Paediatric eye examination and red-reflex assessment — branching viva

Opening question

Examiner: Take me through how you would perform the red-reflex test on this infant, and what you would expect to see normally. [4]

Candidate (model): I would perform it in a dim room, with the direct ophthalmoscope on the largest, brightest spot at zero or low magnification, standing about an arm's length away, roughly 30 to 45 centimetres, and I would examine both pupils together so I can compare the two reflexes side by side — that simultaneous comparison is the Bruckner test. I would repeat it at a couple of distances and gaze positions and document the result for each eye. Normally I expect a symmetric orange-red or reddish glow of equal brightness, colour and size in both pupils. The glow is light that has passed through the clear cornea, aqueous, lens and vitreous, reflected off the retina and the blood-rich, melanin-rich choroid, and returned out. [4] [1]

Branch 1 — interpretation

Examiner: You have found a dense white reflex on the right. What does that mean, and how does it differ from an absent reflex and an asymmetric reflex? [3]

Candidate (model): A white reflex — leukocoria — means a pale lesion is sitting in the visual axis and reflecting light back white instead of letting it reach the choroid; the cause may be retinoblastoma, a congenital cataract, persistent fetal vasculature, Coats disease, toxocariasis or a retinal detachment. An absent or dull reflex means something is blocking the light — a dense cataract, vitreous haemorrhage or a retinal detachment — so the reflex is lost rather than replaced. An asymmetric reflex, where one side is darker but present, is the Bruckner sign of anisometropia or strabismus — that eye is out of focus or turned. A white reflex is the most urgent of the three. [3] [6]

Branch 2 — differential and triage

Examiner: Give me your prioritised differential and your immediate management. [6]

Candidate (model): I lead with retinoblastoma (the primary intraocular malignancy of childhood, fatal beyond the eye) and congenital or developmental cataract (common, treatable, time-critical for amblyopia), then persistent fetal vasculature, Coats disease, ocular toxocariasis, retinal detachment and coloboma. My immediate management is urgent ophthalmology referral, the same day for this infant; I would apply no drops, ointment or home remedy before specialist review; I would document the finding for each eye; and I would safety-net the family — explaining the urgency in plain language without using a diagnostic label, and confirming the appointment was kept. A dense infantile cataract must be operated on within weeks to avoid irreversible amblyopia. [6] [11]

Branch 3 — the normal darker variant

Examiner (probe): A colleague tells you the reflex in an Asian infant looks dark and calls it abnormal. What is your view? [4]

Candidate (model): A heavily pigmented fundus, common in children of Asian, African or First Nations heritage, gives a darker, duskier but still present reflex that is a normal variant, provided it is symmetric. The critical test is comparison between the two eyes: equal brightness, colour and size means normal. The trap is calling a normal symmetric darker reflex abnormal; the fix is always to compare both eyes, and to repeat in a dim room with correct technique. [4] [3]

Branch 4 — age-adapted assessment and screening

Examiner: How would your vision assessment differ for a newborn, a toddler and a preschool child, and what is the role of instrument-based screening? [7]

Candidate (model): For a newborn or young infant I assess fixation and following — by two to three months an infant should fix on a face or light and follow to the midline and beyond. For a preverbal child I use preferential-looking acuity cards (Teller) and observe for steady maintained fixation, alongside the red reflex and instrument-based photoscreening or autorefraction from about one to three years, which flags amblyopia risk factors from a captured image. For a preschool child, about three to five years, I add monocular distance acuity with HOTV or LEA symbols and a stereopsis test (Randot or Titmus), applying the AAPOS age-specific referral thresholds. From school age I use a Snellen or LogMAR chart. The red reflex itself is done at every well-child visit from the newborn check, under the joint AAP, AAO and AAPOS policy. [7] [4]

Branch 5 — pitfalls

Examiner (probe): What are the main false negatives of this test? [6]

Candidate (model): A small or peripheral tumour may not reach the axis, a lesion visible only in certain gaze positions may be missed head-on, a pigmented fundus can mask asymmetry, and mucus or a tear-film deposit can temporarily dull a truly normal reflex. A single normal reflex does not exclude a small retinoblastoma, which is why a parental concern or an abnormal photograph is referred even when the bedside test looks normal. I correct technique and repeat before I record a reflex I have not properly seen, and an uncooperative child gets an examination under anaesthesia rather than a guessed normal. [6] [4]