Dense vitreous haemorrhage in a healthy young man six weeks before examinations
A 24-year-old male university student presents with a two-day history of floaters followed by painless loss of vision in the left eye. He has no history of trauma, no known diabetes, no sickle cell disease and takes no medication. Visual acuity is 6/36 in the left eye and 6/6 in the right. The left anterior segment is quiet with no cells, the intraocular pressure is 14 mmHg, and a dense vitreous haemorrhage obscures all fundal detail. B-scan ultrasonography shows an attached retina, no posterior vitreous detachment, no mass and no choroidal thickening. The right fundus is normal. Random blood glucose is 6.1 mmol/L, blood pressure 118/74 mmHg, and full blood count and coagulation screen are normal. He is due to sit final examinations in six weeks and asks whether he can simply wait and be reviewed afterwards.
1. What is your differential diagnosis for the vitreous haemorrhage, and how does the normal B-scan change your immediate approach?
Read the worked answer
In a healthy 24-year-old with no diabetes and a normal fellow fundus, the differential is narrow but each item changes management:
- Eales disease (idiopathic retinal periphlebitis) is the leading candidate in this age group and in regions where it is common. It typically causes recurrent vitreous haemorrhage from peripheral periphlebitis and neovascularisation, often in otherwise healthy young men.
- Retinal tear or detachment - less likely without myopia or posterior vitreous detachment on B-scan, but not excluded once the view clears.
- Haematological disease - leukaemia, thrombocytopenia, sickle cell retinopathy, bleeding diathesis; the normal full blood count makes these less likely but does not exclude them.
- Unrecognised trauma, and rarely an intraocular tumour or a retinal vascular anomaly.
- Terson syndrome should be considered only with a relevant history; there is none here.
The normal B-scan is the decisive finding. An attached retina with no detachment, no mass and no choroidal thickening means there is no surgical emergency and I can safely observe while the blood settles. Had there been a retinal detachment or a suspicious mass, I would proceed to vitrectomy without waiting. I would also check the intraocular pressure at each visit, since a large haemorrhage can produce ghost-cell glaucoma, and I would examine the fellow eye carefully with indirect ophthalmoscopy and scleral depression for peripheral lesions.
The plan is serial B-scans and clinical review every 2–3 weeks, with the explicit instruction to return immediately for pain, further visual loss or new floaters.
2. How would you investigate and treat him, and how do you answer his question about waiting until after his examinations?
Read the worked answer
His question deserves a direct answer: the retina is attached and there is no reason for emergency surgery, so a short period of observation is safe and he can sit his examinations. But 'wait and be reviewed afterwards' is not acceptable, because the cause must be found and untreated neovascularisation can bleed again or produce tractional detachment.
Investigation should proceed now, not in six weeks: repeat full blood count and film, HbA1c, sickle cell screen, coagulation studies, blood pressure, chest radiograph, and tuberculosis testing (Mantoux or interferon-gamma release assay) with an angiotensin-converting enzyme level if sarcoidosis is suspected. Eales disease remains a diagnosis of exclusion, so a careful systems review for Behçet disease, inflammatory bowel disease and sarcoidosis is part of the work-up.
Management while the view is opaque: serial B-scan ultrasonography every 2–4 weeks, pressure checks, and no anticoagulant or antiplatelet treatment unless medically indicated. I would not prescribe empirical systemic corticosteroids for presumed Eales disease before neovascularisation is demonstrated, because steroids can worsen some infective and malignant mimics and have not been shown to alter the natural history reliably.
Once the vitreous clears, examine the periphery with scleral indentation and perform fluorescein angiography. Peripheral non-perfusion or neovascularisation is treated with panretinal photocoagulation; active periphlebitis may be treated with local or systemic corticosteroid after infection is excluded. I would proceed to vitrectomy if the haemorrhage fails to clear by about 8–12 weeks, if there is tractional or rhegmatogenous detachment, or if ghost-cell glaucoma develops. Because he is uniocular for driving and study, I would counsel on eye protection and on the warning symptoms of rebleeding.