Article Text
Abstract
Electrophysiological studies in primates indicate that the eye fields of the cerebral hemispheres control gaze in three-dimensional space, and contain neurons that encode both conjugate (versive) and vergence eye movements. Two patients with epilepsy who exhibited disconjugate contraversive horizontal eye movements are described, one during electrical stimulation of the frontal eye fields and the other during focal seizures. We postulate that these eye movements resulted from combined contralateral version and vergence, and suggest that human cortical eye fields also govern visual search in a three-dimensional world, shifting the point of fixation between targets lying in different directions and at different depths.
Statistics from Altmetric.com
During visual search, the fovea of the eye, wherein resides the highest density of photoreceptors, is pointed at features of interest.1 Since the visual environment is three-dimensional, it is necessary to use a combination of conjugate (versive) and vergence eye movements to move the fovea of each eye between targets lying in different directions and at different depths. Studies in macaque indicate that both the lateral intraparietal area,2 corresponding to the parietal eye fields (PEF) in humans,1 and frontal eye fields (FEF)3 contain populations of neurons that encode three-dimensional gaze shifts. In the case of the FEF, stimulation studies have identified discrete areas concerned with vergence,4 as well as saccades and smooth pursuit.5
Based on the findings from lesion, imaging and stimulation studies, human FEF, which are thought to lie at the caudal end of the middle frontal gyrus,6–9 are known to control contralateral versive eye movements, including saccades and smooth pursuit. We describe two patients with epilepsy who exhibited disconjugate contraversive horizontal eye deviation during electrical stimulation of the FEF or as a feature of focal seizures. We postulate that these eye movements represent combined contralateral version and vergence, and are consistent with the notion of governance of gaze in three-dimensional space by the cortical eye fields.
CASE REPORTS
Case No 1
A 25-year-old man with a 20 year history of refractory partial seizures was admitted for evaluation for epilepsy surgery. His seizures were preceded by an epigastric aura and characterised by unresponsiveness with dystonic posturing of the right arm. Post-ictally, he had expressive aphasia and, on one occasion, a right-sided Todd’s paresis. Secondary generalised motor seizures were rare. The seizures occurred up to 4 times per day, despite treatment with multiple anticonvulsants. His neurological examination was normal, apart from a left Horner’s syndrome. Brain MRI showed mild volume loss and increased signal in the left hippocampus, with increased signal in the adjacent neocortex. The frontal lobes and brainstem appeared normal. Video-EEG monitoring showed independent interictal left and right frontal sharp waves, and seizure onset in the left fronto-temporal region (electrodes FT7, SP1, F9, or T7).
Subdural electrode arrays were placed over the left fronto-temporal cortex and electrical stimulation was performed to define eloquent cortex and guide the margins of surgical resection. Brain MRI (MP RAGE sequence with voxel dimensions of 0.9×0.9×2.0 mm) was performed 1 day later so that the location of each electrode could be determined. The centre of artefact created by each electrode was identified and a 2 mm sphere representing the electrode tip was placed on a reconstructed three-dimensional image of the patient’s brain. Electrical stimulation consisted of 5–10 s trains of 50 Hz bipolar square wave pulses of 0.3–0.4 ms duration. Electrical activity was monitored in the stimulated and surrounding electrodes before and immediately after stimulation. Stimulation was commenced at 1 mA and increased by 0.5–1 mA steps until the patient had symptoms or signs, discharges were elicited or a maximum of 15.5 mA was reached. Stimulation of electrode 6, which was positioned over the caudal end of the middle frontal gyrus (fig 1), resulted in disconjugate contraversive (rightward) horizontal eye deviation (video 1; see supplementary data online) and inability to make voluntary saccades. The amount of left eye adduction was always greater than the amount of right eye abduction. At lower stimulation currents, only a small amount of adduction of the left eye was noted, and he described difficulty focusing while attempting to read. As the stimulation current was increased, the eye deviation became more pronounced. The ability to pursue a moving object was retained. No discharges were noted in any of the adjacent electrodes following stimulation, even with the higher currents, and no eye movements were seen when adjacent electrodes in the array were stimulated, suggesting that the eye movements were elicited by activation of the cortex underlying or in the immediate vicinity of electrode 6.
The electrode arrays were removed and he underwent a left temporal pole lobectomy and amygdalo-hippocampectomy. Neuro-ophthalmological examination 3 days following the surgery revealed no new abnormalities, and he has remained seizure-free following the surgery.
Case No 2
A 29-year-old man with a 2 year history of partial seizures and prior illicit drug use was admitted for evaluation for epilepsy surgery. His seizures were preceded by an aura of left hemibody paraesthesiae, followed by decreased hearing on the right, and then inability to speak. He would prepare himself for a seizure by putting away his spectacles and lying flat in bed, and he would then develop an altered level of consciousness, head deviation (usually to the right, but occasionally to the left) and generalised clonic movements. The seizures lasted for up to 4 min, were followed by confusion and occurred 2–3 times per week despite treatment with multiple anticonvulsants. His neurological examination, including ocular motility, was normal. Brain MRI revealed increased signal in the left hippocampal tail. The frontal lobes and brainstem appeared normal. Non-invasive monitoring suggested that the epileptogenic zone was either in the left frontal or left temporal region. Evaluation with depth electrodes suggested a left Sylvian seizure origin but the exact location of the epileptogenic zone could not be determined.
He subsequently underwent video-EEG monitoring after subdural electrode arrays were placed over the left frontal, temporal and parietal cortices. The ictal focus was found to be in the left supero-posterior Sylvian bank. Ictal EEG changes progressed to involve adjacent regions of the hemisphere, including the frontal regions, prior to becoming generalised. The progression of EEG changes was highly stereotyped, as was the seizure semiology. During the period of invasive video-EEG monitoring, he developed disconjugate contraversive (rightward) horizontal eye deviation during one of his seizures. During this seizure, the left eye adducted more than the right eye abducted (video 2; see supplementary data online). The EEG changes during this seizure involved all of the subdural electrodes covering the left frontal lobe. However, the plate was centred over the Sylvian fissure and, consequently, its upper border did not cover the FEF or PEF.
DISCUSSION
We have described two patients who developed disconjugate contraversive horizontal eye movements in response to electrical stimulation of the frontal cortex or during focal seizures. In the first patient, the region of stimulation corresponded to the presumed location of the human FEF (fig 1).6–9 In the second, the EEG changes during the seizure most probably involved the FEF region, as well as other parts of the ipsilateral hemispheric cortex. Consequently, we cannot exclude the possibility that the eye movements arose due to stimulation of a cortical area remote from the FEF. To the best of our knowledge, however, disconjugate gaze deviations such as those seen in our patients have not been previously described as a feature of cortical stimulation or seizures in humans although in a classic human cortical stimulation study convergence was observed with FEF stimulation “on one occasion”.10 In our patients, the amount of ipsilateral eye adduction was greater than the amount of contralateral eye abduction and, as neither had a baseline ocular motor deficit or imaging abnormality to suggest a brainstem lesion, we postulate that these disconjugate eye movements arose because of combined contralateral version and vergence.
A role for cortical eye fields in the generation of version and vergence eye movements has been suggested by the findings of electrophysiological studies in macaque. Thus regions of FEF concerned with saccades, pursuit and vergence have been identified.1 3–5 Furthermore, electrical stimulation in identified vergence regions produces short latency increases in both vergence and lens accommodation.4 While studies of human FEF indicate that they have an important role in the control of voluntary saccadic and smooth pursuit eye movements,6 7 our observations suggest that human FEF also control vergence. It is likely, however, that other cortical areas also participate in the control of vergence eye movements. The findings of transcranial magnetic stimulation studies in humans suggest that the PEF and dorsolateral prefrontal cortex play an important role,11–13 and it is probable that the supplementary eye fields are also involved in the planning of the versive and vergence eye movements required to scan the three-dimensional visual environment.14 In conclusion, we suggest that the role of the human cortical eye fields be expanded to include control of all voluntary eye movements, including vergence.
REFERENCES
Footnotes
See Editorial Commentary, p 585
Funding: Supported by NIH grant EY06717, the Office of Research and Development, Medical Research Service, Department of Veterans Affairs, and the Evenor Armington Fund.
Competing interests: None.
Ethics approval: Ethics approval was obtained.
Linked Articles
- Editorial commentary