kajola, M., Knuutila, J. and Simola, J. Activation of the human posterior parietal cortex by median nerve stimulation. Exp. Brain Res, 1994, 99: 309– 315. Ghez, C. and Pisa, M. Inhibition of afferent transmission in cuneate nucleus during voluntary movement in the cat. Brain Res., 1972, 40: 145–151. Jiang, W., Chapman, C.E. and Lamarre, Y. Modulation of somatosensory evoked responses in the primary somatosensory cortex produced by intracortical microstimulation of the motor cortex in the monkey. Exp. Brain Res., 1990, 80: 333–344. Jones, E.G. Connectivity of the primate sensory-motor cortex. In: E.G. Jones and A. Peters (Eds.), Cerebral Cortex, Vol. 5. Plenum Press, New York, 1986, pp. 113–183. Jones, S.J., Halonen, J.-P. and Shawkat, F. Centrifugal and centripetal mechanisms involved in the ‘gating’ of cortical SEPs during movement. Electroenceph. clin. Neurophysiol., 1989, 74: 36–45. Kakigi, R. Somatosensory evoked magnetic fields following median nerve stimulation. Neurosci. Res., 1994, 20: 165–174. Kakigi, R., Koyama S., Hoshiyama M., Watanabe S., Shimojo, M. and Kitamura, Y. Gating of somatosensory evoked responses during active finger movements: magnetoencephalographic studies. J. Neurol. Sci., 1995, 128: 195–204. Kawamura, T., Nakasato, N., Seki, K., Kanno, A., Fujita, S., Fujiwara, S. and Yoshimoto, T. Neuromagnetic evidence of pre- and post-central cortical sources of somatosensory evoked responses. Electroenceph. clin. Neurophysiol., 1996, 100: 44–50. Kujirai, T., Sato, M., Rothwell, J.C. and Cohen, L.G. The effect of transcranial magnetic stimulation on median nerve somatosensory evoked potentials. Electroenceph. clin. Neurophysiol., 1993, 89: 227–234. Lu¨ders, H. The effects of aging on the wave form of the somatosensory cortical evoked potential. Electroenceph. clin. Neurophysiol., 1970, 29: 450–460. Mauguie`re, F. and Desmedt, J.E. Focal capsular vascular lesions can selectively deafferent the prerolandic or the parietal cortex: somatosensory evoked potentials evidence. Ann. Neurol., 1991, 30: 71–75. Mauguie`re, F., Desmedt, J.E. and Courjon, J. Astereognosis and dissociated loss of frontal or parietal components of somatosensory evoked potentials in hemispheric lesions. Brain, 1983, 106: 271–311. Rushton, D.N., Rothwell, J.C. and Craggs, M.D. Gating of somatosensory evoked potentials during different kinds of movement in man. Brain, 1981, 104: 465–49l. Slimp, J.C., Tamas, L.B., Stolov, W.C. and Wyler, A.R. Somatosensory evoked potentials after removal of somatosensory cortex in man. Electroenceph. clin. Neurophysiol., 1986, 65: 111–117. Towe, A.L. and Jabbur, S.J. Cortical inhibition of neurons in dorsal column nuclei of cat. J. Neurophysiol., 1961, 24: 488–498. Tsumoto, T., Nakamura, S. and Iwama, K. Pyramidal tract control over cutaneous and kinesthetic sensory transmission in the cat thalamus. Exp. Brain Res., 1975, 22: 281–294.
Enhanced amplitude reduction of somatosensory evoked potentials by voluntary movement in the elderly T. Touge a ,*, H. Takeuchi a, I. Sasaki a, K. Deguchi a, N. Ichihara b a
Third Department of Internal Medicine, Kagawa Medical University, 1750-1, Ikenobe, Miki-cho, Kita-gun, Kagawa, 761-07, Japan b Department of Neurology, Takamatsu National Hospital, Kagawa, Japan Accepted for publication: 15 January 1997
Abstract We studied the effects of aging on modification of the median nerve somatosensory evoked potentials (SEPs) by voluntary movement in 17 aged (66.5 ± 8.9 years, mean ± SD) and 12 young normal humans (27.5 ± 5.0 years). The amplitudes of cortical SEP components were generally larger in the aged group than in the young group. Following isometric contraction of the thenar muscle, the aged group showed significant attenuation of the prerolandic P22-N28-P45 and the postrolandic P24-N30-P45, while the young group only demonstrated significant reduction of the prerolandic P22-N28 amplitude. In the prerolandic N28-P45 and the postrolandic P24-N30 and N30-P45, amplitudes reduced by voluntary movement (gated amplitude) significantly correlated with amplitudes at rest (resting amplitude) and with the age of subjects. The effects of stimulus intensity and frequency on gating supported the correlative changes between gated and resting amplitudes. These results suggest that the magnitude of gating depends on SEP amplitudes at rest, and that augmented gating in the aged group is a result of enlarged SEPs. Since the cervical and Erb’s potentials were not changed by movement, and passive movement did not significantly affect the SEPs, a centrifugal mechanism is probably responsible for gating in this study. 1997 Elsevier Science Ireland Ltd. Keywords: Somatosensory evoked potentials (SEPs); Gating; Aging; Voluntary movement; Stimulus intensity; Centrifugal mechanism; Stimulus frequency
1. Introduction Modification, usually attenuation, of somatosensory evoked potentials (SEPs) during or even before the initiation of voluntary movement is known as ‘gating’. Previous studies in animals revealed that gating can occur in the somatosensory ascending pathway at the dorsal column nuclei (Towe and Jabbur, 1961), medial lemniscus (Ghez and Pisa, 1972; Coulter, 1974), thalamus (Tsumoto et al., 1975), and cortical sensory areas (Chapin and Woodward, 1981). Possible mechanisms of gating include influences of sensory feedback from peripheral receptors such as the muscle spindle, joint, or skin and modification of cortical electrophysiological events before and during movement (Jones et al., 1989). For example, activation of cortical motor neu-
* Corresponding author. Tel.: +81 878 985111; fax: +81 878 910115.
0168-5597/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0921-884X(97)9613 6-1
rons affects SEPs in humans and animals (Jiang et al., 1990; Kujirai et al., 1993). Furthermore, Angel et al. (1984) have reported that inhibition of SEPs correlated to decreased discrimination of muscle load sensation during voluntary movement. This result suggests that gating may regulate sensitivity to external sensory stimuli during movement and may be important in the execution of precise movements. No prior study has determined the effects of aging on this gating mechanism. Here, we examined SEP changes during isometric voluntary contraction of the thenar muscles in aged and young normal humans.
2. Methods and materials Seventeen aged (66.5 ± 8.9 years, mean ± SD) and 12 young normal volunteers (27.5 ± 5.0 years) participated in this study. No subjects had any neurologic abnormalities.
EEP 96136
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
Fig. 1. SEPs induced by the stronger (A) or weaker stimulus intensity (B) in a representative elderly subject. Each component of the SEP is labeled. Thick lines show SEPs during rest and thin lines show those during movement.
All subjects gave informed consent to participate in this study. The experimental protocol was approved by the ethics committee of our institution. Subjects were placed in a supine position on a couch in a dark room and were instructed to close their eyes and relax their jaw muscles during recordings. To elicit SEPs, electric rectangular pulses of 200 ms duration were delivered percutaneously to the right wrist at a rate of 1/s through Ag/AgCl surface electrodes 3 cm apart along the median nerve; the cathode was positioned proximally. A stimulus intensity (SI) between 1 and 5 mA was employed to produce the minimum muscle twitch. SEPs were recorded at the pre- and postrolandic areas, 2 cm anterior (C3′) or posterior (P3′) to the vertex and 7 cm lateral to the midsagittal line; at the second vertebral position (C2), just inferior to the processus spinosus of the axis; and at Erb’s point through Ag/AgCl surface electrodes using the linked earlobe as a reference. The raw data were amplified with a bandpass between 10 Hz and 2 kHz and averaged
109
248 times by a Medelec Mystro MS25. Contaminating artifacts and overranged data were automatically or manually rejected from averaging. In the first test trial, SEPs were recorded at rest. During the second recording trial, subjects isometrically pushed a saucer-like pressure transducer of 2 cm diameter which was fixed at their little finger by flexing their ipsilateral thumb. To keep the isometric contraction at 20–30% of the maximum force of the thenar muscle, the pressure transducer was simultaneously monitored on an oscilloscope during SEP recording. When the force was improper, correcting instructions were given orally to subjects by operators. Furthermore, in 10 subjects, SEPs were recorded during passive contraction of the thumb. The thumb was fixed by surgical tape at the same posture as during voluntary contraction. Special care was taken to maintain the wrist position to avoid displacing the stimulus electrodes. Additionally, effects of stimulus intensity on gating were studied in 7 aged normals (66.7 ± 11.8 years, mean ± SD). Subjects in which the maximum amplitudes of the prerolandic P22-N28 or N28-P45 or the postrolandic P24-N30 or N30-P45 were larger than 5 mV were chosen because gating is difficult to detect with small amplitude SEPs. Two SIs were employed. Stronger SI (7.3 ± 3.3 mA, n = 7), which produced minimum muscle twitch of the thenar muscle, evoked the prerolandic P22-N28 or N28-P45 or the postrolandic P24-N30 or N30-P45 beyond 5 mV, while weaker SI (3.7 ± 1.4 mA), which elicited no muscle response, produced responses beyond 3 mV. Under those stimulus conditions, the SEPs were recorded during rest as well as during voluntary contraction of the thenar muscle as described above. Furthermore, SEPs during rest and voluntary contraction of the thenar muscle were recorded using stimulus frequencies (SF) of 0.2 and 1 Hz with SI producing minimum muscle twitch of the thenar muscle in 5 young normals (33.3 ± 7.8 years, mean ± SD). 2.1. Data analysis The SEP peaks were defined according to the polaritylatency convention. The P14, N17, P22, N28, P45, and N60 in the prerolandic leads and P14, N17, P24, N30, P45, and N60 in the postrolandic leads were identified (Fig. 1A); the neighboring peak-to-peak amplitudes and the peak latencies were measured. We used the term N28 which is equivalent to N30 in previous reports to differentiate between this potential and the parietal N30. The potentials at C2, P10N13, and Erb’s point, P8-N10, were also analyzed. Amplitudes and latencies of each SEP component were compared statistically using the Mann-Whitney U test for unpaired data and Wilcoxon signed-ranks test for paired data. Differences of effects of muscle contraction on the SEP components between groups were assessed by the repeated measures analysis of variance (ANOVA). Relationships between SEP parameters or age and gating
110
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
Table 1 Comparisons of SEPs during rest between the aged and young groups Prerolandic lead Amplitudes of cortical SEPs (mV) P14-N17 N17-P22 P22-N28 N28-P45 P45-N60 Latencies of cortical SEPs (ms) P14 N17 P22 N28 p45 N60
Aged group
Young group
Postrolandic lead
2.27 4.71 4.97 6.73 8.81
± ± ± ± ±
l.43 2.86 2.90 3.89 2.91
2.03 2.33 2.67 3.21 4.24
± ± ± ± ±
0.97 1.68* 1.16** 1.59** 2.01†
P14-N17 N17-P24 P24-N30 N30-P45 P45-N60
14.02 18.18 22.66 28.85 44.14 66.17
± ± ± ± ± ±
0.83 l.42 l.42 2.19 6.13 5.21
14.05 17.43 21.75 28.23 41.51 55.95
± ± ± ± ± ±
l.26 l.29 2.01* 3.84 l.99 4.58††
P14 N17 P24 N30 P45 N60
Aged group
Young group
3.77 7.48 5.32 5.73 9.73
± ± ± ± ±
l.31 4.23 3.76 3.00 4.24
2.85 3.43 1.71 2.46 3.91
± ± ± ± ±
1.18* 1.92** 0.52† l.33† l.92††
14.19 19.04 24.27 31.15 43.46 66.74
± ± ± ± ± ±
0.82 l.16 l.43 l.76 5.96 6.24
14.03 18.23 22.54 28.32 40.83 55.27
± ± ± ± ± ±
l.25 l.22 2.05* 4.87 3.06 4.02††
Values are shown as the mean ± SD (n = 17 in the aged group and n = 12 in the young group). *P , 0.05; **P , 0.01; †P , 0.001; ††P , 0.0001, Mann-Whitney U test.
were analyzed using Pearson’s correlation coefficient or simple linear regression. Data are expressed as the mean ± SD.
3. Results 3.1. Comparisons between groups of SEP amplitudes and latencies during rest The amplitude of every cortical SEP component except for the prerolandic P14-N17 were significantly larger in the aged group than in the young group (Table 1). The P10-N13 amplitude at C2 was significantly larger (2.98 ± 0.88 mV) in the young group than in the aged group (1.76 ± 0.57 mV, P , 0.01, n = 29). There was no significant difference of P8-N10 amplitudes between the aged and young groups. The latencies of the prerolandic P22, the postrolandic P24 and the pre- and postrolandic N60 were significantly larger in the aged group compared to the young group (Table 1). The latencies of the P8, N10, P10, and N13 were similar between the groups. 3.2. Effects of voluntary or passive contraction of the thumb on SEPs In the young group, isometric voluntary contraction of the thenar muscles significantly attenuated the prerolandic P22N28 amplitude (Fig. 2). The aged group had significantly smaller amplitudes of the prerolandic P22-N28, N28-P45 and the postrolandic P24-N30 and N30-P45 during voluntary muscle contraction than during rest (Fig. 2). The amplitudes of cervical and Erb’s potentials did not significantly change following movement. Movement did not significantly affect SEP latencies in either group. We compared ratios of the difference between SEP amplitude during rest and movement (gated amplitude)/
SEP amplitude during rest (resting amplitude) in the two groups. Ratios of gated/resting amplitudes in the postrolandic N30-P45 were significantly larger in the aged group (0.13 ± 0.33) as compared to the young group (−0.27 ± 0.65, P , 0.02), and those in the prerolandic P45-N60 were significantly smaller in the young group (−0.14 ± 0.22) than in the aged group (0.03 ± 0.34, P , 0.05). In addition, we evaluated the effects of voluntary muscle contraction on SEP components in both groups. SEP amplitude changes following voluntary muscle contraction differed significantly between the groups in the prerolandic N28-P45 (df = 1,27, F = 4.945, P , 0.05), the postrolandic P24-N30 (F = 7.029, P , 0.02) and N30-P45 amplitudes (F = 7.062, P , 0.02) (Fig. 2). Passive contraction of the thumb did not have any significant effect on amplitudes and latencies of any SEP component in the aged group. 3.3. The effect of age or the relationships between SEP resting and gated amplitudes We studied the relationship between resting and gated amplitudes in components which showed significant gating. In the aged group, the gated and resting amplitudes significantly correlated in the prerolandic N28-P45 (P , 0.002, r = 0.68, n = 19) and the postrolandic P24-N30 (P , 0.0005, r = 0.756) (Fig. 3) and N30-P45 (P , 0.00l, r = 0.720). In the prerolandic P22-N28, ratios of gated/resting amplitudes correlated significantly with resting amplitude in the aged group (P , 0.05, r = −0.537). There were no significant correlations between gated and resting amplitudes or between ratios of gated/resting amplitudes and resting amplitude in the young group. The age of subjects significantly correlated to the gated amplitude of the prerolandic N28-P45 (P , 0.05, r = 0.372, n = 29) and the postrolandic P24-N30 (P , 0.0l, r = 0.475) and N30-P45 (P , 0.05, r = 0.385).
111
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
ger SI condition and the prerolandic P22-N28 amplitude in the weaker SI condition (Fig. 1, Table 2). The prerolandic N28-P45 and the postrolandic N30-P45 had significantly larger gated amplitudes in the stronger SI condition compared to the weaker SI (Table 2). We evaluated the effects of voluntary movement on each SEP component during the two conditions. Significant differences between the two SIs were observed only in the postrolandic N30-P45 (df = 1,8, F = 6.669, P , 0.05). 3.5. Effects of stimulus frequency on the gating in young normals
Fig. 2. Changes of SEP amplitudes (mean ± SEM) during rest and movement. SEP amplitudes during rest and voluntary movement are shown as thick lines in the aged group and thin lines in the young group. Dotted lines refer to SEP amplitudes during rest and passive movement in the aged group. Significant differences of SEP amplitudes between rest and voluntary movement (oblique bars) or effects of voluntary movement on SEP amplitudes between the groups (vertical bars) are shown as *P , 0.05 and **P , 0.01.
3.4. Effect of stimulus intensity on gating in aged normals A stronger SI produced amplitudes of 7.78 ± 2.12 mV (n = 20) during rest and 6.42 ± 1.83 mV during movement in the prerolandic P22-N28, N28-P45 and the postrolandic P24-N30, N30-P45. A weaker SI induced those SEP components with amplitudes of 4.81 ± 1.18 mV during rest and 4.37 ± 1.15 mV during movement. Movement significantly attenuated the prerolandic P22-N28 and N28-P45 amplitudes and the postrolandic N30-P45 amplitude in the stron-
The prerolandic P22-N28 and N28-P45 amplitudes tended to be smaller with SF of 1 Hz compared to SF of 0.2 Hz, and the postrolandic N30-P45 and P45-N60 amplitudes were significantly smaller with SF of 1 Hz than SF of 0.2 Hz (Table 3). The amplitudes of the other SEP components and the latencies of all SEP components did not vary significantly between the SF conditions. With SF of 0.2 Hz, voluntary movement significantly attenuated the prerolandic P22-N28 and N28-P45 and the postrolandic N30-P45, and the prerolandic P22-N28 amplitude was significantly smaller during movement than during rest with SF of 1 Hz (Fig. 4, Table 3). Gated amplitudes in the prerolandic N28-P45 and the postrolandic N30-P45 tended to be larger with SF of 0.2 Hz than SF of 1 Hz, but the differences between the SF conditions were not significant. There were no significant differences of effects of voluntary movement on SEP components between the SF conditions. None of the SEP components showed significant changes in latency by movement under each SF condition.
4. Discussion In this study, our purpose was to determine the effects of aging on SEP modification by voluntary movement. We demonstrated enlarged and prolonged cortical SEPs and
Table 2 Effects of stimulus intensity (SI) on gating in aged normals Prerolandic lead: P22-N28 amplitude
Stronger SI Weaker SI
Prerolandic lead: N28-P45 amplitude
Rest
Movement
Gated amplitude
Rest
Movement
6.91 ± 2.14 4.39 ± 1.10
5.83 ± 2.31* 3.75 ± 0.95*
1.08 ± 0.67 0.64 ± 0.38
9.30 ± 2.35 5.48 ± 1.59
6.84 ± 1.79* 4.36 ± 1.53
Postrolandic lead: P24-N30 amplitude
Stronger SI Weaker SI
Gated amplitude 2.46 ± 1.23** 1.12 ± 0.86
Postrolandic lead: N30-P45 amplitude
Rest
Movement
Gated amplitude
Rest
Movement
Gated amplitude
7.64 ± 2.46 4.54 ± 1.07
7.02 ± 2.39 4.34 ± 0.91
0.62 ± 0.77 0.19 ± 0.66
7.28 ± 0.97 4.82 ± 0.91
6.00 ± 0.91* 5.04 ± 1.05
1.27 ± 1.05** −0.22 ± 0.76
Values are shown as the mean ± SD in mV (n = 5). Amplitudes of SEP components between rest and movement conditions or gated amplitudes between stronger and weaker SI conditions were statistically compared by Wilcoxon signed-ranks test. *P , 0.05, rest vs. movement. **P , 0.05, stronger vs. weaker SI.
112
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
Table 3 Effects of stimulus frequency (SF) on gating in young normals SF (Hz)
Prerolandic lead: P22-N28 amplitude Rest
Movement
0.2 1
2.80 ± 0.96 2.52 ± 1.19
2.50 ± 1.03* 1.85 ± 0.52*
SF (Hz)
Postrolandic lead: N30-P45 amplitude
0.2 1
Prerolandic lead: N28-P45 amplitude Gated amplitude 0.30 ± 0.15 0.67 ± 0.73
Rest
Movement
4.10 ± 0.58 3.09 ± 0.79
2.72 ± 0.50* 2.55 ± 0.83
Gated amplitude 1.38 ± 0.83 0.54 ± 1.31
Postrolandic lead: P45-N60 amplitude
Rest
Movement
Gated amplitude
Rest
Movement
Gated amplitude
3.84 ± 0.73** 2.57 ± 1.03
2.97 ± 0.62* 2.62 ± 1.03
0.87 ± 0.38 -0.05 ± 0.77
4.88 ± 2.09** 3.27 ± 1.47
4.37 ± 2.12 3.79 ± 1.68
0.51 ± 0.52 −0.52 ± 0.85
Values are shown as the mean ± SD in mV (n = 5). Amplitudes of SEP components between rest and movement conditions or amplitudes of SEP components during rest between the two SF conditions were statistically compared by Wilcoxon signed-ranks test. *P , 0.05, rest vs. movement. **P , 0.05 between the two SF conditions.
diminished cervical potentials in the aged group compared with the young group. Those SEP changes in aged normals are consistent with results from previous studies (Lu¨ders, 1970; Desmedt and Cheron, 1980). Our study revealed that the aged group had significantly larger amplitude reductions of the prerolandic P22-N28 and N28-P45 amplitudes and the postrolandic P24-N30 and N30-P45 amplitudes following movement. On the other hand, the young group showed significant attenuation of the prerolandic P22-N28 by movement. In a preliminary study in aged subjects, the reduction of the prerolandic N28-P45 and the postrolandic N30-P45 amplitudes by movement disappeared with decreasing amplitude of the potentials achieved by reducing SI, whereas attenuation of the prerolandic P22-N28 remained. In young subjects, increasing SF from 0.2 Hz to 1 Hz caused attenuation of the pre- and postrolandic potentials as previously reported (Abbruzzese et al., 1990). According to the attenuation of SEPs, the reduction of the prerolandic N28P45 and the postrolandic N30-P45 amplitudes by movement disappeared despite retaining attenuation of the prerolandic P22-N28. Cohen and Starr (1987) have reported reductions of the prerolandic P22, N30, P45, and the postrolandic P27 amplitudes in median nerve SEPs following ballistic flexion of the thumb. Cheron and Borenstein (1987) have described attenuations induced by rapid finger movement of the parietal P27, frontal P22 and N30. In a study by Abbruzzese et al. (1981), isometric or isotonic movement of the thumb significantly attenuated the parietal N20-P25. Rushton et al. (1981) also have described reduction of the parietal N20-P30 and P45-N55 amplitudes during tracking movements of the thumb or finger. Although our experimental conditions differed from these prior studies, our results for gating correlated well with the findings described above. In this study, ANOVA analysis demonstrated differences in gating between the two groups. The attenuation of the prerolandic N28-P45 and the postrolandic P24-N30 and N30-P45 induced by movement was larger in the aged
group than in the young group. The aged group had larger SEP amplitudes and more marked cortical SEP gating effects compared to the young group. These findings suggest some relationship between the magnitude of gating and SEP amplitudes. In the correlation analysis, the gated amplitudes of SEPs significantly correlated with the resting amplitudes of the prerolandic N28-P45 and the postrolandic P24-N30 and N30-P45 in the aged group. We also observed correlative changes between the SEP amplitudes and the magnitude of gating in a preliminary study of the effects of stimulus intensity and frequency on gating. These results suggest that gated amplitudes of cortical SEPs depend on SEP amplitudes at rest. This conjecture does not contradict the view that the gating plays a role in screening out irrelevant sensory signals during movement (Coquery, 1978), or that the cortex filters out irrelevant information (Chapin and Woodward, 1981). Excess neural activities in the cortex responding to somatosensory afferent volleys may be cut off in relation to voluntary movement. Additionally, we found significant correlations between the age of subjects and the gated amplitudes. This increased magnitude of gating may be an effect associated with enlarged cortical SEPs in the elderly. Furthermore, the prerolandic P22-N28 was significantly attenuated by movement in the aged and young groups, but the gated amplitude and the resting amplitude of this potential did not correlate statistically. This finding suggests that central gating primarily occurs in the prerolandic P22-N28 and there is a functional difference among the SEP components which had significant gating. In prior studies, surface and depth recordings of SEPs revealed that the precentral and parietal SEPs originated in the cortical somatosensory area including area 3b and area 1 (Allison et al., 1991). In studies using magnetoencephalography, the somatosensory evoked magnetic fields (SEFs) were localized in the primary somatosensory cortex 20 and 35 ms following electrical stimulation of the median nerve (Forss et al., 1994; Kakigi, 1994; Kawamura et al., 1996). Those authors concluded that SEFs of 20 ms correspond to the N20 in the scalp SEPs and
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
113
The mechanism of the augmented gating effect in aged subjects is not clear. In our study, holding the 2 cm round pressure transducer with a stable muscle strength requires sensory feedback from muscle spindle, joint, and cutaneous receptors. However, the amplitudes of cervical and Erb’s potentials did not significantly change following movement, and passive movement did not have any significant effect upon any SEP parameters. In addition, SEPs attenuated with increased movement speed, but not with increased muscle loading (Ghez and Pisa, 1972; Rushton et al., 1981). Anesthesia of skin or joint related to the movement also does not affect gating (Rushton et al., 1981). Jones et al. (1989) have concluded that sensory feedback from muscle spindle, joint and cutaneous receptors distal to the stimulation did not play a significant role in gating because attenuation of SEPs was at a maximum when the stimuli were delivered at the onset of movement. Therefore, we think that the gating in our study was predominantly due to the centrifugal mechanism as inhibitory discharges from the cortical motor system attenuated incoming afferent flow to
Fig. 3. Relationships between the prerolandic N28-P45 amplitude (A) or the postrolandic P24-N30 amplitude (B) during rest and amplitudes of those components reduced by voluntary movement (gated amplitude). Circles represent the aged group and darkened squares are for the young group. Regression lines in the aged group also are shown in scatter diagrams (A: y = 4.511 + 1.457x, P , 0.005; B: y = 3.005 + 2.453x, P , 0.0005).
those of 35 ms correspond to the N35 or N30. However, clinical studies have demonstrated different features of the prerolandic P22 and N30 as compared to the postrolandic potentials (Mauguie`re et al., 1983; Slimp et al., 1986; Dinner et al., 1987; Mauguie`re and Desmedt, 1991). Desmedt and Bourguet (1985) have postulated that P22 originated in area 4 and the origin of N30 (N28) was in the supplementary motor area. Kawamura et al. (1996) have reported that SEFs that occur approximately 28 ms following electrical stimulation of the median nerve are located in the precentral cortex, area 4. Since we hypothesized that different potential origins can explain these functional differences, our results support the view that the prerolandic P22-N28 originates in the cortical motor areas.
Fig. 4. SEPs induced by the stimulus frequencies of 0.2 Hz (A) or 1 Hz (B) in a representative young subject. Thick lines show SEPs during rest and thin lines show those during movement.
114
T. Touge et al. / Electroencephalography and clinical Neurophysiology 104 (1997) 108–114
the primary sensory cortex. The primary motor area has direct interconnections with areas 1, 2, and the anterior part of area 5, and indirectly with areas 3a and 3b. Furthermore, the supplementary motor area is also interconnected with the parietal sensory areas (Jones, 1986). Through such interneuronal connections, the inhibitory or excitatory influences from the cortical motor areas may be able to regulate neuronal activities in the cortical sensory areas receiving somatosensory afferent volleys (Kakigi et al., 1995). We demonstrated enhanced gating associated with enlargement of SEPs in the elderly. When we evaluate the gating of SEPs in humans, it is necessary to consider the effects of aging.
References Abbruzzese, G., Ratto, S., Favale, E. and Abbruzzese, M. Proprioceptive modulation of somatosensory evoked potentials during active or passive finger movements in man. J. Neurol. Neurosurg. Psychiatry, 1981, 44: 942–949. Abbruzzese, G., Dall’Agata, D., Morena, M., Reni, L., Trivelli, G. and Favale, E. Selective effects of repetition rate on frontal and parietal somatosensory evoked potentials (SEPs). In: P.M. Rossini and F. Mauguie`re (Eds.), New Trends and Advanced Techniques in Clinical Neurophysiology (EEG Suppl. 41). Elsevier, Amsterdam, 1990, pp. 145– 148. Allison, T., McCarthy, G., Wood, C.C. and Jones, S.J. Potentials evoked in human and monkey cerebral cortex by stimulation of the median nerve. Brain, 1991, 114: 2465–2503. Angel, R.W., Boylls, C.C. and Weinrich, M. Cerebral evoked potentials and somatosensory perception. Neurology, 1984, 34: 123–126. Chapin, J.K. and Woodward, D.J. Modulation of sensory responsiveness of single somatosensory cortical cells during movement and arousal behaviors. Exp. Neurol., 1981, 72: 164–178. Cheron, G. and Borenstein, S. Specific gating of the early somatosensory evoked potentials during active movement. Electroenceph. clin. Neurophysiol., 1987, 67: 537–548. Cohen, L.G. and Starr, A. Localization, timing and specificity of gating of somatosensory evoked potentials during active movement in man. Brain, 1987, 110: 451–467. Coquery, J.-M. Role of active movement in control of afferent input from skin in cat and man. In: G. Gordon (Ed.), Active Touch. Pergamon Press, New York, 1978, pp. 161–169. Coulter, J.D. Sensory transmission through lemniscal pathway during voluntary movement in the cat. J. Neurophysiol., 1974, 37: 831–845. Desmedt, J.E. and Bourguet, M. Color imaging of parietal and frontal somatosensory potential fields evoked by stimulation of median or posterior tibial nerve in man. Electroenceph. clin. Neurophysiol., 1985, 62: 1–17. Desmedt, J.E. and Cheron, G. Somatosensory evoked potentials to finger stimulation in healthy octogenarians and in young adults: wave forms,
scalp topography and transit times of parietal and frontal components. Electroenceph. clin. Neurophysiol., 1980, 50: 404–425. Dinner, D.S., Lu¨ders, H., Lesser, R.P. and Morris, H.H. Cortical generators of somatosensory evoked potentials to median nerve stimulation. Neurology, 1987, 37: 1141–1145. Forss, N., Hari, R., Salmelin, R., Ahonen, A., Ha¨ma¨la¨inen, M., Kajola, M., Knuutila, J. and Simola, J. Activation of the human posterior parietal cortex by median nerve stimulation. Exp. Brain Res, 1994, 99: 309– 315. Ghez, C. and Pisa, M. Inhibition of afferent transmission in cuneate nucleus during voluntary movement in the cat. Brain Res., 1972, 40: 145–151. Jiang, W., Chapman, C.E. and Lamarre, Y. Modulation of somatosensory evoked responses in the primary somatosensory cortex produced by intracortical microstimulation of the motor cortex in the monkey. Exp. Brain Res., 1990, 80: 333–344. Jones, E.G. Connectivity of the primate sensory-motor cortex. In: E.G. Jones and A. Peters (Eds.), Cerebral Cortex, Vol. 5. Plenum Press, New York, 1986, pp. 113–183. Jones, S.J., Halonen, J.-P. and Shawkat, F. Centrifugal and centripetal mechanisms involved in the ‘gating’ of cortical SEPs during movement. Electroenceph. clin. Neurophysiol., 1989, 74: 36–45. Kakigi, R. Somatosensory evoked magnetic fields following median nerve stimulation. Neurosci. Res., 1994, 20: 165–174. Kakigi, R., Koyama S., Hoshiyama M., Watanabe S., Shimojo, M. and Kitamura, Y. Gating of somatosensory evoked responses during active finger movements: magnetoencephalographic studies. J. Neurol. Sci., 1995, 128: 195–204. Kawamura, T., Nakasato, N., Seki, K., Kanno, A., Fujita, S., Fujiwara, S. and Yoshimoto, T. Neuromagnetic evidence of pre- and post-central cortical sources of somatosensory evoked responses. Electroenceph. clin. Neurophysiol., 1996, 100: 44–50. Kujirai, T., Sato, M., Rothwell, J.C. and Cohen, L.G. The effect of transcranial magnetic stimulation on median nerve somatosensory evoked potentials. Electroenceph. clin. Neurophysiol., 1993, 89: 227–234. Lu¨ders, H. The effects of aging on the wave form of the somatosensory cortical evoked potential. Electroenceph. clin. Neurophysiol., 1970, 29: 450–460. Mauguie`re, F. and Desmedt, J.E. Focal capsular vascular lesions can selectively deafferent the prerolandic or the parietal cortex: somatosensory evoked potentials evidence. Ann. Neurol., 1991, 30: 71–75. Mauguie`re, F., Desmedt, J.E. and Courjon, J. Astereognosis and dissociated loss of frontal or parietal components of somatosensory evoked potentials in hemispheric lesions. Brain, 1983, 106: 271–311. Rushton, D.N., Rothwell, J.C. and Craggs, M.D. Gating of somatosensory evoked potentials during different kinds of movement in man. Brain, 1981, 104: 465–49l. Slimp, J.C., Tamas, L.B., Stolov, W.C. and Wyler, A.R. Somatosensory evoked potentials after removal of somatosensory cortex in man. Electroenceph. clin. Neurophysiol., 1986, 65: 111–117. Towe, A.L. and Jabbur, S.J. Cortical inhibition of neurons in dorsal column nuclei of cat. J. Neurophysiol., 1961, 24: 488–498. Tsumoto, T., Nakamura, S. and Iwama, K. Pyramidal tract control over cutaneous and kinesthetic sensory transmission in the cat thalamus. Exp. Brain Res., 1975, 22: 281–294.