What Is an Afferent Pupil Defect? Causes and Indications

An afferent pupil defect, more precisely called a relative afferent pupillary defect (RAPD), is a condition in which one eye’s pupil responds less briskly to direct light than the other, signaling that the nerve pathway carrying visual information from that eye to the brain is damaged or impaired. It is not an eye disease itself but a physical sign that something is wrong with the optic nerve, retina, or occasionally deeper brain structures on the affected side. Clinicians often call it a “Marcus Gunn pupil,” after the Scottish ophthalmologist who first described it in the late 1800s, and it remains one of the most informative findings a doctor can pick up with nothing more than a penlight.

How the Swinging Flashlight Test Works

The standard way to detect an RAPD is the swinging flashlight test. In a dimly lit room, the examiner shines a bright light into one eye for a couple of seconds, then swings it to the other eye. In a healthy person, both pupils constrict equally regardless of which eye the light enters, because the signal from either eye reaches both sides of the brain. When one eye has an afferent defect, swinging the light to that eye produces a paradoxical response: instead of constricting further or holding steady, the pupil actually dilates. It looks as though the affected eye is “rejecting” the light, when what is really happening is that the brain receives a weaker signal from that eye and interprets the overall light level as having dropped.

This works because the pupillary light reflex is a loop. Light hits the retina, travels along the optic nerve through slow-conducting fibers in the optic tract, and synapses in a region of the midbrain called the pretectal area. From there, neurons project to the Edinger-Westphal nucleus, which sends the constriction signal back out to both pupils through the parasympathetic nerves.¹ Because the signal crosses to both sides, shining light into one eye constricts both pupils. The test exploits this symmetry: if the input from one eye is weaker, quickly alternating the light between the two eyes reveals the mismatch as a visible dilation on the affected side.

Optic Neuritis and Multiple Sclerosis

The single most classic cause of an RAPD is optic neuritis, an inflammation of the optic nerve that damages the myelin sheath insulating its fibers. Optic neuritis is frequently the first recognizable symptom of multiple sclerosis (MS), and the RAPD it produces is so characteristic that finding one in a patient with sudden vision loss and eye pain strongly nudges the diagnosis toward demyelinating disease. In one study using ultrasound-based pupillometry, every patient with acute optic neuritis showed a clear RAPD, and the test perfectly discriminated affected eyes from healthy controls.²

What makes the RAPD particularly useful in MS is that it can persist even after vision subjectively recovers. Patients who have had optic neuritis often regain good visual acuity on an eye chart, yet their pupil responses remain sluggish on the affected side. Researchers studying recovery in optic neuritis patients found that both the constriction amplitude and the latency of the pupil response stayed abnormal long after visual performance improved.³ The RAPD in this scenario is essentially a fingerprint of past nerve damage, detectable years later even if the patient feels fine.

More striking still, sensitive pupillometry can uncover subclinical damage in MS patients who never had a recognized episode of optic neuritis. The same ultrasound pupillometry study detected an RAPD in about one in six MS patients with no history of optic nerve inflammation, none of whom had shown an RAPD on standard swinging flashlight testing.² In other research, RAPD appeared only in eyes that also showed thinning of the retinal nerve fiber layer, reinforcing the idea that it tracks real structural loss rather than temporary inflammation.⁴

Ischemic Optic Neuropathy

If optic neuritis is the textbook cause in younger adults, ischemic optic neuropathy fills that role in people over 50. This happens when blood supply to the optic nerve head is interrupted, essentially a “stroke” of the optic nerve. It comes in two forms. The non-arteritic type (NAION) typically strikes overnight and presents with painless vision loss, a swollen optic disc, and a visual field cut that respects a horizontal boundary. The arteritic type (AAION) is driven by giant cell arteritis, an inflammatory disease of the blood vessel walls, and tends to cause more severe vision loss with a pale, swollen disc.

Both forms commonly produce an RAPD. In a case of giant cell arteritis causing posterior ischemic optic neuropathy, for example, the affected eye showed a clear RAPD along with reduced visual acuity.⁵ In another case of arteritic anterior ischemic optic neuropathy, the involved pupil was dilated and showed no direct light response at all, with only a slow consensual response remaining.⁶ The presence of an RAPD in an older adult with sudden vision loss and a swollen disc is a red flag that demands urgent evaluation for giant cell arteritis, because the other eye can go blind within days if the condition is not treated with high-dose steroids.

Glaucoma

Glaucoma might seem like an unlikely cause, since it typically affects both eyes. But it rarely progresses at exactly the same rate in each eye, and when the damage is sufficiently lopsided, an RAPD appears in the worse eye. This makes the test valuable as a quick bedside indicator of how asymmetric the disease has become.

A landmark study using automated perimetry quantified this relationship precisely: patients who had an RAPD showed an average difference in visual field sensitivity between their two eyes of about 48%, whereas those without an RAPD had only about a 6% difference. There was no overlap between the two groups; every patient whose interocular sensitivity difference reached 13% or more had a detectable RAPD.⁷ The depth of the RAPD correlated with visual field asymmetry but, interestingly, not with asymmetry in cup-to-disc ratio, suggesting that functional damage to the nerve fibers matters more than the structural appearance of the optic disc.

Retinal nerve fiber layer measurements reinforce this picture. The RAPD in glaucoma reflects genuine loss of retinal ganglion cells and their axons on the worse side.⁸ Newer handheld pupillometers can now detect and quantify even small RAPDs in glaucoma patients, correlating the findings with structural and functional asymmetry measured on imaging and perimetry.⁹ For a disease that progresses silently for years, having a sign that can be checked in seconds with a flashlight is clinically powerful.

Retinal Disease and Macular Degeneration

The optic nerve is not the only structure whose damage produces an RAPD. Widespread retinal disease can do it too, because the retina is where the afferent signal originates. Large retinal detachments, central retinal artery or vein occlusions, and advanced macular degeneration can all reduce the total light signal enough to create an asymmetry between the eyes.

In age-related macular degeneration, researchers have measured the RAPD using automated pupillometry and correlated it with the size and type of macular lesions on imaging.¹⁰ The clinical relevance is that when you see an RAPD in a patient who “only” has macular disease, it tells you the damage is substantial. It can also help clinicians decide how aggressively to treat or how urgently to intervene.

Optic Tract and Midbrain Lesions

Most causes of an RAPD sit in or near the eye. But the pathway extends well behind the eyeball, and damage anywhere along it can, at least in theory, produce one. Optic tract lesions, where the nerve fibers have already partially crossed at the chiasm, are a notable example. Because roughly 53% of optic nerve fibers cross to the opposite side at the chiasm and only about 47% stay on the same side, a lesion in one optic tract disrupts more crossed fibers (from the opposite eye’s nasal retina) than uncrossed fibers. The practical result is an RAPD in the eye contralateral to the lesion, even though visual acuity in both eyes is perfectly normal.

In a series of four patients with complete homonymous hemianopia from optic tract lesions, all four had a detectable RAPD in the eye with the temporal visual field loss, despite good acuity in both eyes.¹¹ Further investigation showed that this RAPD reflects the difference in light sensitivity between the intact nasal and temporal hemifields, and its side is predictable based on the greater proportion of crossing pupillomotor fibers.¹² Compression of the midbrain can produce a similar finding, with the RAPD appearing contralateral to the side of compression.¹³

These cases are rare but important, because an RAPD in someone with normal visual acuity and no obvious eye disease can be the first clue to a brain tumor or other intracranial mass. The finding prompts neuroimaging rather than a purely ophthalmic workup.

The Cataract Paradox

A common misconception is that a dense cataract should cause an RAPD in the eye behind it, since it blocks light from reaching the retina. The reality is counterintuitive. A dense unilateral cataract actually produces a measurable RAPD in the opposite, fellow eye. In a study of ten patients with a cataract dense enough to reduce vision to counting fingers or worse in one eye, every patient showed an RAPD in the other, clear-sighted eye, averaging about 0.44 log units. Once the cataract was surgically removed, the RAPD vanished.¹⁴

Why does this happen? The leading explanation is that scattered light within the cataractous eye actually stimulates a broader area of retina than the focused beam entering the clear eye, generating a paradoxically stronger afferent signal. This is clinically significant: if you find an RAPD in a patient with a dense unilateral cataract, do not automatically assume the clear eye has optic nerve disease. The cataract itself may be responsible, and the way to confirm this is to recheck after cataract surgery.

Traumatic Optic Nerve Injury

Trauma, especially to the face and skull, can damage the optic nerve where it passes through the bony optic canal. This traumatic optic neuropathy (TON) tends to produce sudden, severe vision loss on the affected side and a prominent RAPD. In a large Canadian trauma center, TON was found in about 0.4% of all trauma admissions and in roughly 2.3% of patients with significant head injuries. Motor vehicle accidents were the leading cause, followed by falls.¹⁵ In emergency settings, checking for an RAPD can be done in seconds even in a patient who cannot communicate, making it an invaluable triage tool for detecting optic nerve damage that might otherwise be missed among more dramatic injuries.

Quantifying the Defect

The traditional way to grade an RAPD is by holding neutral density filters of increasing darkness over the better eye until the two pupils respond symmetrically. The filter strength (measured in log units) at which the defect “balances out” gives a rough numerical score. This technique is simple but subjective, and research has shown it correlates well with differences in visual field sensitivity between the two eyes.¹⁶

Modern infrared pupillometers offer a more objective alternative. These handheld or tabletop devices record both pupils simultaneously while alternating a calibrated light stimulus between the eyes, essentially an automated swinging flashlight test. By measuring constriction amplitude, velocity, and onset latency, they can detect and quantify RAPDs too subtle for the human eye to catch.¹⁷ Calibration studies using neutral density filters of known attenuation have validated these devices against the clinical standard, allowing the pupillometer output to be converted into the familiar log-unit scale.¹⁸

Portable pupillometry using infrared recording has also proved reliable at simulating and detecting RAPDs in healthy volunteers using dimmed light as a stand-in for nerve damage, confirming that the devices work across a clinically meaningful range.¹⁹ The trend in clinical practice is moving toward automated pupillometry for any setting where tracking the RAPD over time matters, such as monitoring glaucoma progression or following recovery after optic neuritis.

When the RAPD Is Absent and Should Not Be

One of the most underappreciated clinical uses of the RAPD test is the information carried by a negative result. If someone reports profound unilateral vision loss, with the affected eye seeing no light at all, and yet the pupils respond perfectly symmetrically to the swinging flashlight, the optic nerve and retina are almost certainly intact. This mismatch raises strong suspicion for non-organic visual loss, sometimes called functional or psychogenic vision loss, where the problem lies in the brain’s processing or the patient’s psychological state rather than in any structural damage to the eye or nerve.²⁰

This is not a trivial distinction. Patients with non-organic visual loss can present convincingly with dramatic symptoms, and expensive imaging and invasive testing sometimes follow before someone thinks to check the pupils. The absence of an RAPD in the face of claimed unilateral blindness is a simple, free, and highly reliable way to redirect the workup. It does not mean the patient is “faking” in any pejorative sense; non-organic visual loss is a recognized neuropsychiatric condition. But it does mean the eye itself is not the problem.

Pitfalls in Children and in Amblyopia

Testing for an RAPD in young children presents practical challenges: the room must be dim, the child must look straight ahead, and the examiner needs to swing the light smoothly and consistently. Fussy or uncooperative children can produce unreliable results. Still, the test is one of the few objective measures of visual pathway integrity available in patients too young to read an eye chart or describe their symptoms.

A trickier question is whether amblyopia, the common “lazy eye” condition of childhood, produces an RAPD. In a study carefully designed to minimize observer bias and account for the difficulty of testing pupils in young children, only 4 out of 45 amblyopic patients showed a mild afferent pupillary defect.²¹ This means amblyopia rarely produces an RAPD, and when it does, the defect is slight. The practical takeaway is that a clear, easily detectable RAPD in a child should not be dismissed as “just amblyopia.” It warrants further investigation for optic nerve or retinal pathology, even if amblyopia is also present. Conversely, the absence of an RAPD does not rule amblyopia in or out, because the condition primarily involves cortical processing rather than the afferent pathway.

Why “Relative” Matters

The word “relative” in RAPD is doing real work. The test only detects a difference between the two eyes. If both optic nerves are equally damaged, the pupils will still constrict symmetrically, and no RAPD will be found, even though both nerves are badly injured. This is a genuine limitation. Bilateral optic neuritis, for example, can leave both pupils sluggish without producing the telltale asymmetry. The same is true of bilateral advanced glaucoma with roughly symmetric damage. In these cases, clinicians rely on other signs, such as poor pupil reactions overall, reduced visual acuity, abnormal color vision, or visual field loss, to identify the problem.

This is also why the test is useless for evaluating problems behind the lateral geniculate nucleus, deeper in the brain’s visual cortex. Damage in those areas causes visual field deficits but does not alter the pupillary reflex, because the fibers carrying the pupil signal leave the visual pathway before reaching the cortex. A person with cortical blindness from occipital lobe damage will have pupils that react perfectly normally to light, even though they cannot consciously see.

Conditions That Can Mimic or Obscure an RAPD

Several situations can create a false positive or false negative. Pharmacological dilation of one pupil, whether from eye drops used during an exam or from accidental exposure to certain plants or medications, makes the dilated pupil unable to constrict normally, mimicking an efferent (motor) rather than afferent (sensory) problem. Anisocoria, a baseline difference in pupil size from unrelated causes, can confuse the picture unless the examiner is experienced. Prior eye surgery, iris damage, or prosthetic intraocular lenses can also affect the pupil’s mechanical ability to respond.

The dense cataract phenomenon described earlier is perhaps the most commonly encountered trap. And in acute settings, swollen or bruised eyelids, blood in the front of the eye, or an uncooperative patient can make the test difficult or impossible to perform reliably. When the standard bedside test is equivocal, automated pupillometry provides a more objective measurement by recording precise pupil diameter changes that the human eye cannot resolve.