Tracking the OCT Revolution

How OCT has transformed the clinical landscape and earned its standing as an established pillar of eye care at all levels.

BY FRANK CELIA, CONTRIBUTING EDITOR

OCT arrived at an opportune moment in medical history… just in time to play handmaiden to the success of anti-VEGF retinal therapy, arguably the most significant vision-care achievement of our time. Because anti-VEGF drugs, and therapies such as lasers and steroids, produce limited durations of effect and better outcomes on early-stage disease, there exists a need for heightened monitoring of existing retinal patients and quicker diagnosis of new ones. OCT excels at both.

OCT also continues to impact glaucoma, though advancement here has been slower. This is likely due not only to glaucoma’s more elusive pathogenesis, but also to the more gradual progression of the disease. Whereas a condition such as diabetic retinopathy develops over a period of months, glaucoma can take years or even decades to diagnose or progress. So, for example, though incremental breakthroughs have been made in glaucomatous OCT progression analysis, thus far, none have involved spectral domain machines. Studies are ongoing.

Outside subspecialties, OCT can help resolve vague, general vision complaints as well. Optometrists tell of patients with no known refractive error who demonstrate 20/25 vision and upon routine OCT examination reveal vitreomacular traction or an incipient macular hole.

In short, this fast, non-invasive imaging technique holds an important place in a wide range of eyecare endeavors. This article explores some of the practical uses for OCT and how they might evolve in the future.

Most Eyes Have It

Like any maturing industry, OCT manufacturers have begun to move toward a state of equilibrium wherein a handful of established brands dominate the market (OPKO Health, Inc. sold its OCT assets to retinal photography company Optos, Inc., in 2011, for example). In such environments, products begin to resemble each other as industry standards emerge. One such standard has been the rise of gaze-tracking systems.

Although they employ widely different methodology, all tracking systems recognize when a patient’s eye move-ment or blink has skewed an OCT scan, and then auto-matically perform an additional scan to correct the error. Eye tracking is often advertised as a “time saver,” but the technology’s true value lies in guarding against operator error. A skilled, attentive OCT operator should notice when eye movement distorts a scan, and then simply re-take the scan. Unfortunately, this doesn’t always happen. By automating this task, eye tracking removes image-quality decisions from human hands, which provides more accurate results and simplifies the training process for staff.

Perhaps the most noteworthy clinical advantage of eye tracking is its proven precision in calibrating the registration of images from one office visit to the next, sometimes down to the micron,¹ creating enormous monitoring potential over successive scans. “If I’m treating a patient with the goal of decreasing retinal thickening, it’s ideal to have each scan image at the exact same location,” says Ronald C. Gentile, MD, FACS, FASRS, professor of ophthalmology at The New York Eye & Ear Infirmary. “This provides a more accurate and quantitative map of the location of the decrease in retinal thickening and also the opposite, indicating where the thickening may be getting worse. This technology can actually tell you in microns how much better or worse the retinal thickening is.”

Gaze tracking holds so much appeal, almost every major OCT manufacturer now offers it, the sole exception being Bioptigen (See “Coordinated Movement,” page 14-S). This was a conscious decision based on Bioptigen’s unique market niche, says CEO Eric L. Buckland, PhD. “There are many numbers thrown around about imaging, but we feel safe saying our machines are unique in their ability to scan at high density, at high speed,” Dr. Buckland explains. “With high-density scans captured without a need for ‘averaging’ images, we better sample pathology that is within our field of view. Typically, in the systems that offer eye tracking, they take some of the computer memory we expend on collecting data and put it toward averaging the images to reduce image noise.” This creates a high-contrast image well suited to a busy private practice, he concedes — but also an image containing less information. “It’s fair to say eye tracking can add real value to a busy clinical practice, but Bioptigen’s research-oriented customers want unbiased data and richer information content.”

In patients with too much movement, eye tracking can fail, but the Bioptigen system continues to collect data, he says. Where others offer eye tracking, Bioptigen provides user-defined protocols for research studies that rely on a multiplicity of scans to visualize pathology, allowing accurate repetition across patients and times within a research program, according to Dr. Buckland.

Carl Zeiss Meditec, with more than 10,000 OCT devices in the field, also hesitated before developing an eye tracking system, wary about tradeoffs in data density and scan time spent re-calibrating and re-shooting. But ultimately the company decided it was worth adding it as an option to achieve better image quality and interscan registration, according to Christine Ritter, senior director of OCT imaging. “Eye tracking gives us some improvement in the reproducibility of the macular thickness measures on the dense cube rasters, on the order of two or three microns — not a huge benefit, but it’s there for difficult cases where the patients do not fixate well,” she says. “In addition to the raster cube improvements, tracking also enables a series of best-of-the-best quality b-scans that can be placed at exactly the same location every visit. When there’s a difficult patient, or when you want to examine the subtle details of a lesion using the highest quality image, that’s the time to use our FastTrac^TM tracking feature.”

Can Extensive Data Help Answer Enduring Questions?

Researchers remain hopeful that OCT technology’s vast computer memory and large population databases may one day help untangle some of the mysteries surrounding glaucoma. Profound questions endure: What metric best defines and measures disease progression? Does progression occur linearly at a fixed rate, or stepwise with great clumps of cells dying at variant speeds? Which computer algorithm is most useful, event-based or trend-based analysis?

Limited progress has been made in answering these questions. For example, trend-based analysis appears to be edging out event-based formulas as the preferred metric. As one well-regarded study put it: “Measuring the rate of change in [retinal nerve fiber layer] thickness [only achievable with trend-based analysis] would serve as a new paradigm for observing and managing patients with glaucoma.”²

The same paper tentatively suggests glaucoma progresses linearly, while conceding that additional and lengthy follow up is needed to confirm that hypothesis.²

The theory that structural damage, as measured by diagnostics such as OCT, precedes functional damage, as measured by visual field testing — another hotly debated glaucoma topic — appears to be gaining ground.³ One recent study suggests “progression as measured by a decrease in RNFL thickness is more noticeable than is progression assessed by visual field measurements in early-stage glaucoma patients, whereas the reverse is true when the disease is more advanced.”³

Flexible Usage

Last year, the FDA cleared Bioptigen’s Envisu Spectral Domain Ophthalmic Imaging System (SDOIS) for use on pediatric patients. The system’s handheld scanner can image patients of any age, from premature and neonatal infants to adults, whether upright or supine, ambulatory or confined. The ability to capture sharp images rapidly and at high density without averaging is what makes handheld OCT feasible, according to the company.

“Our system is enabling new levels of research,” says Dr. Buckland. “There are a number of trials using our system to evaluate how the retina develops right after birth.” A deeper understanding of how healthy retinas grow may help researchers develop ways to treat retinopathy of prematurity (ROP), which continues to be a problem for premature infants, he notes.

The retina undergoes rapid development up to age two, and then significant growth up to age five. Here too OCTs could reveal practical information, says Dr. Buckland: “Certainly a 2-year-old toddler cannot be imaged by a standard clinical lane [OCT] system. And for subjects, young or old, who are under anesthesia, our mobile handheld systems extend the application to the operating theater.”

Raising the Refractive Bar

OCT technology can play a significant role in refractions, especially in cases of vitreomacular traction (VMT), which often begin as vague visual complaints. “Prior to the advent of OCT, VMT was a rare diagnosis, imperceivable using fundoscopic evaluation alone,” says Diana Shechtman, OD, FAAO, an associate professor of optometry at Nova Southwestern University. “Today, OCT aids in the diagnosis of VMT, which is now recognized as the pathogenesis of a spectrum of maculopathies including macular holes and macular edema.”

Co-management opportunities abound. The early and intermediate stages of dry AMD can last for years, even decades, Dr. Shechtman notes. “In the early or even moderate stage of the disease, so long as the optometrist has the capabilities and instruments, we can certainly follow these patients and play a critical role in their management by making the proper nutritional supplementation recommendations and educating them about lifestyle modification.”

Moving Toward Standardization

As trend-based analysis emerges as the likely preferred algorithm for monitoring glaucoma, one physician is glad to see the OCT industry moving toward agreed-upon industry standards and common functionalities such as eye tracking. “To perform trend analysis, we need stable software platforms, and we’re starting to see that with the spectral domain OCTs,” says Nathan M. Radcliffe, MD, director of glaucoma service at Weill Cornell Medical College. Accurate trend analysis requires at least five scans (ideally over many years) with an identical acquisition protocol. “If these software developers change their minds every 6 months, we’ll have to keep going back to square one,” he says. “I’m happy the CIRRUS has had the same data cube since 2007. That’s the kind of stability we need.”■