OCT now integral to clinical practice, physicians call for standardization

Optical coherence tomography occupies a prominent place among revolutionary changes in ophthalmology. With its continual advances in software and technology, it has gradually evolved into an indispensable tool in many areas, from retina to glaucoma to anterior segment practice.

And yet, OCT was not an immediate success. Wolfgang Drexler, PhD, now professor of medical physics at the Medical University of Vienna, was the first master’s student in Europe whose topic was OCT. At some point in his studies, he wondered if he had made the right choice.

“It was in the early ’90s, just after James G. Fujimoto at MIT in Boston and Adolf F. Fercher at the Medical University of Vienna had pioneered this technology, with respectively more emphasis on its use for tomography in the U.S. and biometry here in Europe,” Drexler said. “At a conference in Budapest we presented OCT, and nobody seemed to understand what the point of it was. Nobody needs it, the public objected. We have ultrasound, and we anesthetize the eye anyway.”

Also, from a commercial point of view, the first generations of time-domain OCT systems was a letdown.

“OCT was put too quickly on the market, when the engineering was still poor,” he said.

In 1996, Carl Zeiss Meditec had nearly decided to withdraw it, when a new team of people came on board at the company. At the same time, Heidelberg Engineering developed the Spectralis. Between 1999 and 2000, the technology took off and became widely accepted.

Platform capabilities

Since then, several other companies have produced OCT technology for ophthalmologists.

In a presentation at the Optical Coherence Tomography and Imaging: Panel Experts Review in Las Vegas, Peter K. Kaiser, MD, OSN U.S. Edition Retina/Vitreous Board Member, presented an overview of spectral-domain OCT (SD-OCT) systems.

The Bioptigen Envisu Spectral Domain Ophthalmic Imaging System (SDOIS) has a wavelength of 840 nm, a scanning depth of 2.5 mm, an axial resolution of 3 µm and a scanning speed of 32,000 axial scans per second.

An advantage of this platform, Kaiser said, is the ability to change the head of the scanning system, which makes it adaptable to patients of all ages, from premature infants to adults, and to different head and body positions.

The Carl Zeiss Cirrus HD-OCT has a wavelength of 840 nm, a scanning depth of 2 mm, an axial resolution of 5 µm and a scanning speed of 27,000 axial scans per second.

Kaiser called the Cirrus a ubiquitous device and said that it offers high-resolution noise reduction, the ability to pseudo-register images, and software to look at drusen and atrophy mapping for dry age-related macular degeneration. Ganglion cell layer mapping is also available, as is FastTrac, a retinal tracking system introduced at the American Academy of Ophthalmology meeting in Chicago.

The Heidelberg Spectralis OCT has a wavelength of 870 nm, a scanning depth of 1.9 mm, an axial resolution of 7 µm and a scanning speed of 40,000 axial scans per second.

The Spectralis is a multi-modal system that offers confocal scanning laser ophthalmoscopy (cSLO) fundus imaging. It can be upgraded to include any combination of fluorescein angiography, indocyanine green angiography, blue laser fundus autofluorescence or MultiColor imaging. Because the device allows for multi-imaging with pinpoint registration, Kaiser said he currently uses the Spectralis to explain to his dry AMD patients why they are losing vision.

The Optovue RTVue Vtrac Premier and RTVue XR devices have a wavelength of 840 nm, a scanning depth of nearly 3 mm and an axial resolution of 5 µm. Scanning speed has been improved from 26,000 axial scans per second in the Vtrac Premier model to 70,000 axial scans per second in the XR model. Both Optovue devices have an anterior segment attachment.

Both Canon and Nidek offer high-resolution, fast-speed OCT technology, with scanning speed up to 50,000 axial scans per second.

The Topcon platforms can take color fundus images, allowing for drusen and atrophy mapping. An algorithm is provided for drusen analysis, Kaiser said.

Before committing to a particular platform, ophthalmologists should consider ease of operation, content of test reports and additional features of the device, such as the capabilities of fluorescein angiography and fundus photography, Joel S. Schuman, MD, OSN U.S. Edition Glaucoma Board Member, said.

“A glaucoma-based practice that is interested in the most accurate and reproducible measurements might be interested in a certain device, while a retina-based practice that is mostly interested in qualitatively visualizing the macula may find precise quantitation of tissue thickness of lesser importance,” Schuman said.

Evolution of OCT imaging

There were a number of milestones in the development of OCT, Drexler said. First was the significant improvement in resolution, and specifically in depth, thanks to new light sources such as titanium sapphire and broadband with superluminescent diodes.

The high resolution that allowed clear delineation of each retinal layer was developed around 2000. A further advancement was made 3 to 4 years later, when the integration of Fourier domain technology dramatically increased the image acquisition speed of OCT devices from the 400 axial scans per second of time domain to the 40,000 axial scans per second of spectral domain. Speed improvement was essential to the development of OCT, as it allowed bypassing the effects of eye movement and minimizing measurement time.

“The ability to sample enough points of the optic disc or the macula in a short amount of time having very little motional effect with good resolution, especially depth resolution, enabled OCT to implement the original concept of ‘optical biopsy.’ A quasi-histology was obtained without cutting the tissue, in a noninvasive way, as with the bar scan at the supermarket,” Drexler explained.

Improved resolution and speed were the forward steps that helped OCT gain clinical and medical acceptance and a primary role in ophthalmology, Drexler said. Compared with ultrasound, OCT offered a much higher resolution and helped significantly in the understanding of disease pathogenesis, early diagnosis and therapy monitoring.

OCT advances in glaucoma

The ability of SD-OCT to detect glaucoma progression has increased over time, Schuman said, but limited correspondence still exists with visual field progression.

“Disease progression can be detected with SD-OCT in the peripapillary region when sampled along a circle that is centered around the optic nerve head or when comparing the peripapillary retinal nerve fiber layer thickness throughout the scanning region, not confined to the sampling circle, with prior scans,” Schuman said.

The macula is another area where glaucoma progression can be detected, Schuman said, especially when examining ganglion cell layer thickness.

“When I started in 1990, the definition of glaucoma was focused on optic disc changes shown by topography. OCT has shifted the definition to an atrophy of the nerve fiber layer, more specifically of the circumpapillary nerve fiber layer. What is happening now is that we are not only looking at the optic disc, but also at the ganglion cell layer in the macula,” Drexler said.

The essential pathologic process of glaucoma is the loss of retinal ganglion cells and their axons. The macula is the place where most ganglion cells are and the primary location where they start dying, he explained.

Significant ganglion cell loss can now be detected by OCT several years before visual field deficits occur, allowing for an earlier diagnosis of glaucoma and more accurate tracking of progression.

“Time-domain OCT was already closer to histology compared to scanning laser ophthalmoscopy. With 3-D spectral-domain OCT, the majority of other optical methods, such as [scanning laser ophthalmoscopy] and [Heidelberg Retina Tomograph], have been replaced. In glaucoma, OCT can really do a lot. It is becoming more and more the standard, together, of course, with visual field and measurement of IOP,” Drexler said.

OCT in retinal pathologies

Retinal imaging was the field in which OCT had the earliest and most significant impact. In the past, retinal diagnostics necessitated a long training by indirect ophthalmoscopy and fundus camera. Long years of experience with good teachers and supervisors were required for young ophthalmologists to learn how to do a reliable diagnosis.

“Nowadays, OCT allows them to quickly achieve a high level of expertise. It provides an immediate 3-D visualization of the retina, similar to histology, from which it is easy to learn how to detect a macular hole, a cyst, a swelling, a drusen. It’s a shortcut to years of cumulative experience that helps young clinicians to gain confidence,” Drexler said.

OCT has become the standard for monitoring anti-VEGF therapy for dry or wet AMD. In most clinical trials, the endpoints are defined by macular parameters measured by OCT.

A need for standardization

According to Drexler, there are currently too many OCT platforms on the market. From an engineering point of view, they have very similar characteristics and comparably high performance. However, there are problems for ophthalmologists regarding the lack of standardization in the way quantitative results are extracted.

“If you take one optic disc from one patient and measure it with five systems, you have five different quantitative results. The follow-up with different machines is an impossible challenge. For the sake of the patients, we need standardization. Clinicians should more aggressively address this with the companies,” he said.

Although manufacturers must have a way to protect their investment, Schuman said, a method must also be developed to access all clinical information across all OCT platforms for a particular patient to help reduce the loss of data from prior testing and the additional cost of potential retesting.

“It is indeed possible to create software that can read OCT data from any OCT device and that can compensate for the different units used, so that measurements from one device to another are essentially interchangeable,” he said. “We have developed such software in our laboratory; however, the proprietary nature of the OCT output from different machines makes it impossible to use the software clinically. … We must demand standardization of scan parameters and access to all patient data in a standardized format that can be analyzed across any OCT platform.”

Intraoperative applications

The potential of intraoperative OCT use is growing and will potentially play a significant role in ophthalmic surgery in the future, according to Susanne Binder, MD, OSN Europe Edition Board Member.

“The advantage of using OCT intraoperatively is that you can design your surgical strategy moment by moment and make the appropriate surgical decision while you do the procedure,” she said. “To make just one example, during membrane peeling you might see on the OCT that you are pulling too hard on the retina or you might enlarge a macular hole. Then you can stop using the forceps and cut the membranes with scissors instead.”

For the last 2.5 years, Binder has been using OCT routinely for vitrectomy and in specific cataract surgery cases. She uses a Cirrus high-definition OCT incorporated into the operating microscope.

“Zeiss made this prototype for us in 2008. To use OCT during surgery efficiently it has to be seamlessly integrated into the workflow of the surgeon so that he has both full control of the surgical procedure and full control of the OCT. We are now identifying areas for improvement to support the surgeon in managing the device by himself,” Binder said.

“This is not only important for ergonomics and stability reasons, but also because the sharpness of the microscopic picture provides optimal OCT evaluation and enables exact measurement of retinal thickness,” she explained. “Currently we are evaluating different indications for its usefulness in anterior and posterior segment surgery.”

Anterior segment surgery can benefit from OCT guidance. For several years, Gábor B. Scharioth, MD, PhD, has been using a microscope-mounted prototype developed with Zeiss for corneal refractive surgery and corneal transplantation surgery.

“You can really see how well the edges of an anterior graft are adapted or, in [Descemet’s membrane endothelial keratoplasty], how good the endothelial transplant adheres to the inner surface of the cornea,” he said.

Scharioth uses OCT in canaloplasty to help place traction sutures into Schlemm’s canal.

“Studies suggest that a tighter stretching suture results in lower IOP postoperatively. One option is to use handheld [ultrasound biomicroscopy] intraoperatively, but then one of your hands is fixed to the handpiece. What is nice about microscope-mounted OCT is that both hands are free and you can carry out your surgical maneuvers while you watch the image,” he said.

Intraoperative use can be extended to phakic IOL implantation, in which OCT imaging can help to control the unfolding and positioning of the lens and measure distances, making sure that no contact is created with 
either the endothelium or the crystalline lens.

The transition to femtosecond laser-assisted cataract surgery will also encourage the use and development of anterior segment OCT as an intraoperative tool.

Femtosecond laser platforms with integrated OCT have been produced by OptiMedica (Catalys) for cataract and by Ziemer (Femto LDV Z6) for corneal refractive and transplantation procedures.

Further advances

Further advances are needed in the technology of anterior chamber OCT, Scharioth said. The most popular platform is still the Visante (Carl Zeiss Meditec), which is a time-domain system with limited resolution.

Some spectral-domain systems, such as the Spectralis, offer an adaptation for anterior segment use. The Tomey SS-1000 Casia is the only anterior segment OCT with 3-D imaging.

“Though resolution is not optimal, Tomey shows that 3-D is the way to go. You can have different views, see specific areas and tissues in connection with the surroundings. Once resolution is improved, we’ll be able to look into the angle, turn the 3-D and have a nice view of structures which we cannot see at the slit lamp, without a gonioscope and completely non-contact. Another very promising indication, once we manage to increase the resolution, will be for glaucoma to look at trabecular meshwork and Schlemm’s canal,” Scharioth said.

A different wavelength, specifically adapted to anterior segment tissues and structures, is needed to increase imaging depth, he said, and speed should be increased.

“We need an [anterior chamber] OCT that is specifically [anterior chamber], rather than an adaptation of the posterior segment technology.”

Speaking about OCT technology in general, Drexler said it should aim at being faster.

“Measurements take 2 to 3 seconds, which is far too long. We must go up to at least 100,000 to 200,000 scans per second. It is possible, it is affordable, and it is necessary to sample larger areas and detect small, subtle changes for early diagnosis,” he said.

A longer wavelength, around 1060 nm, would also allow going beyond the retina and looking into the choroid, which is involved in many pathologies.

Long-term considerations include OCT capabilities for functional information, such as blood flow and tissue oxygenation, as well as imaging of single cells in the retina.

“To look at cells, we need to provide transverse resolution, and to do this, we need to correct the aberration of the eye. By combining OCT with adaptive optics, many groups, including mine, have demonstrated that you can see single photoreceptors, not only cones but also rods. Cellular resolution information will be extremely important together with functional information,” Drexler said. – by Michela Cimberle and Ashley Biro

 

Would you consider OCT optic nerve changes as an indication for glaucoma surgery in absence of visual field changes?

Benefits of early surgery when OCT provides evidence of optic 
nerve damage

The main advantage of OCT over visual field testing relies on the fact that OCT is an objective test whereas visual field is a subjective test. Modern OCTs are becoming extremely accurate, whereas visual field testing has not gone through revolutionary changes during the last 2 decades. Furthermore, elderly patients find visual field testing annoying, and many of them lose concentration or simply fall asleep during testing. It is not uncommon to find nearly normal visual fields in patients with advanced glaucomatous damage in the optic nerve.

Visual field damage can progress in a “cliff” pattern, especially when optic nerve damage is between the 80% and 90% marks. This is the time in which visual field testing can be inconclusive, whereas OCT changes provide solid evidence of glaucoma. The glaucoma surgeon should then rely on OCT changes rather than on dubious visual field findings.

OCT provides at least three independent measures to support the diagnosis of glaucoma: optic nerve cupping, retinal nerve fiber layer and retinal thickness. Although these three measures have their limitations, the average glaucoma specialist can easily diagnose or disprove glaucoma in the majority of cases when presented with a modern OCT examination. The same cannot be said for visual fields.

When in doubt, the glaucoma surgeon should opt for surgery because IOP target achieved by surgery only (complete success) is far better than IOP target achieved by a cocktail of several anti-glaucoma medications. Therefore, when OCT changes indicate that surgery is warranted, the patient should not be denied the benefits of it because the visual field results are inconclusive. This is especially true when the glaucoma surgeon is familiar and experienced with minimally invasive techniques such as nonpenetrating surgery or the Ex-Press device (Alcon). New minimally invasive glaucoma surgery procedures are currently being developed and will provide more surgical options earlier in the course of the disease.

Elie Dahan, MD, is an OSN Europe Edition Board Member. Disclosure: Dahan has no relevant financial disclosures.

Decisions cannot be based on a single element

Decision making in glaucoma therapy is rarely based on a single point. The patient is the key element, and we have to take in account age, life expectancy, tolerance to topical IOP-lowering agents, his/her own wish regarding surgery and ability to understand potential complications. The second point is how well glaucoma is controlled. In some cases there is an agreement between structural and functional tests. Unfortunately, in some other cases there is a discrepancy, and then we have to decide whether to trust function or structure. Visual field is not a truly scientific test, and the artifacts are well known. Moreover, we need at least six visual fields to document progression. The same could be said for imaging, in which we all know the errors related to poor signal strength, saccades, misalignment, tilted disc etc. The third element of the decision is the type of glaucoma based on gonioscopy, which is critical to determine the type of procedure.

So my answer to the question would be no. It is not a negative answer against OCT in itself because we are all aware of the helpful yield of imaging in our practice. It is no because the decision would be based on a single argument. At the beginning of our career, we were all dreaming of easy, straightforward cut-off criteria for decision making in glaucoma. Some authors have called this (poor) reasoning “the red or the green,” referring to the colors displayed on the layouts of the OCTs. With more experience, medicine and life tell us that the reality is more complex.

Alain Bron, MD, is an OSN Europe Edition Board Member. Disclosure: Bron has no relevant financial disclosures.