Search Results

You are looking at 1 - 10 of 20 items for

  • Author or Editor: Kyousuke Kamada x
  • Refine by Access: all x
Clear All Modify Search
Restricted access

Geert-Jan Rutten and Nick Ramsey

Dissociated language functions are largely invalidated by standard techniques such as the amobarbital test and cortical stimulation. Language studies in which magnetoencephalography (MEG) and functional magnetic resonance (fMR) imaging are used to record data while the patient performs lexicosemantic tasks have enabled researchers to perform independent brain mapping for temporal and frontal language functions (MEG is used for temporal and fMR imaging for frontal functions). In this case report, the authors describe a right-handed patient in whom a right-sided insular glioma was diagnosed. The patient had a right-lateralized receptive language area, but expressive language function was identified in the left hemisphere on fMR imaging– and MEG-based mapping. Examinations were performed in 20 right-handed patients with low-grade gliomas (control group) for careful comparison with and interpretation of this patient's results. In these tests, all patients were asked to generate verbs related to acoustically presented nouns (verb generation) for fMR imaging, and to categorize as abstract or concrete a set of visually presented words consisting of three Japanese letters for fMR imaging and MEG.

The most prominent display of fMR imaging activation by the verb-generation task was observed in the left inferior and middle frontal gyri in all participants, including the patient presented here. Estimated dipoles identified with the abstract/concrete categorization task were concentrated in the superior temporal and supra-marginal gyri in the left hemisphere in all control patients. In this patient, however, the right superior temporal region demonstrated significantly stronger activations on MEG and fMR imaging with the abstract/concrete categorization task. Suspected dissociation of the language functions was successfully mapped with these two imaging modalities and was validated by the modified amobarbital test and the postoperative neurological status. The authors describe detailed functional profiles obtained in this patient and review the cases of four previously described patients in whom dissociated language functions were found.

Restricted access

Takahiro Ota, Kensuke Kawai, Kyousuke Kamada, Taichi Kin, and Nobuhito Saito

Object

Intraoperative monitoring of visual evoked potentials (VEPs) has been regarded as having limited significance for the preservation of visual function during neurosurgical procedures, mainly due to its poor spatial resolution and signal-to-noise ratio. The authors evaluated the usefulness of cortically recorded VEPs, instead of the usual scalp VEPs, as intraoperative monitoring focusing on the posterior visual pathway.

Methods

In 17 consecutive patients who underwent microsurgical procedures for lesions near the posterior visual pathway, cortical responses were recorded using 1-Hz flashing light-emitting diodes and subdural strip electrodes after induction of general anesthesia with sevoflurane or propofol. The detectability and waveform of the initial response, stability, and changes during microsurgical manipulations were analyzed in association with the position of electrodes and postoperative changes in visual function.

Results

Initial VEPs were detected in 82% of all patients. The VEPs were detected in 94% of patients without total hemianopia in whom electrodes were placed sufficiently near the occipital pole; in these cases the recordings were not significantly affected by anesthesia. The detectability rates of the negative peak before 100 msec (N1), positive peak ~ 100 msec (P100), and negative peak after 100 msec (N2) were 36, 50, and 100%, respectively. The mean latencies and amplitudes of N1, P100, and N2 were 90.0 ± 15.9 msec and 61.0 ± 64.0 μV, 103.9 ± 13.5 msec and 34.3 ± 38.6 μV, and 125.7 ± 12.2 msec and 44.9 ± 48.9 μV, respectively, showing great variability. In 11 patients, the initial waveforms of VEP remained stable during microsurgical procedures, and the visual status did not change postoperatively, while it disappeared in 2 patients who presented with postoperative hemianopia.

Conclusions

Direct recording from the visual cortices under general anesthesia achieved satisfactory detectability of the visual response to a light-emitting diode flashing light. Although the initial waveforms varied greatly among patients, they were stable during microsurgical procedures, and the changes were consistent with postoperative visual function. Intraoperative cortical VEP monitoring is a potentially useful procedure to monitor the functional integrity of the posterior visual pathway.

Restricted access

Kyousuke Kamada, Yutaka Sawamura, Fumiya Takeuchi, Kiyohiro Houkin, Hideaki Kawaguchi, Yoshinobu Iwasaki, and Shinya Kuriki

✓ Letter-perception centers are not held in as high regard as motor- and language-related cortices during planning of neurosurgical procedures, and there have been no reports suggesting cortical reorganization of reading ability. The authors describe a patient with a left mesial temporal glioma in whom two letter-perception centers (the anterior portion of the left superior temporal gyrus and the left fusiform gyrus) were successfully localized before surgery by performing magnetoencephalography (MEG) during reading tasks. Control MEG examinations of 15 healthy volunteers were also performed to assist in a careful interpretation of patient results. Although a radical resection of the mesial temporal glioma, which involved the left fusiform gyrus, caused severe dyslexia, the patient's impaired reading skills improved gradually during a 1-year postoperative period. In the meantime, the spared left superior temporal gyrus displayed an overshot recovery of MEG responses. During the postoperative period there was no obvious recovery in MEG signals and no compensatory activity in the contralateral fusiform gyrus. This case demonstrates that lexicosemantic centers involved in the reading process can be noninvasively localized using MEG and that the results obtained are highly reliable for surgical planning. The results of the repeated MEG reflected sequentially the patient's recovery from dyslexia. This is the first report in which MEG studies have been shown to predict preoperatively the risk of dyslexia and demonstrate its serial physiological recovery.

Full access

Kyousuke Kamada, Hiroshi Ogawa, Christoph Kapeller, Robert Prueckl, Satoru Hiroshima, Yukie Tamura, Fumiya Takeuchi, and Christoph Guger

OBJECTIVE

Recent neuroimaging studies suggest that intractable epilepsy involves pathological functional networks as well as strong epileptogenic foci. Combining cortico-cortical evoked potential (CCEP) recording and tractography is a useful strategy for mapping functional connectivity in normal and pathological networks. In this study, the authors sought to demonstrate the efficacy of preoperative combined CCEP recording, high gamma activity (HGA) mapping, and tractography for surgical planning, and of intraoperative CCEP measures for confirmation of selective pathological network disconnection.

METHODS

The authors treated 4 cases of intractable epilepsy. Diffusion tensor imaging–based tractography data were acquired before the first surgery for subdural grid implantation. HGA and CCEP investigations were done after the first surgery, before the second surgery was performed to resect epileptogenic foci, with continuous CCEP monitoring during resection.

RESULTS

All 4 patients in this report had measurable pathological CCEPs. The mean negative peak-1 latency of normal CCEPs related to language functions was 22.2 ± 3.5 msec, whereas pathological CCEP latencies varied between 18.1 and 22.4 msec. Pathological CCEPs diminished after complete disconnection in all cases. At last follow-up, all of the patients were in long-term postoperative seizure-free status, although 1 patient still suffered from visual aura every other month.

CONCLUSIONS

Combined CCEP measurement, HGA mapping, and tractography greatly facilitated targeted disconnection of pathological networks in this study. Although CCEP recording requires technical expertise, it allows for assessment of pathological network involvement in intractable epilepsy and may improve seizure outcome.

Full access

Yukie Tamura, Hiroshi Ogawa, Christoph Kapeller, Robert Prueckl, Fumiya Takeuchi, Ryogo Anei, Anthony Ritaccio, Christoph Guger, and Kyousuke Kamada

OBJECTIVE

Electrocortical stimulation (ECS) is the gold standard for functional brain mapping; however, precise functional mapping is still difficult in patients with language deficits. High gamma activity (HGA) between 80 and 140 Hz on electrocorticography is assumed to reflect localized cortical processing, whereas the cortico-cortical evoked potential (CCEP) can reflect bidirectional responses evoked by monophasic pulse stimuli to the language cortices when there is no patient cooperation. The authors propose the use of “passive” mapping by combining HGA mapping and CCEP recording without active tasks during conscious resections of brain tumors.

METHODS

Five patients, each with an intraaxial tumor in their dominant hemisphere, underwent conscious resection of their lesion with passive mapping. The authors performed functional localization for the receptive language area, using real-time HGA mapping, by listening passively to linguistic sounds. Furthermore, single electrical pulses were delivered to the identified receptive temporal language area to detect CCEPs in the frontal lobe. All mapping results were validated by ECS, and the sensitivity and specificity were evaluated.

RESULTS

Linguistic HGA mapping quickly identified the language area in the temporal lobe. Electrical stimulation by linguistic HGA mapping to the identified temporal receptive language area evoked CCEPs on the frontal lobe. The combination of linguistic HGA and frontal CCEPs needed no patient cooperation or effort. In this small case series, the sensitivity and specificity were 93.8% and 89%, respectively.

CONCLUSIONS

The described technique allows for simple and quick functional brain mapping with higher sensitivity and specificity than ECS mapping. The authors believe that this could improve the reliability of functional brain mapping and facilitate rational and objective operations. Passive mapping also sheds light on the underlying physiological mechanisms of language in the human brain.

Restricted access

Kyousuke Kamada, Fumiya Takeuchi, Shinya Kuriki, Tomoki Todo, Akio Morita, and Yutaka Sawamura

✓Dissociated language functions are largely invalidated by standard techniques such as the amobarbital test and cortical stimulation. Language studies in which magnetoencephalography (MEG) and functional magnetic resonance (fMR) imaging are used to record data while the patient performs lexicosemantic tasks have enabled researchers to perform independent brain mapping for temporal and frontal language functions (MEG is used for temporal and fMR imaging for frontal functions). In this case report, the authors describe a right-handed patient in whom a right-sided insular glioma was diagnosed. The patient had a right-lateralized receptive language area, but expressive language function was identified in the left hemisphere on fMR imaging–and MEG-based mapping. Examinations were performed in 20 right-handed patients with low-grade gliomas (control group) for careful comparison with and interpretation of this patient’s results. In these tests, all patients were asked to generate verbs related to acoustically presented nouns (verb generation) for fMR imaging, and to categorize as abstract or concrete a set of visually presented words consisting of three Japanese letters for fMR imaging and MEG.

The most prominent display of fMR imaging activation by the verb-generation task was observed in the left inferior and middle frontal gyri in all participants, including the patient presented here. Estimated dipoles identified with the abstract/concrete categorization task were concentrated in the superior temporal and supramarginal gyri in the left hemisphere in all control patients. In this patient, however, the right superior temporal region demonstrated significantly stronger activations on MEG and fMR imaging with the abstract/concrete categorization task. Suspected dissociation of the language functions was successfully mapped with these two imaging modalities and was validated by the modified amobarbital test and the postoperative neurological status. The authors describe detailed functional profiles obtained in this patient and review the cases of four previously described patients in whom dissociated language functions were found.

Restricted access

Keisuke Maruyama, Kyousuke Kamada, Masahiro Shin, Daisuke Itoh, Shigeki Aoki, Yoshitaka Masutani, Masao Tago, and Takaaki Kirino

Object. In the radiosurgical treatment of critically located lesions, the effort to minimize the risk of complication is essential. In this study the integration of diffusion-tensor (DT) imaging—based tractography was clinically applied to treatment planning for gamma knife surgery (GKS).

Methods. Seven patients with cerebral arteriovenous malformations located adjacent to the corticospinal tract (CST) underwent this technique. Data provided by DT imaging were acquired before the frame was affixed to the patient's head and the CST of the DT tractography was created using our original software. Stereotactic three-dimensional imaging studies were obtained after frame fixation and then coregistered with the data from DT tractography. After image fusion of the two studies, the combined images were transported to a GKS treatment-planning workstation. The spatial relationship between the dose distribution and the CST was clearly demonstrated within the 2 hours it took to complete the entire imaging process. The univariate logistic regression analysis of transient or permanent motor complications revealed a significant independent correlation with the volume of the CST that received 25 Gy or more and with a maximum dose to the CST (p < 0.05).

Conclusions. The integration of DT tractography into the GKS treatment planning was highly useful in confirming the dose to the CST during treatment planning.

Restricted access

Kyousuke Kamada, Tomoki Todo, Yoshitaka Masutani, Shigeki Aoki, Kenji Ino, Tetsuya Takano, Takaaki Kirino, Nobutaka Kawahara, and Akio Morita

Object. The aim of this study was better preoperative planning and direct application to intraoperative procedures through accurate coregistration of diffusion-tensor (DT) imaging—based tractography results and anatomical three-dimensional magnetic resonance images and subsequent importation of the combined images to a neuronavigation system (functional neuronavigation).

Methods. Six patients with brain lesions adjacent to the corticospinal tract (CST) were studied. During surgery, direct fiber stimulation was used to evoke motor responses to confirm the accuracy of CST depicted on functional neuronavigation. In three patients, stimulation of the supposed CST elicited the expected motor evoked potentials. In the other three, stimulation at the resection borders more than 1 cm away from the supposed CST showed no motor response. All patients underwent appropriate tumor resection with preservation of the CST.

Conclusions. Integration of the DT imaging—based tractography information into a traditional neuronavigation system demonstrated spatial relationships between lesions and the CST, allowing for the avoidance of tract injury during lesion resection. Direct fiber stimulation was used for real-time reliable white matter mapping, which served to adjust for any discrepancy between the neuronavigation system data and potentially shifted positions of the brain structures. The combination of these techniques enabled the authors to identify accurate positions of the CST during surgery and to accomplish optimal tumor resections.

Restricted access

Kyousuke Kamada, Kiyohiro Houkin, Yoshinobu Iwasaki, Fumiya Takeuchi, Shinya Kuriki, Kenji Mitsumori, and Yutaka Sawamura

Object. To identify the primary motor area (PMA) quickly and correctly, the authors used magnetic resonance (MR) axonography, including anisotropic diffusion-weighted (DW) MR imaging and three-dimensional anisotropic contrast (3DAC) imaging, which was performed to visualize the corticospinal tract mainly originating from the PMA.

Methods. All studies were obtained in 10 normal volunteers and in 17 patients with brain tumors affecting the central motor system. Data sets of anisotropic DW imaging and anatomical and functional (f)MR imaging were acquired while the participants executed simple hand movements. Offline processing of 3DAC MR axonography images was subsequently done to extract only the anisotropic components of the tract fibers. Somatosensory evoked fields (SSEFs) and intraoperative cortical somatosensory evoked potentials (SSEPs) were recorded after electrical stimulation of the median nerve.

Conclusions. In normal volunteers, anisotropic DW imaging, 3DAC imaging, fMR imaging, and magnetoencephalography consistently localized the PMA in both hemispheres. In contrast, fMR imaging and SSEFs failed to identify the PMA in seven and one of the 17 patients, respectively, because of cortical dysfunctions due to brain tumor. The anisotropic DW imaging data acquired within 30 seconds with no patient tasks successfully identified the PMA in 12 patients, and failed in five patients because of the lesions involving the frontal lobe. The anisotropic axonal components were distinctly visualized on 3DAC images and indicated the PMA location, which was confirmed on intraoperative SSEPs in all 17 affected hemispheres. Swift and noninvasive PMA identification by rapid scanning with MR axonography is a promising method for routine clinical use and is especially beneficial for patients who have severe cortical dysfunction in the PMA.

Restricted access

Kyousuke Kamada, Tomoki Todo, Yoshitaka Masutani, Shigeki Aoki, Kenji Ino, R.T., Akio Morita, and Nobuhito Saito

Object

There is continuous interest in the monitoring of language function during tumor resection around the fron-totemporal regions of the dominant hemisphere. The aim of this study was to visualize language-related subcortical connections, such as the arcuate fasciculus (AF) by diffusion tensor (DT) imaging–based tractography.

Methods

Twenty-two patients with brain lesions adjacent to the AF in the frontotemporal regions of the dominant hemisphere were studied. The AF tractography was accomplished by placing initiation and termination sites (seed and target points) in the frontal and temporal regions, which were functionally identified by using functional magnetic resonance (fMR) imaging in conjunction with a verb generation task and magnetoencephalography (MEG) in conjunction with a reading task. The combination of fMR imaging and MEG data clearly demonstrated the hemispheric dominance of language functions, which was confirmed by an intracranial amobarbital test (Wada procedure). In all 22 patients, the authors were able to consistently visualize the AF by DT imaging–based tractography, using the functionally identified seed and target points and a fractional anisotropy value of 0.16. In two of 22 cases investigated, the functional information, including the results of AF tractography, fMR imaging, and MEG, was imported to a neuronavigation system and was validated by bipolar electric stimulation of the cortical and subcortical areas during awake surgery. The cortical stimulation to the gyrus that included the area of activation identified in fMR imaging with the language task evoked speech arrest, while the subcortical stimulation close to the AF reproducibly caused paranomia without speech arrest. Postoperative AF tractography showed that the distances between the stimulus points and the AF were within 6 mm.

Conclusions

The combination of these techniques facilitated accurate identification of the location of the AF and verification of the language fibers.