OCTA Technology & Principles

OCTA Technology & Principles

OCTA maps blood flow by detecting the motion of red blood cells — no dye, no injection, no referral. It does this by scanning the same retinal location multiple times in rapid succession and comparing the signals. Stationary tissue produces identical signals across scans. Moving blood cells change the reflected signal. That change is the angiographic data.

If you have already worked through Tier 1 and Tier 2, you understand what OCT images: tissue reflectivity. OCTA uses the same optical platform but extracts a different variable — flow. Everything about how OCTA is acquired, interpreted, and used clinically follows from this distinction.

Motion Contrast: How OCTA Works

At each retinal position, OCTA acquires between 4 and 8 B-scans in rapid succession — typically within a few milliseconds of each other. The retinal tissue between scans is effectively stationary: the B-scan signals are nearly identical. Red blood cells flowing through retinal vessels move substantially even in that brief window.

The acquisition software computes the decorrelation (or variance) between the repeated B-scans at each voxel. Where nothing moves, decorrelation is near zero — it appears dark on the en face projection. Where blood cells move, decorrelation is high — it appears bright. The result is a three-dimensional map of blood flow through the retinal and choroidal vasculature.

This motion contrast approach has a fundamental property that distinguishes OCTA from fluorescein angiography: it images flow, not leakage. FA detects fluorescein escaping from vessels through a disrupted blood-retinal barrier. OCTA detects moving blood cells inside intact vessels. These are different physiological phenomena — each has clinical scenarios where it performs better.

Why B-scan repetition rate matters: The ability to detect slow flow (as in capillaries) versus fast flow (as in arterioles) depends on how quickly successive B-scans are acquired. Too slow, and even stationary tissue produces apparent "flow" signal. Too fast, and slow capillary flow doesn't produce enough change to be detected. Modern OCTA platforms are calibrated to detect capillary-level flow — the clinical floor of interest for detecting retinal disease.

Scan Protocols: 3×3mm, 6×6mm, and 12×12mm

Every OCTA acquisition covers a defined retinal area, and that choice determines both what you can see and how reliable the image is. The three standard protocol sizes each serve a different clinical purpose.

3×3mm: Centered on the fovea, this protocol provides the highest spatial resolution for macular OCTA. The scan density is greatest here — the system allocates more A-scans per area unit, producing the sharpest visualization of the FAZ contour, the capillary density in the SCP and DCP, and subtle early non-perfusion. For FAZ metric measurement (area, perimeter, circularity), the 3×3mm protocol is required. For early diabetic retinopathy monitoring where DCP capillary changes are the target, this is the default choice.

6×6mm: The workhorse protocol. It covers the full macular region including the temporal and nasal arcades, making it suitable for AMD monitoring (CNV can arise away from the foveal center), NVE detection in proliferative DR, and standard disease monitoring. Most published OCTA normative databases and disease studies use 6×6mm as the reference scan size. Signal quality is generally good, and motion artifact frequency is manageable for most patients.

12×12mm: Widefield OCTA captures the full posterior pole and extends into the mid-periphery — critical for mapping nonperfusion areas in DR and RVO, detecting peripheral NVE, and characterizing the full extent of ischemic damage in CRVO. The tradeoff: the acquisition window is longer, so motion artifact frequency increases. Patient cooperation is more important with widefield protocols. The Maestro2 platform is capable of 12×12mm OCTA, making it one of the few instruments that provides true widefield OCTA without additional hardware.

Signal Generation: Decorrelation and OMAG

Different OCTA platforms use different mathematical approaches to extract the flow signal from repeated B-scans. While the underlying physics is identical across platforms — all detect the motion of red blood cells — the specific algorithm shapes the noise characteristics and sensitivity profile of the resulting image. Clinicians need enough understanding to interpret images reliably; deep mathematical knowledge is not required.

Split-Spectrum Amplitude Decorrelation Angiography (SSADA): Used by Optovue (AngioVue) and adopted by Zeiss (Angioplex). The OCT signal is split into multiple sub-bands by frequency; decorrelation is computed independently in each sub-band and averaged. This approach increases the signal-to-noise ratio by averaging out speckle noise, which is the dominant noise source in OCT amplitude images. The output is a decorrelation index (0 = no flow, 1 = maximum flow) displayed as a grayscale or false-color en face image.

Optical Microangiography (OMAG): Used by Topcon (including the Maestro2) and also incorporated in Zeiss Cirrus HD-OCT. OMAG analyzes the complex OCT signal — both amplitude and phase — rather than amplitude alone. The phase component carries information about the direction and velocity of scatterer motion, which makes OMAG more sensitive to slow-flow detection, including flow in choriocapillaris-level vessels. For the clinician, the output looks similar to SSADA — a flow map by retinal depth — but OMAG's sensitivity at the choriocapillaris level is a meaningful advantage for detecting early geographic atrophy.

Clinical equivalence: Despite algorithmic differences, the clinical outputs of major OCTA platforms are substantially equivalent for the pathologies encountered in optometric practice. CNV, NPA, FAZ, NVE — all are identifiable on any modern OCTA platform. When reviewing literature or comparing images from different platforms, understand that the algorithms differ, but the biological phenomenon they detect is the same. Interpretation principles transfer across platforms.

OCTA vs Fluorescein Angiography

Fluorescein angiography has been the gold standard for retinal vascular imaging for over 50 years. Understanding where OCTA excels over FA — and where FA still has advantages — defines the clinical context in which OCTA delivers its most value.

What OCTA does that FA cannot:

• Layer-selective imaging: OCTA separates the SCP, DCP, outer retina, and choriocapillaris into discrete en face views. FA projects all layers simultaneously — a Type 1 CNV beneath the RPE is obscured by the overlying retinal vasculature. OCTA isolates each layer.
• No dye, no adverse reactions: FA requires IV injection of sodium fluorescein, which causes nausea in approximately 5% of patients and carries a small but real risk of severe allergic reaction. OCTA eliminates this entirely.
• OD-performed: Because OCTA requires no dye injection, optometrists can perform it without physician oversight or referral. This is the practice-enabling capability that makes OCTA transformative for optometry. CNV detection, capillary non-perfusion mapping, FAZ measurement — all now accessible at the point of care without referral delay.
• Serial monitoring: OCTA can be performed at every visit without any patient burden. Comparing CNV vessel networks or FAZ area visit-over-visit requires no dye, no setup, and minimal additional exam time.

Where FA still has advantages:

• Leakage detection: FA images the breakdown of the blood-retinal barrier directly — hyperfluorescence from leaking vessels is highly visible on FA. OCTA does not detect leakage. Disc leakage in uveitis, diffuse macular leakage in diabetic macular edema, and cystoid leakage in RVO are more directly characterized on FA.
• Wider peripheral field: With widefield lenses, FA can capture the mid- and far-periphery in detail. Widefield OCTA is improving but is currently limited to roughly 12×12mm on the best platforms.
• Dynamic imaging: FA shows the arteriovenous transit — the temporal pattern of dye filling from arteries to veins — which can characterize arteriovenous malformations and choroidal perfusion dynamics not visible on OCTA.