Spectral-domain OCT Eclipses Time-domain in Speed and Resolution

Spectral-domain OCT Eclipses Time-domain in Speed & Resolution


Optical coherence tomography (OCT) is a noninvasive diagnostic imaging tool that has rapidly become indispensable in the management of retinal disease. By measuring the echo-time delay and intensity of backscattered light, this technology allows real-time imaging and examination of epiretinal, intraretinal, and subretinal morphology and can document subtle retinal disease that is often undetectable with conventional clinical examination alone.1,2

The vast majority of retinal specialists own or have access to one of the time-domain (TD) OCT systems, and the general consensus is that, in combination with clinical evaluation, these systems are effective at determining anatomic response to treatment. Approximately 5% of these retinal specialists also own or have access to spectral-domain (SD) OCT, which represents the newest generation of OCT technology. Several SD-OCT systems have become available in recent months, and common among all of them is that they are faster and provide better resolution than TD systems. However, each SD-OCT system has distinguishing characteristics and those of the Carl Zeiss Meditec ([CZM], Dublin, CA) Cirrus HD-OCT will be reviewed in this article.


Basic advantages of the CZM Cirrus system include a user interface designed for efficiency and fast electronics, which means chair time is shortened; a compact footprint, which enables it to fit into even a crowded clinic; and mouse-driven alignment that facilitates simple image capture, which some suggest enables fledgling technicians to conquer the learning curve fairly quickly. The system has 90° orientation, so the operator can easily see the patient during setup and scanning, and an "iris viewer" enables precise registration, which also helps with patient setup, making the entire process more efficient.

David M. Brown, MD, was among the first retinal specialists in the country to have the Cirrus OCT system in his clinic. Dr. Brown directs the Greater Houston Retina Research Center in Houston and is a clinical assistant professor in the department of ophthalmology at the Methodist Hospital, Weil College of Medicine, in Houston. He was also instrumental in the MARINA and ANCHOR intravitreal ranibizumab (Lucentis, Genentech) trials, which relied on OCT in conjunction with clinical examination to evaluate response to vascular endothelial growth factor (VEGF)-blocking agents. "The Cirrus gives us 27 000 A-scans per second instead of 400 A-scans in 1.25 seconds, which makes the Cirrus much less dependent on patient fixation," says Dr. Brown. "These older patients who have age-related macular degeneration [AMD] are often shaky and have poor vision and they can't really fixate, so the faster we can get the scan, the better," he adds.

The Cirrus includes automated patient recall technology that assures that the patient position and the instrument settings are repeated from the previous visit, ensuring reproducibility with overlay of the OCT fundus image from the previous scan for verification of scan position. Another important and useful feature embodied in the Cirrus is the simultaneous capture of the OCT and the scanning laser ophthalmology fundus image, which enables precise registration between the OCT scan and the fundus image.


In Dr. Brown's practice, every patient gets both the Stratus (CZM) and Cirrus OCT, and on follow-up any patient who has AMD, branch retinal vein occlusion, central retinal vein occlusion, cystoid macular edema, diabetic macular edema, or epiretinal membrane gets an OCT before their clinical evaluation. In cases where he cannot bill for it, Dr. Brown does it for free. "I do it for the science," says Dr. Brown. "OCT wasn't even a billable procedure in my state until this year, but I do it because it's better care for the patient and because it makes the visit quicker," says Dr. Brown, adding, "The information helps you focus in on your clinical exam so that you don't miss anything."

Dr. Brown says having every patient undergo both a Cirrus and Stratus OCT is not redundant; rather it serves to show whether the newer system provides additional information. "We do spectral- and time-domain OCT on every patient because we want to see whether the Cirrus actually helps. What we have found is that about 10% to 20% of the time I see changes on the Cirrus that I missed on the Stratus (Figure). We can't be sure if that makes a clinical difference, but it probably does," he says.

Dr. Brown says his experience as first author on the ANCHOR trial results and second author on the MARINA trial results tells him that "OCT enables earlier detection of subtle disease, and we think that makes a difference in the long-term outcome of patients with macular degeneration treated." He and his fellow investigators are finding that among the study subjects, who are now 2 to 3 years out, those who have gotten fluid a couple of times have lost some of the gains that were achieved during the trials. "This tells me that any fluid in the retina isn't a good thing, so although we are not giving automatic monthly [anti-VEGF] injections, we're certainly giving more injections to keep people from getting recurrent fluid, and that paradigm is very dependent on catching any fluid early, which is where highresolution OCT comes in," he says.

The enhanced resolution of the Cirrus literally magnifies the benefits of standard OCT, according to Dr. Brown. "We're starting to see things that look like the histopathology of the eye. We're starting to see cellular layers within the retina, which we couldn't do with the Stratus. In those cellular layers, SD technology is enabling us to gain better visibility of the external limiting membrane and, in some patients, the line between the inner and outer photoreceptors as well," he said.


Frederick L. Ferris III, MD, director of clinical research at the National Eye Institute (NEI), is also among the early adopters of SD-OCT. Dr. Ferris says that standard TDOCT can meet the needs of most retina practices, but that the advances in speed and resolution with the Cirrus are impressive. "I think you could manage patients fairly well with TD-OCT, but the resolution is nowhere near as good as with SD. Also the resolution probably is going to end up improving the reproducibility. This is still up for debate, but I suspect that it is going to turn out that [SD-OCT] is much more reproducible than the earlier-generation OCT," says Dr. Ferris. "The Cirrus recently arrived at the NEI, and it is being used in exactly the same capacity that the Stratus was. We're in the process of understanding all of the various things that we can do with this instrument that are the same and different from its TD counterpart. So far, we're all very happy with it," he adds.

Dr. Ferris points out that TD-OCT is pervasive throughout modern retina practices and that there probably will not be an overnight switch to the second-generation system. "The early OCT systems worked really well. Everyone got one and everyone uses one. Physicians are making judgments about when to treat based on TD systems, and I think the vast majority of these decisions would not be changed if they had the Cirrus rather than the Stratus. However, the Cirrus is clearly better. I believe that it's going to show us things that we could not see on the old one," he says. "It's analogous to computer technology. As computers get faster, people move to a newer models because of speed and convenience and sometimes improved capabilities. Similarly, it's inevitable that we're going to move in the direction of SD, but we shouldn't lose sight of the fact that TD-OCT is adequate for most things."

Figure (a) A 63-year-old woman with neovascular AMD and disciform OD and multiple soft drusen (OS) presented with blurred visual acuity and a little hemorrhage on clinical exam. (b) The Stratus OCT scan (6 cuts) shows completely normal pathology and no evidence of fluid. (c) The Cirrus OCT shows obvious fluid. (d) An angiogram confirms the Cirrus findings.

The earliest adopters of new technology such as the Cirrus OCT tend to be large practices that frequently participate in clinical trials. Having the newest OCT technology may end up becoming a de facto requirement of clinical trial participation so that all participants are measuring outcomes with the same instruments. "Within a clinical trial there are issues about having different instruments in different clinics. I think for a lot of good reasons the trials will want to have the same instrument in all of the participating clinics," says Dr. Ferris. "If you're looking at retinal thickening and you could show that 2 different systems work equally well at demonstrating that outcome, then I think you can have different machines at different clinics. We are wrestling right now with what we need to do with regard to these instruments in our ongoing clinical trials because not everybody has the new versions. I think we are all anxious to use these new versions because we believe they are probably better and probably more reproducible," he added.

The issue of clinical trial participation is a factor that will drive the evolution of SD-OCT adoption. Another is the desire on the part of academics who are eager to discover how much more information this new technology can provide about retinal anatomy in the face of advanced disease and treatment. "Academic centers will be driven by the extra things that we can see on these machines," says Dr. Ferris. "For example, we have traditionally used retinal thickness, cysts, and subretinal fluid as outcome variables. Perhaps we can now add characteristics of individual retinal layers, such as outer segment length."

On a related note, Dr. Ferris points out that, although many in the retina community consider the use of OCT in clinical trials as an outcome variable, it is insufficient. "OCT is not a satisfactory outcome variable for approving a drug — at least not yet — because retinal thickness and thickening are only poorly associated with visual function, so it's rather obvious for that reason that you have to be careful about using OCT information as an outcome variable to approve a drug," he says. However, says Dr. Ferris, "using OCT as an outcome variable to prove that the drug is doing something is terrific because you can find that out with relative immediacy. It's very useful to be able to show whether a drug is doing something, but it doesn't show whether the drug is clinically meaningful. Suppose you had a drug that decreased retinal thickness but didn't change visual acuity. Patients don't care what their retinal thickness is, but they do care about their visual function. So at this point OCT is not useful as a primary outcome variable, but is very useful as a secondary outcome variable," he explains.

Gauging changes in the outer segments of the retina is an outcome variable in which Dr. Ferris is particularly interested. "If we had a drug that we thought might be effective for retinitis pigmentosa, we might be able to, in a short period of time, show that the drug is doing something. While that won't be adequate to get the drug approved, having these proofs of principle is hugely important to the sponsoring companies. Our ability to look in detail at the retinal structure with the Cirrus OCT will serve us very well as outcome variables," he says.

Allen C. Ho, MD, professor of ophthalmology at Thomas Jefferson University and an attending surgeon of the Retina Service of Wills Eye Institute in Philadelphia, is also among the select pool of retinal specialists whose clinic has already made a home for the Cirrus OCT. "The most practical clinical application of either the Cirrus or the Stratus is diagnosing and monitoring treatment of AMD." he said. "The mechanism of data acquisition is, however, different between the Cirrus and the Stratus. The SD system generates a cube of data in the macula; it is not limited to 6 separate cross-sections of the back of the eye as is the Stratus. Because the Cirrus provides a cube of data, it is richer and allows for precise registration of anatomic landmarks that can be compared over time — so you know that where you set your fovea is where your fovea is going to be," he explains. The spectral system enables the surgeon to accurately compare a certain point in the macula in its response to treatment with the same point in the macula at a different point in time, he points out. "So it allows us to assess therapies more effectively and accurately, but it may not necessarily reduce the number of patient visits," he says.

Dr. Ho continues, "Another difference is that the Stratus axial resolution is about 10 μm, whereas the Cirrus resolution is down to 5 μm, so you get more detailed and vivid images and accuracy of the anatomy," he says.


Another benefit afforded by the Cirrus is its speedy software. Apparently the currently available software is much quicker than what was available with the earlier model. "The initial SD software took a long time to save each image and that's the rate-limiting factor on how fast they could be performed," says Dr. Brown. "The new software allows us to do exciting things. For instance, we segment out just the retinal pigment epithelial layer so that we can show our patients their drusen. Eventually we'll be able to quantitate that and show them that their drusen is increasing or decreasing. This becomes particularly important in some of the trials where we are testing different agents for dry AMD," he explains.

Two other important distinctions, according to Dr. Ho, are the smaller, more compact size of the Cirrus and the fact that it is not joystick controlled. "This second factor may actually be a barrier to initial adoption because the technicians who use it may not be as comfortable and familiar with the newer and easier technology," he says. "With the Stratus, the technicians used a joystick to focus on the back of the eye and to place the scans at a certain location at the back of the eye. With the Cirrus, there's no joystick. Our technicians initially preferred the older machine; we found that they were going to the Chevrolet instead of the Ferrari because they were used to how the Chevrolet handled. After some additional training they began to see the beauty of the Cirrus and found that it is actually quite user friendly and forgiving."

Dr. Brown also has experienced a similar response among technicians in his practice. "We're a busy 5-office practice, and we do thousands of OCTs a year. We have found in the past that our technicians who are 19 to 25 years old [and grew up playing video games] find it much easier to pick up the Stratus than our technicians who are 40 to 50, though the older technicians are well seasoned, experienced, and otherwise world class," says Dr. Brown. "With the Cirrus the playing field is leveled and all the technicians are able to identify the pathology I want to see. I think it's safe to say," he adds, "that you can get more useful images with the Cirrus regardless of the experience of your staff, and this is becoming even more important because finding qualified help tends to be among our ongoing challenges." RP


  1. Schuman JS, Puliafito CA, Fujimoto JG. Optical Coherence Tomography of Ocular Diseases.Thorofare, NJ: SLACK Inc.; 2004.
  2. Ozdek SC, Erdinc MA, Gurelik G, et al. Optical coherence tomographic assessment of diabetic macular edema: comparison with fluorescein angiographic and clinical findings. Ophthalmologica. 2005;219:86-92.

David Brown, MD, is a consultant for CZM. Allen Ho, MD, is a member of the early access program for the Cirrus.