Lecture: Fluorescein Angiography

This live webinar will start with an introduction/refresher to Fluorescein Angiography basics. Then we will go into detail about its importance to certain conditions, and actual cases will be shown. Even in the age of OCT there are still important reasons to order a Fluorescein Angiography.

Lecturer: Dr. Doug Rett, OD, FAAO, Veterans Affairs Medical Center, Boston, USA


DR RETT: Hello. I love that introduction. My name is Doug Rett, an optometrist here in Boston, Massachusetts. Let me click on some stuff while I introduce myself. I gave an Orbis lecture about a year ago. It was on herpetic keratitis. I want to talk today about fluorescein angiography. The reason I wanted to do this topic is because I feel like it’s a little bit in danger of being forgotten about. We’re doing OCT more and more. It’s a relatively cheaper test, quicker, less invasive. It works. But some conditions still need a fluorescein angiography. There are things that the OCT can provide about anatomical structure, which you can glean, and you can use that information to determine information about perfusion of the tissue. But still, in some conditions, and I hope to show this later in the talk, it’s still vital to see this information about perfusion of the tissue, and this kind of information that only fluorescein angiography can provide. So fluorescein angiography will evaluate the integrity of the retinal and the choroidal circulation. Conditions like diabetic retinopathy, macular degeneration, you can determine factors about them by OCT. But for some things like, say, neovascularization, or capillary non-perfusion, a fluorescein angiography is still kind of the gold standard, and still is in my opinion superior to the OCT in a lot of ways. Certain things like macular degeneration, CNMV activity, you can determine if it’s slow or fast leaking, how active it is, while you’re watching the dye go through the neovascular membrane, and we’ll talk about these unique sine qua non factors from fluorescein. So the talk will be set up with me giving a bit of an introduction to the technology, and then talking about the anatomy of the retina, using fluorescein angiography, and we’ll get to some cases. First a poll question. The first one of the day. I want to see where everyone is. If you wanted to order a fluorescein angiography for a patient, would you be able to have it done within a week? Go ahead and click in. I’ll waste time and talk and stall. There are a lot of people on this talk from all different parts of the world. Some of us have a lot of experience with retinal disease and fluorescein angiography, and some of us are just getting started, so I just wanted to see where we are as a group. So 71% of us would be able to get a fluorescein angiography within a week. Right? So that’s a pretty substantial chunk of us, in my opinion, that wouldn’t be able to get one. Question two. How comfortable would you be interpreting a fluorescein angiography? Some of us have done it in the past, just need a refresher, and some of us have barely even heard about it. I want you, at the end of this talk, to not be perhaps an expert on it, but just get some clinical pearls on it and look it up at your leisure and get comfortable with it as needed. So we’re right in the middle. There’s room for improvement. At least nobody said I have no idea. So that’s a start. How does this work? When I think of fluorescein angiography, I tend to think of the older style camera, with a white light flash. Most of the fluoresceins that are done nowadays, especially where I practice, are done with a laser, with an OCT machine. It’s not a true camera with film in it that you think about. But I find it easier to think about it in terms of these older styles. So it works the same way that fluorescein does when you put it on the conjunctiva and look at the anterior segment surface. You put this reddish, orangish dye either on the conj, or inject it into the veins, shine a blue light on it, and it emits this glow, this fluorescence, in this yellowish-greenish color. You have the wavelengths in terms of nanometers that helps guide you. So people talk about excitation and barrier filter. All they’re talking about is, when the fluorescein dye is in the patient’s fundus, you flash white light of a camera flash with an excitation filter in front of it, that’s basically just blue. Every company has a proprietary wavelength they think is best, that helps excite the fluorescein molecule at its greatest, but the filter is gonna be between 465 and 490 nanometers. The blue light enters the eye, it hits the whole fundus, but it also hits the fluorescein molecules, it excites them, and it emits light at a different wavelength, which is 520 to 530 nanometers. So the light coming back to the camera is a mix of the reflected blue light that you flashed at the patient and this greenish yellowish light emitted from the fluorescein molecule. This brings us to the barrier filter, which is in front of the camera sensor or the film of the camera, and it’s going to block out all the blue light that you flashed into the patient’s fundus. So the only thing hitting the film or the camera sensor is the light emitted at 520 to 530 wavelength. This greenish yellow color. So when you think of a fluorescein angiography picture, you are seeing a picture that’s painted, essentially, only in fluorescein. So when you see something that’s light on that picture, it seems simplistic, but it’s not white or even bright in real life. It simply contains a lot of fluorescein. And when you see something that’s dark, it’s not that it’s dark or missing. It just doesn’t contain fluorescein. Kind of a simplistic concept, but it helps me. That brings us to the concept of autofluorescence. When you see something that’s white in fluorescein angiography, it either contains fluorescein, or just happens to emit light at a wavelength similar to excited fluorescein. The things that do that in the funds are disc drusen, lipofuscin, et cetera. So why do we even use fluorescein? There are a lot of chemicals in the world. People have been using it for a hundred years, and the reason we do it is one, it has a really nice glow that you can pick up either with the naked eye or certain types of cameras. And two is the size of the molecule. It’s small enough that it will fit through the capillaries of the retina, but it’s large enough that it will be contained in the capillaries, provided they’re healthy, and not slip out through the tight junctions. In a healthy patient without any pathology, the fluorescein dye will go through the retinal arteries, through the capillaries, and back into the veins without leaking out. Which is one of the main reasons we use fluorescein. So what are the risks of this? The patient’s skin will turn yellowish orange. Their urine will also turn yellowish orange, which is a very important concept to tell your patient, at least before they go to the bathroom. Otherwise they’ll freak out. 5% get nauseous and will vomit. This happens early on. We’ll talk about how to prevent it. 2% will develop a hive type of reaction. Blotchy marks on their skin, maybe a systemic itchiness. You can also have pain on injection. And this is a picture below of a patient who had injection of dye not in the antecubital vein like you’re supposed to, but the antecubital artery. The fluorescein leeched into the capillaries of the fingers. Very painful. The pH is around 9, so if it gets to where it’s not supposed to, the patient will have this burning sensation. A more common injury is where you perforate the vein and don’t inject the fluorescein into the vein itself, but into the surrounding tissue around the vein, and they’ll have a burning sensation in that area. So be careful when you’re injecting. Serious risks are rare. It’s hard to find real numbers. One of the numbers that you’ll see in the literature is one in a million fluorescein angiographies can result in death. Death typically comes from anaphylactic shock, which is one of the reasons this is typically done in a Medical Center, as opposed to a smaller private practice that doesn’t have access to full health care. If the patient has renal disease, it increases all the risks. This serious anaphylaxis risk, and also the smaller risks. So typically if the glomerular filtration rate is not at a certain level, we won’t obtain the fluorescein. In this country, at least, you need an informed consent, because it does carry risks. Fluorescein does not have any cross allergies, per se. Like, say, if they’re allergic to shellfish, you should stay away from giving fluorescein. ICG has cross allergies to iodine, so if your patient has iodine allergy, you might consider not doing that. These hive type reactions are typically treated with oral antihistamines. The major reactions need urgent care stat. If you suspect the patient is going into anaphylaxis, you need to get them to urgent care immediately. They’ll give atropine and steroids and epinephrine for the anaphylaxis. So how do you do this? In this country, you typically obtain a prediluted solution, titrated to a 10% solution, and inject 5 mils at a time. You inject this as a bolus over 5 seconds. If you pushed all 5 mils at once, you would get nice fluorescein, nice transit, but the patient would be nauseous if it hit all at once. So you’re supposed to inject it slowly, over 5 seconds. We use a 23 gauge needle with a butterfly, so you can tape it to them. You need a two person system. You need one person doing the injection, monitoring the patient for safety concerns, giving the Band-Aid, applying pressure to the puncture, and then the camera — the photographer, setting the patient up, focusing the device, and getting everything ready. We’ll talk later in a couple slides that you start to see dye 8 to 10 seconds after injection. If you’re taking 5 seconds just to do the injection, and then monitoring the patient, applying the Band-Aid, et cetera, you have no time to bring the patient over to the camera and refocus them. So it’s really a two person system. This is a bit of an introduction on how you do it. We’ll talk about the anatomy now. The first thing the fluorescein hits is the choroid. If you think about it, you’re injecting into a vein. It hits the heart, back to the arteries, ophthalmic artery, the first branch of the ophthalmic artery is the short posterior ciliary artery. That’s why the choroid lights up first. It’s the first thing the fluorescein hits. The deeper vessels in the bottom of the choroid are non-fenestrated, impermeable to fluorescein. But above, in the choriocapillaris, you have fenestrated capillaries. So the fluorescein will be retained by the deeper, larger choroidal vessels, but eventually you’ll get a choroidal flush, as they call it, just a kind of soup of fluorescein, that the main larger choroidal vessels are traveling within. The choroidal system is an open system, with watershed areas. It’s kind of like a tree. The main trunk of the central retinal artery, and then you get these branches from it. The central retinal artery, branching, branching, branching, in this tree-like pattern, until all the retinal capillaries are covered. The choroid is different. There’s different lobules, here and here. And they all light up at once, and then ooze their fluorescein and ooze their blood everywhere and eventually fill everything up. But there are area in between two lobules that are known as watershed areas, that just eventually get filled. This is as opposed to the retinal circulation, which is a closed system — unless it’s damaged. So you’ll see the fluorescein and you’ll see the blood being retained in the vasculature. We all know about the foveal avascular zone, right? When you see fluorescein, you’ll see this darkening in the macula. The fluorescein avascular zone is only a third of this. You’ll see a darkening in the macula that’s much bigger than the FAZ. And the reason you see that darkening is because of a block. Most of our macula contains these pigments, these xanthophylls, that act to tamp down the choroidal flush and sometimes the retinal capillary flush. And also a word to the wise — remember that the larger retinal vessels are on the surface of the tissue, and the smaller vessels, the capillaries, are deeper. This is important to think about, when we’re looking at these cases, and you’re trying to figure out what layer you’re in. You probably notice this on your circle scans, your OCTs. These are very superficial. The capillaries are deeper down. Then there’s the RPE. The RPE acts as a fluid barrier. The choriocapillaris has this soup of fluorescein, but it doesn’t get to the retina, because of the tight junction. It also acts as an optical barrier. If it wasn’t there, you would see the choroid flush so much brighter. But particularly if the patient has heavily pigmented RPE, it tamps down the view of the choroid, except in areas where the RPE is missing or depigmented. You can see that windowing defect into the choroid. So there are some things you just have to memorize. This slide is one of them. The times that fluorescein will hit the different structures in the eye. So these numbers will change, depending on the age of the patient, depending on the health of the patient. For vascular paths, these numbers are gonna be later and delayed. Choroidal flush happens 10 seconds after injection. You’ll see differences in these numbers, but it’s important to know a general time frame. You’re expecting the fluorescein to hit the choroid in 10 seconds, 2 seconds later, the retinal arteries, 3 seconds later, laminar flow, 3 seconds after that, it’s in the veins. All of them will be different for vascular paths, but they’ll be delayed relative to each other. Every stage will be delayed. You don’t want to see a normal arterial phase time and delayed venous phase time and blame the patient’s age or diabetes on just the delaying of venous phase. If you see normal arterial time and delayed venous time, it’s probably a vein occlusion. It’s necessarily that you can blame the delay there just on the patient’s age. Try to memorize these times. Consider the late phase anything after 2 minutes. After 40 to 60 seconds, it’ll be what we call equilibrium, where there is the most amount of fluorescein in the retina that will appear there. And this is a good time to look at the capillary beds, the retinal capillary beds, to check for non-perfusion. This is the time where you want to measure how big the foveal avascular zone is. About one minute in. Then you start losing it. These next slides are the same patient, taken from a fluorescein angiography series that I think is cool, that shows the stages relative to each other. So here’s the choroidal flush. You don’t see any dye in the retinal vessels. They’re only in the choroidal vessels. You can appreciate the lobular nature of the choroid. Up in the left of the image it’s relatively bright, a decent amount of fluorescein in the choroid, but superotemporal to the macula, you see relative darkening of the choroid. Probably because it’s an area that doesn’t have a choroidal lobule there, and eventually in the next slide, you’ll see the choroid — we’ll ignore the retina for now — but the choroid is much more evenly filled, isn’t it? Still a little bit dark in this area, superotemporal to the macula, but eventually things are filling in, in the choroid. The fluorescein is in the retinal arteries now. But not in the veins. The next phase is laminar flow. Laminar flow is not just the dye is starting to come back into the veins, in laminar flow, but it’s not just a term that we use in fluorescein angiography. This is a term in fluid dynamics for the most efficient flow of a liquid through a tube or a cylinder. It flows in these sheets and starts off in the periphery of the tubes. The sheets are why they call it laminar flow. If it hits an obstruction, say a clot, it’ll turn into turbulent flow, which is not a very efficient way for fluid to move down a tube. So when you’re in this stage of laminar flow, don’t just look at the veins and identify laminar flow and move on. Look for crossing changes, like you see in the picture down and to the right. You see an artery crossing over a vein. It’s easy to forget that in veins, the flow of blood is moving opposite. Right? It’s moving towards the optic nerve. So this proximal part of the vein, before the crossing change, is in laminar flow, and distal to the crossing change, it’s also in laminar flow. But if there was some kind of compression there, where the patient is at risk for vein occlusion or thrombus formation, you might see laminar flow proximal to the crossing change and then filling in of fluorescein in the middle of the vessel distal to the crossing change. So look for the loss of laminar flow around crossing changes when you’re suspecting a vein occlusion. After laminar flow is venous phase. This is typically 20 to 30 seconds after injection. And you see the dye completely filling the veins. Late stage is anything after two minutes. You see the disc much brighter than the retina here. This is not because the disc is leaking. It’s because the optic disc stains. It diffuses and picks up dye as the study goes along. So in the late stages, you’ll see that staining of the optic nerve. I’ll show you some pictures later about the difference between staining and leakage. Some terminology. When we talk about OCTs, we’re talking about reflectance. Right? Hyperreflectance? Because this is essentially just B scan ultrasounds. When we’re talking about fluorescein, we’re talking about fluorescence. Hypofluorescence and hyperfluorescence. So you kind of get your terminology down, when you’re assessing fluorescein. Don’t talk about — this is really white and this is really black. You’re talking about hypofluorescence and hyperfluorescence. The main reason to get hypofluorescence is blocking problems. Oftentimes hemorrhages. On the bottom left, you see a patient with some kind of retinopathy, whether it’s vein occlusion or diabetic retinopathy. You see a mix of flame-shaped hemorrhages and dot and blot hemorrhages. The flame-shaped hemorrhages are obviously superficial. That’s where they get their name, that’s where they get their shape. And you can see some of these flame-shaped hemorrhages obscuring the vessels underneath it, blocking the fluorescence and causing hypofluorescence in that area. So you know this is a flame-shaped, one because it’s that shape, but two, it makes sense why it’s that shape and blocking everything underneath it. As opposed to the hemorrhages on the right hand side of the shot — are deeper, and you see some of the retinal vessels traveling over this blocking area. So it’s blocking the capillary perfusion, the retinal capillary perfusion, but it’s not blocking the larger superficial vessels. So you know the hemorrhage here, the blocking lesion, is deeper into the vein. But you know it’s within the retina, as opposed to this middle fluorescein, or at least the one on the bottom right hand side. It’s a much different shape, right? It’s in the shape of a subretinal hemorrhage, which is a pooling of blood under the retina, but above the RPE. And you can tell that it’s under the retina, because this large retinal vessel goes right over it. So this hyperfluorescence is blocking your view of the choroidal flush, but it’s not blocking your view of these major retinal vessels. So you can tell what tissue you’re talking about in fluorescein angiography one by the shape of the lesion that you’re looking at, but two, by the characteristics of what it does and does not block. So keep in mind the anatomy that we talked about. The other way you can get hypofluorescence in fluorescein angiography is a filling defect. Right? Something that’s supposed to be bright, but is not bright. This is a nice shot in a patient with capillary dropout from an ischemic branch retinal vein. This is supposed to be filled with fluorescein in the retinal capillary beds, but it’s not. The opposite of hypo is hyperfluorescence. Here’s a patient with a window defect. The same patient, these pictures were all taken on the same day. The colored photo shows a somewhat obvious RPE problem in the middle, but in my opinion, the fluorescein angiography and the autofluorescence does a better job of capturing the exact dimensions of the spot of dropout, window defect. In the fluorescein angiography, down and to the left of your screen, you see a window defect. You see the choroidal flush, in this area in the macula, where you don’t in other areas. This choroidal flush is happening all over the choroid. Right? It’s just that it’s blocked in the non-macular areas by the optical barrier of the RPE, versus when the RPE is either missing or without pigment, you’re able to see that choroidal flush so much better. This fundus autofluorescence, the down and to the right picture, we don’t have a lot of time to get into fundus autofluorescence, but I think it’s kind of fascinating. Essentially in a nutshell, fundus autofluorescence is the imaging of lipofuscin or autofluorescence in the RPE, in the retina, right? Autofluorescence is one of those things — lipofuscin fluoresces at that wavelength, even though there’s no dye. So if you’re looking at autofluorescence, you’re looking at the level of lipofuscin in the patient, and lipofuscin is stored in the RPE. So when you’re looking at this patient down and to the right, you’re seeing relatively healthy RPE everywhere else, but missing RPE in the middle, where there is no RPE. So you can get a clear definition of those things here. Other ways to get autofluorescence is diffusion, leaking, and pooling. So it’s important to remember what layer you’re in. Diffusion of the dye typically refers to that fluorescein that’s in the swamp of the choriocapillaris. It’s touching the bottom of Bruch’s membrane, which eventually is gonna suck it up, and the dye is gonna be diffused into Bruch’s membrane. Sometimes there are drusen, they’ll suck up the dye, it’ll diffuse into that, and you’ll see the dye manifesting more in drusen, later in the study. So drusen don’t necessarily leak fluorescein, but they absorb fluorescein. And you can see this on the left. The middle one is cystoid macular edema. This makes sense, because in cystoid macular edema, the capillaries are damaged, and the smallish fluorescein molecule is escaping out and you get this classic petaloid appearance. But it’s escaping out into the tissue, as opposed to the picture on the right, which shows diabetic retinopathy and neovascularization. Where it’s escaping above the retina, into the vitreous. Which makes sense with what we know about neovascularization. You also can tell that it’s in the vitreous, because it’s obscuring everything below it. You know it’s leaking above the retina, because in this area where it’s leaking, you can’t see anything underneath, as opposed to the fluorescein in the middle, where you can see some retinal vessels underneath it. So this brings us to poll question number three. What does this fluorescein show? Subretinal pooling? Drusenoid staining? Intraretinal leakage? Or preretinal leakage? So what do you think it looks like here? Subretinal pooling typically happens in an active and acute central serous retinopathy, where there’s misfiring pumps in the RPE, and it’s actively pumping out the fluorescein dye from the choriocapillaris to the subretinal space. That’s not really what you see here, right? Drusenoid staining is not really what you see here. Thankfully, not a lot of you clicked on drusenoid staining. It’s between intraretinal and preretinal leakage. But the answer is number four, preretinal leakage. You can tell it’s preretinal leakage because the fluorescein dye is uncontained. It’s starting to leak into the vitreous and it’s not contained by anything, and it’s gradually getting bigger and bigger, almost like if you dropped a drop of blood into a glass of water. The blood is not gonna be contained. It’s gonna diffuse everywhere. Just like if you dropped fluorescein into a liquid vitreous chamber, it’s gonna diffuse everywhere and not be contained in the retina. You can also tell it’s preretinal because it’s blocking everything under it. If it was intraretinal leakage, you would see some of the retinal vessels going over the leaking tissue and the damaged tissue. Third, you can tell it’s preretinal leakage just from what you know about the rest of the retinal anatomy. You can see on the left hand side of the film a dramatic area of capillary non-perfusion. And on the left side of the film, capillary perfusion. And where does neovascularization happen? On the border of perfused and non-perfused. This is a large area of neovascularization that’s starting to leak into the vitreous. Preretinal. Okay. So it’s a bit of an introduction. And an anatomy lesson on fluorescein angiography. Let’s talk about some interesting cases. I don’t think I mentioned this before, but there’s a Q and A session afterwards. I’m gonna try to save all the questions to the end. I apologize if you’re asking questions right now. But I’ll be around after the talk. And we’ll try to get to all the questions that we can. Case number one is Jimmy. He’s a 54-year-old Black male. Just here for a routine eye exam. Doesn’t really have any complaints. I’ll save his medical history for the end. But he’s seeing great. He’s 20/20 in each eye. On fundoscopy, I see this. So it’s relatively clear media. The disc looks distinct and flat. The vessels seem a little unusual, superior to the macula, right? You see what at first blush to me look like collateral vessels. Then I looked a little bit temporally, and saw a little bit more collateral vessels, but then something very unusual. This large ring of exudates. And then two fibrovascular areas, smack-dab in the middle of the exudates. So my first thought was… I think this guy had a vein occlusion and had collaterals developed from that. But then temporally, I saw this large ring of retinal exudates, and one of the things I think when I think about that, is a retinal arteriolar issue. I tried to get slides of this. It didn’t come out, it was off-axis, and there was so much edema I couldn’t figure out what the vessels were that were feeding it. In the middle of this ring it looks like one vessel coming up from the bottom and one coming down from the top. So I wanted to determine whether those vessels were arteries or veins. Because it would help me to determine whether it was an arteriolar macroaneurysm or some kind of venous problem. Here’s a shot of the left eye. Relatively normal. Clear media, good optic nerves. Macula is fine, RPE changes superotemporal to the macula, but spoiler alert, it doesn’t have anything to do with the case. How can we tell what’s going on with the right eye? We can get a fluorescein angiography. Here’s a shot 22 seconds in. What phase are we in? Laminar flow, since your mics are muted. I’ll just answer for you. You can see some of these unusual vessels that I was talking about. They are venous vessels. I’m not sure if you can see my mouse. You see this hyperfluorescent artery coming over this way, and you see these unusual vessels going under it. Right? So you can see these are collateral vessels coming from the vein, and I think you can see it a little bit better in this shot. Just two seconds later, it’s so much brighter. You see so much fluorescein into the eye. And these unusual vessels are all coming from this vein. If you travel it back, you see this occlusion. Just superotemporal to the disc. This occlusion of the retinal vein. That’s blocked. And this is why all these collaterals are happening. These are vein to vein collaterals, crossing the horizontal raphe, and making their way, letting blood exit the eye, getting around this occlusion. So that kind of confirms our suspicion that this is a branch retinal vein occlusion in the supratemporal arcade. You can see if you go very close to the disc — you can see where the pinch point occlusion is happening. You see the artery here, the fluorescein is quite bright in the artery, this is a bifurcation here, and right at the bifurcation there’s a trifurcation of the supratemporal retinal vein. And this is where the pinch point is happening. And this is where the occlusion is happening. And that makes sense, right? Vein occlusions happen where arteries cross over veins. So we see collaterals crossing the horizontal raphe, we see evidence of the vein occlusion that was hard to see on the fundus picture. But what’s happening temporally? This is 20 seconds in. We’re laminar flow, getting close to venous phase. Here’s 29 seconds in. You can see spots of fibrovascular changes there. And remember the question was: Are these arteries or veins? Let me back up. 24 seconds in, you can see the vessel supplying this fibrovascular lesion coming from the top. The question is: Is it an artery or a vein? And we can see it’s a vein. Do you believe me here? You can see the superotemporal artery coursing and crossing over this area, which is a five-way bifurcation, and a vein coming over, crossing over that, and crossing out of area temporal to it. So the superior part of this lesion is supplied by the vein, and if you come to the next slide, you can see that the inferior portion is also supplied by a vein. Both of these are kind of confusing, and I couldn’t really tell, live, because you kind of get lost in the retinal vasculature, when you’re out there. But this is a 55 degree, somewhat wide angle fluorescein, which really helps in these lesions. If you’re just looking at the macula, you can just do regular fluorescein angiography, but the wide field angiographies are very helpful for peripheral things, and it gives you just an overall view of what’s happening. So I think the next shot is a nice one. Really lights up everything. And you can tell about what’s happening in our circinate exudate lesion. This is a collateral vessel. Just like there’s collateral vessels kind of all over, when you really look at the vessels here. There’s a lot of collateralization. It’s just that the collaterals temporal to the macula are poorly compensating and leaking so dramatically that the ring of exudate is almost threatening it. Here’s a late stage looking up. Where you see more collateralization, and then the final shot here. You can see the blocking, a little bit of blocking, of the retinal capillaries and the choroidal flush of those hard exudates, right? The lesion is leaking into the retina. You know, when I was in school, I was taught that neovascularization leaks, but collaterals don’t leak, which is true for the most part, right? But sometimes when it’s a bad enough case, collaterals will leak a little bit, but they won’t leak into the vitreous. They’ll leak into the surrounding tissue, like this is doing. Left eye, relatively normal vasculature. You see a little bit of mottling temporal to the macula, but no big deal. So the diagnosis here is an old branch retinal vein occlusion, with an overloaded or poorly compensating collateral vessel. Johnny has a lot of vascular problems, but primarily has this uncontrolled hypertension. His blood pressure has been bad for about two decades. And right before I saw him, he stopped taking his meds completely, and his blood pressure that day was 205/100. We brought him to the emergency room and got his blood pressure lower, and talked to him about long-term care, and he got under much better control, and this eye is still doing well. Okay. Save those questions to the end. Case number two. 63-year-old White male. He complains: When he plays golf, he loses the ball too quickly. He’s had cataract surgery in his left eye, but it’s the right that has poor vision. The entrance tests were unremarkable. These are the fundus photos. To me, the media are relatively clear. A little bit blurrier in the right eye, but that’s to be expected. Optic nerves look fine. The vasculature looks okay too. Maybe increased light reflex and some crossing changes, but it’s the macula that looks a little bit concerning in the left eye, doesn’t it? You don’t see the foveal light reflex that you would expect in a pseudophakic patient. So what do we do? Get a fluorescein. See what’s happening. This is 21 seconds in. What phase are we in? Silence. Arteriolar phase. Everything seems pretty normal. The choroid seems a little patchy. Just three seconds later, you’re starting to see the dye come back, you see laminar flow, and you see the capillaries, so much better illuminated. Better lit up. I think if anything you notice the capillary vessels more on the nasal side of the macula, versus the temporal side. When you go to the one 38 seconds in, it’s nice. I love this shot. The retinal capillaries seem relatively perfused, besides the macula. We’re in the venous phase. This is around equilibrium, where you can start to judge capillary non-perfusion, and around the fovea you see these highly irregular capillary vessels in the fovea, and some of them appear like they’re starting to leak, especially on the temporal side. A minute, 18 seconds in, same thing. A little bit darker. You’re starting to eventually lose dye from the eye, but where you’re not losing dye is this temporal area. It’s leaking, but it’s leaking into the retina, not into the vitreous. At least, not yet. But let’s follow it. 3 minutes, 10 seconds in, the dye is leaving, but it’s not leaving the temporal area. Six minutes in, you see even more leakage on the nasal side, right? That wasn’t as evident in the early stages. But now you see in the later stages that there’s leakage into the retinal tissue, kind of everywhere, 360, but it’s worse temporally. How do I know it’s not in the vitreous? Because you can see retinal vessels going over this area of hyperfluorescence, so you know it’s at least under the superficial part of the retina, and the shape and the characteristics of the outline of the fluorescein shows to you that it’s in the retinal tissue, as opposed to, say, under the retina. What does the next eye look like? Here’s a minute into the fluorescein, and you start to see changes in this eye too that I didn’t really pick up on. A very astute observer might notice this right angled venule just temporal to the macula, that seems to be purposefully draining this irregular area temporal to the retina. Three minutes in, you see the same style of leakage that you saw in the left eye, and in a late stage, you realize that the edema is even more so. It’s worse on the temporal side, but also happening on the nasal side, just like in the left eye. These pictures were taken several years later, as this condition got worse, but it’s the same patient. You can see the same right angled venule, and this classic sine qua non type of finding of intraretinal flecks, crystalline deposits, and this ILM drape that you see on OCT. So it brings us to the final poll question, I think. What is the diagnosis for this patient? Is this diabetic macular edema? There is macular edema, right? And he’s diabetic. Macular telangiectasias? Ocular ischemic syndrome? Remember, it’s bilateral, you have a bilateral macular edema. Or retinal vein occlusions? It would be a little bit weird if it was bilateral retinal vein occlusions, and it would be a little bit weird if I did two retinal vein occlusions in a row, but anything’s possible! All right. Vote early and vote often. Most of us think it’s macular telangiectasia, and most of us are right. Diabetic macular edema is a good guess. Let’s talk about why it’s mac tel. It’s this slide right here. You see this zoomed in, in close-up, that you simply did not see in the color photo. That’s why I talk about the benefits of fluorescein angiography. It gives us data that structural and anatomic OCT can’t give us, and it illuminates these microscopic capillaries that you can’t see on fundus photography or fundus examination. Maybe you could see some of these on fundus examination, but not nearly as dramatic as you see here on the fluorescein. So the reason why this isn’t macular edema is because you don’t see a lot of capillary ischemia. Or, frankly, diabetic retinopathy, in this patient. But you see evidence of all these irregular or telangiectatic capillaries around the macula. And this is all classic for mac tel. It’s worse on the temporal side, you see right angled venules, you see the ILM drape on the OCT, and the classic intraretinal flecks. Mac tel is interesting. For the sake of time, we’ll skip this slide, but if you want to learn more about mac tel, check it out. I think it’s really kind of fascinating. But this will give us time to get to our last case. Case number three is a 26-year-old White male. Never had an eye exam before, but he’s concerned. He’s had blurry vision in his left eye for two days. He comes in to you. Medical history unremarkable. Healthy 26-year-old, a little bit of asthma. But his left eye is down to 20/40, whereas his right eye is 20/15 or even better. When you look at the fundus in the right eye, you see clear media, sharp, distinct nerve, very healthy. The left eye is different. It’s relatively clear media, but you see these indistinct margins of the optic nerve in the left eye, and you see some retinal folds, just temporal to the optic nerve. The vessels seem normal to me. But in the macula, you see this unusual change, almost like at the level of the RPE. So you see some swollen nerve, some changes in the macula, and some changes elsewhere, that I’ll be silent on for now. What is the diagnosis here? I don’t know. Let’s get a fluorescein. 14 seconds into the fluorescein, in the left eye, we’re in arteriolar phase. Seems pretty normal. And this is about where you would expect a patient to be in a fluorescein angiography in a healthy patient. 13, 14 seconds, artery phase. 4 seconds later, 18 seconds in, we’re in laminar flow, a little bit delayed in the vein on the nasal side. I think that’s maybe just an anatomical variation. But the vessels seem good, the macula still seems good in my opinion. 25 seconds in, we’re in full venous phase here. Right? Still everything kind of seems okay. But when you have the patient look a little bit nasally, we’re 46 seconds in, you see some kind of change here. And the more you look at this fluorescein, the more changes you see, and they’re all at the level of the RPE. Can I convince you that this is a deep hyperfluorescence that you’re seeing in these kind of splotchy areas? You see the splotchy areas more nasally. Probably simply because there’s less dye over there. If you look around the nerve, especially nasally to the nerve, you see it’s all splotchy there. One minute, 22 seconds in. Similar, but you’re starting to see some leakage at the optic nerve, nasal spot of the optic nerve. Four minutes in, kind of the same thing. The nerve is leaking a little bit more. About six minutes in, the nerve is leaking. The fluorescein is mostly out of the eye. He’s a relatively healthy patient, it’s draining out quickly, but it’s leaking around the nerve and still there around the nerve. When he looks nasally, you see those splotches again. Here’s the right eye. The unaffected eye. Midstage, this is kind of what a fluorescein angiography is classically supposed to look like, everything is fine, and a late stage here. I put this in, because I wanted to show you staining of the optic canal. This is a normal amount of staining. The optic sheath is just taking up the dye, as opposed to this shot, which is late stage, which is more actively leaking, right? So what’s the diagnosis for this guy? It’s a tough one. But it’s one that the fluorescein can help us find. You had a hard time seeing it on the color fundus photo, right? This is MEWDS. Multiple evanescent white dot syndrome. Fluorescein angiography is one of the best ways to differentiate all the white dot syndromes from each other, and the reason why it’s important to differentiate which white dot syndrome it is is because they’re treated differently. MEWDS is classically unilateral, like our patient. Young person from 15 to 40. And often female. Our patient wasn’t female, but two out of the three. And you have this acute multifocal inflammation of the outer retina or RPE, like you saw here, and you see in this window that I have right there. A mild vitritis, which our patient didn’t have. Maybe slightly when he came back for follow-up. And unilateral disc edema. So the classic finding for MEWDS which separates it from the other white dot syndromes is that it stains early and it stains late. It was hyperfluorescing a few seconds into the fluorescein. There are other white dot syndromes that block early and you have hypofluorescence, and as it goes into the mid- and late stages, it’ll pick up stain and you’ll see hyperfluorescence. So these white dots have a reflex staining, which is kind of like a donut-shaped staining. And these patchy areas, which are kind of — I blew one up to convince you maybe that it looks like a wreath or a donut. MEWDS is self-limiting. We did not treat this patient. He got better each week he came back. But there are some white dot syndromes you want to rule out with the fluorescein, that are treated, sometimes with IV steroids, with hospitalization. So getting the right diagnosis is important. All right, what did we learn? Look at the whole fluorescein transit. Right? Look at the times. Don’t just jump to the end of the stage and see what’s leaking and where the hot spots are. Look to see if there’s delay, look to see if there’s delay in each phase relative to each other. Look at the superior fundus versus the inferior fundus and see if it matches up or if there’s asymmetry there. Is there asymmetry between the right eye and the left eye? Is this spot that you see and that you’re concerned about leaking into the retina, under the retina, or over the retina? You can tell that, based on what we learned in fluorescein characteristics. Is the dye pooling, in between two tissues, or is it leaking into the tissue, or is it staining and just kind of absorbing the dye? So we’re not gonna abandon the OCT. We’re not arguing for that. But I’m just arguing to save fluorescein angiography, and to keep learning about it and keep refreshing about it, because there are some things you can tell with the fluorescein, like this, the presence of a cilioretinal artery. Is it, is it not? You can tell on fluorescein angiography. Is there silent choroid sign, like you see in Stargardt’s disease, this storage disorder, the lipid builds up and blocks the view of the choroid? So you don’t see it in Stargardt’s disease. And the white dot symptoms that we talked about. It’s very important if you have one, if you suspect one, to get a fluorescein to differentiate between them. All right. So we did it! We did it in under an hour. Thank you for your attention. I’m gonna hit this stop share video. We’re gonna go to the question session. Pull up some questions here. So feel free. I’ll be on for a while. We’ll just talk and answer questions. If you have to go, you have to go. If you want to stay and listen to the questions, feel free. Question number one comes from Sunanda. I’m sorry in advance if I mispronounce your name. Question number one comes from Sunanda. In a case of PPCRA and Eales disease and toxoplasma scar, what would be the classic sign in fluorescein angiography? It doesn’t matter how to spell it, but Eales disease is peripheral ischemia of the retina. So you’ll see capillary non-perfusion in the periphery. And it’s often bilateral. Right? So it’s a diagnosis of exclusion. If you see a patient with capillary non-perfusion in the periphery, you want to rule out all the classic things like tuberculosis, sarcoid, toxoplasma maybe is one of them, and if you rule everything out and it’s bilateral and the patient fits the profile for Eales disease, you make that call. But it’s one of the reasons why wide field fluorescein angiography is important to get out there. I think we’re seeing it more in diabetic retinopathy and in a lot of these vascular diseases, that peripheral ischemia and peripheral capillary non-perfusion is much more common than we thought it was. Toxo is much different than Eales disease. You’ll just see these one or two hot spots, where it’s active disease, it’s leaking into the retina and into the vitreous, and then eventually it kind of becomes quiescent and it just scars over. Peripapillary CRA is interesting. We didn’t get into this, but the difference between CRA is kind of like a window defect, right? The difference between a window defect and something that’s actively leaking is you look at the fluorescein as the study goes on. A window defect will not increase in its intensity of hyperfluorescence. It’ll stay the same and die off and go away. Whereas something that’s actively leaking, like an active toxo scar or neovascularization will increase and crescendo in its hyperfluorescence. Good question. Next one is from SM Abdullah. How to differentiate hyperfluorescein from pseudofluorescein? This is a good question. Hyperfluorescence from autofluorescence, right? Say you see a spot in the fluorescein that is hyper. How do you know it’s a disc drusen that’s automatically doing that, or something that’s picking up the stain? The answer here is the same as I just gave for the chorioretinal atrophy. Early in the fluorescein, sometimes even before the dye has entered the eye, you still should see autofluorescence, because you’re shining this excitation light into the eye and receiving the autofluorescence back. So you’ll see autofluorescence in the seconds where the photographer is capturing pictures of the eye before fluorescein even goes into the eye. So autofluorescence is seen at all stages of the fluorescein. It doesn’t even matter if dye is in the patient. So that’s how you tell the difference there. Dmytro shouts out — hello, Dmytro. Mohammed Amine asks if I could explain the difference between diffusion, pooling, and leaking. Diffusion is just dye being absorbed. Almost like if you put a sponge next to water. The water isn’t actively leaking into that sponge. It’s just being absorbed by it. The example we used for that is Bruch’s membrane. Right? Bruch’s membrane is right on top of the choriocapillaris. And the fluorescein in the choriocapillaris is gonna be absorbed by Bruch’s membrane, and the drusen that are in Bruch’s membrane are gonna absorb the fluorescein. So at the end stages of the fluorescein, you’ll see the drusen light up. It doesn’t mean the drusen are leaking fluorescein. It just means they’re soaking up the fluorescein. Versus pooling — pooling and leaking are kind of the same thing, but pooling is a term that we use when the fluorescein is going kind of in between tissues. Like, say, maybe under the vitreous, but over the retina. It’s pooling in this area. Or classically, under the retina but above the RPE. It’s pooling in that area. Whereas leaking is the same thing, but it’s going into a tissue. I hope that helps. Logan Vk asks: We link fundus photo, OCT, and fluorescein/ICG to come to a diagnosis. How do you link FAF with FFA? That’s a good question. I wish we could talk more about fundus autofluorescence. How do we do it at the hospital I work at? If we’re thinking about a diabetic retinopathy case or that Eales disease case, something like that, I don’t really care about fundus autofluorescence. It’s something that I care about for RPE diseases. Right? So classically, I use fundus autofluorescence for central serous. At the hospital I work at, we used to have a lot of cases of central serous or macular degeneration. These are classically RPE diseases. If we shoot a fundus autofluorescence and a fluorescein angiography, we’ll typically do some shots before, fundus autofluorescence before and after. So we’ll see what’s leaking on the fluorescein angiography, and we’ll compare it to those shots of the fundus autofluorescence. Alexandru asks: Do you recommend any blood tests before performing fluorescein angiography? This is a great question. And I would recommend kidney function tests. The fluorescein is filtered by the kidney. And so we typically do… If the patient is… If the patient has a kidney function test within the last month and doesn’t have any history of kidney disease, then we’ll just let it ride. But say the patient does have a history of kidney disease, or you just don’t know much about the patient, and you suspect kidney disease, that’s what I would be looking for. So I would say no, besides kidney function tests. And kidney disease is a serious consideration. And we’ve stopped or withheld fluorescein angiography because of concern of kidney tests. Sagarika asks: Is it indicated in pregnant women? I have to admit that I don’t know the answer to this. I would say that it’s probably a category C, if I thought about it. I work at a veteran’s hospital. We don’t see a lot of pregnant women. But I would imagine most clinicians would not perform it on pregnant women. Because it’s rarely a test that is that vital. Right? So I wouldn’t think that most people would do it on pregnant women. Dmytro asks: What types of artifacts are most common in performing of fluorescein angiography? What types of artifacts? That’s a good question. Probably the answer is the same artifacts that would happen in fundus photography. Cataracts. When you’re doing it with a bright camera flash, kind of like the old school way to do it, you get a lot of light into the eye. So a little bit of cataract maybe isn’t — the image isn’t taken down by a little bit of cataract. But some of the newer styles have this kind of autofluorescent laser, which sometimes is blocked more or less by media. The artifacts are sometimes a really bright light, and so the patient will just kind of freeze up, and squeeze up, when you’re flashing it. But I would say media is the most common artifact. Cataracts. Asdrubal asks: Is it possible to do a good quality fluorescein with only 1 CC of fluorescein? That’s a good question. I never thought about that. I would guess it would just be weaker. That’s a good question. I had on my slide that you can even do oral fluorescein, if you’re in a place that doesn’t do injections. But nobody really does this. You don’t get the beginning phases in oral fluorescein. If you ingest fluorescein, it will be several hours until enough of it shows up in your eye that you can take shots of. But it’s just really weak and you’ll really just get the late stages of the fluorescein, where most of the fluorescein is gone from the eye and you see what’s leaking. Can you inject 1 CC of fluorescein? I think you could do it, and just get a really dim fluorescein, and it would be hard to tell and you have to squint your eyes. But you can try it. Dakki Sherpa asks: Will fluorescein help in the era of OCT angiography? How does fluorescein defer from OCT angiography? This is a good question. And a lot of people will say that OCT angiography is the future. And I think it is. I’m excite to get an OCT A. We’re going to get one in the hospital in a couple months. The biggest plus that OCT A can do is capillary non-perfusion. Right? OCT A, and again, we don’t have time to get into this. This would be a nice talk to do, just like FAF and OCT A… OCT A doesn’t necessarily show dye leaking out from a vessel. Right? Whereas fluorescein can show that dye leaking out. OCT A kind of specifies more in, say, capillary non-perfusion. Which is very important, which is a lot of the reason why we do fluorescein angiography. But would it be able to tell the difference between a slow leak from a CNVM? It would be able to show you one, maybe even that it’s leaking a little bit, but would it show you the difference between a fast leak and a slow leak? I’m not sure. So it’s kind of more of a structural change, like in non-perfusion. Samrat asks: What are the findings of early CNVM in fundus fluorescein angiography? And the findings of early CNVM are probably easier to see than late or advanced CNVM. Early CNVM in macular degeneration will light up first, right? This is one of the reasons why it’s important to look at the early stage of the fluorescein angiography. Is if there’s some vessels in the choroid that you suspect are a neovascular membrane, those vessels will light up as the choroid lights up, and not as the retina lights up. So there’s also a difference between occult and classic CNVM. If it’s above or below the RPE, and your ability to see that. If it’s classic and above the RPE, you’ll see those choroidal vessels light up early in the fluorescein. And then the classic findings of CNVM on fluorescein is late stage leakage. In the CNVM, it’s leaking, and often under the retina in the subretinal space. We see subretinal fluid in an OCT. That fluid is actively leaking, and will have fluorescein dye in it. So you look for late stage leakage in fluorescein. Moupiya Das asks… What do we specifically differentiate in OCT A and fundus fluorescein angiography? I think that was the other question. I’ll skip that one. Anonymous attendee asks: Is there a book that you recommend on fluorescein angiography? There are not a lot of books on fluorescein angiography, to be honest with you. I think that you and I, anonymous, should write one together. I have a book over on my bookshelf there, that I hesitate to reach for, because I’ll probably knock my camera over, but it’s from the mid-80s. But it’s still great. It’s called manual of fundus fluorescein angiography by Chopdar. I like that book, even though it’s crazy old now. Because sometimes old books give you information that has been forgotten over the years. So I give this talk maybe once a year, and I always try to reread the beginning chapters of that book, to refresh myself. The CEO of Samsung, Galaxy S7 edge asks… How often should FFA be done? Thank you for the compliment. I’ll answer in terms of wet macular degeneration, because probably the most common reason to perform fluorescein angiography is mention. It varies depending on the clinician, but at my hospital, we perform a fundus fluorescein angiography when we’re first considering an injection. If we find OCT suspicious enough that it would warrant an injection, we get a baseline fluorescein to see if there’s anything that we’re not thinking of. Should we expand our differential diagnosis? That maybe the vasculature can tell us? And after the fluorescein angiography, if it’s still the diagnosis, we do the injection, and we do follow-up exams mostly with OCTs. We don’t repeat the fluoresceins too often. Next question is by anonymous. In your clinic, the fluorescein is performed by a nurse or a doctor? In my clinic, we have a photographer, who is shooting the pictures, and he’s a certified ophthalmic tech. And we have a person doing the injection, and that person frankly changes by the day. Sometimes we have nurses there, and sometimes we grab the resident doctors to do the injection. But the person taking the pictures is always an ophthalmic photographer, and the person doing the injection is some kind of certified nurse or doctor. Depends on how busy we are. Moupiya comes back at us again. Why do we prefer BAT for chloroquine function or any toxic effects? This is like a brightness acuity test maybe for hydrochloroquine function? I’m not sure. I don’t usually do BAT for chloroquine. I just like OCT and fundus autofluorescence for Plaquenil or chloroquine toxicity. Dakki thanks us. Thank you. Isashah asks… Could you please say the name of the book again? I would be happy to. I wonder if this guy is still writing. It’s called Manual of Fundus Fluorescein Angiography. And it’s by Chopdar. I think the first name is Sonjay. And Monica — shout out to Monica, congratulations on the new baby, and thank you for complimenting my coffee mug. All right, guys. Thank you for your attention. I’m gonna end it here, but I hope you have a great day and a happy weekend.

Download Slides


Last Updated: October 31, 2022

5 thoughts on “Lecture: Fluorescein Angiography”

Leave a Comment