Utilizing a Multi-Modality Imaging Instrument
A panel of experts discusses capabilities, efficiency and applications.
Jeffrey S. Heier, MD: Panel members, now that we have the ability to combine spectral domain optical coherence tomography (OCT) with infrared imaging, fluorescein and indocyanine green (ICG) angiography and autofluorescence in the same instrument, how do you think it will affect how you take care of patients?
David M. Brown, MD, FACS: The eye tracking feature, which compensates for eye movement, really makes the Spectralis HRA+OCT stand out. It is difficult to find a talented photographer for private practice and even harder for clinical trials. Less experienced photographers tend not to be able to capture good images when the patient's eye is moving around. However, this instrument's eye tracker and image stabilization makes patient cooperation much less of an issue. It really levels the playing field both for photography and OCT imaging.
Whereas the ICG angiograms and autofluorescence imaging from almost every other imaging systems look fuzzy, the Spectralis ICGs look as crisp as a film fluorescein angiogram. The choroidal vessels are visible in both the early and late images. Because of the eye tracking, this instrument has far less noise to clean up as it is constructing images, and I think this is the reason for the better resolution. Personally, I never found autofluoresence or ICG imaging useful at all until we started using a laser imaging system. Now we use it every day.
With the spectral domain OCT imaging, the specifc histopathologic bands between the different retinal layers are visible in most patients. It's like getting a noninvasive retinal biopsy. With spectral domain OCT, I actually had to review my ophthalmic pathology notes to first identify what I was seeing.
|"Personally, I never found autofluorescence or ICG imaging useful at all until we started using a laser imaging system. Now we use it every day.|
—David M. Brown, MD, FACS
*Quan Dong Nguyen, MD, MSc: One of the spectacular properties of the Spectralis HRA+OCT is the eye tracking system. It can fill a major need in today's practices now that many retina specialists rely on noninvasive OCT technology in making decisions, such as in deciding whether patients with neovascular AMD need re-treatment. It is crucial to know precisely what retinal area is being evaluated from visit to visit, and the eye tracking/image alignment provides that.
Another special feature is the ability to perform autofluorescence and fluorescein angiography simultaneously with spectral domain OCT. Autofluorescence may help us to understand what is occurring in patients whose eyes appear within normal limits on fluorescein angiography but who have suboptimal vision. There is a lot for us to learn about the capabilities of the Spectralis, and I am sure many of us will have a lot of information to report within a year from now.
|*The views expressed by Dr. Nguyen in this article do not constitute or imply any endorsement by the Johns Hopkins University, the Johns Hopkins Hospital or the Johns Hopkins Health System.|
Dr. Heier: Everyone on the panel participates in clinical trials, a situation where OCT reproducibility is crucial for measuring treatment effect. How will multi-modality imaging impact this?
Dr. Nguyen: For certain diseases like diabetic macular edema, it may not be too difficult to see changes from visit to visit because we may be interested only in assessing large areas of intraretinal fluid. However, in other processes such as choroidal neovascularization (CNV) from age-related macular degeneration, we may have a difficult time identifying and detecting changes at different time points within the study, given the complexity of the disease manifestations (eg, subretinal fluid, intraretinal fluid, RPE detachments). I believe a system that allows us to take point-to-point measurements is invaluable in clinical trials. Trial investigators will appreciate it, as we will be at our Retinal Imaging Research and Reading Center in Baltimore, where we always strive to achieve the highest level of detecting changes within a clinical trial, as well as assessing and comparing different measurement systems available for standard clinical care and clinical research.
|"It is crucial to know precisely what retinal area is being evaluated from visit to visit, and the eye tracking/image alignment provides that."|
—Quan Dong Nguyen, MD, MSc
Without the tracking system, one just does not know whether one is getting accurate measurements at different visits in a clinical study. On OCT measurements, when one moves the cursor just a few microns back and forth, one can change the retinal thickness value drastically. If the investigators rely on just the thickness value generated without a precise point-to-point correspondence among visits, they may not have truly valid measurements, and hence, may not be able to fully assess the study questions. Therefore, any system that allows true tracking with precise measurements at specific points will be invaluable in data collection and analysis for any clinical trial, in my opinion.
|"In one-doctor offices, a combination device can be the single instrument, so all of the testing can be done in one place."|
—Scott W. Cousins, MD
ADDING EFFICIENCY TO IMAGING
Dr. Brown: We certainly benefit from the acquisition of higher quality, more accurate images in practice, but I hope obtaining them and evaluating them will not slow down the clinic. We would like to be able to figure out where the pathology is located fairly quickly.
Dr. Heier: We have certainly seen cases in which additional imaging has enabled us to make the diagnosis. For example, a young boy referred to us a month after having commotio retinae from being hit in the eye with a paint ball had 20/200 visual acuity. No one had been able to figure out why. He had been previously evaluated using fluorescein angiography, without a clear-cut explanation for the decreased vision. Multi-modality imaging with autofluorescence and spectral domain OCT revealed extensive RPE damage in the region of the commotio, but the macula was not involved. However, the SD-OCT revealed subtle photoreceptor damage that the Stratus couldn't pick up.
One question many practices have is how they would handle an upgrade to spectral domain OCT. Right now, everyone has a time domain OCT system in the office along with other instruments with other capabilities. How will they transition?
|"Reader software is vital because you do not want to keep running over to the photography area."|
—Peter K. Kaiser, MD
Scott W. Cousins, MD: We are finding that combining fundus imaging and SD-OCT on one platform does not slow us down at all. To keep the day moving, we have our photographers perform specific studies of certain types of lesions. For example, when they see a pigment epithelial detachment they know to perform the ICG angiography. Also, in one-doctor offices, a combination device can be the single instrument, so all of the testing can be done in one place. You can perform the ICG angiography in 20 to 30 seconds because you are interested only in the transit phase. You are not tying up your photography unit for 15 to 30 minutes like you do with the bright flash technique where you are looking at the late frames 15 to 30 minutes later.
I have busy clinics like most of the panel members, but I have to look at the videos myself in a few cases each day. Most of the time, the photographer can point out where the pathology is and bring up that image for me on the digital viewing station. Using the Spectralis HRA+OCT as a workhorse instrument does not slow down my efficiency, but it increases my ability to diagnose and treat patients who fall outside the standard way that we approach common cases.
Dr. Heier: SD-OCT does appear to have significant potential to impact our management of patients today, and its value is likely to grow as we learn more. Imagine, when a patient comes into the clinic, we do a volume scan and find a subtle area of pathology. Now, we can reference that point every time; we absolutely know what effect our treatment is having. We are unlikely to miss anything. If we have an area that we want to keep checking, we know we will be able to check it repeatedly.
Spectral domain technology will allow us to better diagnose different types of lesions, such as polypoidal and retinal angiomatous proliferation, and monitor treatment effect. A recent case of mine comes to mind. A patient was referred to us as a high myope with a little hemorrhage. On time domain OCT we weren't convinced we had localized the hemorrhage. We were able to image the exact point of the hemorrhage and isolate it with confidence since the fundus image was obtained simultaneously with the SD-OCT. That allowed us to determine the hemorrhage was intraretinal; there was no underlying CNV. Also, now we can come back to that exact spot every time.
TRANSITIONING TO NEW TECHNOLOGIES
Dr. Heier: A number of different spectral domain OCT instruments are available now. What are the important points we should be looking for?
Dr. Kaiser: Look carefully at the software. Software is what differentiates these systems. The hardware for most of them is similar, with the exception of the Spectralis spectral cSLO/OCT and the Opko/Ophthalmic Technologies Inc. Spectral SLO/OCT. These two stand out in terms of hardware.
When you are evaluating the systems, find out how easy the software is to use. Does it include reader software? Reader software is vital because you do not want to keep running over to the photography area. Some of the companies have reader software and some do not. Also, much of the software for the various instruments is immature right now and constantly improving. Therefore, find out what costs are involved with upgrading as new versions are released. Are software upgrades free? Are you locked in? Do you need to buy a one-year service contract to get those? That is a big issue.
The question I always ask the manufacturers is what they are going to do with my time domain OCT data. You are likely not going to want to keep your time domain equipment, so some sort of trade-in should be available. Whatever we are going to do with all of our old data, it is an important issue.
When you are considering purchasing one of the new spectral domain OCT devices, another aspect to check on is whether the camera and light source are fixed. When they are fixed, it may be difficult to get all patients, especially larger patients, into the device. With the Spectralis HRA+OCT you can kind of move in any direction, which is an advantage. We do not always consider these types of things when we are focused on the images, but sit down and see how each of these instruments is for patients and the flow in your room. They are a ll different in these respects.
Dr. Heier: Where is this technology going in the future? I heard a recent talk refer to intraocular OCT. We would have an intraocular probe to follow right where we wanted to go inside the eye and know exactly where we were.
Dr. Brown: The more we can see, the more we can do, and the more we can detect subtle differences. I think the best way to compare the different spectral domain machines is to compare details in the outer retina. In other words, how often can you identify the external limiting membrane and define the band between the outer and inner photoreceptors? Whereas these bands are hardly ever imaged with the time domain units, the Spectralis demonstrates these layers routinely even in challenging patients.
In AMD, we have some pretty amazing therapeutics right now, but we really don't know if any of our combination therapies are additive or not. In other words, are the add-ons giving us any incremental improvements? As we won't have any randomized trials to determine this, we have to rely on anatomic markers. Therefore, we need great imaging to really know whether they are helping or hurting. I think with spectral domain technology, we are "off to the moon" in this area, so to speak.
Audience Member: Can anyone comment about the differences between the instruments in regard to imaging the vitreoretinal interface?
Dr. Brown: Spectral domain OCT imaging is clearly superior to the time domain system when it comes to imaging the vitreoretinal interface. I have cases of traction that were not visible with time domain technology, but I could see very well with spectral domain OCT. The motion video is much better at demonstrating vitreoretinal separation and traction prior to surgical intervention than the individual cuts of a time domain unit.
|"This technology likely will play a role in driving our advances over the next several years."|
—Jeffrey S. Heier, MD
Dr. Kaiser: Most of the spectral domain OCTs are similar in their ability to pick up the vitreoretinal interface. Some of the 3-D volume views are very interesting in how they show the vitreous actually coming down into the retina. They provide a good idea of how to manage these problems.
Dr. Heier: At this point, many physicians are wondering exactly what impact the capabilities of spectral domain OCT will have on practice. Having seen what we have from OCT, I do not think it is a stretch to imagine that the capabilities are going to have a significant impact on how we treat patients going forward. It is up to us to determine what the impact will be and apply it to our pharmacological and surgical interventions. This technology likely will play a role in driving our advances over the next several years.