OCTA in AMD: Practical Scan and Segmentation Strategies for Real-World Clinical Scenarios

AMD presents with diverse structural changes and vascular complications. OCTA provides non-invasive, depth-resolved visualization of retinal and choroidal blood flow, but interpretation depends on the clinical scenario, scan size, segmentation boundaries and artifact recognition.

Objective: To offer practical imaging protocols that help clinicians select suitable OCTA scan areas, cross-sectional views and en face slabs for common AMD presentations.

• Non-invasive vascular imaging: OCTA detects flow-related decorrelation signals without intravenous dye injection.
• Depth-resolved assessment: Cross-sectional OCTA overlays flow signals on structural B-scans; en face images display flow within a selected slab.
• Scan-size trade-off: Larger fields cover more retina, but image detail generally declines as scan size increases. The review commonly recommends 3 × 3 mm or 6 × 6 mm scans when high-resolution macular assessment is needed.
• Important limitation: OCTA demonstrates flow, not leakage. Structural OCT remains essential to assess fluid and other signs of exudation.

• ORCC slab: Covers the outer retina through choriocapillaris region and can capture neovascular components both below and above the RPE.
• RPE–RPEfit slab: Can improve contrast between type 1 MNV and surrounding tissue when segmentation is reliable and above-RPE components are not the main concern.
• Choriocapillaris (CC) slab: May be useful in geographic atrophy when RPE atrophy disrupts conventional segmentation; the slab can sometimes be manually adjusted using Bruch’s membrane as a reference.
• Cross-sectional OCTA first: Localize flow relative to the RPE, Bruch’s membrane, subretinal hyperreflective material (SHRM) or an outer-retinal lesion before using en face images to understand network morphology.

Type 1 MNV: Suspect flow between the RPE and Bruch’s membrane, often in association with a shallow, irregular pigment epithelial detachment (PED) or a thick double-layer sign. Use cross-sectional OCTA, followed by en face ORCC or RPE–RPEfit images. With a tall serous PED, segmentation may fail and fluid can obscure the lesion; examine the PED margins carefully.

Type 2 MNV: SHRM should raise suspicion, but is not specific for neovascularization. Confirm flow within SHRM on cross-sectional OCTA; the ORCC slab is recommended for en face assessment because it encompasses the outer-retinal lesion and may capture components beneath the RPE.

Type 3 MNV: Assess whether an outer-retinal hyperreflective lesion contains flow. Cross-sectional images are central to confirmation; en face morphology may be less distinctive. Exclude projection artifacts from the superficial or deep vascular complexes and distinguish these lesions from deep retinal age-related microvascular anomalies (DRAMAs).

In geographic atrophy (GA), structural OCT remains the primary modality. OCTA can help identify MNV complicating atrophy and examine choriocapillaris perfusion around the atrophic area. Standard segmentation can be distorted by atrophy; ORCC or a carefully adjusted CC slab may provide alternatives, while each remains susceptible to artifacts or structural distortion.

In early or intermediate AMD, non-exudative type 1 MNV may appear as a shallow irregular PED with a double-layer sign but no clear structural OCT exudation. Non-exudative type 3 MNV may appear as an outer-retinal hyperreflective focus without definite intraretinal fluid. OCTA can identify flow and inform closer follow-up, but treatment is not typically indicated solely for non-exudative MNV; monitoring should be individualized.

Acquired vitelliform lesions: These can resemble SHRM. OCTA should assess for flow within the material or in a flat irregular PED beneath it. The review favors ORCC in many cases because it can include both regions. Fundus autofluorescence may provide additional differentiation: acquired vitelliform material is typically hyperautofluorescent, whereas exudative SHRM is often iso- or hypoautofluorescent.

Intraretinal fluid without obvious MNV: OCTA is useful to exclude MNV, particularly type 3, and to assess possible DRAMAs. Such anomalies can contain detectable flow and may be associated with fluid. Correct classification matters because treatment implications differ.

Non-exudative subretinal fluid: Fluid may occur with large drusenoid PEDs. Structural OCT helps distinguish drusenoid PEDs, which typically have higher internal reflectivity, from serous PEDs, which tend to be more hyporeflective. With serous PED, examine the margins for a shallow irregular PED and possible type 1 MNV. In large serous PEDs, OCTA segmentation may be inaccurate, so fluorescein angiography or indocyanine green angiography may still be needed.

Degenerative intraretinal pseudocysts/cavitations: These may occur in areas of outer retinal atrophy. Structural OCT and the absence of peri-cystic perfusion can support a degenerative explanation, but OCTA remains important for excluding MNV or retinal microvascular anomalies when clinically suspected.

• Projection artifacts: Apparent flow in an outer-retinal lesion may be projected from superficial or deep retinal vessels. Check the same location across layers and interpret cross-sectional images carefully.
• Segmentation errors: PEDs, atrophy and distorted anatomy may misplace slab boundaries. Inspect segmentation and manually adjust where the device permits.
• Absent flow is not always definitive: High-flow turbulence in aneurysmal lesions can fall below OCTA detection thresholds; some lesions may therefore lack a detectable signal.
• OCTA does not establish exudation: Evaluate structural OCT for intraretinal fluid, subretinal fluid, SHRM and related changes. FA or ICGA may remain useful when OCTA is inconclusive.

This article is an expert consensus-based review integrating literature and clinical experience; it does not report a newly enrolled patient cohort, randomized comparison, pooled effect estimate or formal diagnostic-accuracy analysis. The authors note that devices may use different segmentation strategies and may not all permit manual boundary adjustment, creating heterogeneity that can limit direct transfer of a protocol between platforms. Quantitative OCTA metrics remain methodologically heterogeneous, and the review does not establish a single universal quantification standard.

OCTA is most useful when its scan area, cross-sectional assessment and en face slab are selected for the suspected lesion and interpreted alongside structural OCT. The practical sequence is to identify the clinical question, inspect cross-sectional flow, choose the slab that best captures the relevant anatomy, verify segmentation and artifacts, and use structural OCT to judge exudation. Tailored imaging is more reliable than applying one slab or scan strategy to every AMD presentation.