Similar retinal signs can follow very different causes. Radiation dose estimates are context-dependent; Valsalva often resolves but does not guarantee full vision; Purtscher evidence is weak; Terson requires neurological coordination and individual surgical timing.
Terminology and draft status
Terminology review is not complete. Selected actual WHO UMD queries and their limits are recorded in terminology.md; they do not approve the whole translation.
01Start with the pattern and the systemic history
Microaneurysms reflect focal vessel-wall change; dot/blot and flame-shaped haemorrhages help localise bleeding. Cotton-wool spots (CWS) reflect axoplasmic transport interruption in the nerve-fibre layer, often from focal ischaemia; hard exudates are lipid-rich deposits from leakage. None identifies one cause alone. 1 3
Ask about diabetes, blood pressure, medication, trauma, radiation/cancer treatment and systemic symptoms. In a young person without diabetes, consider inflammatory, haematological and hyperviscosity causes such as leukaemia when supported by the presentation; the original automatic uveitis-screening rule for every such patient is too broad. 1
02Read radiation injury with the actual dosimetry
Radiation retinopathy is delayed vascular injury after external-beam or ocular brachytherapy, including treatment of brain/head-and-neck tumours or choroidal melanoma with ruthenium applicators. Obtain treatment field, retinal dose, fractionation and latency; local dose, vascular comorbidity and treated volume matter. Diabetes, hypertension and carotid/ocular ischaemia can mimic or coexist with it, without a universal ranking of aggressiveness. 1 2
The original mean 18-month latency,15 Gy low-dose example,30–35 Gy threshold,60 Gy/50% and 70–80 Gy/90% risk figures lack their precise cohorts and dosimetry here. Preserve them as historical source claims, not universal thresholds or individual risk estimates. Brachytherapy is not invariably earlier than external-beam injury. 1
Anti-VEGF is used for vision-threatening radiation macular oedema with indication-specific off-label considerations; selected steroids carry pressure/cataract risks. In ARRT,39 eyes were randomised between aflibercept schedules,30 completed 1 year: overall BCVA change+4.3 letters (P=.087), while CRT fell 484.4→326.5 µm. Both arms had active treatment, with attrition and no observation comparator; anatomical improvement is not proof of a clear visual benefit or of superiority over no treatment. 2
Do not prescribe PRP for every non-perfused region or reserve it only for iris rubeosis. Assess retinal/iris neovascularisation, visibility, visual potential and ischaemic burden with a retina specialist. Laser has field and tissue costs; the original assertion that it routinely accelerates atrophy in all irradiated eyes is not established here. 1
03Locate Valsalva blood before selecting observation or drainage
Straining, heavy lifting or forceful coughing can abruptly increase thoracic/venous pressure and rupture retinal capillaries. Premacular blood may lie beneath the posterior hyaloid, the internal limiting membrane(ILM), or both. OCT helps localise it; a shiny surface or the original mobile “hourglass” description alone is not a validated anatomical discriminator. 1 7
Many selected Valsalva haemorrhages clear with observation over weeks to months. The original “up to 3 months” and eventual acuity 1.0 are not a compulsory wait or guaranteed outcome. Consider duration, location, density, visual need and alternative causes; massive, non-clearing or complicated bleeding may warrant specialist Nd:YAG membranotomy or PPV, with retinal injury and persistent vitreous haemorrhage risks. 1 7
04Distinguish Purtscher findings from an asserted histological depth
Purtscher retinopathy follows major trauma such as thoracic compression or head injury; Purtscher-like describes similar findings with systemic disease, including pancreatitis, renal failure, SLE or amniotic-fluid embolism. Complement activation and leukocyte aggregation are proposed mechanisms, not a proven identical sequence in every case. Assess and treat the underlying illness. 1 3
Peripapillary polygonal retinal whitening (Purtscher flecken), CWS and haemorrhage support the clinical pattern. The original “deep infarct versus superficial CWS” and visible-overlying-vessel rule should not be treated as a definitive histological diagnosis from colour alone. Commotio retinae/Berlin oedema is a trauma differential, but accompanying haemorrhage or other injuries can coexist and do not automatically exclude it. 1 3
A systematic review collected 168 cases from 114 reports; trauma accounted for 39.88%, SLE 13.1% and pancreatitis 11.9%. This published-case pool does not establish pancreatitis as the commonest non-traumatic cause in every setting. Low-quality reports, selection and inconsistent outcomes prevent a reliable comparative treatment or prognosis claim; spontaneous improvement does not prove a therapy worked. 3
High-dose corticosteroids have been discussed for Purtscher-like disease, but the available selected case literature does not establish a reliable visual benefit. Treat the underlying systemic disorder; neither all cases are bilateral nor a particular surface colour proves the depth of the infarct. 3 8
05Coordinate Terson care with the neurological illness
Terson syndrome is intraocular bleeding associated with intracranial haemorrhage, classically subarachnoid bleeding(SAH), but also other intracranial haemorrhages including traumatic contexts. A new severe headache, reduced consciousness or neurological deficit is an emergency; do not treat this solely as a self-limited ocular haemorrhage. The original subdural-haemorrhage context is therefore retained within the broader intracranial differential. 1 4 5 6
The original one-third/33% prevalence is not a universal estimate. A 2026 Japanese administrative cohort recorded 860 diagnoses among 104576 aneurysmal-SAH survivors (0.8%); coded diagnosis and survival selection can miss cases and differ from systematic eye screening, so this is not a replacement estimate for all SAH patients. 1 6
Observe suitable resolving haemorrhage or consider PPV for persistent dense bleeding, bilateral functional need, complications or rehabilitation goals. The original 4–6 week “sweet spot” and inevitable haemosiderosis/white organisation after that interval are not validated universal deadlines. Coordinate medical stability and retinal findings; longer observation is not automatically safe either. 1 4 5
Retrospective series illustrate uncertainty:117 eyes (46 observed/71 PPV) had different starting vision,25 surgical eyes had retinal breaks associated with delay>3 months; another 24 eye/19 patient series found no clear final-acuity difference for≤100 versus>100 days and had 1 postoperative endophthalmitis. Nonrandom selection, small samples and absent significant differences do not prove equivalent or risk-free timing. 4 5
The original subretinal rt-PA/TPA through a 41 Gneedle followed by gas refers to a separate selected submacular-haemorrhage technique, including macroaneurysm-related bleeding, not standard treatment for every Terson eye. Do not infer its indication, route or safety from the presence of vitreous blood; specialist source, operative and product verification remains necessary. 1
The original epidural-haemorrhage example belongs to the broader intracranial differential; ocular bleeding may accompany more severe neurological illness. This association is not proof that the eye bleed causes the poor neurological outcome. Assess neurological urgency first and coordinate eye treatment with systemic stability. 8 6
The original 25µg rtPA through a 41 G needle is preserved as an unverified, selected subretinal technique example. It does not establish a standard Terson dose, route, indication or preparation, and should not be extrapolated from a vitreous haemorrhage diagnosis. 8
06Choose imaging and preserve the emergency pathway
Document pupils and a relative afferent pupillary defect(RAPD) before dilation when feasible. FA can map leakage and non-perfusion when the fundus is visible and the result informs care; it is not mandatory for every acute haemorrhage. Use OCT for the affected compartment and B-scan when the view is blocked, remembering that a negative scan cannot exclude every small retinal tear. 1
Use the table as a differential prompt, not a pathognomonic checklist or fixed timetable. Prognosis depends on macular/optic-nerve damage, systemic disease, bleeding location and complications; a named syndrome alone does not guarantee good vision. 1
Select blood-pressure, blood-count/coagulation and other systemic investigations from the history and findings. Pancreatic enzymes are appropriate when pancreatitis is suspected, not an obligatory panel for every CWS or haemorrhage. Iris/angle neovascularisation with raised pressure requires urgent pressure and retinal-ischaemia management; rubeosis can require action before pressure rises. 8
| Condition | Main finding | Differential/context | Treatment principle | Prognosis and limit |
|---|---|---|---|---|
| Radiation retinopathy | Microangiopathy/CWS/ischaemia | Actual radiation field/dose;15–30 Gy is no universal threshold | Selected anti-VEGF/off-label; laser according to neovascularisation, not rubeosis alone | Progression possible; no universal 60 Gy/50% or 70–80 Gy/90% risk |
| Valsalva retinopathy | Premacular sub-ILM/subhyaloid blood | Trigger/context and OCT; shine alone not definitive | Observation selected;3 months not a mandatory wait; selected Nd:YAG/PPV | Often improves, not guaranteed 1.0 |
| Purtscher/Purtscher-like | Peripapillary flecken/CWS/haemorrhage | Trauma or systemic cause; commotio may coexist | Treat cause; selected observation; comparative treatment uncertain | Variable; low-quality case evidence |
| Terson syndrome | Intraocular haemorrhage with intracranial bleeding | Neurological assessment; original 33% not universal | Observation or PPV individually;4–6 weeks not a universal deadline | Visual recovery possible, with retinal/systemic/surgical limits |
What are the main learning points?
Similar retinal signs can follow very different causes. Radiation dose estimates are context-dependent; Valsalva often resolves but does not guarantee full vision; Purtscher evidence is weak; Terson requires neurological coordination and individual surgical timing.
When is assessment urgent?
Sudden visual or field loss, severe headache, trauma or neurological signs needs urgent assessment; consider intracranial bleeding when appropriate. Severe pain, marked inflammation, pressure rise or suspected infection is also urgent. Examine pupils and RAPD before dilation when feasible without delaying emergency care.
References and access status
- 1Immutable original export; vascular retinopathies lecture; examiner attribution unverified
Historical 136-vascular-retinopathies appraisal reused; not new full-source access. Complete export read; original attribution and any recording/transcript or appointment not independently verified.
- 2ARRT.Randomised aflibercept treat-and-extend for radiation retinopathy,2023.PMID 37351877
Historical 136-vascular-retinopathies appraisal reused; not new full-source access. Complete primary NCBI abstract independently read 15 November 2026; full report, tables, figures, protocol and supplements not appraised.
- 3Purtscher/Purtscher-like systematic review of 168 cases,2024.PMID 39240905
Historical 136-vascular-retinopathies appraisal reused; not new full-source access. Complete primary NCBI abstract independently read 15 November 2026; full report, tables, figures, protocol and supplements not appraised.
- 4Terson surgery/observation retrospective 117 eye series,2024.PMID 38740960
Historical 136-vascular-retinopathies appraisal reused; not new full-source access. Complete primary NCBI abstract independently read 15 November 2026; full report, tables, figures, protocol and supplements not appraised.
- 5Terson vitrectomy retrospective 24 eye series,2024.PMID 39457648
Historical 136-vascular-retinopathies appraisal reused; not new full-source access. Complete primary NCBI abstract independently read 15 November 2026; full report, tables, figures, protocol and supplements not appraised.
- 6Japanese national retrospective aSAH-survivor Terson study,2026.PMID 42397666
Historical 136-vascular-retinopathies appraisal reused; not new full-source access. Complete primary NCBI abstract independently read 15 November 2026; full report, tables, figures, protocol and supplements not appraised.
- 7Valsalva retinopathy 24 eye retrospective management series,2021.PMID 32586109
Historical 136-vascular-retinopathies appraisal reused; not new full-source access. Primary indexed design/conclusion text located 15 November 2026; complete abstract/full report not yet appraised. No numerical efficacy claim taken from it.
- 8Immutable original export; own radiation/Valsalva/Purtscher/Terson lecture; attribution unverified
Complete export read; original attribution and any recording/transcript or appointment not independently verified.
Original lecture link
For patients
Bleeding or pale retinal patches need assessment of both the eye and the possible underlying illness. Some clear naturally; others need treatment and planned follow-up.
Examination and selected tests guide individual treatment and follow-up; this educational text cannot replace your own assessment.
Sudden visual or field loss, severe headache, trauma or neurological signs needs urgent assessment; consider intracranial bleeding when appropriate. Severe pain, marked inflammation, pressure rise or suspected infection is also urgent. Examine pupils and RAPD before dilation when feasible without delaying emergency care.
Sudden visual or field loss, severe headache, trauma or neurological signs needs urgent assessment; consider intracranial bleeding when appropriate. Severe pain, marked inflammation, pressure rise or suspected infection is also urgent. Examine pupils and RAPD before dilation when feasible without delaying emergency care.