Publications
Text queries can be conducted by Author, Title, or Keyword.
Optical intrinsic signal imaging in a rodent seizure model
Source: Neurology 2000 Jul;55(2):312-315.
Author: Chen JW, O'Farrell AM, Toga AW.
PubMed ID: 10908916
Abstract:
ABSTRACT: The authors studied seizure activity with optical intrinsic signal (OIS) imaging in a rat seizure model. OIS, which measures vascular and metabolic effects associated with neuronal activity, showed cortical reflectance changes: in the contralateral study, 0.66 ± 0.55% vs. -0.09 ± 0.06% (seizure vs. baseline, mean ± s.e.m., n = 4); in the ipsilateral study, 1.72 ± 1.12% vs. -0.22 ± 0.33% (n = 5). Furthermore, OIS changes often preceded initial EEG spikes. These observations suggest that cerebral perfusion is well coupled with seizure activity, and may provide sensitive cues for seizure detection. To design better treatments for seizure disorders, it is crucial to understand the mechanisms underlying seizure induction, propagation, cessation and epileptogenesis. Functional neuro-imaging techniques, such as fMRI, PET, SPECT and optical intrinsic signal imaging provide tools to spatially and temporally characterize perfusion related seizure activities. Most functional neuro-imaging techniques depend on detecting changes in perfusion related signals, and assume a tight coupling between perfusion and neuronal activity. OIS represents activity-related changes in the reflectance of cortex, and co-localizes with changes in regional cerebral blood volume, hemoglobin absorption, and light scattering. OIS responses to seizure activity have been observed in the isolated brain preparation of guinea pig,and in human and monkey visual cortex. We investigated seizure activities triggered by topically applied penicillin in a rat model with simultaneous OIS and EEG recording. OIS changes were noted during pre-ictal, ictal and post-ictal periods. To our knowledge, this is the first report in the literature of pre-ictal OIS changes. The pre-ictal change may provide important information of the aura phenomena. In addition, OIS might further our understanding of cerebral perfusion during seizure and be applied to other areas of epilepsy research, such as seizure induction, propagation, focus detection, mirror focus induction and seizure cessation. METHODS: Adult male Sprague Dawley rats were prepared under halothane anesthesia. Two holes were drilled in the right parietal bone, for EEG depth-electrode insertion, and for penicillin application through an injection needle. After surgery, the animal was transferred to the imaging stage; halothane was discontinued and replaced with 1-1.5% enflurane or intermittent intravenous injection of urethane. The animal was allowed to recuperate for at least one hour before data collection. The cortex was epi-illuminated with white light from a voltage-regulated Cuda I-150 source, the images were filtered at 850 nm, and collected with a cooled charge coupled device camera. 64 images were acquired at a time and stored digitally. Data were collected before, during, and for up to several hours after seizure induction. In the ipsilateral study, OIS images were taken from the hemisphere ipsilateral to the penicillin injection site, and vice versa for the contralateral study. The EEG electrode and injection needle were mounted on a stereotaxic manipulator and positioned at a depth of 1 to 2 mm in the cortex. The reference electrode was inserted in the skin at the midline near the surgical incision on the skull. The ground electrode was placed on the abdomen. EEG signal was amplified with a Grass Amplifier, and collected with Labview on a personal computer. Labview was used to synchronize collection of the electrophysiological and optical data in digital format. Penicillin G, a convulsant that induces seizure by inhibiting the neurotransmitter GABA pathway, was dissolved in normal saline and injected at a range of dosages. For each set of 64 optical images, a ratio analysis was performed. A control image was collected before seizure induction, and subtracted from all images. Then, a pixel by pixel ratio was calculated. RESULTS: Electrographic and subsequent clinical seizures were induced by local penicillin injection in the cortex. During clinical seizures, focal seizure was observed first. Seizure commonly presented with pupil dilatation, twitching of the whiskers and face contralateral to the injection hemisphere. When seizure activity escalated to involve both sides, intermittent back arching and twitching of the whiskers and limbs were observed. During seizure, decreased light reflectance was observed in the hemispheres ipsilateral and contralateral to the seizure induction site. The OIS changes in the hemisphere contralateral to the penicillin injection site probably represent spread of seizure activities to this side. The reflectance changes from seizures were reproducible with repeated penicillin injection. The reflectance change can be quantified in the average pixel-by-pixel subtraction ratio: in the contralateral study, 0.66 ± 0.55% vs. -0.09 ± 0.06% (seizure vs. baseline, mean ± s.e.m., n = 4); in the ipsilateral study, 1.72 ± 1.12% vs. -0.22 ± 0.33% (n = 5). An interesting finding is that OIS changes were often noted to precede, up to 1 minute, the initial seizure spikes on EEG. DISCUSSION: In OIS imaging, decreased reflectance correlates with neuronal activation. Millisecond temporal resolution and micron spatial resolution have been attained with the present technique. In this study, OIS changes showed good correlation with seizure activity. Furthermore, early OIS changes preceded the initial EEG spikes. Similar observations were reported in a BOLD fMRI study, in which cortical activation was detected before the onset of clinical or electrographic seizures. These findings suggest that OIS, and other perfusion related imaging techniques, might provide sensitive tools for seizure detection, if adequate temporal resolution can be achieved. The current understanding of seizure is closely related to our understanding of the EEG. It is known that EEG, which detects field potential changes secondary to flip of the cortical dipoles, is limited in its sensitivity. The magnitude of the potential measured by an EEG electrode is determined by 1) the size of dipole layer generating the source; 2) the distance of the electrode from the dipole layer; 3) the solid angle, Omega, subtended by the dipole surface to the electrode; and 4) the dielectric constant of the medium. EEG, using surface electrodes, detects field potential changes in the superficial cerebral cortex, and is limited in detecting subcortical seizures. The depth-electrode EEG can detect activity from a deeper dipole generator, but may suffer from a smaller Omega angle. The OIS changes before the onset of clinical and electrographic seizure might represent physiological changes in the seizure induction phase or the aura phenomenon, an area that is not well defined by EEG. However, the physiological implications of these pre-ictal changes on fMRI and OIS are still unclear, and would warrant future investigations. Seizures might spread along preferred pathways, suggested by the observations in this study. It can be postulated that the seizure waves also transmit along established and preferred neuronal connections, in contrast to the biophysical view of volume conduction in seizure propagation. The cortical neurons are organized in vertical columns, with horizontal collaterals connecting to surrounding neurons. Also, the myelinated axons of cortical neurons extend to subcortical nuclei. It can be imagined that during seizure propagation, in addition to the recruitment of surrounding neurons through slower unmyelinated horizontal collaterals, cortical-subcortical volleys might also underlie fast seizure propagation. The fast propagation pathways are probably important for transmitting seizure activity to distant areas. These early patchy areas might represent the targets of the fast pathways. This observation may imply that the cortical-subcortical volleys underlie the fast seizure propagation, and are followed by expanding horizontal recruitment of surrounding neurons. The surrounding area of the active seizure region showed increased reflectance, which might represent decreased excitatory or increased inhibitory neuronal activity in these regions. This observation suggests that cortical brain tissue might have different seizure thresholds at different regions. In addition, the seizure threshold is probably modulated by neuronal activity. The induction of seizure at a locus might reciprocally induce an inhibitory surround by raising the seizure threshold. OIS imaging is an excellent tool for investigating these questions, either in animal or human subjects during epilepsy surgery.
Source: Neurology 2000 Jul;55(2):312-315.
Author: Chen JW, O'Farrell AM, Toga AW.
PubMed ID: 10908916
Abstract:
ABSTRACT: The authors studied seizure activity with optical intrinsic signal (OIS) imaging in a rat seizure model. OIS, which measures vascular and metabolic effects associated with neuronal activity, showed cortical reflectance changes: in the contralateral study, 0.66 ± 0.55% vs. -0.09 ± 0.06% (seizure vs. baseline, mean ± s.e.m., n = 4); in the ipsilateral study, 1.72 ± 1.12% vs. -0.22 ± 0.33% (n = 5). Furthermore, OIS changes often preceded initial EEG spikes. These observations suggest that cerebral perfusion is well coupled with seizure activity, and may provide sensitive cues for seizure detection. To design better treatments for seizure disorders, it is crucial to understand the mechanisms underlying seizure induction, propagation, cessation and epileptogenesis. Functional neuro-imaging techniques, such as fMRI, PET, SPECT and optical intrinsic signal imaging provide tools to spatially and temporally characterize perfusion related seizure activities. Most functional neuro-imaging techniques depend on detecting changes in perfusion related signals, and assume a tight coupling between perfusion and neuronal activity. OIS represents activity-related changes in the reflectance of cortex, and co-localizes with changes in regional cerebral blood volume, hemoglobin absorption, and light scattering. OIS responses to seizure activity have been observed in the isolated brain preparation of guinea pig,and in human and monkey visual cortex. We investigated seizure activities triggered by topically applied penicillin in a rat model with simultaneous OIS and EEG recording. OIS changes were noted during pre-ictal, ictal and post-ictal periods. To our knowledge, this is the first report in the literature of pre-ictal OIS changes. The pre-ictal change may provide important information of the aura phenomena. In addition, OIS might further our understanding of cerebral perfusion during seizure and be applied to other areas of epilepsy research, such as seizure induction, propagation, focus detection, mirror focus induction and seizure cessation. METHODS: Adult male Sprague Dawley rats were prepared under halothane anesthesia. Two holes were drilled in the right parietal bone, for EEG depth-electrode insertion, and for penicillin application through an injection needle. After surgery, the animal was transferred to the imaging stage; halothane was discontinued and replaced with 1-1.5% enflurane or intermittent intravenous injection of urethane. The animal was allowed to recuperate for at least one hour before data collection. The cortex was epi-illuminated with white light from a voltage-regulated Cuda I-150 source, the images were filtered at 850 nm, and collected with a cooled charge coupled device camera. 64 images were acquired at a time and stored digitally. Data were collected before, during, and for up to several hours after seizure induction. In the ipsilateral study, OIS images were taken from the hemisphere ipsilateral to the penicillin injection site, and vice versa for the contralateral study. The EEG electrode and injection needle were mounted on a stereotaxic manipulator and positioned at a depth of 1 to 2 mm in the cortex. The reference electrode was inserted in the skin at the midline near the surgical incision on the skull. The ground electrode was placed on the abdomen. EEG signal was amplified with a Grass Amplifier, and collected with Labview on a personal computer. Labview was used to synchronize collection of the electrophysiological and optical data in digital format. Penicillin G, a convulsant that induces seizure by inhibiting the neurotransmitter GABA pathway, was dissolved in normal saline and injected at a range of dosages. For each set of 64 optical images, a ratio analysis was performed. A control image was collected before seizure induction, and subtracted from all images. Then, a pixel by pixel ratio was calculated. RESULTS: Electrographic and subsequent clinical seizures were induced by local penicillin injection in the cortex. During clinical seizures, focal seizure was observed first. Seizure commonly presented with pupil dilatation, twitching of the whiskers and face contralateral to the injection hemisphere. When seizure activity escalated to involve both sides, intermittent back arching and twitching of the whiskers and limbs were observed. During seizure, decreased light reflectance was observed in the hemispheres ipsilateral and contralateral to the seizure induction site. The OIS changes in the hemisphere contralateral to the penicillin injection site probably represent spread of seizure activities to this side. The reflectance changes from seizures were reproducible with repeated penicillin injection. The reflectance change can be quantified in the average pixel-by-pixel subtraction ratio: in the contralateral study, 0.66 ± 0.55% vs. -0.09 ± 0.06% (seizure vs. baseline, mean ± s.e.m., n = 4); in the ipsilateral study, 1.72 ± 1.12% vs. -0.22 ± 0.33% (n = 5). An interesting finding is that OIS changes were often noted to precede, up to 1 minute, the initial seizure spikes on EEG. DISCUSSION: In OIS imaging, decreased reflectance correlates with neuronal activation. Millisecond temporal resolution and micron spatial resolution have been attained with the present technique. In this study, OIS changes showed good correlation with seizure activity. Furthermore, early OIS changes preceded the initial EEG spikes. Similar observations were reported in a BOLD fMRI study, in which cortical activation was detected before the onset of clinical or electrographic seizures. These findings suggest that OIS, and other perfusion related imaging techniques, might provide sensitive tools for seizure detection, if adequate temporal resolution can be achieved. The current understanding of seizure is closely related to our understanding of the EEG. It is known that EEG, which detects field potential changes secondary to flip of the cortical dipoles, is limited in its sensitivity. The magnitude of the potential measured by an EEG electrode is determined by 1) the size of dipole layer generating the source; 2) the distance of the electrode from the dipole layer; 3) the solid angle, Omega, subtended by the dipole surface to the electrode; and 4) the dielectric constant of the medium. EEG, using surface electrodes, detects field potential changes in the superficial cerebral cortex, and is limited in detecting subcortical seizures. The depth-electrode EEG can detect activity from a deeper dipole generator, but may suffer from a smaller Omega angle. The OIS changes before the onset of clinical and electrographic seizure might represent physiological changes in the seizure induction phase or the aura phenomenon, an area that is not well defined by EEG. However, the physiological implications of these pre-ictal changes on fMRI and OIS are still unclear, and would warrant future investigations. Seizures might spread along preferred pathways, suggested by the observations in this study. It can be postulated that the seizure waves also transmit along established and preferred neuronal connections, in contrast to the biophysical view of volume conduction in seizure propagation. The cortical neurons are organized in vertical columns, with horizontal collaterals connecting to surrounding neurons. Also, the myelinated axons of cortical neurons extend to subcortical nuclei. It can be imagined that during seizure propagation, in addition to the recruitment of surrounding neurons through slower unmyelinated horizontal collaterals, cortical-subcortical volleys might also underlie fast seizure propagation. The fast propagation pathways are probably important for transmitting seizure activity to distant areas. These early patchy areas might represent the targets of the fast pathways. This observation may imply that the cortical-subcortical volleys underlie the fast seizure propagation, and are followed by expanding horizontal recruitment of surrounding neurons. The surrounding area of the active seizure region showed increased reflectance, which might represent decreased excitatory or increased inhibitory neuronal activity in these regions. This observation suggests that cortical brain tissue might have different seizure thresholds at different regions. In addition, the seizure threshold is probably modulated by neuronal activity. The induction of seizure at a locus might reciprocally induce an inhibitory surround by raising the seizure threshold. OIS imaging is an excellent tool for investigating these questions, either in animal or human subjects during epilepsy surgery.