Everywhere I turn recently, it seems that someone is demonstrating, lecturing, or asking about optical coherence tomography angiography (OCT-A). My inbox is flooded with invitations to OCT-A conferences, my fellows are focused on finding great OCT-A case examples, and OCT-A research projects abound. OCT-A images of the parafoveal capillaries are often stunning in their ability to reveal the architecture of the smallest retinal vessels (Figure 1). But what is the true value of these images? Can we use our improved ability to view the retinal vasculature to guide patient care? Or will OCT-A be a passing fad that provides pretty pictures without evidence-based benefit?
Since its introduction in the 1990s, standard OCT imaging of the retina has revolutionized the field. OCT retinal thickness maps and B-scan cross-sections of the neural retina are widely used to guide management recommendations for diabetic retinopathy, age-related macular degeneration, retinal vein occlusions, and a host of other retinal diseases. But current OCT parameters are not enough—retinal thickness alone is only modestly associated with present or future visual outcomes. Many of us use other morphologic traits, such as the presence of intraretinal or subretinal fluid, to also help guide treatment decisions, but these abnormalities have not been linked precisely to visual outcomes and they require time-consuming review of individual B-scans. Thus, the search continues for additional OCT parameters that are more reliably associated with visual acuity.
Will OCT-A provide such new measures that can be used to guide patient care? I believe that it has the promise to do so. OCT-A technology generates an en face map of blood flow by detecting changes in reflectance from light-scattering particles, such as blood cells, over multiple frames acquired of the same area over time. The method that produces this high-resolution visualization of the retinal vasculature is quickly performed and easily tolerated by most patients. Moreover, the technique provides a view of the smallest retinal capillaries that is similar to or even better than fluorescein angiography, but without the need for invasive dyes, avoiding adverse effects such as injection-related pain, lightheadedness, nausea, or allergic reactions. An unprecedented feature of OCT-A lies in its ability to distinguish between superficial and deep capillary layers. OCT-A images can also be used to identify alterations in outer retinal and choriocapillaris flow, such as those associated with choroidal neovascular membranes. Quantitative estimates of capillary density and nonperfusion area can also be obtained with the use of automated algorithms.
There are limitations of the OCT-A technique that are important to recognize. Structures with slower blood flow may not be apparent on motion contrast images. Microaneurysms that are clearly visible on fluorescein angiography or color fundus photographs are not present on OCT-A if changes in reflectance due to blood flow in these structures falls below threshold values. OCT-A images also do not reveal vessels that are no longer perfused. In addition, leakage from abnormal retinal vascular permeability that can easily be appreciated with the use of angiographic dyes cannot be visualized on OCT-A.
So what are the best uses of OCT-A right now? My own use of OCT-A in clinical practice is still evolving. My patients always love seeing pictures of their own eyes, and OCT-A images are yet another tool to help them better understand how their eyes are affected (or unaffected) by retinal abnormalities. In the case of patients with diabetes, showing fundus images including, but not limited to, OCT-A can be highly motivating to encourage improvements in systemic glycemic, hypertensive, and cholesterol control to avoid further retinal damage. Demonstrating a good treatment response or illustrating the recurrence of abnormalities through individual images also helps to motivate patients to follow up regularly and receive treatments when needed. I’ve used OCT-A to identify the onset of choroidal neovascular membranes and to document growth or regression of ocular neovascularization.
As gratifying as it is to acquire these beautifully detailed images of the retinal vasculature, I’m still unclear how to use all of the data that can be generated. Accompanying measures of capillary density and nonperfusion are readily available for each image set, but there is a lot of individual variability in these measures and there are no well-established normative databases with which I can compare an individual patient’s values. We really need prospective, well-designed clinical studies to help us understand whether and how specific changes on OCT-A imaging are correlated with anatomic and visual outcomes across the spectrum of retinal diseases. Large-scale, multicenter studies of OCT-A findings in relation to diabetic eye disease are getting underway and hopefully results from these and other studies that are ongoing will help to refine the use of this nascent technology.
OCT-A is one of the hottest developments in in the field of retina right now, and yet we don’t currently have well-defined strategies to optimize its use in clinic or research settings. Only time will tell as to whether this technology is truly transformative or just another toy in the imaging suite. I, however, am optimistic that ongoing clinical studies will help us uncover new and more effective ways to use OCT-A to better the lives of our current and future patients.