Ictal Lid Movements: Blinks and Lid Saccades

ABSTRACT

Two types of lid movements, blinks and lid saccades, have discrete kinematic properties and physiology. These differences are reflected in distinct phenomenology of disorders affecting their neural substrate. Proof of this principle was seen in two patients, one with parietal lobe epilepsy and the other with temporal lobe epilepsy. The lid movements in the patient with parietal lobe epilepsy were rhythmic, yoked, and had a rapid upward component that instantaneously followed a slow downward drift. These cyclic movements strikingly resembled nystagmus, but unlike typical eye nystagmus, the rapid upward component was pathological and seemed to involve a saccadic mechanism. We suggest the terms “ictal lid saccades” or “ictal lid nystagmus” to describe such a phenomenon. In contrast, the patient with temporal lobe epilepsy had ipsilateral lid movements with rapid downward trajectories resembling reflex or spontaneous blinks. The term “ictal blink” is appropriate for this phenomenon.

Introduction

The eyelid movement is one of the simplest motor functions that involves three active forces, one passive force, two skeletal muscles, and one smooth muscle. The skeletal muscles provide dominant forces; the tonically active levator palpebrae (LP) keep the eyelids retracted whilst orbicularis oculi (OO) remains silent. A burst of discharge of OO, simultaneous inhibition of LP, and passive forces close the eyelid. LP and OO are therefore functionally “reciprocal”. However, their motor control is fundamentally different from conventional motor systems that obey Sheringtonian principles for reciprocal innervation. For example, in a typical motor system the commands are generated in the same cortical region that excites agonists and inhibits antagonists. These agonists and antagonists also have analogous innervation patterns and laterality. In contrast, LP and OO are innervated by separate cranial nerves, motor and premotor nuclei, and cortical substrates. LP is innervated by the third cranial nerve, while the seventh nerve innervates OO. Electrical stimulation of the frontal and parietal eye fields in healthy subjects triggers vertical eye saccades and results in synchronous activation of both LP, while OO remains silent. The activation of LP is yoked. These lid movements, synchronised with the eye saccades, are called lid saccades.^1,2 The active upward trajectory of the lid saccade is faster than their passive downward movement.¹ In contrast, the activation of OO is typically seen with reflex or spontaneous blinks. Stimulation of the trigeminal nerve (the afferent limb of the blink reflex) activates the ipsilateral OO, but inhibits bilateral LP.^3–5 The blinks are not yoked and the downward trajectory of the blink is faster than its upward movement.¹ Figure 1 summarises the physiological and kinematic differences between lid saccades and blinks.

Figure 1.

Schematic organisation of the underlying neural circuit for lid saccades (a) and blinks (b). The caricature of the trajectory of lid saccades (c) and blinks (d). In the caricature of lid saccades, vertical lid position is schematised in blue, while the corresponding lid velocity is schematised in brown. The bottom row schematises EMG activity from levator palpebrae (green) and orbicularis oculi (light brown). In the schematic in panels (c) and (d) the x-axis always represents time and the schematised EMG activity is temporally matched with the schematised lid position and lid velocity

Abbreviations: FEF (frontal eye fields); PEF (parietal eye fields); SC (superior colliculus); riMLF (rostral interstitial nucleus of the medial longitudinal fasciculus); INC (interstitial nucleus of Cajal); CCN (central caudal nucleus); LP (levator palpebrae), OO (orbicularis oculi); EMG (electromyogram).

The functional and anatomical discrepancies in the mechanisms of lid saccades and blinks vouch for the discrete phenomenology of diseases affecting their neural correlates. Different ictal lid movements that accompany various seizure types establish the precedent. Unilateral “ictal blinking” is typical of temporal lobe epilepsy.^6–8 In contrast, “ictal lid saccades” could have precise localising value for frontal or parietal eye field seizures.

We describe two patients. The first patient had rhythmic bilateral lid saccades that followed (passive) downward drift of both eyelids. This correlated with rhythmic epileptiform discharges from the right parietal eye field on EEG. The other patient presented with unilateral ictal blinking due to ipsilateral temporal lobe seizures. These examples demonstrate that the kinematics of the yoked, saccadic lid movements that accompany the eye-field seizures are fundamentally different from ictal blinking seen with temporal lobe seizures.

The cases

Case 1: Periodic epileptiform discharges from parietal eye fields causing rhythmic lid saccades

A 77-year-old right-handed woman had periodic “lid twitches” and altered mental status. She had a past medical history of bilateral occipital infarcts (causing cortical blindness), a subacute right parietal infarct, and stenosis of both distal vertebral arteries, the proximal basilar artery, the right internal carotid artery, and the left middle cerebral artery. The patient also had chronic complete bilateral supranuclear ophthalmoplegia without ptosis as can be explained by a lesion of the vertical burst generators.

The patient was alert and orientated to time, but not to place or person. Her pupils were 2 mm in size and reactive to light. There was no afferent pupillary defect. Her gaze was restricted to the primary position. However, the vestibulo-ocular reflex (VOR) was present on passive head impulses in the horizontal and vertical canal planes. There was no ptosis. She could blink spontaneously. There were rhythmic yoked lid movements, which were fast in the upward direction followed by slow downward drifts (see Supplementary material video 1). These movement patterns strikingly resembled nystagmus. Similar to nystagmus, there was a quick phase with saccadic characteristics (the rapid upward movements) and a slow phase (relatively slow downward drifts). Occasionally, the patient had left-sided facial twitching with slurred speech. The examination of the appendicular motor system revealed strength at three out of five on the Medical Research Council scale in the left upper and lower extremities; the strength on the right side was normal. The patient appropriately responded to tactile and noxious stimuli. The deep tendon reflexes were normal on the right side and brisk on the left.

Magnetic resonance imaging (MRI) with diffusion weighted imaging showed changes consistent with an acute ischaemic stroke in the right parieto-occipital cortex (Figure 2). A surface electroencephalogram (EEG) with a conventional 10–20 montage revealed periodic lateralised epileptiform discharges (PLEDs) at electrodes P8 and C4, emerging from the structural injury in the right parietal cortex (Figure 3). We simultaneously recorded a surface electromyogram (sEMG) from LP. The sEMG activities of both LP were synchronous bilaterally and temporally correlated with the right parietal eye-field PLEDs (Figure 2).

Figure 2.

Diffusion weighted imaging (a) and corresponding apparent diffusion coefficient correlate (b) suggestive of an acute infarct in the right pareito-occipital cortex where the parietal eye field is located

Figure 3.

Electroencephalogram showing periodic lateralised epileptiform discharges (PLEDs) from the right parietal region (maximum at P8) that temporally correlated with the synchronous eyelid activity. The frequency of PLEDs and eyelid movements were correlated, occurring approximately once every 1–2 seconds

Here we learned that epileptiform discharges from the eye fields can cause spontaneous, rhythmic lid movements that are yoked on both sides. The rapid upward components of these movements followed the kinetic profile that is typical for lid saccades. These rapid upward lid movements typically followed slow downward drifts, appearing like lid nystagmus. Hence, the terms “ictal lid saccades” or “ictal lid nystagmus” might appropriately describe this phenomenon.

Case 2: Unilateral ictal blinking during temporal lobe seizures

Case 2, featuring typical unilateral ictal blinking during temporal lobe seizures, contrasts the kinematic features of the ictal lid blinking and ictal saccades described in Case 1. A 22-year-old right-handed man had seizures after head trauma. His seizure semiology was characterised by staring spells, followed by blinking of the right or left eyelid and clonic movements of the right upper extremity progressing to generalised clonic seizures. Occasionally, the patient had dialeptic seizures progressing to right arm clonic and then generalised tonic-clonic seizures, or dialeptic seizures progressing to complex motor seizures. The inter-ictal neurological examination was normal. An EEG during right eye blinking revealed a right temporal lobe seizure pattern with maximal activity at the right sphenoidal electrode. The left eye blinks correlated with a seizure pattern maximal at the left sphenoidal electrode. The “ictal blinking” in this patient closely correlated with the discharge frequency during the seizure (Figure 4). The trajectories of the ictal lid movements (the ictal blinks) comprised rapid downward movements followed by rapid upward movements (see Supplementary material video 2). These lid movements were always ipsilateral to the electrographic seizure activity.

Figure 4.

Electroencephalogram showing left temporal lobe seizure activity that correlated with left-sided blinks in a typical bipolar montage. The Fp1 lead was placed above the eyebrow, over the frontalis muscle on the left side. Fp2 was placed on the same location but on the right side. The activity in Fp1 lead represents electromyographic activity from the orbicularis oculi, levators, and frontalis. The figure depicts correlating myographic activity of these facial nerve innervated muscles responsible for blink. This diagram suggests the possibility of ipsilateral ictal blinking, putatively a consequence of direct electrical stimulation of the orbital branch of the trigeminal nerve

It was clear that ictal blinking, as seen in Case 2, had distinct kinematic properties contrasting with the lid saccades in Case 1. The features of ictal blinking were similar to that of reflexive and spontaneous physiological blinks, i.e., rapid downward lid movements immediately followed by rapid upward movements (Figure 1d).

Discussion

We report two cases, one with a parietal epileptogenic zone and the other with temporal lobe epilepsy. In both, the cortical electrical activity correlated with the discrete types of pathological lid movements. Here we learned that patients with parieto-occipital epilepsy can have ictal lid movements that are yoked and characterised by rapid upward movements instantaneously followed by slow downward drifts. These cyclic movements strikingly resemble nystagmus, but unlike typical jerk nystagmus of the eye, the rapid upward phase (the quick phase) of these lid movements is pathological. The quick phase seems to be triggered by the activity in the parietal eye field and therefore, must involve a saccadic mechanism and share features of the upward lid saccades. In our opinion, the terms “ictal lid saccade” or “ictal lid nystagmus” are appropriate to describe this phenomenon. In contrast, patients with temporal lobe epilepsy have ipsilateral lid movements with rapid downward trajectories resembling reflex or spontaneous blinks.

Blinks and lid saccades have discrete physiology and anatomical correlates. In the subsequent section we will review the relevant anatomy and physiology of the neural circuits underlying blinks and lid saccades. Subsequently, we will discuss the pathophysiology of ictal lid blinks and unilateral ictal saccades in the context of epileptic activity in the eye fields and temporal lobes, respectively.

Lid saccades

Lid saccades have different kinematics, motor control, and physiological purpose as compared to blinks.¹ Lid saccades, synchronised with vertical eye movements, simultaneously move the lids with the eyes to protect the cornea and to prevent visual obstruction. Like vertical eye movements, lid saccades are yoked. The upward lid saccade is an active phenomenon, which requires burst-tonic activity of LP.¹ The downward lid saccade is a passive phenomenon; LP activity ceases during this movement but ligament pull facilitates the process.¹ Gravity and OO have no influence on downward lid saccades.^1,9,10 As a result, typical upward lid saccades are faster than downward lid saccades.^1,9,10

The motoneurons innervating the LP are localised in the central caudal nucleus (CCN) of the oculomotor nucleus.^11,12 The eyelid motoneurons on the both sides are intermixed in the CCN.¹³ This anatomical organisation is critical for yoking of the lid saccades. The steady state firing of these motoneurons and the subsequent tonic input to LP keeps the eyes open during the awake state; decrease in this input, during drowsiness, lowers the lids.¹⁴ The pre-motor neurons are the source of dynamic excitatory burst signals to the CCN motoneurons.² In primates, the pre-motor neurons for the lid saccades are in the M-group, the nucleus medial to the pre-motor burst generators for the vertical saccades in the rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF).^15,16 For upward saccades, the upward burst neurons in the riMLF and the M-group receive commands from the medial superior colliculus (Figure 1a).¹⁵ As a result, upward lid and eye saccades are synchronised.^2,16,17 The commands for downward eye saccades are projected from the lateral superior colliculus to the riMLF but without excitatory collaterals to the M-group (Figure 1a).¹⁵ Furthermore, downward bursts in the riMLF drive inhibitory GABA mediated projections from the interstitial nucleus of Cajal (INC) to inhibit M-group neurons responsible for upward lid saccades.^13,15 As a result, downward lid saccades, as passive phenomena, are slower than upward lid saccades. Deep layers of the primate superior colliculi receive excitatory projections from the frontal and parietal eye fields.¹⁸ This principle supports the hypothesis that pathological spontaneous discharges from the frontal and parietal eye fields can cause ictal lid movements that are yoked, spontaneous, upward directed lid saccades. These active upward lid movements can follow passive slow downward drifts. Case 1 provides the proof of this principle.

Ictal blinks

The neural substrate and the physiology of blinks are distinct from lid saccades. Unlike with lid saccades, OO is active during blinks. The cumulative discharge of OO (the area under the curve) correlates with the amplitude of the blink, while its peak firing rate correlates with the maximal blink velocity. Unlike during lid saccades, there is minimal increase in the activity of LP during the upward lid movement of a blink (Figure 1c,d).^19,20

The physiology of blinks is also remarkably different from that of lid saccades. Stimulation of the supra-orbital branch of the trigeminal nerve, either by mechanically stimulating the corneal surface or by electrical stimulation, triggers the blink reflex. Peri-ocular and corneal sensory projections, the afferent limb of the blink reflex, reach the principal trigeminal nucleus.^3,4,21 The trigeminal neurons then project to the ipsilateral facial nucleus innervating the OO.²¹ The blink-associated trigeminal neurons also inhibit the motoneurons in the caudal most aspect of the oculomotor nucleus that innervates LP. The laterality of this inhibition varies amongst subjects; it could be ipsilateral, contralateral, or bilateral.^5,22 Hypothetically, this projection is supposed to facilitate the inhibition of LP when OO is excited.^5,15,16 The cerebral cortex is unlikely to play a direct role in eliciting the blink reflex.

To our knowledge, this is the first report of “ictal lid saccades (or “ictal lid nystagmus”). This is a unique case from several perspectives. The patient had complete ophthalmoplegia but with sparing of lid movements. Such pathology assured us that the rapid upward lid movements, that we addressed here as “lid saccades,” were an active phenomenon and not passive lid movements that accompanied vertical nystagmus or eye saccades. Bilateral pontine and midbrain lesions explained the complete ophthalmoplegia.²³ The supranuclear vertical gaze palsy can be attributed to the involvement of the riMLF. Loss of adduction, but intact lid movements can be described by an infarct involving the oculomotor nucleus but sparing its caudal pole, the CCN (Figure 1a). The pseudo-abducens palsy can be described by the lesions affecting midbrain thalamic junction resulting in interruption of the inhibitory vergence pathways.²⁴ Severe stenosis of the basilar artery and both vertebral arteries are consistent with an ischaemic vulnerability of the affected area. MRI may lack sensitivity to identify infarcts in such anatomically compact regions. An intact VOR during the head impulse test suggests that the ocular motor nuclei, cranial nerves, and orbital muscles were unaffected.

There are a few reported cases of ipsilateral ictal blinking as a lateralising sign of temporal lobe epilepsy.^6,8 Lesser and colleagues proposed that unilateral ictal blinking is an unnatural, non-physiological phenomenon resulting from the spread of seizure-related electric potentials to the trigeminal nerve fibres accompanying pial vessels and dura in the anterior and middle cranial fossae.^25,26 Indeed, intracranial stimulation of trigeminal fibres evokes ipsilateral blinking in humans.²⁷

In summary, these two cases show us an important distinction between two types of lid movements, each specific to a particular epileptogenic zone. Yoked, rhythmic, rapid upward movement of the lids (lid saccades) correlated with epileptiform discharges from the frontal or parietal eye fields. In contrast, rapid downward, unilateral lid movements (blinks) correlate with ipsilateral temporal lobe seizures. We must emphasise that the differences in trajectory of the two types of lid movements were made based on clinical observations in two patients. Future studies involving a larger patient cohort and higher resolution, intracranial electrographic recordings are desirable to further support these pilot observations.

Funding Statement

This work was supported by the American Parkinson Disease Association [George C Cotzias Memorial Fellowship]; NIH [U01NS090405](Lhatoo); DMRF [Research Grant]; AAN [Career Award]; U.S. Department of Veterans Affairs [I01CX002086-01A2](Shaikh).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/01658107.2021.1900280.