How OCT became a 'game changer'

Listen to what Joel Schuman, MD, tells  J.C. Noreika, MD, MBA, about the use of optical coherence tomography (OCT) to diagnose and follow glaucoma with patients, as well as what the future may bring for the specialty. Dr. Schuman begins by describing the early influences in his career that led to his current post. 

Editor’s Note: Welcome to the latest installment of “Sight Lines,” a feature in which J.C. Noreika, MD, MBA, an ophthalmologist in Medina, OH, discusses trends in ophthalmology, medicine, and health care with key leaders in their fields. In this issue, Dr. Noreika talks with Joel Schuman, MD, chairman, Department of Ophthalmology, University of Pittsburgh Medical Center.

Dr. Noreika: Dr. Schuman, you are internationally recognized not only as a glaucoma specialist but also for your role in developing optical coherence tomography (OCT), which was a game changer in ophthalmology. Can you tell us how it came to be developed?

Dr. Schuman: OCT is a technology that almost wasn’t. It’s the result of teamwork between engineers, scientists, clinician-scientists, and physicists. It also is an example of why it is important to be aware of what is around you, especially in an innovation-rich environment.

I became involved through serendipity. I was working in the laser lab at Mass Eye and Ear during my fellowship. I wasn’t doing a laser fellowship, but Carmen Puliafito, MD, who ran the laser lab, was kind enough to let me use the technologies however I wanted.

I was studying the gradient of resistance flow from Schlemm’s canal through the sclera. To do that without damaging the tissue with each slice, I was using an excimer laser. The laser was a laboratory excimer laser and it only coincidentally has to do with this story because they were developing a technology in the room next door that would measure the thickness of the cornea. The technology was called Optical Coherence Domain Reflectometry (OCDR). The idea was that you need the feedback for refractive surgery. It occurred to me the technology might be able to get to the retina, in which case we would be able to measure the thickness of the retina in glaucoma and macular disease. Bill Stinson, MD, a retina specialist who was a pre-residency fellow with me in the lab, also was involved in these discussions.

We ended up going to MIT where Jim Fujimoto, an expert in high-speed laser physics and engineering, was developing OCDR. I brought a bag of calf eyes and with David Huang, who was an MD-PhD student working with Jim, we cut them in half and looked at the back half underneath the OCDR beam to see if there would be a signal. There was! 

This told us it could be done. After that, a lot of work by many people went into developing what we have today as OCT. And the idea actually of OCT itself, of creating a tomogram, came from Dr. Huang. He said that if we can do these A-scans, why not just move the beam transversally and we could do a B-scan, and then interpolate between the points and create a tomogram.

Dr. Noreika: How do you use OCT in your practice?

Dr. Schuman: The early iterations of OCT were most useful in terms of ruling out glaucoma as opposed to identifying people who do have it. I still think that is an important use of OCT technology. The second key role for OCT today is in identifying people who have parametric glaucoma, and identifying the areas of abnormality and degree of damage to the retinal nerve fiber layer (RNFL). The third use is to help guide treatment. Being able to assess if there has been a statistically significant change in RNFL thickness is extremely helpful.

One of the keys is knowing when OCT is helpful and when it may be giving a false sense of confidence. . . . If a patient has an abnormal OCT reading but normal visual field and normal pressures, we are very suspicious about glaucoma. If the abnormality is consistent with the location and pattern that I would expect in glaucoma, if it is an arcuate abnormality that is emanating from the supra-temporal or infra-temporal portion of the nerve, that helps discriminate between a non-glaucomatous abnormality and glaucoma.

You can tell if the abnormality in the RNFL is glaucomatous by looking at the macula and the ganglion cell inner plexiform layer or the ganglion cell complex. Very early on in the disease course, the two may not correspond, but before there is a visual field defect, you should be able to find correspondence between the nerve fiber layer measurement and the macular measurement. That is an internal check I use all the time.

Dr. Noreika: If this internal check is positive, do you expect to see something on perimetry? 

Dr. Schuman: No. Early in the disease, you will have abnormalities on OCT without a field defect. We published a paper called the “Tipping Point” in 2012 and in the Shaffer Lecture I gave at the Academy in 2013, I talked about longitudinal assessment of a population and found a tipping point.

Before this tipping point, there is a poor relationship between structure and function. There will be a structural abnormality but no corresponding functional loss and likely no visual field defect. After the tipping point, the relationship between structure and function becomes quite strong. A change in OCT and the visual field is likely to occur around the same time. Early in the disease, all you can use is OCT in someone who has glaucoma damage but normal field and normal pressures. OCT lets me have a high level of certainty before I diagnose glaucoma.

I would consider a patient a glaucoma suspect in the case that I just talked about, in which there is an early abnormality in the nerve fiber layer or macula. . . . If they are progressing, if the location and pattern of the damage is glaucomatous, even without a field defect or an elevated pressure, I would make the diagnosis and start treatment.

The tipping point is at about 75 µm as measured with the Zeiss OCT Cirrus (with other spectral-domain OCTs, the number is slightly different). The 95% confidence interval is about 6 µm in either direction. I would not expect a field defect in somebody who has inner fiber layer thickness in the 80s.

As devices and algorithms improve, that bottoming-out effect will occur at a lower thickness value. But right now, if someone has a mean nerve fiber layer thickness in the low 50s, it will seem to be stable even if the patient is getting worse. That is when you need to go by visual field and not get a false sense of confidence from a stable OCT reading.

Dr. Noreika: So under 50 µm, there is no point in continuing to use OCT? You would really have to use perimetry determine if there is progression?

Dr. Schuman: Right. When nerve fiber layer thickness is greater than 75 to 80, (again, these are Cirrus units), the visual field is unlikely to be very helpful. If it is thinner than about 55, the nerve fiber layers are unlikely to be very helpful. On both extremes, the tissue that is not helpful is likely to give you a false sense of security. 

Dr. Noreika: You have confirmed a very important concept, which is that these tests go hand in hand but we need to know how to use them appropriately. The biggest change I have made in my practice due to using Cirrus OCT is I do not overtreat as much. When you can verify the presence of a thick nerve fiber layer, you feel more comfortable.

Dr. Schuman: Yes, the ability to be sure a patient does not have disease was the major benefit early on of OCT and still is a huge benefit today.