- Design
- retrospective observational study at a tertiary eye care centre, independent reading of OCT and fundus examination
- Population
- 10,819 eyes of 6,539 patients assessed for cataract surgery, 2018 to 2023
- Primary outcome
- detection of macular abnormalities and effect on surgical planning
- Effect
- sensitivity 80%, specificity 95% vs fundus examination; plan modified in 0.7% of all eyes
A tertiary Indian eye centre reviewed 10,819 eyes of 6,539 patients assessed for cataract surgery between 2018 and 2023, all of whom had undilated macular OCT alongside fundus examination, with retina specialists reading each independently.
Fundus examination found abnormalities in 18% of eyes and macular abnormalities in 10%. OCT found macular abnormalities in 7%, comprising 57% of the eyes already abnormal on fundus examination and 2% of the 8,863 eyes with clinically normal fundi. Against fundus examination, OCT had sensitivity 80% (95% CI 0.769 to 0.825) and specificity 95% (0.949 to 0.957), with a 5.8% misclassification rate. OCT findings changed the surgical plan in 9% of eyes with an abnormal scan - which is 0.7% of all eyes assessed. Postoperative visual outcomes did not differ between eyes with abnormal and normal scans (P = 0.12).
The authors' conclusion is the honest one, and it runs against the direction of travel in busy cataract practice: a careful dilated fundus examination remains the cornerstone, and universal preoperative OCT adds diagnostic yield without changing much. The 2% pick-up in clinically normal fundi is the number worth arguing about - it is not nothing, and some of those eyes matter for setting expectations about postoperative vision even if the operation itself is unchanged. The defensible position is selective OCT: scan where the fundus view is poor, where the history suggests macular disease, where the patient is diabetic, or where a premium or multifocal lens is planned and the visual expectation is high. Scanning everyone instead of examining everyone properly is the failure mode this paper describes.
- Keep a careful dilated fundus examination as the primary preoperative assessment
- Order macular OCT selectively: poor view, suggestive history, diabetes, or a premium lens planned
- Use an abnormal scan mainly to set visual expectations, since it rarely changes the operation
- Do not treat a normal OCT as replacing fundus examination - sensitivity was 80%
- Audit locally how often your preoperative scans change anything before making them routine
Why it matters
Universal preoperative OCT is becoming routine on the assumption it changes management, and in 10,819 eyes it changed 0.7%.
Don't overread it
Retrospective, single-centre, and using fundus examination as the comparator rather than an independent reference standard.
The statistics, in plain English
Sensitivity and specificity here are measured against fundus examination rather than against a true reference standard, so they describe agreement between two imperfect tests, not accuracy. The figure that matters for the decision is 0.7% - the share of all assessed eyes whose management changed - because a test's diagnostic yield is only worth the cost if it changes something. P = 0.12 for postoperative outcomes means no difference was demonstrated, in a retrospective series where eyes with abnormal scans may have been managed differently.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for ophthalmology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free