ULTRA-WIDEFIELD IMAGING

Ultra-Widefield Imaging Q&A

BY SUSAN WORLEY, CONTRIBUTING EDITOR

Widefield imaging captures more of the retina than ever before, paving the way for more complete understanding of the extent of retinal vascular disease and how to treat these conditions. New Retinal Physician recently spoke with Charles C. Wykoff, MD, PhD, to discuss his experience with widefield imaging technology.

Q: When did you first begin using widefield imaging?

A: Widefield imaging has been around for a while. With early retinal cameras, of course, it was possible to take multiple images that could be montaged to create a sort of pseudo-widefield image. The earliest single-capture approach involved use of the Staurenghi lens, a large lens that was very effective for capturing a wide image of almost the entire retina, using fluorescein angiography (FA) on a Heidelberg-based imaging platform. It expanded the field of view from 30 degrees to approximately 150 to 160 degrees.

I became comfortable with this approach when we used the Ocular Staurenghi 230SLO Retina Lens during the RAVE trial.¹ In this prospective trial, we performed serial widefield angiographic imaging on 12 eyes with ischemic central retinal vein occlusions (CRVO) over a 3-year time period, which allowed us to observe extensive retinal non-perfusion at baseline that increased while the eyes received anti-VEGF therapy with ranibizumab. That was my initial exposure to the potential value of widefield imaging, and, subsequently, I became very interested in using the Optos imaging platform.

Q: What are the primary modalities used today?

A: The most broadly used widefield imaging platform is from Optos, which many prospective trials are using as the standard of care, and the Heidelberg Engineering system, which is also very good. These platforms have largely replaced the Staurenghi Lens, which is a good lens but can be somewhat cumbersome. The Optos imaging system comprises a standalone camera in one of three platforms — one that takes just fundus photographs, another that enables color fundus photography with FA, and a third (the California) that allows both FA and indocyanine green angiography.

The other option is manufactured by Heidelberg and offers a field that is not quite as wide. Both systems are very good; however, I seem to get better resolution and more detailed imaging of capillaries in the far periphery with the Optos system. We have both in our practice, and although I use the Optos system more frequently, I regularly use Heidelberg for comparative imaging. There are a number of handheld modules, such as the Retcam (Clarity Medical Systems), that are also widely used, mainly for pediatric retina.

Q: Could you talk a bit more about your use of widefield imaging in research?

A: After the RAVE study, I became interested in the development of de-warping mechanisms that might allow one to account for the distortions that occur when a 3-D structure is projected onto to a 2-D image. We struggled with this for a long time, and examined various ways of correcting for the distortions. We worked closely with Optos while developing successful software.^2,3

Another major paper on de-warping retinal images was published by SriniVas R. Sadda, MD, and colleagues.⁴ Dr. Sadda is at the forefront of imaging, and one the most prominent experts on de-warping. Our approaches were similar. More recently, we are working toward trying to understand the clinical implications of accurate retinal ischemia quantification.

Widefield imaging is allowing a more complete understanding of the extent of retinal vascular disease, in conditions such as diabetic retinopathy (DR) and retinal vein occlusion (RVO). Previously, we had no way to quantify the true surface area of retinal non-perfusion and were limited to quantifying pixels, which may not accurately or fully represent the true extent of pathology. The hope is that accurately quantifying areas of pathology and retinal non-perfusion may lead to better correlation with prognostication and treatment variables.

We’re analyzing data from two prospective trials, WAVE and DAVE, both of which are examining patients with extensive retinal ischemia due to RVO and diabetes, respectively, with the goal of creating a model for prognostication and correlation with outcomes.

Q: Can you mention another example of notable research using widefield imaging?

A: The Diabetic Retinopathy Clinical Research Network (DRCR.net) has a 5-year observational study in progress called DRCR-AA,⁵ which involves widefield imaging of patients annually and correlation of DR at baseline with outcomes. The goal is to be able to predict disease progression based on peripheral pathology — that is, does the use of clinical information obtained from the periphery change prognostication from what can be obtained from clinical data contained in the posterior pole? This study is essentially updating a portion of the EDTRS, which was conducted using just seven standard fields that captured about 90 degrees of retina, or about 30% of the total retina surface area. In comparison, with Optos imaging, about 80-90% of the retinal surface area can be captured. DRCR-AA is fully enrolled, and results will be forthcoming when the trial is complete. Lloyd Paul Aiello, MD, PhD, and colleagues have published reports evaluating the value of widefield retinal findings, and have shown that they are of substantial clinical value for prognostication.⁶

Montaged widefield fluorescein angiogram demonstrating multifocal zones of retinal non-perfusion and vascular leakage secondary to diabetic retinopathy.
Image taken using the 200Tx by Optos

Q: Could you tell us about your use of widefield imaging in clinical practice?

A: Major categories of pathology for which widefield imaging is valuable include retinal vascular diseases, inflammatory diseases, and mass lesions, where precise longitudinal area measurements may be valuable. The classic example in my clinical practice is the use of widefield imaging in eyes with DR.

For patients who are asymptomatic, or when there is only mild non-proliferative DR, or when I am not considering treatment for the patient, I don’t typically perform FA. However, if a patient has DME that needs to be treated or has proliferative DR, I will obtain baseline widefield FA and use the extent of retinal capillary nonperfusion as a guide to determine how sick the eye is, and what the patient’s risk might be going forward. There is good data from anti-VEGF treatment trials in eyes with DME causing vision loss showing that repeated injections slow the loss of retinal capillaries, and may improve vascular status in some eyes.⁷ It’s one thing to be able to see and quantify the extent of retinal non-perfusion with widefield imaging, but another to be able to do something about it.

We need more data about the role of using anti-VEGF and steroid injections to modify the underlying disease process in DR — progressive retinal ischemia. Trials addressing this issue are under way, including DRCR-W and PANORAMA.

Widefield fluorescein angiogram demonstrating far peripheral retinal non-perfusion temporally with normal posterior pole perfusion.
Image taken using the 200Tx by Optos

Q: What are some of the limitations of this technology?

A: Even with widefield imaging, it is challenging and rare to capture 100% of the retina in an image. The Optos elliptical mirror allows better imaging of the temporal and nasal retinal peripheries than of the inferior and superior, so you may obtain beautiful images of temporal and nasal retina but not as far into the periphery inferiorly or superiorly. The gold standard for peripheral imaging would be the ability to capture 360 degrees to the ora serrata, and we don’t currently have that capability. Optos has software that permits the construction of a composite image from multiple different images, and allows you to capture a larger percentage of the retina than is possible with a single capture. But ideally, it would be nice to have complete retinal imaging with a single capture.

Another important limitation is that a 2-D image, even of the entire retina, lacks the context and information that a 3-D image provides. It’s still necessary to use indirect ophthalmoscopy with dynamic sclera depression to adequately evaluate the peripheral retina. This is still critical for evaluating retinal pathologies such as retinal breaks, something you might otherwise miss with a compressed 2-D image.

Some of the earlier limitations of widefield imaging are less problematic now. The need for highly trained photographers in retina practice is less critical. With a camera that captures 200 degrees, it is much easier to obtain a decent picture, and with digital imaging, you have immediate feedback on the quality of the images. Back when we used 35-mm film, we wouldn’t get results for a week, so highly skilled photographers were critical to ensure the images were correctly obtained the first time.

As I mentioned earlier, we’ve also found ways to account for and de-warp the distortion inherent in peripheral retinal imaging. We have commercially available software that de-warps images, so that limitation has been successfully addressed.

Q: What would you say is on the horizon with regard to widefield imaging?

A: To begin with, I think it’s essential to clearly define the clinical value of any new imaging modality. The imaging field is moving very rapidly, and with each step forward, we must ensure that we objectively determine how this progress impacts clinical practice, whether it is from a diagnostic, prognostic, or treatment perspective. We must continually ask ourselves: ‘How is this useful for the patient in front of me?’ Widefield application of OCT angiography, swept source OCT, and adaptive optics technology would certainly be great steps forward. NRP

REFERENCES

Dr. Wykoff is director of clinical research at Retina Consultants of Houston, deputy chair of ophthalmology at the Blanton Eye Institute at Houston Methodist Hospital, and co-director of the Greater Houston Retina Research Foundation.