| Sign In to gain access to subscriptions and/or personal tools. |
Somatosensory-evoked Fields for Gingiva, Lip, and Tongue
1 Dental Division, Miyagi National Hospital, Miyagi, Japan; Correspondence: * corresponding author, nak{at}kohnan-sendai.or.jp
To localize the oral primary somatosensory cortex, we measured somatosensory-evoked fields for the lip, gingiva, and tongue in six healthy subjects. The latency of the first peak of the posterior-oriented current in the contralateral hemisphere was 50.9 ± 8.3 ms for the gingiva, significantly shorter than those for the lip and tongue peaks. The equivalent current dipole was localized on the central sulcus. The gingival dipole was localized significantly inferior to the lip dipole but not different from the tongue dipole. The moment of the gingival dipole was significantly smaller than that of the lip dipole but not different from that of the tongue dipole. Differences in the above parameters were negligible between the left and right, anterior and posterior, and upper and lower locations within the same organ, except that the dipole location for the anterior upper tongue was significantly inferior to that for the lower tongue.
Key Words: magnetoencephalography • somatosensory evoked potential • gingiva • lip • tongue
The primary somatosensory (SI) cortex is known to incorporate a somatotopic functional organization called the "homunculus," a representation of brain areas related to all parts of the human body (Penfield and Boldrey, 1937). The SI cortex representing the oral organs is relatively large in comparison with its size in proportion to the rest of the human body, suggesting its great importance for oral function in humans. Sensory information from the oral organs is integrated into the bilateral SI cortices, located in the posterior bank of the central sulcus of the cerebrum. Functional brain-imaging techniques—such as positron emission tomography (PET) and functional magnetic resonance (fMR) imaging—have identified the sensorimotor areas corresponding to the oral organs (Petersen et al., 1988; Grafton et al., 1991; Wildgruber et al., 1996; Corfield et al., 1999; Lotze et al., 2000; Fox et al., 2001). However, PET or fMR imaging indicates that the activated areas include multiple and extended parts of the brain. In addition, the limited temporal resolution of these techniques cannot separate the pure SI area from the secondary somatosensory cortex, higher sensory cortices, or even the motor cortex. The somatotopy of the oral SI cortex has been demonstrated by direct cortical stimulation during awake craniectomy surgery (Penfield and Boldrey, 1937; Picard and Olivier, 1983). For example, the face, upper lip, lower lip, tongue tip, and back of the tongue areas are located from superior to inferior in the SI cortex (Penfield and Boldrey, 1937). However, oral SI somatotopy is variable in individuals (Penfield and Boldrey, 1937; Picard and Olivier, 1983). SI somatotopy can also be studied by intracranial recording of somatosensory-evoked potentials (SEPs) (Baumgartner et al., 1992; McCarthy et al., 1993). SEP recording showed general agreement in the medial-to-lateral (superior-to-inferior) representation in the somatosensory cortex of the hand, chin, upper lip, lower lip, tongue, and palate (McCarthy et al., 1993). However, the oral somatotopy was variable and overlapping, and polarity inversion of potentials across the sulcus was a less reliable criterion for trigeminal SEPs. For example, the palatal SI cortex was spatially separated or distributed, as well as variable among subjects. Moreover, these intracranial methods can be used only during brain surgery, and not for normal subjects or patients with oral diseases. Magnetoencephalography (MEG) is a non-invasive method whereby brain functions can be localized. MEG measures the weak magnetic fields produced by neuronal currents. MEG has excellent temporal resolution (milliseconds) similar to that achieved with electroencephalography (EEG). Moreover, the spatial resolution of MEG is higher than that of scalp EEG, because little distortion is caused by inhomogeneous head conductivity. Somatosensory-evoked magnetic fields (SEFs) for hand and foot stimuli are now well-established and are applied clinically (Sutherling et al., 1988; Baumgartner et al., 1991; Gallen et al., 1993; Kakigi, 1994; Nakasato et al., 1996; Ohtomo et al., 1996; Kawamura et al., 1996; Nakasato and Yoshimoto, 2000; Kumabe et al., 2000; Iwasaki et al., 2001). SEFs for oral organs have been studied only for limited stimulation points, such as the tongue (Karhu et al., 1991; Hari et al., 1993; Yamashita et al., 1999), lip (Hari et al., 1993; Mogilner et al., 1994; Hoshiyama et al., 1996, 1997; Schnitzler et al., 1999; Yamashita et al., 1999; Nagamatsu et al., 2001), and buccal mucosa (Yamashita et al., 1999). The SI somatotopy remains little known (Hari et al., 1993; Mogilner et al., 1994; Hoshiyama et al., 1996; Yamashita et al., 1999). The present study measured SEFs for the lip, gingiva, and tongue. Stimulation points for each organ were systematically selected to include left vs. right, upper vs. lower, and anterior vs. posterior parts of each organ. The neural source was estimated at the peak latency for each response according to a single equivalent current dipole (ECD) model, which assumes that the measured magnetic field can be most appropriately explained by a single current at one point in the brain. Latency, ECD moments, and ECD location of the SEFs were statistically compared between the stimulation points for evaluation of the functional organization of the oral SI cortex in time and space.
Six right-handed healthy subjects, two males and four females, aged 20–47 yrs (mean, 26.8 yrs), participated in this study. Written informed consent, reviewed and approved by the ethical committee of Kohnan Hospital, was obtained from the subjects after explanation of the experimental procedure.
We induced SEFs by electrical stimulation of the lip, gingiva, and tongue at a total of 24 points using our homemade clip electrode (Fig. 1A
The SEF signals were measured with a whole-head MEG system (Fig. 1C  ) consisting of an array of 102 identical sensor elements distributed over the whole head except the lower face and immersed in liquid helium (Vector View; Neuromag Ltd., Helsinki, Finland). Each sensor element was comprised of two orthogonal planar-type first-order gradiometers with a 13-mm baseline. The MEG measurements were performed in a magnetically shielded room. The subjects lay supine, and we determined the exact location of the head, with respect to the sensor elements, with 3 fiduciary markers by measuring magnetic signals produced by indicator currents passed through coils placed at known locations on the scalp. The positions of the coils with respect to the outer landmarks of the head were obtained by a three-dimensional digitizer. Three-dimensional magnetic resonance (MR) images (Signa Horizon LX ver 8.2; GE Medical Systems, Milwaukee, WI, USA) were obtained in all subjects, with 3 fiduciary markers formed by small oil-containing capsules placed at the same locations on the scalp. The MEG signals were recorded from 50 ms before to 300 ms after the trigger point, filtered from 0.1 to 300 Hz, and digitized at about 1000 Hz. The resulting data were averaged based on recordings from 200 stimuli for the lip, gingiva, tongue, and median nerve. Bilateral SEF waveforms were recorded, and the contralateral response at the peak latency of 30–100 ms with posterior orientation was used for further analysis. A single dipole model for the SEF data in the contralateral hemisphere to the stimulus side was used for evaluation of the source moment and position. We also used sequential ECD analysis, from –5 ms to +5 ms around the peak at 1-ms intervals, to verify stable ECD estimation within 1-mm distance for each session. The position of the ECD was superimposed on the MR images for correlation with the central sulcus. The relative angles of ECDs were calculated between the oral SEFs and median nerve N20m in the coronal plane of the best-fitting sphere for each subject. This study accepted only ECDs accounting for > 90% of the field variance and with confidence volume < 1 cm3. The single-ECD assumption is well-known to account for peak signals in previous MEG studies of SI response (Karhu et al., 1991; Gallen et al., 1993; Hari et al., 1993; Kakigi, 1994; Kawamura et al., 1996; Nakasato et al., 1996; Ohtomo et al., 1996; Iwasaki et al., 2001; Nagamatsu et al., 2001). Therefore, we used the single-ECD model for the hemispheric data contralateral to the stimulus side. Two statistical analyses were performed. The latency, ECD moments, and relative ECD angles of all subjects were compared among the anterior-upper, anterior-lower, posterior-upper, and posterior-lower stimulus points for each organ. The dipole parameters were averaged over 8 stimulus points, and the overall averages of dipole parameters were compared among the lip, gingiva, and tongue. The paired t test was used, and data were considered significant at P < 0.01.
SEF responses were observed bilaterally for all stimulus points (n = 24) in all subjects (n = 6). A typical example is shown in Fig. 2  . All ECD dipoles were estimated to be located on the central sulcus, based on the MR images of individual subjects.
The latency and ECD parameters were compared among the 3 organs, as well as between stimulus points within each organ (Fig. 3 ). Significant differences were found as follows: The latency of the gingival SEFs was shorter than the latencies for the tongue and lip; the dipole moment of the gingival SEFs was smaller than that of the lip; the ECD position of the lip SEFs was superior to those of the tongue and gingiva; and the ECD position for the anterior upper stimulus of the tongue was lower than those of the anterior lower and posterior lower stimuli.
This study of SEFs for gingival stimuli corresponds to a previous study of cortical SEP for palatal gum (upper alveolus) stimuli (McCarthy et al., 1993). The latency of the first posterior-oriented current in the gingival SEFs was 50.9 ± 8.3 ms (mean ± standard deviation) in the present study, corresponding to the latency of the prominent SEP peak of around 40 to 60 ms (McCarthy et al., 1993). We have named this peak response "cP55m", according to the nomenclature system in previous reports of SEFs (Kakigi, 1994; Hoshiyama et al., 1996; Kawamura et al., 1996; Iwasaki et al., 2001; Nagamatsu et al., 2001). "cP55m" is defined as follows: c, contralateral cerebral hemisphere to the stimulus side; P, positive current toward the parietal area; 55, intermediate value of the peak latency in ms; and m, magnetic counterpart of the electrical response. The ECD of the gingival cP55m was localized on the central sulcus, significantly inferior to the lip cP55m but not significantly different from the tongue cP55m in the present study. A similar trend was observed for the SEP distribution (McCarthy et al., 1993). Comparison of the current orientation of SEFs and SEPs is difficult, because SEFs represent tangential currents to the scalp, whereas SEPs mainly represent radial currents. The first SI response of the lip SEF, cN15m, has an anterior-oriented current with a peak latency around 15 ms (Nagamatsu et al., 2001). The cN15m has small amplitude, so previous studies of trigeminal SEFs have focused on the later components with posterior currents and peak latencies of 20 to 60 ms (Karhu et al., 1991; Hari et al., 1993; Mogilner et al., 1994; Hoshiyama et al., 1996). The present study also used cP55m for comparison, since the cN15m was not observed among the gingival SEFs. The latency of the gingival cP55m was significantly shorter than those of the lip or tongue cP55m. Latency differences in these trigeminal SEFs can be explained by the peripheral conduction time, including transmission time, at the peripheral mechanoreceptors; the central conduction time from the pons to the SI cortex; and the central processing time within the cortex, i.e., the inter-peak latency between cN15m and cP55m. The peripheral conduction time in lip SEFs was estimated at about 2.1 ms, which cannot explain the shorter latency (about 6 ms) of the gingival cP55m (Nagamatsu et al., 2001). Central conduction time from the pons to the SI cortex must be the same for all trigeminal systems. Therefore, we believe that the central processing time must be shorter for gingival SEFs than for lip or tongue SEFs. In addition, our subjects, including the two dentists among the authors (HN and SM), felt a tooth-tap sensation during stimulus of the gingiva, suggesting that both the mucosa and the periodontium were stimulated. Somatosensory information from the gingiva or periodontium may undergo rapid processing within the SI cortex in humans. The dipole moment of the gingival cP55m was significantly smaller than that of the tongue cP55m. The small moment of the gingival SEFs may indicate that only a small number of neurons is excited. However, the cortical representation was unexpectedly large for the palatal gum stimulus compared with the lip or tongue stimulus in humans (McCarthy et al., 1993). This observation and our present results suggest a lower density of excited neurons over the widely distributed gingival SI cortex. All cP55m dipoles for the gingiva, lip, and tongue were localized on the central sulcus. The lip dipole was significantly superior to the gingival and tongue dipoles, whereas the gingival and tongue dipoles were very close. Differences in ECD positions were negligible between anterior and posterior stimuli, and between upper and lower stimuli within the same organ, except that the ECD for the anterior upper tongue stimulus was significantly lower than the ECD for the lower tongue stimulus. The somatotopy of the oral SI cortex is generally agreed to locate the lip SI superior and medial to the tongue SI cortex (Penfield and Boldrey, 1937; Hari et al., 1993; McCarthy et al., 1993; Yamashita et al., 1999). However, the findings of the somatotopy differ for the other oral organs as well as the different parts of each organ. The well-known "homunculus" map of the SI cortex shows that the upper lip area is drawn superior to the lower lip area (Penfield and Boldrey, 1937). However, the relative functional anatomy between the upper and lower lip areas was not described. Moreover, the original mapping of the cortical stimulation points (Penfield and Boldrey, 1937) shows variable and wide distribution of the oral sensory responses. Previous SEF studies (Mogilner et al., 1994; Hoshiyama et al., 1996, 1997; Yamashita et al., 1999) have suggested that the upper lip SI is superior to the lower lip SI. However, these studies provided no statistical evidence. Our present results, together with those from a previous study (Hari et al., 1993), indicate that there is no fixed organization of the oral SI somatotopy except for general agreement for the lip and tongue areas. The present study shows that MEG provides a non-invasive method for the investigation of function in human intra-oral structures, including the gingiva, with high spatial and temporal resolutions.
This study was partly supported by a grant from the Ministry of Health, Labour and Welfare, Japan. We thank Dr. Norio Taira and the staff of Kohnan Hospital Ryogo Center for their continuous support of this work. A preliminary report was presented at the 81st General Session of the International Association for Dental Research, Göteborg, Sweden, June 25–28, 2003. Received for publication August 6, 2003. Revision received January 22, 2004. Accepted for publication January 22, 2004.
Journal of Dental Research, Vol. 83, No. 4,
307-311 (2004) This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
