Video Website powered by Einstein IndustriesOptovue, Inc.

DR. JENNIFER LIM: Thanks, Michael. It's my pleasure to be here today to discuss the use of OCT in retinal disease. The OCT machine is I think an outstanding example of applied physics and its usefulness in every day medicine for clinical care. As most of you probably know the OCT was developed by David Huang, who's sitting right over here in the front, as well as Dr. Fujimoto back in 1991. And it was based upon Michelson interferometry, which basically is the principle that if you pass a light through the eye different reflections from different layers will give you different signals, and you can use the interference pattern that is caused by the reflections from these tissues and compare that with a reference meter to generate your images. So currently the OCT machines evolved from initially what is known as time domain OCT to Fourier domain and then most recently to Ultrahigh Resolution OCT, and the ultrahigh I will not concentrate on today that's mostly for research purposes and is not widely clinically available.

So if we look here at the slide we'll see that first of all the OCT machine first came out by Zeiss in 1996 and at that time it was only able to scan about 100 A-scans per second. Within four years the capability quadrupled to being able to scan 400 A-scans per second, and you can see that in the Zeiss Status machine which most of you in this room are familiar with. And then there was a leap in 2006 with the Fourier domain OCT units which were able to scan 26,000 A-scans per second. So you can image with that speed then you eliminate motion artifact and you can get finer resolution and higher detail. And indeed David has compared this to the evolution of airplanes going from a biplane machine, which he likens to the Zeiss Stratus, to that of a jetliner, which he likens to the current Optovue systems and the like Fourier domain systems. And just to emphasize then the RTVue, just like other Fourier domains, can have 65 times the speed and two times the resolution of the Stratus OCT.

If we look closely here you will see a scan done here with the dime domain OCT and with the time domain you can only get one pixel at a time in contrast with the Fourier domain you can actually get 2,048 pixels at one time, again eliminating the motion artifact. This results in higher speed, higher definition and higher signals. If you'll look closely because of this higher resolution you can now see the small blood vessels within the retina in contrast to the time domain OCT, and in addition you can actually see the inner segment outer segment junction of the photoreceptors which is a highly reflective signal. And if you look at the time domain it's very unclear exactly where that is and then you can see perhaps it's here and then there's the RPE, whereas here with the Fourier it's extremely clear above the RPE. And then just to emphasize also motion artifact appears as ripples in the time domain now absent in the Fourier domain, and lastly you can also see coronal vessels with the Fourier domain OCT.

This Fourier domain technology has resulted in a boom with very many companies now producing Fourier domain OCTs. The Optovue system is shown here. There is also the Topcon system, there's the Heidelberg Spectralis system, there's the Zeiss Meditech as well as the OTI Spectral OCT/SLO, optipol caperticus [phonetic] and Bioptigen 3D SDOCT. The Fourier domain, as most of you know because of this higher resolution can also allow us to do 3D imaging, and I think we'll be hearing from that from both Vik and David later on in the glaucoma and the cornea sections. The resolutions now are much better. We can range from three to eight microns in contrast to ten microns just as recently as only six years ago.

So what does a retina look like when we use the OCT? Just for reference you can look here at these blown up pictures and then at the regular Fourier domain OCT at the bottom and you can see that the inner retina nerve fiber layer and the ganglion cell layers are bright as shown here. In contrast nuclear layers are dark so the inner nuclear is dark here the outer nuclear layer is dark. And then as I alluded to earlier, the inner segment outer segment junction is bright but the photoreceptors themselves are hyper-reflective. And lastly the RPE is also hyper-reflective. And again you can see the - - in this Fourier domain OCT.

Most recently the Optovue has also changed their algorithm and has now been able to do frame averaging which increases the fine detail. So you look at the top here's an original OCT frame of the Optovue and then at the bottom is frame averaging. I think with the resolution shown here something with the brightness of the computer it kind of drops out but there are actually two layers here in the RPE that you can actually see which you cannot as clearly see on the non-frame averaged OCT.

So in retinal diseases how do we use the OCT? Well I found it very useful for three specific areas, first of all in the diagnosis of certain conditions, secondly in the management of retinal diseases, and lastly in research applications. First of all in the diagnosis of retinal disease it's extremely useful in detecting mechanical traction that perhaps are not evident to the naked eye. There is also the capability of seeing how much traction is actually caused by an epiretinal membrane and I will show you these examples. You can also see areas of intraretinal edema, presence of subretinal fluid that again may not be detectable to clinical exam, as well as other lesions in atrophic structures.

So if we look at this example this is actually a third generation or time domain OCT taken with the Stratus machine, and if you look here on this eye you can see that the vitreous is attached to the foveal area and that this is causing vitreous foveal traction and this could explain that patient's decreased vision. On the bottom slide here you'll see this fine epiretinal membrane again causing focal areas of adherence to the retina and again focal traction.

This next example is a patient with diabetic macular edema who had seven sessions of focal laser treatment but failed to respond to the therapy. The OCT showed in this particular patient that there was adhesion of the vitreous to the fovea causing this traction and therefore this patient was actually helped by doing a pars plana vitrectomy, and when you looked at this patient clinically it was difficult to detect this and if you look at the antigram there wasn't much in terms that you could do for further focal laser treatment.

Now what about macular degeneration and other degenerative diseases such as epiretinal membranes? Let's look first here on the slide on the left and you can see in this time domain OCT that there's a suggestion of bumps perhaps in the RPE but its not very clear. There's also sort of undulation of the entire retina here and this is due to the motion artifact. Now if we look at the Fourier domain we clearly see these distinct bumps, and I call your attention to the fact that the superior surface of the retina is not corrugated so there is no motion artifact in this Fourier domain OCT. Instead we can clearly see these areas of drusen and below we actually see Brooks membrane as being continuous. Second of all, this patient also had another aging problem that is of an epiretinal membrane. If we look at the time domain we see this hyper-reflective layer here but not as clearly as on the Fourier domain. Here you can see the epiretinal membrane with focal areas of the retina pulled up causing little peaks in that eye and here it is in lower magnification.

What about pigment epithelial detachments? Again, if we look here on the third generation OCT we see an area of cornea neovascularization. We see the RPE. We see subretinal fluid and suggestion of fluid below the RPE. If we look at the Fourier domain we clearly see the different retinal structures and more readily can appreciate these areas of thickening and cystic formation within the retina, the cornea neovascular membrane and the RPE elevation.

Here's another example again of an age-related macular degeneration eye. If we look here you can see that in this time domain OCT that you can't really see Brooks membrane quite as easily, but if you look down here now at the Fourier domain you can clearly see the RPE. You can see Brooks. You could again see the internal segment outer segment junction of the photoreceptors and the external limiting membrane quite nicely.

What about cornea neovascularization? Can we see cystic formation and the answer is yes. You can see it with both time domain, of course, as well as Fourier domain, but again with the higher resolution scans you can now detect that this area here is actually subretinal as opposed to a pigment epithelial detachment and when you look at the time domain it might not be really clear. You might look at this and say well there are intraretinal cysts. There appears to be perhaps a PED or subretinal fluid but you can't really tell for sure, but again when you look at the Fourier domain you can clearly see that it is subretinal fluid and not a PED because you can see the RPE below it over here.

What about geographic atrophy? If you look at the time domain scans you can see that there's a hyper-reflective area here and so that is suggestive of geographic atrophy because of the increased transmission. If we now look at the Fourier domain scan you can see that the inner segment outer segment layer stops right here then it gain picks up over here. So this part is discontinuous and that area is a geographic atrophy. There's loss of photoreceptors. There is also thinning of the RPE in that region.

I find the OCT to be particularly helpful, however, in the management of retinal diseases particularly in the management of cornea neovascularization as well as the management of diabetic macular edema. And the reason for this is you can actually get quantitative measurements of how thick is the retina and how much of a response you've had to your treatment. In addition, you can also see tiny areas of subretinal fluid that you might not be able to see by clinical examination alone and you can also document the mechanical change, which I find is useful when you're discussing your patient's situation with them. They like to be able to see that their retina is getting thinner and that the cysts are going away. I will caution you, however, that the OCT is not a replacement for fluorescein angiography. So it's useful for the management but fluorescein still has a role in the management of retinal diseases.

So here's an example of one of my patients that previously had been treated with another anti-VEGF and then was switched over to Bevacizumab which is off label use of Bevacizumab for cornea neovascularization. And on the third generation OCT the visual acuity was 20/200, the thickness was 534 and you can clearly see this patient had a large PED. There were areas of intraretinal cystic change and perhaps some subretinal fluid here. After treatment this patient's vision improved to 20/60 and the OCT thickness decreased to 203 and you can see here that the cysts are still present, although markedly decreased, but more dramatically this PED has flattened out. You can still see the cornea neovascularization present.

Now if you look at the fluorescein - - and perhaps there is an RPE rip here, there is still some leakage on the - - but if you looked at this you wouldn’t be able to see this - - dramatic an effect of your treatment when you're trying to explain that to your patient. And I show you here both a vertical scan and a horizontal scan from that same patient.

What about in the treatment of cystoid macular edema? Again, prior to treatment multiple cysts within the retina, after treatment normalization of retinal structure. Again, extremely useful when you're discussing this with your patient. Lastly I find the OCT to be very useful in research when we're looking at specific retinal diseases and for example today I'd just like to show you some work that was done while I was at Doheny looking at retinal dystrophy eyes and Fourier domain OCT. And what specifically we wanted to know was could the OCT offer a live biopsy of these eyes with retinal dystrophy. Given that the resolution is five microns could we determine structural changes in the retina early on as well as late in these disease and that these correlate with visual acuity. So I'll just show you the structural changes that we're able to see.

This is a normal retina for reference and what we did was we drew boundaries and a lot of this work was done by Dr. Tomitan [phonetic] as well as David and myself, and what we did was we drew these layers the inner nerve fiber layer shown here, the outer inner plexiform layer, the outer plexiform layer and then the inner layer of the RPE. And then we defined what is inner retina and what is outer retina and we said inner retina thickness is the nerve fiber layer thickness plus the ganglion cell layer thickness and the inner plexiform layer. We then took the total retinal thickness which you can calculate from these scans and subtracted the inner retinal layer to give us the outer retinal thickness. And obviously the outer retinal thickness would be able to help us determine how thick was the RPE, how thick was the photoreceptor layer which we believe were involved in these dystrophy eyes.

So here's an example of these boundaries overlaid in a normal eye and the profiles of the total retinal thickness in microns, the outer retinal thickness in microns and the inner retinal layer thickness in microns. Here's a representative example of a retinal retinitis pigmentosa patient, and if you look you'll probably say well gosh you know the ISOS junction is missing here and I would say that you're right. It is missing in a lot of these dystrophy eyes you can't resolve it and for that reason you can't just measure that thickness directly. And if you look here at the scans you'll see the retinal thickness is slightly decreased compared to normal, but what really jumps out at you is that the outer retinal layer thickness is markedly decreased compared to normal - - but I will show you examples of this. So here's the retinitis pigmentosa eye and you can see here that in comparison to normal, no ISOS junction and marked thinning in these outer retinal layers. If you map out what is normal for the outer retinal layer thickness and you map each patient in the same scale you'll see that the outer retinal layer is indeed thinner than what normative value should be.

Now what about cone dystrophy eyes? We also looked at cone dystrophy eyes and we saw that the cornea capillaries here above the - - was also thinned and this was different from what we had found in the RP patients. If we mapped out the thickness profile in cone retinal, of cone-rod dystrophy we also saw that it was markedly decreased as compared to normative data.

What about Stargardt's disease early Stargardt's disease and late Stargardt's disease? If we look at early Stargardt's disease even when the vision is preserved we begin to see thinning of the retina centrally, and then if we look late we see more generalized atrophy and thinning of all retinal layers and we can map this out. Early on there's more thinning centrally and later on all layers are thinned compared to normal which is shown in gray.

So, overall using the Fourier domain OCT we're able to show that retinal dystrophy eyes had small that is 16% decreases in the inner retinal layer thickness but very severe thinning of the outer retinal layer with 48% decreasing compared to normal measurements. So this higher resolution and the definition of Fourier domain OCT actually facilitated measurements of the thickness of the retinal sub-layers and enabled us to answer our question. Because the OCT Fourier domain has higher resolution it results therefore in less motion artifacts and higher resolution. It enabled us to have its usefulness in diagnosis and management of retinal diseases and also was useful for research. So for those reasons I think OCT really has revolutionized the care of the retinal patient and I'll submit to you that almost every retinal practice in this country uses some form of OCT and that it is extremely useful in the clinical day-to-day applications. So, thanks for your attention.