Coma is a prolonged state of unconsciousness in which the patient does not respond to external stimuli, regardless of the presence or absence of noxious stimuli. It is caused by severe brain damage and has various etiologies including trauma, stroke, brain tumor, severe blood glucose abnormalities, prolonged cerebral hypoxia, epileptic seizures, and drug intoxication.
Comatose patients lose the ability to think and perceive their surroundings. Many gradually recover, but some progress to prolonged disorders of consciousness (vegetative state) or death. The outcome depends largely on the cause, severity, and location of the neurological damage.
The Glasgow Coma Scale (GCS) is widely used to assess the severity of coma. It evaluates eye opening, verbal response, and motor response; a score below 8 generally indicates a comatose state. However, the GCS does not include an examination of eye movements.
The Full Outline of Unresponsiveness (FOUR) score is an alternative method that assesses eye response, motor response, brainstem reflexes, and respiration. Both the GCS and FOUR score are excellent predictors of outcome.
The main purpose of neuro-ophthalmologic examination in comatose patients is to help identify the cause, localization, and severity of coma. Findings of the eyelids, pupils, eye movements, and fundus reflect the level of brainstem dysfunction. Loss of the pupillary light reflex is one of the predictors of poor prognosis in traumatic brain injury.
The GCS evaluates three components: eye opening, verbal response, and motor response, but does not include eye movements. The FOUR score evaluates four components: eye response, motor response, brainstem reflexes, and respiration, reflecting brainstem function in more detail. Both are useful as predictors of outcome.
Comatose patients are unconscious and cannot report subjective symptoms. Before the onset of coma (in the early stage of decreased consciousness), headache, nausea/vomiting, decreased vision, diplopia, and clouding of consciousness may precede.
Neuro-ophthalmological findings in comatose patients are broadly classified into pupillary findings, spontaneous eye movements, eye position and movements, eyelids, and fundus findings.
The pupillary light reflex is easy to test and directly aids in localizing the cause of coma. First, observe pupil size in light and dark conditions to check for anisocoria. Shine light into one eye and record the response (brisk, sluggish, or no reaction) in both eyes. Use the swinging flashlight test to detect a relative afferent pupillary defect (RAPD).
If anisocoria is present, determine the affected eye by the magnitude and speed of the light response. The eye with no or slower response is the affected eye.
Unilateral Mydriasis and Fixation
Hutchinson pupil: Compression of the oculomotor nerve due to brain herniation (uncal herniation). An emergency finding.
Ruptured posterior communicating artery aneurysm: Compressive lesion along the oculomotor nerve.
Pharmacologic mydriasis: Mimicked by atropine eye drops, scopolamine patches, etc. Can be differentiated with 1% pilocarpine.
Bilateral miosis (pinpoint)
Damage to the dorsal pons: The pupillary dilator pathway is interrupted, e.g., in pontine hemorrhage. Pupils constrict to about 1 mm, but the light reflex is preserved.
Medullary paramedian reticular formation: Abnormal excitation due to disruption of the inhibitory pathway to the Edinger-Westphal nucleus.
Bilateral small pupils with light reflex
Metabolic/drug-induced: In the absence of structural brain damage, suggests a metabolic or systemic drug-related cause.
Diencephalic injury and many toxins/drugs also present similar findings.
Horner syndrome
Unilateral miosis, ptosis, anhidrosis: Damage to the ipsilateral sympathetic pathway (hypothalamus → brainstem posterolateral → spinal cord → sympathetic trunk → eye).
Confirmation: Diagnosis with cocaine or apraclonidine eye drops.
Argyll Robertson pupils exhibit light-near dissociation, where the light reflex is absent but the near reflex (accommodation-convergence) is preserved. This is caused by lesions in the dorsal midbrain (pretectal area) and can occur in neurosyphilis, as well as diabetes, cerebrovascular disease, and demyelinating diseases.
To assess voluntary eye movements, the eyelids are passively opened and purposeful movements are observed. If purposeful eye movements are seen in an unresponsive patient, consider catatonia, locked-in syndrome, or feigned coma.
Suspect Hutchinson pupil. This is an emergency sign caused by oculomotor nerve compression due to brain herniation (uncal herniation). However, drug-induced mydriasis from atropine eye drops or scopolamine patches can present similarly, so differentiation using 1% pilocarpine eye drops is important. For details, see the “Clinical Findings” section.
The causes of coma are broadly classified into structural lesions and systemic/metabolic causes.
The pupillary light reflex test is the most basic and important examination method. In a semi-dark room, shine a penlight into one eye and observe the direct response (constriction of the illuminated eye) and the indirect response (constriction of the non-illuminated eye). Assess whether the response is brisk or sluggish, and whether the degree of constriction is sufficient or insufficient.
The swinging flashlight test is used to detect a relative afferent pupillary defect (RAPD). Alternately stimulate each eye with a penlight and observe changes in pupil size. If the constriction of the illuminated eye cannot be maintained and the pupil dilates, it is considered RAPD positive. This is positive in cases of unilateral optic nerve damage or severe retinal damage.
The 1% pilocarpine eye drop test is useful for differentiating pharmacologic mydriasis. An atropinized pupil does not respond to pilocarpine and does not constrict, allowing differentiation from oculomotor nerve palsy.
The apraclonidine eye drop test is used to diagnose Horner syndrome. After instilling drops in both eyes and observing 30 to 45 minutes later, the unaffected eye shows no change, while the affected eye shows dilation and improvement of ptosis.
Perform after confirming cervical spine stability. With the eyelids open, quickly rotate the head left and right and observe eye movements.
This is performed after confirming that the eardrum is normal. A small amount of water that is 7°C higher (warm water) or lower (cold water) than body temperature is injected into the external auditory canal.
A positive response is bilateral blinking when a tactile stimulus such as a cotton swab is applied to the corneal surface. A positive response indicates that cranial nerves V and VII are intact bilaterally. Loss of the reflex suggests lesions of cranial nerves V and/or VII and/or the pons.
A hand is moved quickly toward the patient’s eyes from different directions to elicit a blink reflex. This helps assess the presence of afferent input (preserved vision).
A negative finding indicates deep metabolic coma, vestibular damage, or brainstem dysfunction at the midbrain to pontine level. Loss of the vestibulo-ocular reflex has been reported to predict brain death with 100% accuracy, making it a crucial finding for prognosis. For details, see the section on “Vestibulo-ocular reflex testing.”
Neuro-ophthalmological findings in comatose patients can help estimate the site of the lesion. Characteristic patterns of findings are shown for different hemorrhage sites in cerebrovascular disease.
| Finding | Pontine hemorrhage | Thalamic hemorrhage | Subarachnoid hemorrhage |
|---|---|---|---|
| Pupil | Constriction | Small, anisocoria | Various |
| Gaze palsy | Ipsilateral | Contralateral | None |
| Vertical eye movement | Present | Absent | Absent |
Focal findings (hemiparesis, aphasia, conjugate gaze deviation, unilateral reflex abnormalities) suggest an organic lesion.
The pupillary light reflex consists of an afferent pathway (retina → optic nerve → pretectal area → Edinger-Westphal nucleus) and an efferent pathway (Edinger-Westphal nucleus → oculomotor nerve → ciliary ganglion → pupillary sphincter muscle). The efferent pathway for the near response is shared with the light reflex, but the supranuclear fibers to the Edinger-Westphal nucleus run more ventrally than the midbrain pretectal area and posterior commissure, through which the afferent fibers of the light reflex pass. Therefore, a lesion in the pretectal area causes light-near dissociation.
The ratio of neurons involved in the light reflex to those involved in accommodation in the ciliary ganglion is said to be 3:97. Due to this ratio, even if the light reflex is impaired, pupillary constriction during the near response is likely to be preserved.
In pontine lesions, especially pontine hemorrhage, the pupil constricts to about 1 mm in diameter. An ascending pathway from the paramedian pontine reticular formation that inhibits the Edinger-Westphal nucleus is hypothesized; damage to this inhibitory fiber leads to abnormal excitation of the Edinger-Westphal nucleus, resulting in severe miosis. The light reflex and near response are preserved.
Classical ocular bobbing is typically associated with ventral pontine lesions. Damage to the paramedian pontine reticular formation (PPRF) and omnipause neurons disrupts saccadic suppression, resulting in uncontrolled vertical saccades 1). Impaired modulation by the cerebellum (fastigial nucleus) also contributes to the imbalance between fast and slow phases 1).
The clinical significance of each variant differs. Classical ocular bobbing suggests irreversible brainstem pathology, whereas variants associated with metabolic encephalopathy are often transient and reversible 1).
In the ciliary ganglion, the ratio of neurons involved in the light reflex to those involved in accommodation is 3:97, so the accommodation and near response pathways are likely to be spared even if the light reflex pathway is selectively damaged. Furthermore, because the supranuclear fibers for the near response to the Edinger-Westphal nucleus run more ventrally than the light reflex pathway, a light-near dissociation occurs in lesions of the pretectal area of the midbrain.
Multiple variants of ocular bobbing and their clinical localization have been reported1). Classic ocular bobbing is associated with structural pontine lesions (infarction, hemorrhage) and reflects damage to the vertical saccade pathway and gaze-holding center. Inverse ocular bobbing is more common in metabolic encephalopathy (hepatic, uremic) and indicates diffuse cortical dysfunction rather than focal brainstem lesions. Ocular dipping is frequently associated with hypoxic-ischemic brain injury and reflects suppression of cortical function with relative preservation of brainstem reflexes. Reverse ocular dipping is rare and seen in severe metabolic or cerebellar disorders.
Analysis of consecutive PPG cases showed a median age of 60 years, with males accounting for 12/14 cases. The PPG cycle ranged from 1.5 to 6.5 seconds1). Main causes were acute ischemic stroke, postictal state, and hypoxic-ischemic encephalopathy. 88.9% of full-field PPG had similar bilateral cerebral hemisphere damage, while 80% of hemifield PPG had unilateral or markedly asymmetric bilateral cerebral hemisphere damage. Neurological recovery, transition to vegetative state, and death have been reported.
A detailed eye movement recording of a case of ocular bobbing after pontine hemorrhage has been reported1). Pendular oscillations accompanied the ocular bobbing, suggesting complex interactions between different neural mechanisms in the brainstem. The importance of comprehensive eye movement recording is emphasized.
