OCTA Illustrates Shortcomings of Basing AMD Treatment on Subretinal Fluid
By Paul E. Tornambe, MD, FACS
Using optical coherence tomography angiography (OCTA) daily in my practice, I’m struck by its potential to improve current treatment paradigms in AMD. Because the technology enables direct monitoring (non-invasively) of blood particle flow through the vessels in choroidal neovascular membranes (CNVMs), the decision to start or stop treatment with a VEGF inhibitor (VEGF-I) can be based on the unique morphology of each CNVM. This is, of course, quite different than the current approach of relying on clinical trial data and the surrogate parameters of subretinal fluid detected by OCT B-scans or leakage on fluorescein angiography.
OCTA findings suggest that subretinal fluid is a late marker — and sometimes a false marker — of disease activity that requires treatment. In other words, the presence of subretinal fluid on OCT B-scan can be misleading as a sole determinant of the need for treatment. Although the OCT B-scan is important to review, OCTA findings give us valuable different information to consider as we manage AMD patients. The ability to closely monitor over days or weeks as a CNVM shrinks or grows, becomes non-perfused after a VEGF-I treatment, or re-perfuses after one or more VEGF-I treatments, provides a much more accurate picture of disease activity.
With OCTA, it’s apparent that reperfusion of a CNVM frequently precedes the re-accumulation of subretinal fluid. Fluid is a later finding. But, subretinal fluid doesn’t necessarily indicate an actively leaking CNVM. It may indicate a poorly functioning or non-functioning retinal pigment epithelium (Figure 1). Conversely, the lack of subretinal fluid on OCT B-scan doesn’t necessarily mean a lack of disease activity. A CNVM may be growing and even leaking, but the RPE may be effectively pumping out the fluid (Figures 2 and 3). In general, a recurrence of fluid following a VEGF-I treatment is a sign of actively leaking vessels, while the persistence of fluid despite treatment is a sign that the RPE is unhealthy or the currently employed VEGF-I agent is ineffective.
An additional insight that can be confirmed by OCTA is that not every CNVM that reperfuses will leak. In the case of mature membranes, accumulation of subretinal fluid won’t follow reperfusion. Mature CNVMs don’t leak, nor do they respond to VEGF-I therapy.
Including OCTA in the management of AMD patients is an opportunity to improve care by individualizing it to the specific CNVM. Doing so allows us to avoid delivering VEGF-I treatment that wouldn’t be effective, as in the case of a dysfunctional RPE with no associated CNVM or a mature and, thus, non-leaking CNVM. It also enables consideration of earlier treatment, at the time of CNVM reperfusion, prior to accumulation of subretinal fluid, which may serve to minimize retinal damage and loss of vision. Using OCTA for my AMD patients has allowed me to detect disease activity earlier as well as determine whether a CNVM is mature. The latter gives me a reason to consider stopping VEGF-I treatment and observing. As a result, I’ve been able to give fewer injections.
Figure 1. Subretinal fluid on the OCT B-scan, in the absence of OCTA evidence of a CNVM, indicates RPE dysfunction, not an actively leaking membrane, as the reason for the fluid. A VEGF-I is not indicated even though there is fluid.
Figure 2. The OCTA (top row) clearly shows a type 2 CNVM. The OCT B-scan (bottom row, right) shows no fluid. The RPE bump (red arrows) is the area of the CNV complex. Although the FA shows definite leakage (Figure 3), there is no subretinal fluid on the OCT B-scan. This demonstrates that the absence of fluid on an OCT B-scan may falsely imply there is no CNVM leakage. The RPE is effectively pumping out fluid as quickly as it is produced.
Figure 3. The FA shows leakage in the nasal macula secondary to a CNVM.
Dr. Tornambe, a frequent author and ophthalmic innovator, is founder and president of Retina Consultants San Diego and director of the Retina Research Foundation of San Diego.