- Email: [email protected]
Localization of the optic tract by using subcortical visual evoked potentials (VEPs) in cats
Clinical Neurophysiology 110 (1999) 92–96
Localization of the optic tract by using subcortical visual evoked potentials (VEPs) in cats Kenji Sugiyama*, Tetsuo Yokoyama, Takamichi Yamamoto, Kenichi Uemura Department of Neurosurgery, Hamamatsu University School of Medicine, Handa-cho, Hamamatsu-city, 431-3192, Japan; accepted for publication: 1 July 1998
Abstract Objectives: This study was aimed to clarify the relationship between the amplitudes of visual evoked near field potentials and distance from the optic tract, and to determine the adequate filter settings to record these potentials from the optic tract separately from the far field potentials. Methods: The visual evoked near field potentials from the optic tract were consecutively recorded through intracerebral electrodes in 6 cats’ brains. Different filter settings were tried and the amplitudes of visual evoked potentials (VEPs) were compared with the distance from the optic tract. Results: The filter settings of 100 Hz to 1 kHz were the best to obtain only the near field potentials separately from the far field potentials. Histological sections revealed that the potentials of the surface of the optic tract showed sudden increase of amplitude, above the 50% of the maximum VEPs amplitudes. Conclusions: The optic tract can be identified using these methods. These results can be applied to localize the optic tract during such an operative procedure as postero-ventral pallidotomy. 1999 Elsevier Science Ireland Ltd. All rights reserved Keywords: Visual evoked potentials; Optic tract; Near field potentials; Pallidotomy; Band-pass filter; Potentials
1. Introduction The method to localize the deep seated nerve fibers such as the optic tract within the brain are not well clarified. Visual evoked potentials (VEPs) have been tried to identify the optic tract during the posteroventral pallidotomy for patients with Parkinson’s disease to prevent a postoperative complication of quadrant hemianopsia, because Laitinen et al., reported that the major complication of the posteroventral pallidotomy was the injuring of the optic tract and that six out of 42 patients (14%) had the lower quadrant hemianopsia after surgery (Laitinen et al., 1992a,b; Yokoyama et al., 1997). However, the adequate factors including bandpass filters or the relationship between the actual amplitude of VEPs and the distance from the optic tract are not known.
Such potentials have been reported to follow the Coulomb’s law or the dipole equation that the amplitudes and the square of the distance have a converse relationship (Nunez, 1981). It is not known whether these equations can be applied to the VEPs from the optic tract because the impedance of the surrounding structures around the optic tract, such as cerebrospinal fluid, nuclei and passing fibers, are not even. Also, the N45 or P50 of the VEPs which can be seen at 40–60 ms after visual stimulation using the band-pass filter setting of 1–100 Hz are known as far field potentials. They are diffusely distributed not only over the skull but also within the wide area of the brain (Schroeder et al., 1992). We studied the subcortical VEPs in cats to determine the adequate recording parameters for obtaining only the near field potentials from the optic tract and to clarify the relationship between the amplitude of the consecutive VEPs and the distance from the optic tract.
* Corresponding author. Tel.: +81 53 4352283; fax: +81 53 4352282; e–mail: [email protected]
1388-2457/99/$ - see front matter 1999 Elsevier Science Ireland Ltd. All rights reserved PI I S0168-559 7(98)00050-1
CLINPH 97168
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
2. Methods 2.1. Preparation of cats Experiments were performed on 6 cats weighing 3.0–4.5 kg. Cats were anesthetized by intraperitoneal injection of the pentobarbital (3.3 mg/kg body wt i.p). Anesthesia was maintained with periodic injections of pentobarbital (1.5 mg/h) and was determined to be adequate by the absence of corneal and paw withdrawal reflexes and spontaneous movements. Rectal temperature was monitored and maintained near 36.6°C with a circulating water heating pad. Cats were positioned in a stereotactic apparatus equipped with ear and teeth bars. A midline scalp incision was made and the skin and underlying muscle were retracted. Through craniectomy an electrode was introduce into the brain. Electrode penetrations were made within the following stereotactic coordinates in relation to interaural zero: anteriorposterior, +11.0 to +13.0; medial-lateral, +4.0 to +7.0; horizontal, +3.0 to +7.0 (Jasper, 1954). 2.2. Stimulation and recording VEPs were recorded with the epoxy-coated tungsten electrode having a 1 mm diameter and 1 mm bared tip. The reference and the ground Ag-AgCl electrodes were placed on the skull separately. The cats’ eyes were stimulated 1/s with a red flash of light from a LED (light emitted diode) of a goggle type photic stimulator at a distance of 2 cm from the cat’s eye. The luminance of LEDs were kept at 230 cd/ m2 and the chromaticity was x = 0.712 y = 0.284 in chromaticity diagram. The duration of each flash light was 10 ms. An artificial pupil was not used. At the beginning of the experiment, the cats’ pupils were dilated using 1.0% atropine sulfate, the position of area centralis was determined and the stimulation was centered on the area centralis. Saline drops were applied to the tested eye between each VEP trial. The responses up to 250 ms were amplified, filtered (1–100 Hz or 100 Hz–1 kHz) and averaged 50 times using a signal processor of 4 channels (Compact four, Nichole, USA). VEPs were consecutively recorded from 10 mm above the surface of the optic tract in 0.5 mm steps. Other filter settings for obtaining the VEPs were also tried in two cats. The VEPs of 4 different filter settings were simultaneously obtained in this experiments. High frequency filters (HFFs) were changed from 500 Hz to 10 kHz and low frequency filters (LFFs) were changed from 1 to 300 Hz (Fig. 3, A–H). 2.3. Histological procedures In selected penetrations, the locations of the recording sites or the top of the electrode tract were marked with electrolytic lesions (100 mA, 10–20 s). Cats were overdosed with pentobarbital at the end of the experiment. The brains were removed and fixed in 10% formalin, and placed in
93
paraffin. Serial sections were made with a 5 mm thick slice every 100 mm. Tissue sections containing the electrode track and lesions were mounted on the slides and stained with cresyl violet. The electrode tract and lesions were reconstructed with the use of microprojector, and the recording sites were determined with histological coordinates of the electrode track and lesions.
3. Results Fig. 1 shows an example of the consecutive VEPs using the two different filter settings at each depth. In the VEPs with a filter setting of 1–100 Hz (Fig. 1, A, right) large negative far field potentials with peak latencies of 75 and 100 ms were continuously observed, but the near field potentials of the optic tract were not distinct. On the other hand, the VEPs with a filter setting of 100–1000 Hz (Fig. 1, A, middle) only the near field potentials of the optic tract were clearly observed. These near field potentials had tri- to multi-phasic waves with the onset latency of 24 ms after the optical stimulation and lasted for 30 ms. VEPs were recorded from 1.0 mm above the surface of the optic tract (Fig. 1, A, left and middle, black arrow head), and their amplitudes suddenly increased when the electrode tip reached the surface of the optic tract (Fig. 1, A, left and middle, white arrow head). When the amplitudes of the VEPs were converted to the percentage of the maximum amplitude responses (Fig. 1, A, middle, double arrow heads), they suddenly increased to 64.3% at the surface of the optic tract from 29.2% at 0.5 mm above it (Fig. 1, B). The first 6 consecutive percent amplitude responses from maximum to the background was plotted in Fig. 1C. The mathematical function that best fits to the percent amplitude responses was a logarithmic curve (R 2 = 0.951) (Fig. 1C), which indicated that the relationship between the amplitude of the VEPs and the distance from the optic tract fit to the Coulomb’s law. The mean and standard deviation of R 2 value to the logarithmic curve in 6 cats was 0.95 ± 0.010. The mean and standard deviation of the percent amplitude responses of VEPs in 6 cats were 75.9 ± 12.1% on the surface of the optic tract and 34.1 ± 4.5% at 0.5 mm above it (Fig. 2). These results show that the optic tract can be identified by the sudden increase of the VEP amplitude above the 50% of the maximum VEPs amplitudes. Fig. 3 shows the consecutive VEPs using 7 other different filter settings compared with the filter settings of 100 Hz–1 kHz. When the LFF was lowered to less than 30 Hz, large negative potentials with peak latencies of 75 and 100 ms were continuously observed, and the near field potentials of the optic tract were not distinct. When the HFF was increased up to 10 000 Hz, multiple artificial noises were mixed with VEPs. The filter setting of 100–1000 Hz was the best to see only the near field potentials of the optic tract (Fig. 3D, black arrow heads).
94
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
4. Discussion Our studies showed that a proper filter setting is essential to obtaining near field potentials so as to localize the surface
of the optic tract. The relationship between the amplitude of the near field potentials and its distance from the optic tract could be also used to exactly identify the surface of the optic tract.
Fig. 1. VEPs recorded at successive depths during vertical penetration through the basal ganglia into the optic tract. (A): left, photomicrographs of a coronal section stained with cresyl violet. The lesioning sites were marked by asterisks and the each recording site was pointed out by an arrow. Middle, VEPs recorded by using the low frequency filter 100 Hz and high frequency filter 115 Hz. Right, VEPs recorded by using the low frequency filter of 1 Hz and high frequency filter of 100 kHz. (B): The percent amplitude response at each depth of the near field VEPs shown in (A) middle. The amplitudes were converted to the percentage of the maximum amplitude responses of the VEPs. As shown in (A) left, 0 mm indicated the lesioning site within the optic tract and the surface of the optic tract was located at 1 mm above it, where the amplitudes of the VEPs suddenly increased from 29.2% to 64.3%. (C): The mathematical function that best fits to the percent amplitude responses was a logarithmic curve (R 2 = 0.951).
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
Fig. 2. The percent amplitude histogram of the VEPs in each distance from the optic tract in 6 cats. The percent amplitudes, which were 34.06 ± 4.53% at 0.5 mm above the surface of the optic tract, became 75.89 ± 12.10% on the surface of the optic tract.
The factors affecting the near field potentials from nerve fibers such as the optic tract include (1) the strength of the electrical source, (2) the distance from the electrical source, (3) the impedance of electrodes and the volume conductor surrounding the electrical source and (4) the band-pass filter settings (Nunez, 1981). The potentials have been well known to follow the Coulomb’s law or dipole equation;
95
that is the amplitude of the potentials are inversely proportional to the square of the distance from the electrical source if the strength of the electrical source, the impedance of the electrode, and the volume conductor were constant (Nunez, 1981). Therefore, the coefficient between the amplitude and the distance from the electrical generator (i.e. how abruptly or gradually the potentials decay with following Coulomb’s law when the electrode tip is moved away from the electrical source) is determined by the other 3 factors in each situation. In our experiment the electrical generator is the optic tract and the volume conductor is a cat brain. The strength of the electric generator and the impedance of the volume conductor are a constant value. The impedance of the electrode is fixed to a low level of 10 kQ. Our experiment identified the remaining two factors for the purpose of identifying the optic tract using VEPs. The first is that a proper filter setting is essential to discriminate the near field potentials from the far field potentials, because only the near field potentials from the optic tract are useful for the purpose of localizing the optic tract. The band-pass filter setting in almost all of the commercially available VEP programs are fixed with a low frequency filter to 1 Hz and a high frequency filter to 100 Hz (Sollazzo, 1985; Cedzich et al., 1987; Harding et al., 1990, 1996). As shown in our study, only the far field potentials of VEPs are recorded with this filter setting, thus unable to localize the nerve fibers. The
Fig. 3. The successive VEPs with different filter settings. The filter setting of 100 Hz–1 kHz was the best to see only the near field potentials of the optic tract (D: black arrow heads). Scale bar: vertical bar, 2.5 mV; horizontal bar, 35 ms.
96
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
low frequency filter of 100 Hz and the high frequency filter of 1 kHz were the best to discriminate the near field potentials from the far field potentials. The other important factor is the coefficient mentioned above. Whereas the VEPs in the cat’s brain are already visualized 1 mm away from the optic tract, the amplitude of the VEPs are less than 50%. The optic tract can be located with these relationships. The amplitudes of the consecutive VEPs suddenly rise above the 50% of the maximum when the electrode tip touches the optic tract. Our results can be applied to localize the deep seated nerve fibers such as the optic tract during the brain operation. We already applied this method for 21 patients to identify the optic tract during postero-ventral pallidotomy. The optic tracts were successfully identified using these methods, and the stereotactic lesions by thermocoagulation could be made preventing the optic tract injury. These details were described elsewhere (Yokoyama et al., 1997). Recently Tobimatsu et al., reported the VEPs during stereotactic pallidotomy (Tobimatsu et al., 1997). They used the single axis bipolar electrode a 30 mm apart and the band-pass filter settings at 0.5–200 Hz. They showed 3 consecutive near field responses from optic tract and reported that the maximal amplitude responses were observed at the ventralmost of globus pallidus interna (GPi) and the anza lenticularis. The major advantage of their method is that the large far field potentials were not seen even using the band-pass filter settings of 0.5–200 Hz. However, we think that this method has some problems with the accuracy in localizing the optic tract. The ventral border of GPi, where they obtained the maximal optic response, is ordinally 2–4 mm above the optic tract (Schaltenbrand and Wahren, 1977: Lozano et al., 1996). This might be caused due to the fact both of the bipolar electrodes were simultaneously activated. As shown in our data, the maximal responses using our methods were clearly seen when the electrode was on or in the optic tract.
The proper setting of the filter enables us to obtain the near field potential and localize the potential source of the nerve fibers.
References Cedzich, C., Schramm, J. and Fahlbusch, R. Are flash-evoked visual potentials useful for intraoperative monitoring of visual pathway function? Neurosurgery, 1987, 21: 709–715. Harding, G., Bland, J. and Smith, V. Visual evoked potential monitoring of optic nerve function during surgery. J. Neurol. Neurosurg. Psychiatry, 1990, 53: 890–895. Harding, G., Odom, J., Spileers, W. and Spekreijse, H. Standard for visual evoked potentials 1995. Vis. Res., 1996, 36: 3567–3572. Jasper, H.H. and Ajmone-Marsan, C. A Stereotaxic Atlas of the Diencephalon of the Cat. National Research Council of Canada, Ottawa, 1954. Laitinen, L.V., Bergenheim, A.T. and Hariz, M.I. Ventroposterolateral pallidotomy can abolish all parkinsonian symptoms, Stereotact. Funct. Neurosurg., 58 (1992a) 14–21. Laitinen, L.V., Bergenheim, A.T. and Hariz, M.I. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease, J. Neurosurg., 76 (1992b) 53–61. Lozano, A., Hutchison, W., Jiss, Z., Tasker, R., Davis, K. and Dostrovsky, J. Methods for microelectrode-guided posteroventral pallidotomy. J. Neurosurg., 1996, 84: 194–202. Nunez, P. Electrical Fields of the Brain. Oxford University Press, New York, 1981. Schaltenbrand, G. and Wahren, W. Atlas for Stereotaxy of the Human Brain. Georg Thieme Publishers, Stuttgart, 1977. Schroeder, C., Tenke, C. and Givre, S. Subcortical contributions to the surface-recorded flash-VEP in the awake macaque. Electroenceph. clin. Neurophysiol., 1992, 84: 219–231. Sollazzo, D. Influence of l-dopa/Carbidopa on pattern reversal VEP: behavioral differences in primary and secondary Parkinsonism. Electroenceph. clin. Neurophysiol., 1985, 1: 236–242. Tobimatsu, A., Shima F., Ishida K. and Kato,I. Visual evoked potentials in the vicinity of the optic tract during stereotactic pallidotomy. Electroenceph. clin. Neurophysiol., (1997) 274–279. Yokoyama, T., Sugiyama, K., Nishizawa, S., Ryu, H., Hinokuma, K., Yamamoto, S., Endoh, M., Ohta, S., Yokota, N. and Uemura, K. Visual evoked potential guidance for posteroventral pallidotomy in Parkinson’s disease. Neurol. Med.-Chir (Tokyo), 1997, 37: 257–264.
Localization of the optic tract by using subcortical visual evoked potentials (VEPs) in cats Kenji Sugiyama*, Tetsuo Yokoyama, Takamichi Yamamoto, Kenichi Uemura Department of Neurosurgery, Hamamatsu University School of Medicine, Handa-cho, Hamamatsu-city, 431-3192, Japan; accepted for publication: 1 July 1998
Abstract Objectives: This study was aimed to clarify the relationship between the amplitudes of visual evoked near field potentials and distance from the optic tract, and to determine the adequate filter settings to record these potentials from the optic tract separately from the far field potentials. Methods: The visual evoked near field potentials from the optic tract were consecutively recorded through intracerebral electrodes in 6 cats’ brains. Different filter settings were tried and the amplitudes of visual evoked potentials (VEPs) were compared with the distance from the optic tract. Results: The filter settings of 100 Hz to 1 kHz were the best to obtain only the near field potentials separately from the far field potentials. Histological sections revealed that the potentials of the surface of the optic tract showed sudden increase of amplitude, above the 50% of the maximum VEPs amplitudes. Conclusions: The optic tract can be identified using these methods. These results can be applied to localize the optic tract during such an operative procedure as postero-ventral pallidotomy. 1999 Elsevier Science Ireland Ltd. All rights reserved Keywords: Visual evoked potentials; Optic tract; Near field potentials; Pallidotomy; Band-pass filter; Potentials
1. Introduction The method to localize the deep seated nerve fibers such as the optic tract within the brain are not well clarified. Visual evoked potentials (VEPs) have been tried to identify the optic tract during the posteroventral pallidotomy for patients with Parkinson’s disease to prevent a postoperative complication of quadrant hemianopsia, because Laitinen et al., reported that the major complication of the posteroventral pallidotomy was the injuring of the optic tract and that six out of 42 patients (14%) had the lower quadrant hemianopsia after surgery (Laitinen et al., 1992a,b; Yokoyama et al., 1997). However, the adequate factors including bandpass filters or the relationship between the actual amplitude of VEPs and the distance from the optic tract are not known.
Such potentials have been reported to follow the Coulomb’s law or the dipole equation that the amplitudes and the square of the distance have a converse relationship (Nunez, 1981). It is not known whether these equations can be applied to the VEPs from the optic tract because the impedance of the surrounding structures around the optic tract, such as cerebrospinal fluid, nuclei and passing fibers, are not even. Also, the N45 or P50 of the VEPs which can be seen at 40–60 ms after visual stimulation using the band-pass filter setting of 1–100 Hz are known as far field potentials. They are diffusely distributed not only over the skull but also within the wide area of the brain (Schroeder et al., 1992). We studied the subcortical VEPs in cats to determine the adequate recording parameters for obtaining only the near field potentials from the optic tract and to clarify the relationship between the amplitude of the consecutive VEPs and the distance from the optic tract.
* Corresponding author. Tel.: +81 53 4352283; fax: +81 53 4352282; e–mail: [email protected]
1388-2457/99/$ - see front matter 1999 Elsevier Science Ireland Ltd. All rights reserved PI I S0168-559 7(98)00050-1
CLINPH 97168
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
2. Methods 2.1. Preparation of cats Experiments were performed on 6 cats weighing 3.0–4.5 kg. Cats were anesthetized by intraperitoneal injection of the pentobarbital (3.3 mg/kg body wt i.p). Anesthesia was maintained with periodic injections of pentobarbital (1.5 mg/h) and was determined to be adequate by the absence of corneal and paw withdrawal reflexes and spontaneous movements. Rectal temperature was monitored and maintained near 36.6°C with a circulating water heating pad. Cats were positioned in a stereotactic apparatus equipped with ear and teeth bars. A midline scalp incision was made and the skin and underlying muscle were retracted. Through craniectomy an electrode was introduce into the brain. Electrode penetrations were made within the following stereotactic coordinates in relation to interaural zero: anteriorposterior, +11.0 to +13.0; medial-lateral, +4.0 to +7.0; horizontal, +3.0 to +7.0 (Jasper, 1954). 2.2. Stimulation and recording VEPs were recorded with the epoxy-coated tungsten electrode having a 1 mm diameter and 1 mm bared tip. The reference and the ground Ag-AgCl electrodes were placed on the skull separately. The cats’ eyes were stimulated 1/s with a red flash of light from a LED (light emitted diode) of a goggle type photic stimulator at a distance of 2 cm from the cat’s eye. The luminance of LEDs were kept at 230 cd/ m2 and the chromaticity was x = 0.712 y = 0.284 in chromaticity diagram. The duration of each flash light was 10 ms. An artificial pupil was not used. At the beginning of the experiment, the cats’ pupils were dilated using 1.0% atropine sulfate, the position of area centralis was determined and the stimulation was centered on the area centralis. Saline drops were applied to the tested eye between each VEP trial. The responses up to 250 ms were amplified, filtered (1–100 Hz or 100 Hz–1 kHz) and averaged 50 times using a signal processor of 4 channels (Compact four, Nichole, USA). VEPs were consecutively recorded from 10 mm above the surface of the optic tract in 0.5 mm steps. Other filter settings for obtaining the VEPs were also tried in two cats. The VEPs of 4 different filter settings were simultaneously obtained in this experiments. High frequency filters (HFFs) were changed from 500 Hz to 10 kHz and low frequency filters (LFFs) were changed from 1 to 300 Hz (Fig. 3, A–H). 2.3. Histological procedures In selected penetrations, the locations of the recording sites or the top of the electrode tract were marked with electrolytic lesions (100 mA, 10–20 s). Cats were overdosed with pentobarbital at the end of the experiment. The brains were removed and fixed in 10% formalin, and placed in
93
paraffin. Serial sections were made with a 5 mm thick slice every 100 mm. Tissue sections containing the electrode track and lesions were mounted on the slides and stained with cresyl violet. The electrode tract and lesions were reconstructed with the use of microprojector, and the recording sites were determined with histological coordinates of the electrode track and lesions.
3. Results Fig. 1 shows an example of the consecutive VEPs using the two different filter settings at each depth. In the VEPs with a filter setting of 1–100 Hz (Fig. 1, A, right) large negative far field potentials with peak latencies of 75 and 100 ms were continuously observed, but the near field potentials of the optic tract were not distinct. On the other hand, the VEPs with a filter setting of 100–1000 Hz (Fig. 1, A, middle) only the near field potentials of the optic tract were clearly observed. These near field potentials had tri- to multi-phasic waves with the onset latency of 24 ms after the optical stimulation and lasted for 30 ms. VEPs were recorded from 1.0 mm above the surface of the optic tract (Fig. 1, A, left and middle, black arrow head), and their amplitudes suddenly increased when the electrode tip reached the surface of the optic tract (Fig. 1, A, left and middle, white arrow head). When the amplitudes of the VEPs were converted to the percentage of the maximum amplitude responses (Fig. 1, A, middle, double arrow heads), they suddenly increased to 64.3% at the surface of the optic tract from 29.2% at 0.5 mm above it (Fig. 1, B). The first 6 consecutive percent amplitude responses from maximum to the background was plotted in Fig. 1C. The mathematical function that best fits to the percent amplitude responses was a logarithmic curve (R 2 = 0.951) (Fig. 1C), which indicated that the relationship between the amplitude of the VEPs and the distance from the optic tract fit to the Coulomb’s law. The mean and standard deviation of R 2 value to the logarithmic curve in 6 cats was 0.95 ± 0.010. The mean and standard deviation of the percent amplitude responses of VEPs in 6 cats were 75.9 ± 12.1% on the surface of the optic tract and 34.1 ± 4.5% at 0.5 mm above it (Fig. 2). These results show that the optic tract can be identified by the sudden increase of the VEP amplitude above the 50% of the maximum VEPs amplitudes. Fig. 3 shows the consecutive VEPs using 7 other different filter settings compared with the filter settings of 100 Hz–1 kHz. When the LFF was lowered to less than 30 Hz, large negative potentials with peak latencies of 75 and 100 ms were continuously observed, and the near field potentials of the optic tract were not distinct. When the HFF was increased up to 10 000 Hz, multiple artificial noises were mixed with VEPs. The filter setting of 100–1000 Hz was the best to see only the near field potentials of the optic tract (Fig. 3D, black arrow heads).
94
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
4. Discussion Our studies showed that a proper filter setting is essential to obtaining near field potentials so as to localize the surface
of the optic tract. The relationship between the amplitude of the near field potentials and its distance from the optic tract could be also used to exactly identify the surface of the optic tract.
Fig. 1. VEPs recorded at successive depths during vertical penetration through the basal ganglia into the optic tract. (A): left, photomicrographs of a coronal section stained with cresyl violet. The lesioning sites were marked by asterisks and the each recording site was pointed out by an arrow. Middle, VEPs recorded by using the low frequency filter 100 Hz and high frequency filter 115 Hz. Right, VEPs recorded by using the low frequency filter of 1 Hz and high frequency filter of 100 kHz. (B): The percent amplitude response at each depth of the near field VEPs shown in (A) middle. The amplitudes were converted to the percentage of the maximum amplitude responses of the VEPs. As shown in (A) left, 0 mm indicated the lesioning site within the optic tract and the surface of the optic tract was located at 1 mm above it, where the amplitudes of the VEPs suddenly increased from 29.2% to 64.3%. (C): The mathematical function that best fits to the percent amplitude responses was a logarithmic curve (R 2 = 0.951).
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
Fig. 2. The percent amplitude histogram of the VEPs in each distance from the optic tract in 6 cats. The percent amplitudes, which were 34.06 ± 4.53% at 0.5 mm above the surface of the optic tract, became 75.89 ± 12.10% on the surface of the optic tract.
The factors affecting the near field potentials from nerve fibers such as the optic tract include (1) the strength of the electrical source, (2) the distance from the electrical source, (3) the impedance of electrodes and the volume conductor surrounding the electrical source and (4) the band-pass filter settings (Nunez, 1981). The potentials have been well known to follow the Coulomb’s law or dipole equation;
95
that is the amplitude of the potentials are inversely proportional to the square of the distance from the electrical source if the strength of the electrical source, the impedance of the electrode, and the volume conductor were constant (Nunez, 1981). Therefore, the coefficient between the amplitude and the distance from the electrical generator (i.e. how abruptly or gradually the potentials decay with following Coulomb’s law when the electrode tip is moved away from the electrical source) is determined by the other 3 factors in each situation. In our experiment the electrical generator is the optic tract and the volume conductor is a cat brain. The strength of the electric generator and the impedance of the volume conductor are a constant value. The impedance of the electrode is fixed to a low level of 10 kQ. Our experiment identified the remaining two factors for the purpose of identifying the optic tract using VEPs. The first is that a proper filter setting is essential to discriminate the near field potentials from the far field potentials, because only the near field potentials from the optic tract are useful for the purpose of localizing the optic tract. The band-pass filter setting in almost all of the commercially available VEP programs are fixed with a low frequency filter to 1 Hz and a high frequency filter to 100 Hz (Sollazzo, 1985; Cedzich et al., 1987; Harding et al., 1990, 1996). As shown in our study, only the far field potentials of VEPs are recorded with this filter setting, thus unable to localize the nerve fibers. The
Fig. 3. The successive VEPs with different filter settings. The filter setting of 100 Hz–1 kHz was the best to see only the near field potentials of the optic tract (D: black arrow heads). Scale bar: vertical bar, 2.5 mV; horizontal bar, 35 ms.
96
K. Sugiyama et al. / Clinical Neurophysiology 110 (1999) 92–96
low frequency filter of 100 Hz and the high frequency filter of 1 kHz were the best to discriminate the near field potentials from the far field potentials. The other important factor is the coefficient mentioned above. Whereas the VEPs in the cat’s brain are already visualized 1 mm away from the optic tract, the amplitude of the VEPs are less than 50%. The optic tract can be located with these relationships. The amplitudes of the consecutive VEPs suddenly rise above the 50% of the maximum when the electrode tip touches the optic tract. Our results can be applied to localize the deep seated nerve fibers such as the optic tract during the brain operation. We already applied this method for 21 patients to identify the optic tract during postero-ventral pallidotomy. The optic tracts were successfully identified using these methods, and the stereotactic lesions by thermocoagulation could be made preventing the optic tract injury. These details were described elsewhere (Yokoyama et al., 1997). Recently Tobimatsu et al., reported the VEPs during stereotactic pallidotomy (Tobimatsu et al., 1997). They used the single axis bipolar electrode a 30 mm apart and the band-pass filter settings at 0.5–200 Hz. They showed 3 consecutive near field responses from optic tract and reported that the maximal amplitude responses were observed at the ventralmost of globus pallidus interna (GPi) and the anza lenticularis. The major advantage of their method is that the large far field potentials were not seen even using the band-pass filter settings of 0.5–200 Hz. However, we think that this method has some problems with the accuracy in localizing the optic tract. The ventral border of GPi, where they obtained the maximal optic response, is ordinally 2–4 mm above the optic tract (Schaltenbrand and Wahren, 1977: Lozano et al., 1996). This might be caused due to the fact both of the bipolar electrodes were simultaneously activated. As shown in our data, the maximal responses using our methods were clearly seen when the electrode was on or in the optic tract.
The proper setting of the filter enables us to obtain the near field potential and localize the potential source of the nerve fibers.
References Cedzich, C., Schramm, J. and Fahlbusch, R. Are flash-evoked visual potentials useful for intraoperative monitoring of visual pathway function? Neurosurgery, 1987, 21: 709–715. Harding, G., Bland, J. and Smith, V. Visual evoked potential monitoring of optic nerve function during surgery. J. Neurol. Neurosurg. Psychiatry, 1990, 53: 890–895. Harding, G., Odom, J., Spileers, W. and Spekreijse, H. Standard for visual evoked potentials 1995. Vis. Res., 1996, 36: 3567–3572. Jasper, H.H. and Ajmone-Marsan, C. A Stereotaxic Atlas of the Diencephalon of the Cat. National Research Council of Canada, Ottawa, 1954. Laitinen, L.V., Bergenheim, A.T. and Hariz, M.I. Ventroposterolateral pallidotomy can abolish all parkinsonian symptoms, Stereotact. Funct. Neurosurg., 58 (1992a) 14–21. Laitinen, L.V., Bergenheim, A.T. and Hariz, M.I. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease, J. Neurosurg., 76 (1992b) 53–61. Lozano, A., Hutchison, W., Jiss, Z., Tasker, R., Davis, K. and Dostrovsky, J. Methods for microelectrode-guided posteroventral pallidotomy. J. Neurosurg., 1996, 84: 194–202. Nunez, P. Electrical Fields of the Brain. Oxford University Press, New York, 1981. Schaltenbrand, G. and Wahren, W. Atlas for Stereotaxy of the Human Brain. Georg Thieme Publishers, Stuttgart, 1977. Schroeder, C., Tenke, C. and Givre, S. Subcortical contributions to the surface-recorded flash-VEP in the awake macaque. Electroenceph. clin. Neurophysiol., 1992, 84: 219–231. Sollazzo, D. Influence of l-dopa/Carbidopa on pattern reversal VEP: behavioral differences in primary and secondary Parkinsonism. Electroenceph. clin. Neurophysiol., 1985, 1: 236–242. Tobimatsu, A., Shima F., Ishida K. and Kato,I. Visual evoked potentials in the vicinity of the optic tract during stereotactic pallidotomy. Electroenceph. clin. Neurophysiol., (1997) 274–279. Yokoyama, T., Sugiyama, K., Nishizawa, S., Ryu, H., Hinokuma, K., Yamamoto, S., Endoh, M., Ohta, S., Yokota, N. and Uemura, K. Visual evoked potential guidance for posteroventral pallidotomy in Parkinson’s disease. Neurol. Med.-Chir (Tokyo), 1997, 37: 257–264.