OCT angiography moves first, and it costs you some specificity

Design

retrospective cohort study of event-based progression, with test-retest variability thresholds and guided progression analysis

Population

180 eyes (111 perimetric, 69 preperimetric) from 116 patients in the Diagnostic Innovations in Glaucoma Study; mean age 67.6 years, mean follow-up 5.1 years

Primary outcome

rate and timing of progression detection by OCT angiography vessel density, OCT nerve fibre layer thickness and visual field

Effect

progression in 32.2%, 25.0% and 32.8% respectively; OCT angiography lead time over visual field 2.3 years (95% CI 1.3 to 3.2); specificity 68.9% vs 75.6% for OCT

This retrospective cohort from the Diagnostic Innovations in Glaucoma Study asked a narrow question with practical consequences: when three monitoring modalities are all being used, which one declares progression first? It included 180 eyes - 111 perimetric, 69 preperimetric - from 116 patients with at least two years of follow-up and at least four visits on each modality, imaged between 2015 and 2023.

Over a mean 5.1 years, 107 eyes (59.4%) progressed on at least one modality: 32.2% on OCT angiography vessel density, 25.0% on OCT circumpapillary nerve fibre layer thickness, 32.8% on visual field. Among progressing eyes, the earliest detection was by OCT angiography in 37.4%, by OCT in 29.9% and by perimetry in 24.3%. Mean lead time over visual field progression was 2.3 years for OCT angiography (95% CI 1.3 to 3.2) and 1.5 years for OCT (0.6 to 2.4); between the two structural measures the difference was 0.2 years with an interval crossing zero.

The cost is stated honestly by the authors. In 45 eyes judged stable, specificity was 68.9% for OCT angiography against 75.6% for OCT - so roughly three in ten stable eyes were flagged as progressing by vessel density. In a disease monitored for decades, a modality that calls progression early and is wrong three times in ten will generate treatment escalation that was not needed. This supports a complementary role, which is what the paper claims, and not a replacement for either existing modality.

• Read a vessel density change as a prompt to look harder, not as a decision to escalate treatment.
• Confirm any single-modality progression signal against the other two before changing management.
• Note the specificity gap: 68.9% for OCT angiography against 75.6% for OCT in stable eyes.
• The structural measures did not separate from each other on timing - the 0.2 year difference had an interval crossing zero.
• Cost-benefit of adding a third modality was not assessed here, and matters most where imaging capacity is the constraint.

The statistics, in plain English

Lead time here is measured only among eyes that progressed on both modalities, so it describes the head start when both eventually agree - not how often the early signal turns out to be real. That question is answered by the specificity figures, and 68.9% is low for a monitoring test used repeatedly: applied at every visit over years, it will produce false alarms faster than it produces true ones in a mostly stable population. The comparison between OCT and OCT angiography, 0.2 years with an interval of -0.6 to 0.9, is a genuine null - the two structural measures moved at effectively the same time. With 180 eyes and 45 stable comparators, all these estimates are imprecise.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.