Somatosensory evoked magnetic fields following electric tongue stimulation using pin electrodes

Available online at www.sciencedirect.com

Neuroscience Research 62 (2008) 131–139 www.elsevier.com/locate/neures

Somatosensory evoked magnetic fields following electric tongue stimulation using pin electrodes Hitoshi Maezawa a,b, Kazuya Yoshida a,c, Takashi Nagamine b,*, Jun Matsubayashi b,d, Rei Enatsu e, Kazuhisa Bessho a, Hidenao Fukuyama b a

Department of Oral and Maxillofacial Surgery, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan b Human Brain Research Center, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan c Department of Oral and Maxillofacial Surgery, National Hospital Organization, Kyoto Medical Center, Fushimi-ku, Kyoto 612-8555, Japan d Department of Occupational Therapy, School of Health Sciences, Faculty of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan e Department of Neurosurgery, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan Received 19 April 2008; accepted 11 July 2008 Available online 29 July 2008

Abstract Quantitative evaluation of the sensory disturbance of the tongue is important clinically. However, because the conventional electrophysiological approach to the peripheral nerve cannot be used in the mandible owing to the deep route of the lingual nerve, we applied evoked potentials in the central nervous system. Somatosensory evoked magnetic fields (SEFs) following electric stimulation were recorded in 10 healthy subjects by means of pin electrodes placed on the tongue mucosa. Three or four components (P25m, P40m, P60m, and P80m) were identified over the contralateral hemisphere with unilateral stimulation. Because none of the components were consistently detected in all subjects, we evaluated the root mean square (RMS) of 18 channels over the contralateral hemisphere. To estimate the activated cortical response, we calculated the difference in mean RMS amplitude between 10 and 150 ms and that of the baseline period (aRMS = RMS[10, 150] RMS[ 50, 5]). The aRMS values for right-sided and left-sided stimulation were 10.18  7.92 and 10.99  8.98 fT/cm, respectively, and the mean laterality index, expressed by [(left right)/(left + right)] was 0.025  0.104. This parameter can be useful for evaluating patients with unilateral sensory abnormality of the tongue. # 2008 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved. Keywords: Magnetoencephalography; Primary somatosensory cortex; Root mean square; Tongue; Lingual nerve; Trigeminal nerve; Laterality; Sensory disturbance

1. Introduction In oral and maxillofacial surgery, we sometimes encounter injuries to the unilateral mandibular nerve that cause sensory disturbances in the corresponding area of the tongue. In these

Abbreviations: MEG, magnetoencephalography; SEPs, somatosensory evoked potentials; SEFs, somatosensory evoked magnetic fields; MRI, magnetic resonance imaging; ECD, equivalent current dipole; RMS, root mean square; SI, primary somatosensory cortex; SII, secondary somatosensory cortex. * Corresponding author. Current address: Department of System Neuroscience, School of Medicine, Sapporo Medical University, Sapporo, Japan. Tel.: +81 11 611 2111x2660; fax: +81 11 644 1020. E-mail addresses: [email protected], [email protected] (T. Nagamine).

cases, it is important to investigate the extent and severity of the sensory disturbance to assess prognosis and to select treatment. However, because no method of objectively examining the extent of sensory damage has been established, we can use only subjective methods, such as two-point discrimination thresholds, whose results might not be reliable or reproducible. Moreover, most commonly used electrophysiological approaches to the peripheral nerve, such as nerve conduction studies, cannot be used, because the lingual nerve runs deep within the mandible. However, previous reports have indicated that evoked responses in the central nervous system, such as somatosensory evoked potentials (SEPs) and somatosensory evoked fields (SEFs), can complement other established clinical methods, such as nerve conduction studies and electromyography (Synek, 1987; Shibasaki et al., 1988; Altenmu¨ller et al., 1990; McDonald et al., 1996; Yamashita et al., 1999). Thus, SEFs following tongue

0168-0102/$ – see front matter # 2008 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved. doi:10.1016/j.neures.2008.07.004

132

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139

stimulation might be useful for objectively evaluating somatosensory impairment after injuries of the lingual nerve. The primary somatosensorimotor cortex (SI) is organized in an orderly somatotopic way, which has been termed the ‘‘homunculus’’ representation of brain areas related to the whole human body (Penfield and Rasmussen, 1952). The tongue area represented in the homunculus is large, suggesting its importance for human sensory processing. After the first report by Ishiko et al. (1980) of SEPs following tongue stimulation, there have been several reports on SEPs (Hashimoto, 1988; Altenmu¨ller et al., 1990; Maloney et al., 2000), intracranial recording of SEPs (McCarthy et al., 1993), and SEFs (Karhu et al., 1991; Yamashita et al., 1999; Disbrow et al., 2003; Nakahara et al., 2004; Sakamoto et al., 2008). However, obtaining clear waveforms was difficult because of excessive stimulus artifacts and the induction of unintended muscle activity due to the short distance between stimulating and recording electrodes. This difficulty is mainly produced by the stimulus methods. Therefore, the signal-to-noise ratio might be improved if noise can be reduced through a better stimulation method with newly devised electrodes. The aims of this study were (1) to establish SEFs following tongue stimulation in healthy subjects by lowering the electric intensity with pin electrodes, and (2) to examine the laterality of the cortical activity evoked by unilateral tongue stimulation. The study can provide a new parameter, which can be useful for evaluating unilateral sensory disturbances of the tongue.

2. Materials and methods 2.1. Subjects Ten healthy subjects (10 men aged 27–34 years; mean age, 29.8 years) were studied. All subjects were right-handed. None of the subjects had a history of neurological or psychiatric illness. Informed consent was obtained from all

subjects before the start of the study, which was approved by the Kyoto University Graduate School and Faculty of Medicine, Ethics Committee (Protocol No. E-196).

2.2. Tongue stimulation We stimulated three points on the dorsum of the tongue in the following order: (1) the right side (2 cm from the tip of tongue, 1 cm from the edge), (2) the left side (symmetric to the right side), and (3) the midline (1 cm from the tip of the tongue). Before recording, the three stimulus points were marked with crystal violet. A pair of epoxy resin-coated platinum pin electrodes (0.4 mm in diameter) was used for stimulation. The electrodes were placed 3 mm from each other, and the tip of each electrode was exposed 0.2 mm (Custom made, Unique Medical, Tokyo, Japan) (Fig. 1a). After the tips of the electrodes were placed against the dorsum of the tongue mucosa gently and without causing pain sensation, adhesive tape was used to fix them to the tongue (Fig. 1b and c). The electrical biphasic square pulse waves of a constant voltage were delivered at a rate of 1 Hz using an electrical stimulator (SEN7203, Nihon Kohden, Tokyo, Japan). To reduce the stimulus intensity, a long stimulus duration (0.5 ms for one phase) was used to minimize stimulus artifacts. The stimulus intensity was adjusted to four times the sensory threshold of each subject. During recording, subjects were requested to open their mouths slightly to become relaxed.

2.3. Magnetoencephalography recordings Neuromagnetic signals were measured with a helmet-shaped 306-channel apparatus (VectorView, Elekta Neuromag, Helsinki, Finland) in a magnetically shielded room (Fig. 1b). This device had 102 trios that are composed of a magnetometer and a pair of planar gradiometers oriented orthogonally. We used only 204 planar gradiometers for analysis, which detect the largest signal just above the corresponding generator source (Ha¨ma¨la¨inen et al., 1993). The recording band pass was 0.1–990 Hz for magnetoencephalography (MEG), and the sampling rate for digital conversion was 2997 Hz. The analysis window for averaging was 100 ms before to 500 ms after each trigger signal, and the prestimulus time period from 50 to 5 ms was used as the baseline. Epochs containing MEG signals exceeding 1500 fT/cm were excluded automatically from averaging. At least 600 responses were averaged for each session, and two sessions separated by a rest period were held in each stimulus point. We used the group averaged data of two sessions of each stimulus point for further analysis after confirming reproducibility. The subjects were instructed to sit on a chair and requested to watch a silent movie on a screen in the shielded room to maintain the subjects’ vigilance during the recordings.

Fig. 1. Measurement of SEFs following tongue stimulation. (a) A pair of pin electrodes used for the stimulation of the tongue. The interelectrode distance was 3 mm. (b) Measurement of SEFs using a helmet-shaped MEG system. (c) Electrodes were gently fixed to the dorsum of the tongue mucosa.

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139 To visualize locations of MEG sources, magnetic resonance imaging scans of the head were obtained with a 0.2-T Signa Profile System (General Electric Medical Systems, Milwaukee, WI, USA) from all subjects.

2.4. Data analysis The peak latency of the response was measured from the channel showing the maximal signal. Isocontour maps were constructed at the selected time points using the minimum-norm estimate. The digitized shape of the head of each subject was fitted using a simple spherical head model (Sarvas, 1987). The sources of the magnetic fields were modeled as equivalent current dipoles (ECDs), whose location, orientation, and current strength were estimated from the measured magnetic waveforms. We accepted only ECDs attaining 90% goodness-of-fit (GOF) and a confidence volume smaller than 1000 mm3. We calculated the mean amplitude within 10–150 ms after the stimulus onset for the root mean square (RMS) over the contralateral hemisphere (RMS[10, 150]), where the RMS was calculated from the 18-channel waveforms including the channel of maximum amplitude. To estimate the activation of the mean activated cortical response, we introduced the term ‘‘activated RMS’’ (aRMS) by subtracting the mean amplitude of the baseline (50–5 ms before the stimulus) from RMS[10, 150]. To evaluate the effect of location on the laterality of the RMS following unilateral stimulation, distances between the head origin and ECD locations were compared at the peak latency of the maximum magnitude component over the contralateral hemisphere. Data are expressed as mean  standard deviation (S.D.). Laterality between the right-sided and left-sided stimulation was checked with Wilcoxon signedrank test for sensory thresholds, latency, dipole moment, aRMS, and the distance from the head origin to the ECD location. The level of statistical significance difference was set at 5% (P < 0.05).

3. Results 3.1. Stimulus intensity The sensory thresholds of each stimulated area ranged from 0.11 to 0.45 mA (mean, 0.24 mA) for the right side, from 0.12 to 0.40 mA (mean, 0.23 mA) for the left side, and from 0.11 to 0.30 mA (mean, 0.19 mA) for the midline area. Therefore, the stimulus intensity applied for each subject ranged from 0.44 to 1.80 mA (mean, 0.88 mA). No significant difference in the sensory thresholds was recognized between right-sided and left-sided stimulation (P = 0.93). The stimuli were not painful and did not cause any visible contraction of the facial muscles or blinking. We did not detect tongue muscle movement or twitch in any subject. Subjects felt electric stimulation only in the stimulated area of the tongue and did not feel any taste sensation during the recordings. 3.2. Waveform configuration Clear responses were detected over the bilateral temporoparietal areas in all subjects. A representative subject (subject 1) showed four main deflections peaking at 23, 35, 59, and 75 ms over the right hemisphere following the left tongue stimulation (Fig. 2). Three reflections were observed over the left hemisphere at 36, 63, and 80 ms. We designated the former four responses over the contralateral hemisphere as P25m, P40m, P60m, and P80m and the latter three components over the ipsilateral hemisphere as P40m(I), P60m(I), and P80m(I) according to the nomenclature system employed in previous

133

studies of trigeminal SEFs (Hoshiyama et al., 1996; Nagamatsu et al., 2001; Nakahara et al., 2004; Nguyen et al., 2004; Nevalainen et al., 2006). P25m was identified only in the contralateral hemisphere, but P40m, P60m, and P80m were detected over both hemispheres. 3.3. Contralateral responses following unilateral stimulation Right-sided tongue stimulation elicited P25m over the contralateral temporoparietal area in 7 subjects at a peak latency of 24.7  2.2 ms, and left-sided stimulation evoked P25m in 7 subjects at a peak latency of 24.8  2.1 ms (Table 1). The magnitude of the component ranged from 7.2 to 39.0 fT/cm after right-sided stimulation and from 8.5 to 22.0 fT/cm after leftsided stimulation. This component sometimes appeared only as a shoulder or a notch embedded in the following deflection to make a complex waveform. The latency of P40m was 39.2  4.4 ms for right-sided stimulation and 39.4  4.2 ms for left-sided stimulation (Table 1), and the latency of P60m and P80m are shown in Table 1. The response magnitudes of P40m, P60m, and P80m were 12.2–90.0, 13.5–67.2 and 18.4–150.0 fT/cm for rightsided stimulation and 11.2–84.8, 14.1–78.0, and 20.2–175.5 fT/ cm for left-sided stimulation. The isofield contour maps of each component following leftsided tongue stimulation showed a clear dipolar pattern, and all orientations directed posteriorly in a typical subject (subject 1) (Fig. 3(a)). The ECDs of these components were located around the lower part of the central sulcus (Fig. 3(b)). The strength of the P25m ECDs ranged from 0.8 to 7.0 nAm and from 0.8 to 3.7 nAm after right-sided and left-sided stimulation, respectively. The strength of the P40m, P60m, and P80m ECDs were 2.4–17.3, 3.8–24.0, and 5.6–24.7 nAm for the right stimulation, and 1.4–14.6, 8.4–28.7, and 5.9–27.2 nAm for left-sided stimulation. No significant difference in the latency, magnitude, or strength of ECD was recognized between right-sided and left-sided stimulation in any of the components. 3.4. The RMS of contralateral responses after unilateral stimulation The RMS waveforms over the contralateral area evoked by left-sided stimulation had a similar shape to waveforms after right-sided stimulation in the same subject but they varied from subject to subject (Fig. 4). The across-subjects average of aRMS for right-sided and left-sided stimulation were 10.18  7.92 and 10.99  8.98 fT/cm, respectively. No significant difference in aRMS was recognized between rightsided and left-sided stimulation (P = 0.29). The mean laterality index of aRMS shown by [(left right)/(left + right)] was 0.025  0.104 (Table 2). Despite the intersubject variability of the RMS amplitude, a high intrasubject resemblance was observed between right-sided and left-sided stimulation. The distance between the head origin and the dipole location following right-sided stimulation was 62.4  5.8 mm, and the distance after left-sided stimulation was 63.0  5.0 mm. No

134

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139

Fig. 2. The whole-head magnetic waveforms of SEFs following left-sided tongue stimulation in a typical subject (subject 1). (a) The top view of the SEF recorded by the planar 204-channel recording shows clear responses over the bilateral parietotemporal areas. Each trace started 50 ms before to 300 ms after the stimulus onset. As shown in the expanded waveforms (b and c), 4 and 3 components are identified over the contralateral and ipsilateral hemispheres. (d) The RMS waveforms over the contralateral hemisphere were calculated from 18-channel waveforms including the channel with local maximum amplitude.

significant difference was observed between right-sided and left-sided stimulation (P = 0.72). 3.5. Ipsilateral responses following unilateral stimulation P40m(I), P60m(I), and P80m(I) were detected in 7, 6, and 4 subjects, respectively, after right-sided stimulation and in 7, 7, and 3 subjects, respectively, after left-sided stimulation (Table 1). The latencies of P40m(I) following right-sided and left-sided stimulation were 38.3  4.6 ms and 38.0 

4.4 ms, respectively, and the latencies of P60m(I) and P80m(I) are shown in Table 1. The magnitude of the P40m(I) ranged from 10.5 to 43.4 fT/cm after right-sided stimulation and from 7.2 to 38.2 fT/cm after left-sided stimulation. The magnitude of P60m(I) ranged from 15.6 to 87.2 fT/cm and from 7.6 to 89.5 fT/cm after right-sided and left-sided stimulation, respectively. The magnitude of P80m(I) ranged from 13.8 to 31.0 fT/cm and from 17.1 to 30.1 fT/cm after right-sided and left-sided stimulation, respectively. The isocontour maps of these components showed a dipolar pattern and allowed us to

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139

135

Table 1 SEF latencies over both hemispheres Unilateral stimulation Contra. hemisphere

P25m

rt stim. lt stim. ipsi. hemisphere

P40m

P60m

P80m

n

Latency (ms)

n

Latency (ms)

n

Latency (ms)

n

Latency (ms)

7 7

24.7  2.2 24.8  2.1

8 8

39.2  4.4 39.4  4.2

7 7

57.4  7.3 57.3  5.1

7 6

77.9  9.0 76.3  7.0

P25m n

P40m Latency (ms)

rt stim. lt stim.

P60m

P80m

n

Latency (ms)

n

Latency (ms)

n

Latency (ms)

7 7

38.3  4.6 38.0  4.4

6 7

58.1  6.9 57.4  5.4

4 3

75.0  5.4 76.0  7.8

Midline stimulation Left hemisphere

Right hemisphere

Mid-P25m(L)

Mid-P40m(L)

Mid-P60m(L)

Mid-P80m(L)

n

Latency (ms)

n

Latency (ms)

n

Latency (ms)

n

Latency (ms)

7

26.3  1.5

9

40.4  2.9

8

56.4  4.8

4

77.8  3.8

Mid-P25m(R)

Mid-P40m(R)

Mid-P60m(R)

Mid-P80m(R)

n

Latency (ms)

n

Latency (ms)

n

Latency (ms)

n

Latency (ms)

5

24.3  3.5

9

42.3  4.8

7

56.3  5.9

4

76.5  7.9

contra. hemisphere, contralateral hemisphere; ipsi. hemisphere, ipsilateral hemisphere; rt stim., right-sided stimulation; lt stim., left-sided stimulation.

calculate their ECD locations. The estimated locations were all close to the lower portion of the central sulcus. The strength of the P40m(I) ECDs ranged from 1.6 to 8.2 nAm and from 1.0 to 14.8 nAm after right-sided and left-sided stimulation, respectively. The strength of the P60m(I) ECDs and P80m(I) ECDs were 4.0–18.5 and 4.1–8.6 nAm for right-sided stimulation and 3.8–28.7 and 6.3–13.7 nAm after left-sided stimulation. The latency, magnitude, and ECD strength of eachP40m(I) and P60m(I) did not differ significantly between the sides of stimulation. The P80m(I) component was excluded from the analysis because of the small sample size. 3.6. Midline stimulation When the center of the tongue was stimulated, SEF responses were detected over both hemispheres. The typical subject (subject 1) showed components peaking at 27, 38, 51, and 74 ms over the left hemisphere and at 22, 38, 50, and 72 ms over the right hemisphere. These 4 components in each hemisphere were designated Mid-P25m(L), Mid-P40m(L), Mid-P60m(L), and Mid-P80m(L), respectively, for those in the left hemisphere and Mid-P25m(R), Mid-P40m(R), Mid-P60m(R), and MidP80m(R), respectively, for those in the right hemisphere. MidP25m(L) and Mid-P25m(R) were identified in 7 and 5 subjects respectively. Both Mid-P25m(L) and Mid-P25m (R) were detected in 4 subjects. The latencies of Mid-P40m, Mid-P60m, and Mid-P80m are shown in Table 1. All the ECDs were located around the lower part of the central sulcus. 4. Discussion The present study has demonstrated clear tongue SEFs, which were obtained with the successful reduction of stimulus

artifacts by means of pin electrodes (Fig. 2(a)). Compared with the large intersubject variability of the waveform configuration, the intrasubject similarity was noted in the laterality index. P25m, the first recognizable component in the study, was detected in 7 of 10 subjects only over the contralateral hemisphere. A previous study has demonstrated that the initial component of the lip SEFs by electric stimulation had an anteriorly directed current with a peak latency of 15 ms (Nagamatsu et al., 2001). Judging from the distance from the stimulus site to the recording position, we can reasonably assume similar values for both the lip SEFs and tongue SEFs as for the latency and direction of the initial component. Thus, P25m, which has posteriorly directed current, does not represent the initial component of the tongue SEFs. We can suggest two possible reasons for the failure to detect the initial component. First, the intensity of the electrical stimulation might be insufficient to evoke the initial component, because it is smaller than the stimulus intensity of nine times the sensory threshold employed in a previous report (Nagamatsu et al., 2001). However, Sakamoto reported the initial component at about 19 ms following lingual nerve stimulation at two times sensory thresholds. They used a bipolar electrode, which had a cathode (1.0 mm diameter) and anode (3.5 mm diameter) in a concentric circle. The contact area with tongue mucosa using this electrode was larger than that using pin electrode in the study. There is a possibility that the large contact area in the peripheral enabled to raise such large cortical activation that initial component could be measured. Second, the response evoked by the pin electrodes was smaller than that elicited by the clip electrodes because of the limited area of contact over the tongue mucosa. The deflections of P40m, P60m, and P80m were detected over both hemispheres. The deflections of P40m and P60m corresponded to those recorded in a previous MEG

136

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139

Fig. 3. Isocontour maps and the dipole locations of each component following left-sided tongue stimulation in a representative subject (subject 1). (a) The contour maps were obtained from 4 components (Fig. 2b). The exact timing is shown in each map. The contour steps are 10 fT (P25m), 20 fT (P40m), and 40 fT (P60m, and P80m). Red and blue lines indicate outgoing and incoming magnetic fluxes, respectively. Green arrows show the location and direction of ECDs projected on the skull surface producing the SEF distribution. Arrowheads indicate the negative pole of the ECD. The directions of all ECDs are similarly posterior. (b) All ECDs were superimposed on the slices of magnetic resonance images and surface rendering images of the subject. They were located in the same area close to the lower part of the central sulcus.

study using tactile stimulation: 30 and 55 ms (Disbrow et al., 2003), and the deflection of P60m agreed with that by electrical stimulation: cP55m (Nakahara et al., 2004). Because the ECD originating from the secondary somatosensory cortex (SII) directed upward in previous studies of trigeminal nerve stimulation (Hoshiyama et al., 1996; Nguyen et al., 2004, 2005; Yoshida et al., 2006), it is unlikely that any of these responses (P25m, P40m, P60m, and P80m) with posteriorly directed current had emerged from SII. In the previous study, Karhu et al. (1991) reported that the late component at around 140 ms (N140m), which was originated from SII. However, we did not detect this component whose ECD directed medial–

lateral direction, although we detected a component whose ECD showed anterior–posterior direction in two subjects at around 100 ms. The main objective of the present study was to examine the laterality of the cortical activity evoked by unilateral tongue stimulation. Usually, some components consistently detected in all the subjects can serve as indicators for this purpose. However, because none of the four components (P25m, P40m, P60m, and P80m) detected in the study fitted this condition, we could not adopt their latency, amplitude, and location as reliable parameters for assessing cortical activity. Therefore, to overcome this problem, we employed a two-step analysis

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139

137

Fig. 4. The RMS waveforms calculated from the 18-channel around the channel with local maximum amplitude over the contralateral hemisphere in all subjects. The vertical scale was 50 fT/cm except for subject 1 with 80 fT/cm. The 2 dash lines in each figure show the time points of 10 and 150 ms respectively. The shape and amplitude of the waveforms varied among subjects. The high correlation of the shape and amplitude is observed between the intrasubject hemispheres following either right-sided or left-sided stimulation. rt stim., right side stimulation; lt stim., left side stimulation.

method by using spatial and temporal summation. The first step of spatial summation was conducted with the RMS, which was calculated from the 18 channels over the contralateral area. In the second step of temporal summation, we calculated the increment of the mean amplitude of the RMS. Because we did not encounter any significant responses before or after 10– 150 ms from the stimulus, we calculated the mean amplitude of Table 2 The aRMS in all subjects Subject

aRMS (fT/cm) rt stim.

lt stim.

Laterality index

1 2 3 4 5 6 7 8 9 10

31.5 9.6 12.5 10.2 6.4 8.9 7.1 6.5 5.6 3.5

34.7 13.9 13.1 11.0 8.6 7.8 5.2 5.8 5.4 4.4

0.05 0.18 0.02 0.04 0.15 0.07 0.15 0.06 0.02 0.11

Average S.D.

10.18 7.92

10.99 8.98

0.025 0.104

aRMS means the difference of average RMS amplitude from 10 to 150 ms and that from 50 to 5 ms (aRMS = RMS[10, 150] RMS[ 50, 5]). Laterality index of aRMS was calculated with [(lt stim. rt stim.)/(lt stim. + rt stim.)]. rt stim., right-sided stimulation; lt stim., left-sided stimulation.

this time period (RMS[10, 150]). Net cortical activation was obtained by subtracting the RMS of the baseline noise (RMS[ 50, 5]) from RMS[10, 150] (aRMS = (RMS[10, 150]) (RMS[ 50, 5]). Despite significant intersubject variation in the RMS morphology and the aRMS value, a high intrasubject similarity was observed between the hemispheres, which is similar to the finding of a previous report of median nerve stimulation in which the shapes of SEFs over the contralateral area were similar between the sides of stimulation (Tecchio et al., 2000). Analysis using the RMS has two advantages. First, it does not require a hypothesis regarding the number of generator sources. This analysis can accept any combination of multiple generators. Second, analysis can be performed even when generators cannot provide a high enough goodness of fit or confidence volume due to a small response. On the other hand, RMS analysis also has disadvantages. The most significant one is that the RMS is influenced by the distance between the generators and sensors, which might differ between the hemispheres. In this study, because we found no significant difference in distance between the hemispheres, we could ignore the effect from the relative ECD location. Obtaining the mean amplitude during a certain time range is often effective when the waveform does not have sharp peak components, such as the P300 component (Mecklinger et al., 1998; Shirahama et al., 2004) and is also effective when some constituents have lost their synchronicity and failed to form

138

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139

peaks due to the loss of constant arrival from the tongue to the cortical area. In this study, we tried to stimulate small areas in the oral cavity using pin electrodes. As a result, we could effectively stimulate the tongue with an electric stimulus intensity lower than that used previously. The use of pin electrodes has three advantages. First, the low electric intensity prevented the tongue from being damaged by electric shocks. Second, the low intensity suppressed stimulus artifacts and the twitching of tongue muscles. In the previous study (Sakamoto et al., 2008), suprahyoid electromyogram was monitored to evaluate contamination of magnetic field derived from muscle contraction. However, since no significant activity was detected before P40m in our results, we did not need to check suprahyoid electromyogram. Third, the pin electrodes enabled us to correctly stimulate a small area in any location, which is difficult to do with clip electrodes or tactile stimulating devices that have large contact areas. The ability to stimulate small areas can be very useful, especially for evaluating patients who have damage in a narrow area. On the other hand, the electrical stimulation by the electrodes in this study also has a disadvantage. We cannot specify what kind of receptor or free nerve ending is stimulated, although we believe we have not stimulated either the facial or glossopharyngeal nerve because none of the subjects felt any taste sensation during the recording. As with unilateral stimulation, stimulation of the center of the tongue evoked Mid-P40m, Mid-P60m, and Mid-P80m bilaterally. The components, except for P25m, were essentially the same as those observed following unilateral tongue stimulation in terms of the latency and amplitude. P25m was detected over both hemispheres in 4 subjects, which was in clear contrast with that after unilateral stimulation. This result can be explained by the tip of the tongue being innervated by both the right and left of the lingual nerves. In conclusion, clear SEFs were successfully recorded in response to tongue stimulation using a pair of pin electrodes. Contralateral SEF waveforms following the unilateral tongue stimulation are similar regardless of the side of stimulation. This result shows that the response to unilateral side stimulation of the tongue can be compared with that of homologous contralateral stimulation, and, thus, suggests that the stimulation of the presumably healthy side in individual patients can serve as a control in cases of unilateral nerve injury. This laterality index of the tongue obtained in healthy subjects using pin electrodes can serve as a useful parameter for assessing unilateral damage to the tongue. Acknowledgements We thank Dr. M. Matsuhashi (Kyoto Institute of Technology) for his advice regarding data analysis. Part of this study was presented at the First Conference of International Society for the Advancement Clinical Magnetoencephalography and the 37th Meeting of Japanese Society of Clinical Neurophysiology. This study was supported by Grants-in-Aid for Scientific Research (C) 17592074, (C) 19592292, and for Exploratory Research 17659632 from the Ministry of Education, Culture, Sports,

Science and Technology for KY, and Grants-in-Aid for Scientific Research on Priority Areas (18020014) for HF from the Japan Ministry of Education, Culture, Sports, Science and Technology (MEXT). References Altenmu¨ller, E., Cornelius, C.P., Buettner, U.W., 1990. Somatosensory evoked potentials following tongue stimulation in normal subjects and patients with lesions of afferent trigeminal system. Electroencephalogr. Clin. Neurophysiol. 77, 403–415. Disbrow, E.A., Hinkley, L.B.N., Roberts, T.P.L., 2003. Ipsilateral representation of oral structures in human anterior parietal somatosensory cortex and integration of inputs across the midline. J. Comp. Neurol. 467, 487–495. Ha¨ma¨la¨inen, M., Hari, R., Ilmoniemi, R.J., Knuutila, J., Lounasmaa, O.V., 1993. Magnetoencephalography-theory, instrumentation, and applications to noninvasive studies of the working human brain. Rev. Mod. Phys. 65, 413–497. Hashimoto, I., 1988. Trigeminal evoked potentials following brief air puff: enhanced signal-to-noise ratio. Ann. Neurol. 23, 332–338. Hoshiyama, M., Kakigi, R., Koyama, S., Kitamura, Y., Shimojo, M., Watanabe, S., 1996. Somatosensory evoked magnetic fields following stimulation of the lip in humans. Electroencephalogr. Clin. Neurophysiol. 100, 96–104. Ishiko, N., Hanamori, T., Murayama, N., 1980. Spatial distribution of somatosensory responses evoked by tapping the tongue and finger in man. Electroencephalogr. Clin. Neurophysiol. 50, 1–10. Karhu, J., Hari, R., Lu, S.T., Paetau, R., Rif, J., 1991. Cerebral magnetic fields to lingual stimulation. Electroencephalogr. Clin. Neurophysiol. 80, 459–468. Maloney, S.R., Bell, W.L., Shoaf, S.C., Blair, D., Bastings, E.P., Good, D.C., Quinlivan, L., 2000. Measurement of lingual and palatine somatosensory evoked potentials. Clin. Neurophysiol. 111, 291–296. McCarthy, G., Allison, T., Spencer, D.D., 1993. Localization of the face area of human sensorimotor cortex by intracranial recording of somatosensory evoked potentials. J. Neurosurg. 79, 874–884. McDonald, A.R., Roberts, T.P., Rowley, H.A., Pogrel, M.A., 1996. Noninvasive somatosensory monitoring of the injured inferior alveolar nerve using magnetic source imaging. J. Oral Maxillofac. Surg. 54, 1068–1072. Mecklinger, A., Maess, B., Opitz, B., Pfeifer, E., Cheyne, D., Weinberg, H., 1998. A MEG analysis of the P300 in visual discrimination tasks. Electroencephalogr. Clin. Neurophysiol. 108, 45–46. Nagamatsu, K., Nakasato, N., Hatanaka, K., Kanno, A., Iwasaki, M., Yoshimoto, T., 2001. Neuromagnetic localization of N15, the initial cortical response to lip stimulus. Neuroreport 12, 1–5. Nakahara, H., Nakasato, N., Kanno, A., Murayama, S., Hatanaka, K., Itoh, H., Yoshimoto, T., 2004. Somatosensory-evoked fields for gingiva, lip, and tongue. J. Dent. Res. 83, 307–311. Nevalainen, P., Ramstad, R., Isotalo, E., Haapanen, M.L., Lauronen, L., 2006. Trigeminal somatosensory evoked magnetic fields to tactile stimulation. Clin. Neurophysiol. 117, 2007–2015. Nguyen, B.T., Tran, T.D., Hoshiyama, M., Inui, K., Kakigi, R., 2004. Face representation in the human primary somatosensory cortex. Neurosci. Res. 50, 227–232. Nguyen, B.T., Inui, K., Hoshiyama, M., Nakata, H., Kakigi, R., 2005. Face representation in the human secondary somatosensory cortex. Clin. Neurophysiol. 116, 1247–1253. Penfield, W., Rasmussen, T., 1952. The Cerebral Cortex of Man: A Clinical Study of Localization of Function. Macmillan, New York. Sarvas, J., 1987. Basic mathematical and electromagnetic concepts of the biomagnetic inverse problem. Phys. Med. Biol. 32, 11–22. Sakamoto, K., Nakata, H., Kakigi, R., 2008. Somatosensory-evoked magnetic fields following stimulation of the tongue in humans. Clin. Neurophysiol. 119, 1664–1673. Shibasaki, H., Endo, C., Kuroda, Y., Kakigi, R., Oda, K., Komine, S., 1988. Clinical picture of HTLV-I associated myelopathy. J. Neurol. Sci. 87, 15–24. Shirahama, Y., Ohta, K., Takashima, A., Matsushima, E., Okubo, Y., 2004. Magnetic brain activity elicited by visually presented symbols and Japanese characters. Neuroreport 15, 771–775.

H. Maezawa et al. / Neuroscience Research 62 (2008) 131–139 Synek, V.M., 1987. Role of somatosensory evoked-potentials in the diagnosis of peripheral nerve lesions: recent advances. J. Clin. Neurophysiol. 4, 55–73. Tecchio, F., Pasqualetti, P., Pizzella, V., Romani, G., Rossini, P.M., 2000. Morphology of somatosensory evoked fields: inter-hemispheric similarity as a parameter for physiological and pathological neural connectivity. Neurosci. Lett. 287, 203–206.

139

Yamashita, H., Kumamoto, Y., Nakashima, T., Yamamoto, T., Inokuchi, A., Komiyama, S., 1999. Magnetic sensory cortical responses evoked by tactile stimulations of the human face, oral cavity and flap reconstructions of the tongue. Eur. Arch. Otorhinolaryngol. 256 (Suppl. 1), S42–S46. Yoshida, K., Maezawa, H., Nagamine, T., Fukuyama, H., Murakami, K., Iizuka, T., 2006. Somatosensory evoked magnetic fields to air-puff stimulation on the soft palate. Neurosci. Res. 55, 116–122.