Optical coherence tomography (OCT) has revolutionized eyecare by allowing us to visualize the retina, choroid, and optic nerve in near-histological detail with significant ease and reproducibility. OCT angiography (OCT-A), the next evolution of this technology, allows for the non-invasive imaging of ocular vasculature.
The Eyes On Eyecare 2024 Retina Report found that 52% of optometrists surveyed don’t have or use OCT-A in their practice. For clinics, its successful adoption hinges on understanding coding, reimbursement, clinical utility, and operational integration.
What is OCT angiography?
OCT-A relies on motion contrast to generate images of ocular vasculature. The area of interest is scanned several times in quick succession, and then the OCT-A algorithm detects differences between the scans.
Because intraocular tissue doesn’t move over the course of a few milliseconds, any movement detected by the OCT-A algorithm is assumed to be secondary to the movement of blood within vessels. This motion signal is then used to create three-dimensional maps of ocular vasculature called OCT angiograms.
These OCT angiograms are viewed in en face images called slabs, which can be generated for various portions of the retinochoroidal or optic nerve vasculature, allowing for localization of vascular abnormalities. OCT angiograms are usually analyzed in tandem with structural OCT, allowing for the localization of vasculature to specific structural layers.
Depending on the OCT-A system being used, various sizes of OCT-A scan protocols may be available. Typically, the larger the scan size, the lower the resolution of the scan. For example, an 8x8mm OCT-A scan will provide less detail than a 3x3mm OCT-A scan.
Figure 1: Retina slab of a normal patient in scan protocol sizes 3x3mm, 6x6mm, and 8x8mm. The anterior and posterior borders of the slab are depicted by the dotted purple lines in the structural OCT below each OCT angiogram.
Figure 1: Courtesy of Daniel Epshtein, OD, FAAO.
Interpreting OCT-A scans
Though not a replacement for dye-based angiography such as fluorescein angiography, OCT-A can help detect and differentiate vascular lesions with a quick, non-invasive, in-office image.
When analyzing OCT angiograms, it is important to keep a few things in mind:
- Any movement within a scan is depicted as white or hyper-reflectivity, correlating to vasculature or perfusion
- Lack of movement within a scan is depicted as black or hypo-reflective, correlating to a lack of vasculature or nonperfusion
- Ideally, all en face slabs should be analyzed, but specific slabs may yield greater clinical utility depending on the ocular condition being evaluated.
Table 1: Short reference guide on using OCT-A for various conditions.
| Disease Entity | Vascular Pathology | Area of Concern | OCT-A Slab |
|---|---|---|---|
| AMD | Choroidal neovascularization | Outer retina/anterior choroid | Outer retina and anterior choroid |
| Diabetes | Retinal nonperfusion, neovascularization elsewhere | Retina, vitreoretinal interface | Whole retina, vitreoretinal interface and posterior vitreous |
| Glaucoma | Retinal nerve fiber layer (RNFL) vessel nonperfusion | Peripapillary | Radial peripapillary capillaries |
OCT-A in age-related macular degeneration
One of the most important tenets of AMD management is that early detection and treatment of neovascular AMD (nAMD) will produce better visual outcomes. Though structural OCT may identify nAMD, mild cases are often equivocal and require angiography to confirm the diagnosis.
Clinicians can use OCT-A to identify choroidal neovascularization (CNV) by looking for abnormal vasculature within the outer retina and/or choroid. CNV will most often present as branching bundles of vessels or filamentous vascular complexes. If these OCT-A patterns are noted within the outer retina or choroid and they are associated with subretinal and/or intraretinal fluid, CNV is essentially confirmed.
With the advent of OCT-A, nAMD has been further refined into exudative nAMD and nonexudative nAMD. Any CNV with associated fluid secondary to AMD is considered exudative nAMD, while nonexudative nAMD is diagnosed when nAMD is not associated with any fluid.
Discerning nonexudative and exudative AMD on OCT-A
Nonexudative nAMD is most often detected within shallow, irregular pigment epithelial detachments but may also present as vascularized drusen. Clinically and structurally, nonexudative nAMD often appears as intermediate AMD but requires OCT-A or indocyanine green angiography to visualize the neovascular complexes and make an accurate diagnosis.
In studies of patients with exudative nAMD in one eye and intermediate AMD in the fellow eye, 5.5 to 27% of intermediate AMD eyes were reclassified as nonexudative nAMD with the use of OCT-A.1,2
Though nonexudative nAMD typically doesn’t require treatment, up to 80% of nonexudative nAMD will convert to exudative nAMD within 2 years.2 Consequently, the prompt identification and proper monitoring of nonexudative nAMD is imperative to the early detection of treatable nAMD.
Figure 2: (A) Exudative CNV, (B) nonexudative CNV, and (C) disciform scar. Pigment epithelial detachments are visualized with structural OCT in all three cases, but OCT-A reveals CNV in (A) and (B) only. Note the similarity of the shallow pigment epithelial detachment in (B) and (C).
Figure 2: Courtesy of Daniel Epshtein, OD, FAAO.
OCT-A in diabetes
Though mild nonproliferative diabetic retinopathy (NPDR) is considered the earliest form of clinical diabetic retinal pathology, OCT-A can visualize subtle retinal nonperfusion and foveal avascular zone (FAZ) degeneration before clinically evident microaneurysms or hemorrhages appear.3,4
As diabetic patients progress from preclinical NPDR to severe NPDR, areas of retinal nonperfusion increase, and the FAZ will enlarge and/or lose circularity.5 Detailed imaging of retinal nonperfusion and the FAZ are best achieved with an OCT-A slab that can visualize vessels throughout the whole thickness of the retina.
Figure 3: FAZ degeneration and nonperfusion noted in advancing stages of NPDR. A normal OCT-A scan is provided for reference.
Figure 3: Courtesy of Daniel Epshtein, OD, FAAO.
Progressive diabetic retinal ischemia may result in upregulation of angiogenic factors such as vascular endothelial growth factor (VEGF), leading to neovascularization along the vitreoretinal interface. Similar to nAMD, early identification and appropriate intervention of proliferative diabetic retinopathy will often improve visual outcomes.
Though intraretinal microvascular abnormalities (IRMA), vessel remodeling, or hemorrhage may sometimes complicate the identification of diabetic neovascularization, OCT-A can detect and localize pathological vasculature to specific vitreoretinal locations to simplify diagnoses.
OCT-A slabs imaging the posterior vitreous and retinal surface, which are normally black and void of signal, will have a hyperreflective lacy lesion corresponding to neovascularization on the retinal surface. This same technique can be used for the assessment of neovascularization of the optic nerve head.
Figure 4: Neovascularization elsewhere is observed within the superior-temporal arcades. Note the areas of nonperfusion in the temporal midperiphery.
Figure 4: Courtesy of Daniel Epshtein, OD, FAAO.
OCT-A in glaucoma
Though it is still unknown whether vascular changes in glaucoma occur before, during, or after structural and functional changes, recent evidence has shown that OCT-A may help clinicians manage glaucoma patients. OCT-A can visualize blood vessels that nourish the RNFL, overall optic nerve head, and ganglion cell bodies within the peripapillary, papillary, and macular areas, respectively.
Nonperfusion in all three of these areas has been shown to correlate with structural RNFL and/or ganglion cell-inner plexiform layer (GC-IPL) thinning. OCT-A vessel density measurements of the radial peripapillary capillaries, in particular, have been shown to correlate well with structural OCT and visual field functional glaucomatous loss.
Figure 5: Radial peripapillary capillary (RPC) slab reveals superior-temporal nonperfusion, which correlates with superior-temporal RNFL thinning on the RNFL deviation map. Note the decreased amount of hyper-reflectivity within the red oval as compared to the green oval.
Figure 5: Courtesy of Daniel Epshtein, OD, FAAO.
As glaucoma worsens, OCT-A vessel densities decrease, though at a different rate than structural OCT thickness parameters. OCT-A vessel densities seem to correlate better with visual field sensitivity levels than structural RNFL thickness measurements, possibly providing an objective surrogate to functional visual function in glaucoma patients.6
As with structural OCT thickness measurements, OCT-A vessel densities can be compared over time to monitor for progression. Due to the floor effect of RNFL thickness measurements, where no further RNFL thinning occurs even though visual field loss continues to worsen, structural OCT is often not very useful in advanced glaucoma.
OCT-A parameters seem to have a lower floor effect than structural OCT, providing an objective measurement even in advanced glaucoma.7
Coding and coverage guidelines for OCT-A
CPT code 92137 encompasses bilateral OCT-A imaging of the posterior segment, including interpretation. Typically, payers reimburse for this service when it is used to manage retinal pathologies. On average, Medicare reimburses $57 per test ($23 technical, $34 professional component).
- Covered Diagnoses: May include diabetic retinopathy (E10.31–E11.3599), age-related macular degeneration (H35.31–H35.32), retinal vascular occlusions (H34.0–H34.9), and choroidal neovascularization (H35.05).
- Billing Restrictions: CMS and commercial payers prohibit same-day billing with other scanning computerized ophthalmic diagnostic imaging (SCODI) codes (e.g., 92132, 92133, 92134).
This makes sense when we consider that elements of 92133 and 92134 are included in an OCT-A evaluation. Additionally, Medicare and other payers may limit reimbursement to once monthly per patient for active retinal disease management.
Practice Pearl: For now, OCT-A can be performed on the same day as automated perimetry 92083 and fundus photography 92250.
Reimbursement challenges with OCT-A
If you receive a denial, it may be due to any of the following reasons:
- Prior Authorization Requirements: Some carriers mandate proof of medical necessity, particularly for non-diabetic diagnoses. To this end, it is important to document your clinical suspicion and reason for additional evaluation in your clinical order.
- Frequency Limits: Since FA (fluorescein angiography) is already utilized by retina practices, insurers could adopt fluorescein angiography’s reimbursement policies (e.g., two to five scans per year), creating staff appeals for exceptions.
- Smaller Payer Recognition Implementation: These carriers may lag in updating systems to recognize 92137. This could necessitate manual claim submission.
- Denial Rate: It has been reported that denial rates may range from 15 to 25%, so be prepared!
Practice Pearl: Practices should obtain advance beneficiary notices (ABNs) for Medicare patients when coverage is uncertain.
Also, be prepared to provide comments about some of the following in your Interpretation and Report:
- Vascular density metrics
- Structural change with B-scans and vessels
- Management considerations (e.g., changed patient’s nutraceutical regimen, initiated PMB, referred for anti-VEGF therapy due to new CNV)
Integrating OCT-A into workflows
Technicians typically require 3 to 5 minutes per OCT-A scan. A variety of factors can impact this. These may include ocular surface conditions, cataracts, or even neurological defects, making it difficult for patients to sit or focus.
If there are unreliable or abnormal findings, be prepared to order additional testing such as traditional FA, ICG angiography, or microperimetry.
Return on investment with OCT-A
The initial capital investment for OCT-A systems ($50,000 to $90,000) can be recovered relatively quickly.
Mature Practice:
Assuming 15 weekly scans at $57 reimbursement, practices may recover costs within 18 to 24 months.
- Calculation: 15 x $57 x 48 weeks = $41,040 per year
Emerging Practice:
Assuming 8 weekly scans at $57 reimbursement, practices may recover costs within 36–48 months.
- Calculation: 8 x $57 x 48 weeks = $21,888 per year
Indirect benefits of OCT-A include:
- Enhanced patient retention
- Reduced unnecessary referrals for retinal pathologies. Some indications are that this may reduce referrals by 30 to 40%
- Practice builder from enhanced patient confidence
Conclusion
OCT-A can enhance diagnostic precision in retinal and optic nerve diseases while minimizing unnecessary referrals and invasive testing. Its implementation doesn’t require additional staff time and requires only minimal training.
Ultimately, if a clinic adheres to the coding guidelines and documentation requirements and deploys the technology for many conditions that already exist in the practice, they can and will achieve both clinical and financial success.
