Lecture: Neuroimaging in Ophthalmology

This live webinar covers indications to order CT, CTA, MRI, MRA and other relevant neuroimaging modalities. Numerous scans are shown illustrating many neuro-ophthalmic conditions.

Lecturer: Dr. Karl Golnik, Chairman, Department of Ophthalmology, University of Cincinnati College of Medicine, Cincinnati, Ohio, USA


(To translate please select your language to the right of this page)

DR GOLNIK: Greetings! This is Dr. Karl Golnik, coming to you from the University of Cincinnati in Cincinnati, Ohio, in the United States. Happy new year to everyone. This is my first webinar of the year. And we decided on a topic: Neuroimaging for the ophthalmologist. So I’m hoping to show you lots and lots of neuroimages and try to cover some of the more common things that we see in ophthalmology, at least from a neuroimaging standpoint. One of our fathers of neuro-ophthalmology in the United States is Bill Hoyt. And Bill Hoyt said this about neuro-ophthalmology: Neuro-ophthalmology is that subspecialty where the diagnosis is made upon reinterpretation of allegedly normal scans. So I think this is certainly true. And I routinely look at MRIs or CT scans where they were read as normal, and actually are not normal. And I will show you some number of examples of that, as we go through the webinar today. So my objectives are: When we’re done, you will be able to describe what type of neuroimaging to order, you’ll be able to list some of the pitfalls — pitfalls means some of the difficulties, I guess, in imaging interpretation, and you’ll be able to recognize MRI and CT findings of common neuro-ophthalmology problems. So in the way of an outline, I’m gonna talk just briefly about CAT scan versus MRI. Then we’ll talk about some of the pitfalls, some of the difficulties and problems that can be encountered when you’re ordering these tests. And then we’ll look at just sort of a potpourri of common neuro-ophthalmology conditions, and how the imaging will look. So I’m gonna start just with a polling question here. And this is something that I will read the question, and the question is: MRI is preferable to CT in all of the following, except which of these four conditions? Is it preferable in brain stem lesion? How about a patient with a bitemporal hemianopsia, or how about a patient with thyroid eye disease, or acute stroke? So the question is: Which one of these is MRI not preferable? Is MRI and CT maybe similarly useful? And you can vote by clicking on your computers. For one of these four things. I think you simply click in the circle, next to the poll. And I’m not sure you’re seeing that poll, honestly. Oh, yes, you are. So here’s the responses. A couple people thought maybe a brain stem lesion, a couple people thought bitemporal hemianopsia, where we’re thinking about a chiasmal region problem. MRI is definitely better for the brain stem, definitely better for the chiasma, sella, and super-sellar region. A couple people thought acute stroke and thyroid eye disease. The example is thyroid eye disease. Acute stroke depending on how you define that, but certainly if you define it win the first 24 to 36 hours, a CAT scan may not show any abnormalities in the first 36 hours, in somebody with an ischemic stroke. So a typical ischemic stroke — CAT scan can look normal. And I think we’ll show an example of that as we go through the webinar. So an MRI is definitely preferable in acute stroke. Now, if there’s a hemorrhage in the brain, that’s another story, and either an MRI or a CT would show that. Certainly if there’s a hemorrhagic stroke — and that’s why they might quickly, at least in the United States, if you go in and they think you’ve had a stroke, they want to rule out bleeding in the brain, because that can kill you quickly. Whereas an ischemic stroke probably won’t. So they often will do a quick CAT scan to rule out blood. CAT scans are good for blood. But if it’s ischemic and there’s no blood, then the CAT scan can look totally normal, and we’ll get to that in a moment. But thyroid eye disease, which is an orbital process — in most orbital processes, a CAT scan is just as good as an MRI, and depending on where you live, a CAT scan may be cheaper and it may be easier to get than an MRI. And we’ll go over the pros and cons of that in a moment. So what about CT versus MRI? Well, of course, there may be patient factors that dictate your choice. So if there is any magnetic metal — these days, at least in the US, they use titanium and have been using titanium, which is not magnetic. That said, non-magnetic metal can certainly cause artifacts, both with CAT scan and MRI. I just saw a patient recently who was thought to have an inferior rectus muscle problem because of an MRI that was done for double vision, and she had extensive dental work, and there was a metallic artifact. When we got a CAT scan, although there was metallic artifact, it looked different. You could tell that the inferior rectus muscle was totally normal on the CT scan, whereas it looked abnormal on the MRI, because of the metallic artifact. So the metal, again, if it’s magnetic, you can’t get an MRI. And if it’s non-magnetic, you can. But the CT and the MRI both have different types of artifacts with metal. Magnetic metal is perfectly okay, of course, for the CT scan. And then sometimes patients have claustrophobia. They have trouble with tight spaces. And they’re not going to appreciate the MRI. So when I talk to patients about the MRI, I ask them if they’re claustrophobic. If they say yes, I’m claustrophobic, usually they can’t do the MRI, certainly not without significant sedation. I try to avoid intravenous sedation. I do prescribe Valium or… I’m blanking on the generic name. I might prescribe some sort of an antianxiety agent, if they have mild claustrophobia. Certainly people don’t think — some people don’t think they’re claustrophobic, and they get in the MRI machine, and they realize they are claustrophobic. And MRI, as I say here, it’s better for pretty much anything in the brain or in the skull base. If you’re thinking brain stem, if you’re thinking brain, almost always the MRI is better. Now, the exception to that would be if you were interested in the bone. So in the orbit, I think the MRI and CT are fairly equivalent, but if you’re specifically interested in bone — why would you be? Well, maybe there’s an abnormality that you see on the MRI, and you want to know whether the bone is being involved. Because oftentimes bad things like metastases may erode the bone. That’s very difficult — not very difficult; impossible — to see on MRI, as we’ll see in subsequent slides. The bone is invisible, if you will, to the MRI. So the CT is, I think, as good as MRI in the eye socket. If you’re sure it’s a condition of the eye socket, then CT is probably better. If there’s exophthalmos, with certain exceptions, perhaps the CT is just as good, and if you’re looking at bone, then the CT is far superior. There’s also some talk these days about MRA versus CTA. As you probably know, the MRA, magnetic resonance angiography, versus computed tomographic angiography, are both ways to look at the blood vessels. These are both simply software variations on the MRI and CT machines. It does not require anything extra, really. You just push some different buttons to get the MRA and CTA. And depending on your institution, some institutions just do really good MRAs. Some do really good CTAs. So it doesn’t really matter too, too much, I think, between MRA and CTA, if you know where you’re getting your image, your neuroimaging. You can often ask which is better for — oftentimes we’re thinking about aneurysms, sometimes cavernous sinus fistulas. But if you don’t know, you can always ask your radiologist: Hey, where we are, which is better for detecting aneurysms? Is it our MRA, or is it our CTA? And we’ll look at some examples of those momentarily. I would note that — I didn’t mention earlier on, but if there are questions as we go through this, we’re gonna answer questions at the end. You can type your questions at any time, but they won’t be answered until we stop at the end, and then I’ll try to answer any and all questions that come up. So when do I order what? So for MRI, the head. If the person’s chief complaint is double vision, they have ocular misalignment — of course, it’s binocular double vision. But if they have misalignment of the eye, I want an MRI of the head. Could it be an orbital process? Sure, but it could also be a brain problem. If a person has bilateral optic neuropathy and I’m worried about something in the sellar region, I’m gonna order MRI of the head. Any suspected intracranial pathology — MRI is superior. Now, you may not have MRI. So I’m just talking about what I’m gonna order if I have the choice of both. Obviously if you don’t have MRI, CT is gonna be the next best mode. How about MRI of the orbits? If there’s a unilateral optic neuropathy, I often order MRI of the orbit. Because if it’s unilateral and compressive, then the lesion is gonna be in the orbit, typically. And I’m gonna get really good views of the eye socket, if I order MRI of the orbit. And I will get automatically in the US — if you order MRI of the orbit — will get fat-suppressed views, and I’ll show you what that means in a moment. So if there’s a unilateral optic neuropathy, I’ll order MRI of the orbit. If there’s bilateral optic neuropathy, I’ll order MRI of the head. Now, in the United States, sometimes we’ll be interested in plaques, for demyelination. But they can’t get your orbits without getting your brain. They won’t get detailed views, but they’ll get good enough views to look for plaques in the head. So instead of ordering orbit and head, I just order orbit. In the US, if you order orbit and head, that is two studies, and two studies costs twice as much as one study. In the US, we’re talking about literally thousands of dollars. How about CT of the head? Well, if you really need something fast, and you’re worried about bleeding, as I mentioned, blood in the brain will show up well in a CAT scan of the head. If you’re only concerned about hydrocephalus, then a CT will clearly show you enlarged ventricles. That’s rare, that you’re only concerned about hydrocephalus. But that’s when a CT of the head might be ordered. CT of the orbits — I routinely order one in somebody with Graves’ disease or thyroid eye disease. If I’m quite certain it’s an orbital process, there’s unilateral exophthalmos, extraocular motility problems, I’m not considering a cavernous sinus fistula, which would be different imaging, then a CT of the orbit may well be good enough. That said, I’ve had patients where I find a problem in the CT of the orbit and order MRI to correlate. But I’ve also had the opposite issue. An MRI shows a lesion and the report says correlate with CAT scan to look at the bone. And CTA versus MRA depends on your institution. They can both be excellent ways to look at the blood vessels, rule out aneurysms, look at cavernous sinus fistula, and arteriovenous malformations. So that’s sort of my general philosophy about CT versus MRI. I want to move on to the second part of the webinar, which is pitfalls. And here are the pitfalls I’m gonna consider. Wrong study, no contrast. No fat suppression, no instructions, and no persistence. And I’ll run through each of these briefly. Here’s the case of the wrong study. This is a 60-year-old man. He awoke today with blurry vision and was offbalance. He had a history of diabetes and high blood pressure. On examination — and I saw him actually after he went to the emergency room — he had a right homonymous hemianopsia. We’re down here looking at his cerebellum. This is his CT scan. They sent him from the emergency room to see me because of his visual complaints, and here’s his CT scan, which looks pretty normal. Let’s look at the MRI cut, the same day, looking exactly the same place, and you’ll see he has multiple — hopefully you can see my pointer — he has multiple strokes. These are ischemic strokes. This is not blood. So that’s why I say in MRI within the first 24, maybe 36… I’m sorry, CT, within the first 24 to 36 hours, may look totally normal. And so I sent him back to the emergency room and said… He needs an MRI, because he had a homonymous hemianopsia and further up in the brain in the occipital lope there was a stroke causing the homonymous hemianopsia, and they admitted him for acute ischemic stroke. So stroke may not show on the CT, certainly within the first 24 hours. I frequently see people — the person has an occipital stroke, they have only visual symptoms, go to the emergency room, CT scan is normal. Oftentimes I see them days or even weeks later. I tell them… Gee, you have a homonymous hemianopsia, you have diabetes, high blood pressure, you probably had a stroke. They say oh, no, I didn’t have a stroke. They did a scan when I went to the emergency room that showed no stroke. So MRI better than CT except in orbit or bone. And this is another polling question. And this is just to help me understand a little bit what you all, the audience, has available to you. So the question is: What is your access to neuroimaging? Can you… And read these right away… Can you get MRI the same day if needed? If not, can you get one within a week? If not, can you get one within a month? Or can you only get a CT, or neither available, which would obviously be a big problem? So that’s the question. I’m curious. I think I’ve covered most of the bases. MRI same day if needed. Get it really quick. Or within a week or within a month, or only CT, can’t get MRI within a month. And here’s the results. I’m not sure they can see those results, because they’re not on my main screen. But the results are: About 20% of people can get the MRI the same day if needed. 50% can get an MRI within one week. 15% within one month. 13%, CT only. So the majority of people, really, 70% of you, can get an MRI within one week. That’s pretty good, I think. Pretty good. That helps me a little bit. So here’s another case of what I thought of was the wrong study. So this is a 70-year-old woman who had a droopy lid for one day. She had discharge for a deep venous thrombosis. And her MRI was said to be normal. She did have a past medical history of melanoma that had been excised from her skin. And on exam, the reason I was seeing her other than the ptosis, was that she had a 5 prism diopter right hypertropia when she looked downward. She had slight proptosis and the obvious ptosis. So she has maybe an orbital process. Maybe not. This could be a mild third nerve palsy, or something like that. So her MRI was done, and they ordered an MRI of her brain. And here was one of the cuts of the MRI of the brain. And they interpreted and said… Well, everything looks okay. Here’s another cut of the MRI of the brain. Some more cuts, MRI of the brain. And here is ultimately the study that we ordered, which is the MRI of the orbit. And what you’ll see here… Is, in her right orbit, you’ll notice… Here’s the left superior rectus. And hopefully you can see my pointer okay. Here’s the left superior rectus levator complex. And here’s the right. And what you’ll notice here is that this superior complex is much enlarged compared to the left. So this can be fairly difficult to see on an MRI of the brain, because you may not get good views of the orbit, and you may not get — depending — good coronal images of the orbit. So in this person who has mild proptosis and double vision with ocular motility problems, clearly you’re gonna want to look at the orbits, as well as the brain. And you’re gonna look for things that may be somewhat subtle. So this person actually had — I believe this was a metastasis, melanoma metastasis, to her orbit. And here you can see again on the sagittal view of the orbit, here is that lesion, right up above the superior rectus levator complex and involving it. So here is the lesion in the orbit. So you need to give good instructions and order the correct test of the correct anatomy. Don’t keep it a secret. If the person has double vision, tell the radiologist: What are you looking for specifically? In this case, we said the person has exophthalmos and double vision. We strongly suspect an orbital lesion. So give them as much information as you can give them, to help them, as they get the study. And I know there are some places where really the radiologist determines what study is ordered. That is problematic, I think, and I think it’s much better if you, the ophthalmologist, knows enough to order the right study. Because usually — clearly, at least from a neuro-ophthalmology standpoint, I’m definitely gonna know way more than the radiologist about the best study to order for the problems that I am seeing on my exam. So order the right study. Here’s the second pitfall, second problem. No contrast. So gadolinium, gadopentetate. Not sure what it would be called. No contrast, or an iodine-based contrast in CAT scans. This is a 40-year-old woman. Gradual loss of vision in the left eye. Painless loss of vision. She’s 20/20 or 6/6 in the right eye. 20/60 or 6/18 in the left eye. There is a left relative afferent pupillary defect, and she does not see any of the color test plates with the left eye. So we know that this woman has a left optic neuropathy. A unilateral optic neuropathy, and I mentioned earlier that if I see someone with a unilateral optic neuropathy, I’m gonna order MRI of the orbit. So here’s the non-contrast MRI of the orbit. And you can see here in this case — here’s her right optic nerve, here’s her left optic nerve. The white here is the orbital fat. We’re sort of in the back of the eye socket, where the muscles and the annulus of Zinn are sort of coming in. And there’s nothing really obvious, I don’t think, in this scan. There’s a little asymmetry between the nerves, but it looks like maybe this black — nerve sheath, maybe there’s a little fluid around the nerve sheath. Nothing too obvious in this non-contrast scan. But here is the gadolinium-enhanced scan. And what we see here… Here’s the right optic nerve, right here. No enhancement. Look at the left optic nerve. There is marked enhancement, all around the optic nerve. This person has a left optic nerve sheath meningioma. And this meningioma can be very difficult to diagnose. It’s near-impossible to diagnose unless you give contrast. In this case, in the MRI, gadolinium. So if optic neuropathy is present, I always get it with contrast. There are very few times that I order an MRI where I don’t ask for contrast. Very rare that I would ever do that. So order the gadolinium almost every time. Here’s another pitfall. Another problem. No fat suppression. Here’s a 30-year-old. I saw this person fairly recently. Another Caucasian woman. Gradual loss of vision in the left eye. She’s 20/20 or 6/6 in the right eye. 6/24 or thereabouts in the left eye. Doesn’t see any color plates, definite left relative afferent pupillary defect, visual field test somewhat depressed. So here is an MRI of the orbit with no gadolinium and no fat suppression. So this is all fat in between the muscle and the nerve. Right? All this orbital fat is bright white. This is a T1 image. We’ll talk more about that. The T1 we know because the vitreous, the fluid, is black. And so here’s the same MRI, in a patient with gadolinium. But there’s no fat suppression. So you still see all this very bright white fat. It doesn’t look significantly different than this. The one difference is: There is some enhancement of the extraocular muscles. That’s normal. The extraocular muscles will enhance when you give gadolinium. But you don’t see any obvious lesion here. Here is the MRI, same patient, with gadolinium and with fat suppression. So now the fat — and this is just a software sequence — the fat is gone. It’s black. Because of this fat suppression computer program sequence. And now look back here at the back of the eye socket. Whereas this used to blend in, this enhancement blended in some with the orbital fat, now it’s much more obvious that there is an abnormality in this axial MRI image. We’ve cut right through the nerve, and this is another optic nerve sheath meningioma. So we’re cutting right through the nerve, and all we see is the enhancement on either side of this nerve sheath meningioma. So one of the enhancements I order with a unilateral optic neuropathy and I’m considering there could be something in the orbit, I automatically get fat suppression. In the US, if I order orbit, I get fat suppression. Why? Because there’s fat in the orbit. If I order brain, I don’t get fat suppression. Because unless you’re a fathead, there’s no fat in the brain and you don’t need to do fat suppression. That’s one of the reasons I order orbits. I automatically get fat suppression. If you don’t automatically get fat suppression when you order an orbital MRI, then you need to write it out. When MRIs were becoming more popular in the US years ago, I had to write it out, every time I ordered it. Orbital MRI with fat suppression and with gadolinium. Here is a patient with bad instructions. So here’s a 59-year-old guy. He’s had double vision for the last six weeks. He has high blood pressure. On exam, he has a subtle right 6th nerve palsy. Look at how well his left eye abducts. Compared to his right eye. And he has an esotropia. We always check — it says to turn off the video, I think. Hold on. We always check facial sensation and facial nerve function, motor function, when we see a 6th nerve palsy, because the differential diagnosis is much different, if there’s a V, VI, and VII, versus just a VI. So suffice to say he had mild right sensory problems on the right, mild right 7th nerve findings, and a mild right abduction deficit. So we sent him for MRI of the brain, because he’s got V, VI, and VII. We’re thinking about brain stem area. MRI is gonna be much better than CT. Sure enough, they called me. I think this was one or two days later. Yep, we found the problem. It’s hard to see on the scan, but I’m trying to point it out there. We found a small left pontine lacunar infarct in this hypertensive 60-year-old guy. There’s your answer. I said… Wait a minute. I’m pretty sure he had a right 5th, 6th, and 7th nerve palsy. They said this is where the problem is! Show me the films. They sent me the films. And I said… What’s this? It turned out to be a metastatic lung cancer lesion. Not in the brain stem, but in the cerebellopontine angle, where V, VI, and VII run. This was the first sign he had lung cancer. He had an unknown primary when this was done. It looked metastatic, and at that point he just had a chest x-ray, showing obvious lung cancer. So make sure you give good instructions. I tried to give this person good instructions, but they didn’t follow it very well. And because they found that little left-sided abnormality, they really didn’t focus on the rest of the picture. Which seems in hindsight very obvious. There’s an enhancing lesion where I have this yellow arrow, and not a good-looking one. So give detailed instructions. Try to avoid this problem. It didn’t really help me much in this case. But no secrets. Tell them exactly where you think it is. Oftentimes I get radiologists saying gosh, you knew exactly where the problem was! That’s what neuroophthalmology is all about. So this person — when she looks to the left, she has diplopia. A scan was done because she had double vision. Someone ordered a CAT scan — sorry, an MRI, I guess. Although I’m gonna show you a CAT scan. And the scan was read as normal. But she developed a little esotropia when she looked to the left. Here’s her CAT scan. How do we know it’s a CAT scan? Because all this white stuff is bone. On MRI, this is all invisible. It’s black. On a CAT scan, it’s white. That’s why we say if we want to look at bone, we need to did a CAT scan. So the extraocular muscles are often misread or overlooked by radiologist. I have even had people where I say: I am interested in this particular muscle! And the scan report comes back normal. And here’s this patient, who had mild thyroid eye disease and an enlarged left inferior rectus muscle. So when she was looking to the left, this tethered the eye a little bit. And remember, in thyroid eye disease, it’s not a paralytic process. It’s a restrictive process. So she had an enlarged medial rectus muscle, consistent with thyroid eye disease. If we go back for a moment to look at her, you notice she’s got a little eyelid retraction on the right. As soon as we see a little eyelid retraction on the right and an enlarged muscle on the left, it tells us it’s bilateral, and it tells us that this is thyroid eye disease. And then the fifth problem, the fifth difficulty or what I call pitfall, is no persistence. So this is a young guy who complains of vision loss in the right eye for three months. He has an important past medical history. Ewing’s sarcoma of the femur. And he was thought to be disease-free. He had a right optic neuropathy, and because of the vision problem, before he saw me, he saw his family doctor, who said you had cancer. We’re gonna get an MRI. And his oncologist the same week. They both ordered MRIs. Both MRIs were said to be normal. Why he had two MRIs in a week without saying something, I don’t know, but he was a good patient, I guess, and was compliant. They thought… You must have optic neuritis. You’re a young, healthy guy, other than this history of the sarcoma. So it must be optic neuritis. So he was ultimately, after several months, sent to me. Here’s the scoop. In October of 2002, his vision was 6/12 or 20/40 in the eye. About a month later, five weeks later, in early November, he was 6… Whatever that equals. 24, or 20/80. And by the time I saw him, another month later, he’s count-fingers. So the first question is: Is this the history of optic neuritis? And the answer is clearly not. Optic neuritis typically will occur fairly acutely, and hopefully within one or two weeks, the vision will be at its worst, and over the next few months, the vision will do the opposite of what we’ve seen here, and that is the vision will gradually improve. Clearly this is not the story of optic neuritis. So I said what all good neuro-ophthalmologists say. Show me the films! Show me the MRI that you have done. And here’s the MRI. A little washed out. So again, we’re looking at the right optic nerve. You can see that this is an MRI. There’s the bright orbital fat. Right? So this is a non-suppressed image. There was no gadolinium or fat suppression. Of course, what did we recommend? Although I thought I saw the abnormality — look right here. There’s something a little asymmetric in the optic canal. But when we give gadolinium and we do fat suppression… We can now see the fat. You probably didn’t need fat suppression for this, but the fat is gone, like we would expect, and here with the gadolinium, the muscle is enhanced, like we said, and this lesion, which turned out to be a Ewing’s sarcoma metastasis — ultimately the patient died of this — was causing a compressive optic neuropathy, and there was no persistence. They thought it must be optic neuritis. But he got worse. I think you can make a diagnosis over here, but you definitely want to give gadolinium and do fat suppression in the setting of unilateral optic neuropathy. And here are a bunch of other scans. All of these scans were read as normal by the radiologist. And clearly you all probably have very different levels of expertise in terms of radiology. I was somewhere recently where they had an MRI and had no radiologist. Which really puts the onus on the ophthalmologist. This patient had a right optic neuropathy, the MRI was read as normal, even though it was done with gadolinium. Here is enhancement of the optic nerve into the chiasm. This was read as normal. Another patient, here’s the right optic nerve, as it’s forming the optic chiasm. Here is the left, here is the pituitary gland, which also enhances. If you see a nice bright white pituitary, that means gadolinium was given here. This is the suprasellar area right up here, here is the enhancing distal nerve, read as normal. Over here hopefully everybody sees the problem. I said rule out abnormality of the right inferior rectus muscle. Look at the right inferior rectus muscle compared to the left inferior rectus muscle. Thyroid eye disease! And here’s a patient with double vision, known… I think this was colon cancer. Thought not to be metastasized. With this giant, in my opinion, orbital lesion. Actually a metastasis. Here’s the superior oblique muscle right here. Here’s the met to the superior oblique muscle. Here’s the normal medial rectus right down here. But this was a metastasis to the superior oblique muscle, causing the double vision. I showed the patient the scan and he said… What’s that? I said yeah, that’s the “normal” scan. He was not happy. All of these scans are reasons that you, the ophthalmologist, should be looking at your scans. Because you know exactly where you want to look. So the top five pitfalls are — or the solutions are — get the right study, give contrast to everybody, get fat suppression. If you’re interested in the orbits, get fat suppression. Give specific instructions. And be persistent. All right. Now I’m gonna talk a little bit just about some imaging differences. This is the third polling question. And you can see the MRI here. And it might look familiar, because I just showed it. The MR image below is which of the following? Is it T2-weighted? T1? T1 with fat suppression? Or T1 with contrast and with fat suppression? So let’s see if you were listening. I’ll give you a second to vote for these. T2-weighted, T1, T1 with fat suppression, T1 with contrast and fat suppression. Nobody said T2. Very good. Only a couple people said T1. It is T1. But a third of you said T1 with fat suppression. It’s definitely with fat suppression. But it’s also with contrast. And so what you’ll see here is the fat suppression. It’s T1, because you can look at the globe. If it’s T2, the vitreous will be white. T1, it will be dark. So it’s T1. There’s definitely fat suppression, because all this is fat in the orbit. And if there’s no fat suppression, this is gonna be bright white. And it’s also with contrast. Look at the extraocular muscles. The only two muscles we’re seeing are the medial and lateral rectus. Both enhancing. This was a scan where I showed you there was a little enhancement right here. But you can look at the muscles and say ah-ha, it’s with fat suppression. So this is T1. The vitreous is dark. With fat suppression, the fat is not bright white. And in contrast, the muscles are enhancing. So T1, T2. Here’s a T1 in the upper left. I’m sorry. Yes, the T1 in the upper left. No fat suppression. There’s nice bright white fat. The muscles are not enhancing. So this is T1. T2 — look at that bright white vitreous. So the vitreous will be bright white. The cerebrospinal fluid, see down here, is bright white. The fluid around the optic nerve. Normal fluid in the optic nerve sheath. Bright white. This is T2. T1 with fat suppression. Fat signal goes away. The vitreous is dark. The spinal fluid is dark. The muscles are not enhancing. T1 with fat suppression and contrast, the muscles are enhancing. The pituitary down in here is enhancing. And that’s how you tell the difference. CT versus MRI in bone. As we said, here is CAT scan. It’s obvious. You see the nice thick bone, normal-looking bone, the very thin lamina papyracea. That’s why they call it that. It’s thin as paper. Here’s the CT. Here’s the bone on an MRI. It’s invisible. It’s black. That’s why I said if you want to look at bony detail, you get a CT. Let’s see. I think just some things to point out. Just normal — fairly normal anatomy. Hopefully everyone can identify the extraocular muscles, the optic nerve. This image is a CT scan. You see the nice white bone. What’s this? This is an image — the person is just rotated slightly, so this is the back of the globe. This is not a tumor. You’ll notice that when you look at other images. It’s just the back of the eyeball, because the patient’s head is slightly rotated. Here we have a coronal image further back. Here’s the optic chiasm. Here’s internal carotid artery here. Here’s the supra-sellar area. In this case, this would be a partially empty sella, if you hear that term. It means the pituitary gland is kind of squished down here. You can see a partially empty sella in increased intracranial pressure, but you can definitely see a partially empty sella as a totally normal finding. Totally normal. And here’s a nice sagittal image through the midline with the cerebellum, the pons here, medulla, midbrain. Here’s the sella, partially empty of pituitary gland. It’s all squished down here. Here’s the optic nerve and chiasm right here. This structure right up here is the superior ophthalmic vein. We’ll be looking again at that in a little bit. So let’s look at some common neuro-ophthalmology conditions that you will see. Graves’ disease. We’ve seen this already. This is certainly probably the most common orbital condition that you will see. You might see these nice very enlarged muscles. This is a contrast-enhanced CAT scan, so you see the white bone, the white muscles from the contrast, you see why a compressive optic neuropathy may occur, because back in the back of the orbital apex, you’ve got bone here and here. You’ve got muscle that are big. You’ve got nerve. The nerve loses that battle, because the bone isn’t going anywhere. You can do an orbital decompression to remove bone and give that nerve breathing space. So you can get rid of the optic neuropathy. Here’s a patient of course with the orbits. We like to really look at the muscles. You need a coronal image. So this is another CAT scan. You see the bone, and you see the big muscles, for reasons that are unclear to me — inferior most common, superior second, medial and lateral. A good example of symmetric-looking thyroid eye disease with some of the extraocular muscles enlarged. So a type of inflammatory condition. Here’s a patient — depending on where you live, you may not see much. Central nervous system sarcoidosis. You get this nodular lumpy, bumpy-looking enhancement of the meninges. All of this is inflammation. Here we are further back, where you see just this enhancing… Eugh, inflammation in a patient with neurosarcoid. So that’s what meningeal enhancement may look like. In this case, in sarcoid, it’s often nodular-looking, as opposed to smooth, that you may see with other conditions. Optic neuritis we’ve talked about a little bit already. Certainly most patients — not all patients — most patients with optic neuritis will have enhancement of the optic nerve, which is apparent, again, this is T1. Fat suppressed, contrast given. Look at the muscles. Normal nerve on the left, not enhancing. Enhancing nerve on the right, optic neuritis. Sometimes I’ll get a report back saying: Not optic neuritis. The nerve doesn’t enhance. The nerve doesn’t always enhance. That does not rule out optic neuritis. This is what we’re looking for on MRI. T1, fluid in the ventricles is white. Look at all the plaques, this is contrast enhanced, all the plaques that we see in a patient with multiple sclerosis. So let’s look at some tumors. And I’m gonna ask a question. This patient probably has which of the following? So I will say before I give you the choices: This is a coronal image. Looking through the chiasmal area. This is a sagittal image. So here’s the brain stem. Sellar area. What do you think this patient probably has? Double vision? Bitemporal field defect? Headache, or nystagmus? I’ll give you a few seconds to vote. This one is, I think, an easy one. We’ll see. The answer is: 87% of you said bitemporal field defect. We’ll get rid of the polling. It’s possible this person is double vision. This is a craniopharyngioma. There’s a cyst in it, which is a common characteristic. Here’s the chiasm. It’s pushing up and indenting. Clear bitemporal field defect. These tumors can get into the cavernous sinus on either side. This is internal carotid artery. You can see tumor in the cavernous sinuses. Interestingly, even when there’s tumor in the cavernous sinus in the case of craniopharyngioma, you don’t usually get double vision. So most of you got the correct answer. Here are some other tumors we might see in the eye socket. This is a cavernous hemangioma in a very old CAT scan. Bone, so obviously it’s CT scan, pushing the optic nerve to the side. Here’s a small cavernous hemangioma, causing no symptoms whatsoever. This was incidentally discovered. A non-fat suppressed image. Here’s the fat. In this case, it does outline the benign tumor, the most common tumor we see in the orbit. And this was a patient we followed and did nothing and just observed to make sure it wasn’t getting bigger. This is a patient with an optic nerve glioma. An old image. You can’t separate this tumor from the optic nerve. Here’s another patient with a pituitary tumor. Here’s the pituitary fossa, here’s the suprasellar region, the chiasm draped over the top of this pituitary adenoma, the most common tumor you’ll find in this general area. Here’s another sagittal image, same patient, with this large tumor. This is a non-contrast image, so there’s no enhancement in there. Here’s a patient who had a bilateral optic neuropathy. You can see both optic nerves, heading into this large skull base meningioma. Causing bilateral optic neuropathy. Here’s the same patient, with this olfactory groove meningioma, that’s just huge. The person actually did quite well when this tumor was removed, with good recovery of vision. And here is a smaller pituitary tumor, coming right up and just hitting the distal left optic nerve. This patient presented with a unilateral optic neuropathy. But had a tumor that was just coming up. Here’s the right side of the optic chiasm and distal nerve. The way it was growing gave us a left optic neuropathy. So pituitary tumors can do that. Here’s another patient with a large sphenoid wing meningioma. I’m sorry, this was a tuberculin sella meningioma. Causing multiple cranial nerve palsies. This is a small clinoidal meningioma. This will be the same intensity as brain unless you give gadolinium. If they’re huge, you can see them. If they’re small, you might not. Another big olfactory groove meningioma, causing a bilateral optic neuropathy, as the nerves come back into the skull. And then an incidental finding in a patient, a meningioma, that was growing in, causing pressure, causing no visual symptoms, probably because it was extremely slowly growing. And then tumors — in the nerve sheath meningioma. Optic nerve, sagittal image with tumor above and below, fat suppressed, gadolinium enhanced, and the same tumor on an axial cut. You can see it trying to run through the middle here. And I showed this earlier. It’s very subtle. Tram track-type appearance of a small apical orbital-apical meningioma, and the same tumor here. We saw this earlier on when we looked at pearls and pitfalls. Here is a patient’s CT scan with a tumor right here involving the inferior medial rectus. But importantly, the reason why the CAT scan might be superior in some places — look at this bone here. It’s gone. This is a breast cancer metastasis that has eaten through the bone. This is a bad problem, not a benign problem. This is a patient with both eye sockets filled with breast cancer. Just filled, all of this enhancement — is lung cancer. And then some intrinsic tumors. So these are gliomas of different grades. I think this is a glioblastoma, intrinsic tumor of the brain. And another glioma over here. A glioblastoma in the occipital lobe, causing very few symptoms. And then quickly with the last few minutes, we’ll look at some vascular abnormalities. So here is an MRV, looking at the cerebral venous structures. Here’s the superior sagittal sinus. The lateral sinuses. The transverse, sorry, transverse sinuses. You can see that in this case, there’s some sigmoid sinus here, but there’s a clot here. So bad-looking venous sinus. Here is another — this is an MRA or MRV as well. Here’s the nice-looking superior sagittal sinus. Here’s the normal right transverse sinus, and the narrowed and stenotic left transverse sinus. So this occurs… The pressure goes up on the venous side, and in turn, the pressure goes up in the spinal fluid. It can cause papilledema. This patient has a very large flow void in the cavernous sinus. This is a cavernous sinus aneurysm, seen on MRI. So on MRI, flowing blood in this T2-weighted image… Flowing blood is black. So this big flow void is an aneurysm. Here is the patient’s MRA. A fairly old MRA. This is the carotid artery, the white. But this area here is the aneurysm. This is a patient with a cavernous sinus fistula. Here is this huge superior ophthalmic vein. Here’s the normal size, over here. So this patient has a cavernous sinus fistula, where the blood flow in the venous side is under high pressure, and it’s distended, this vein, and the blood is actually flowing this direction, as opposed to this direction, back into the cavernous sinus. Here’s another patient with a CAT scan that was read as normal. Here is the left side. You see the muscles are slightly enlarged, all of them. But look at the superior ophthalmic vein right here, versus over here, this little dot. So this is a relatively enlarged superior ophthalmic vein. And this can sometimes be the only sign of a cavernous sinus fistula, that even can be seen on a CAT scan. Look for that on the coronal images in particular, on CT scans, if you have any level of suspicion, or cavernous sinus fistula. This is our last audience polling slide. This patient probably has which of the following? This is an MRA. Double vision, bitemporal field defect, headache, or nystagmus. Same choices as the last question, but different neuroimaging study done. I’ll give you a few seconds to vote. And about a third of the people said double vision. About half said headache. A lot of people get headaches. A couple said nystagmus or bitemporal field defect. So here is the Circle of Willis right here. Here is the posterior communicating artery, this artery and this artery. So this is a PCOM aneurysm, causing a third nerve palsy. They may well have a headache, but for this purpose, I was looking for double vision, because of their third nerve palsy. And some other findings. Let’s see. This is a patient who had had a homonymous hemianopsia. To try to just illustrate some sequences. Here’s a T2. See, the vitreous is white, the CSF is white. And you might say… Well, there’s something going on maybe back here. Not that obvious. Here’s a contrast-enhanced MRI, T1. Where now there’s clearly something present going on right where you would expect it, given this person. This is the left occipital lobe, so they have a right homonymous hemianopsia. But this is a diffusion-weighted image. This is something we do routinely in the US. If you think there’s an ischemic stroke, it’s gonna shine just bright white. So the best way to look for an acute ischemic stroke is DWI, or diffusion-weighted imaging. And then finally a few miscellaneous conditions. Here’s one that I try to stump my residents with. Here is a person who’s got downbeat nystagmus. Downbeat nystagmus. The brain structures look pretty good. Here’s the normal pituitary gland, anterior visual system, optic nerve coming into chiasm, brain stem looks pretty good, no obvious lesions, this is T1, CSF is dark, no contrast. But look down here. Here’s the cerebellum. Look what’s happened. That cerebellum has scooted down through the foramen magnum. This is a Chiari malformation, the most common cause of downbeat nystagmus. Not all Chiari malformations are symptomatic. In fact, small ones — this is a big one — small ones are incidental, and not relevant, probably, to the patient’s symptoms. This is a CT scan, obviously. You see the nice bright white bone. This person has fibrous dysplasia. So here’s the normal-looking bone that’s been remodeled. Here is a person with fibrous dysplasia, this abnormal bone deposition. And you can see what’s happened. Look at the volume — or I should say, I guess, the area — of the posterior eye socket. How this is a nice normal-look area. You can imagine that the volume of the eye socket is much smaller. So this person has exophthalmos and some motility deficits. Possibly developed a compressive optic neuropathy, so fibrous dysplasia. Here was a patient who was sent to me — they had a headache or something. Got a scan, and something was going on in the optic nerves, here and here. Calcification. This is a person with typical optic nerve drusen. Optic nerve drusen. And then I think my last image is this. A patient with pain and loss of vision who has some enhancement of the optic nerve sheath. Not as prominent over here. Who had a condition called optic perineuritis. Perineuritis. No, actually, I take that back. I take that back totally. I think this is probably just the normal nerve. What I see that isn’t normal — sorry about that — is something else. And that is: Look at this superior oblique muscle right here. Nice normal, healthy-looking. Look at this one. This is a person with a congenital 4th nerve palsy. And over the years and decades, there’s just atrophy of the superior oblique muscle. So here is the muscle on the side of the superior oblique palsy. The congenital. And here’s the normal muscle. Sorry, got that one wrong. So thank you for your attention. I think we have a few minutes, at least, to try to answer any questions. I think I’m supposed to pull up the Q and A. If that’s what I am, I don’t see any questions. I’m not sure.

>> You can stop sharing your screen.

DR GOLNIK: Okay, stop sharing my screen. Is that better? Or no? Stop.

>> You can bring up the Q and A box. There’s two questions, currently.

DR GOLNIK: Two questions? Okay. The question has to do with dysthyroid optic neuropathy. So I think I answered that question. Because earlier there had been a question about what do you prefer. I think CT, MRI. I think the CT in dysthyroid optic neuropathy, I think, is superior, because again, it gives you that bone that you can’t see on the MRI. And since dysthyroid optic neuropathy means, usually, there’s gonna be a tight orbital apex — how do you know it’s tight if you can’t see the bone? On MRI, you can’t see the bone. You can get a feel for it, certainly, but the CAT scan is gonna show you the bone, the nerve, and the muscles. I hope that answers the second question, which is why is it better. What would be the best imaging for tuberous sclerosis? So you’re looking for tubers in the brain. So MRI of the brain with gadolinium would be the best study for that. Which one is superior in assessing transient visual loss? CTA or MRA? I think that depends on what we talked about. It really depends on your center, or where you’re getting your imaging. If you have good CTA and good MRA, I don’t think it probably matters that much. Of course, if you only have CTA or only MRA, then that’s your choice. But I would ask your radiologist that question. Hey, do you think one is superior to the other for this indication? I would ask them. It may vary from place to place. Hello, Francis. What reasons would determine whether we order T2 or T1 plus fat suppression and contrast? If you order MRI, you’re probably automatically… At minimum, you’re gonna get T2 and T1 images, automatically. If you order a brain MRI, you’re automatically gonna get T1 and T2. You’re not gonna get fat suppression. But if you really think the problem is in the brain, you probably don’t need fat suppression. On the other hand, if you’re not sure if it’s brain or orbit, then you want fat suppression, and you’re gonna have to ask for it. And then the contrast — I see films all the time that are done without contrast. And usually one of two things. One, it’s read as normal, but you get the contrast and you find the answer. Or two, you find an answer, but then the radiologist says… Oh, we could probably tell you better what this is if you give the contrast. So in the United States, at least, I almost always order it with contrast. Certainly if you’re looking for something subtle or small. Anything inflammatory. If you think this could possibly be inflammation, you’ve got to order it with contrast. If you think it’s a big tumor somewhere… You can probably diagnose that without the contrast. But… It’s still, I think… I almost never order a scan without the contrast. Any role for x-rays in neuroimaging? Well, of course, if you don’t have a CAT scan and you don’t have an MRI, then yeah. There’s a role for x-ray. I think you’re not… You’re gonna miss all sorts of things, of course. I think x-rays can probably still tell you if there’s enlargement of the sella. If you have a pituitary tumor, sometimes, not always, that tumor can enlarge the bony sella. With x-rays, you’re just gonna be seeing the bone, more or less. So the indications… I have not ordered an x-ray in decades, personally. And I think if you don’t have CT or MRI… You’re in trouble from a neuro-ophthalmology standpoint, is all I can say. What is the role of x-rays… I’m sorry. We just talked about that. Can grey matter lesion in temporal lobe cause papilledema, or is it coincidence? The answer is no. The only way a grey matter lesion in temporal lobe can cause papilledema is if it is huge and is causing increased volume in your brain, or somehow if it was obstructing CSF outflow, but temporal lobe lesions rarely do that. You would have to have a big tumor. I can tell you, I just had a patient recently, my personal assistant of 25 years, who had headaches. And I said… Oh, let me look at those nerves. Papilledema. And she had an intrinsic brain tumor in more than just her temporal lobe that was so big it was causing papilledema. After orbital trauma, should we advise both CT and MRI of the orbit? I usually don’t. Again, after orbital trauma, you know there’s trauma to the orbit. I’m interested in the orbital structures. I’m interested in the bone. I’m gonna order CAT scan. I don’t order an MRI. If there’s some finding — I don’t know, well… Let’s just leave it there. I don’t order an MRI if there’s orbital trauma. If pituitary adenoma can cause unilateral optic neuropathy, how do we ensure we don’t miss it if we routinely order MRI of the orbit? The reason you won’t miss it is because, if you order MRI of the orbit, you’re gonna get the sellar area and the brain. They don’t just stop at the orbit. They can’t. When you look at axial cuts, it’s possible they won’t go really far back with the coronal images, but you’ll find the tumor in the pituitary area. So you’ll still get pituitary and sellar region if you order MRI of the orbits. Or at least, I do. But if you just look at your next MRI of the orbit, make sure you see the sellar region. Can 16 slices detect CT… Well, the answer is sure. Can 16 slices CT detect… Well, obviously you might miss a small lesion. You’re not gonna miss a big lesion. There’s a little more luck of the draw. Are you gonna be within that? So when I showed you the image of the calcified optic disc drusen. I would never order a scan to look for drusen, because depending on the image slice, I could easily miss those drusen, unless you’re right on the money. And of course it’s way more expensive than ultrasonography or something like that. Is it appropriate to order imaging in all cases of 3rd nerve palsy? Well, the general teaching is yes. I think recently it’s become… You know, we talk about the rule of the pupil and all that stuff. The bottom line is, for a comprehensive ophthalmologist, the safest thing is to probably get an MRI if you can. MRI and MRA, or CT/CTA. In all cases of third nerve palsy. Now, do I order an MRI in every patient? Nope. I see lots of third nerve palsies, and if I see someone over the age of 50 with acute onset of a complete third nerve palsy with pupil sparing, I personally don’t order an MRI/MRA or CTA in that case. But I do order MRI/MRA/CTA otherwise. How can we exclude optic nerve avulsion by CT only? The answer is — certainly a CT would be just as good as an MRI to look for optic nerve avulsion. I don’t think there’s any difference in CT and MRI. For traumatic optic neuropathy, however, the CT and the MRI could be completely normal. I would say the majority of traumatic optic neuropathies I see, the imaging is normal. Now, there may be — on CT, there may be a fracture in the optic canal that indicates to you… There’s been significant force applied. But most of my patients with traumatic optic neuropathy, they may have some broken bones in their face, which you’re only gonna see with the CT. But the traumatic optic neuropathy — clearly the nerve can look completely normal. Optic nerve avulsion, which at least in my practice, relatively rare — I’ve seen a few in 30 years — a CT is gonna show you the orbital optic nerve essentially as well as the MRI. What is your advice if the MRI is contraindicated? Ah-ha. That happens to me quite a bit. They have a pacemaker, for instance. And they have metallic metal in them. Orthodontic metal braces… That is not a contraindication. And by the way, I have patients with orthodontic metal braces, and they just whip them off. I had a patient mean years ago whose mother was an orthodontic technician. She said… Oh, we’ll just take those braces off for the MRI. No big deal. I definitely do that, if the metal braces are gonna cause artifact. At least in the US — I can’t speak for where you are — the metal braces are not magnetic, however. So I clearly see people all the time who have braces who get MRIs with no ill effect. But certainly they can create a lot of artifact. And sometimes the metal braces come off. Clearly, though, people have other metal implants, like as I mentioned, the pacemaker is probably the most common we see. And then you have to go with your next best imaging, which is gonna be a CAT scan. Could orthodontic metal braces cause orbital pain and painful eye movements? No. In the case of cavernous sinus fistula, after embolization, when do you prefer to do a reimaging study? The answer is I wouldn’t do a study at all. I mean, if you embolize a cavernous sinus fistula, now, if you order… If you embolize a cavernous sinus fistula successfully, within hours or days everything will get better. Their intraocular pressure will go down. The chemosis will go away. The conjunctival injection will improve. The exophthalmos will resolve. It should happen pretty quickly. Certainly within a week or two they should be almost back to normal, if it’s successful embolization. So in that scenario, in my opinion, I wouldn’t image them at all. Why bother? That said, the neuroradiologist, interventional radiologist, might well, or the neurosurgeon. And that’s fine. However, if their symptoms aren’t going away, because their intraocular pressure really should improve — I mean, within 24 to 48 hours, or even one hour. It’s almost immediate. If it doesn’t, then you’re gonna have to reimage them, and you’re gonna use the same imaging modality that you diagnosed them probably… Hopefully a CTA or an MRA, to look to see what’s going on. I think that is the end of the questions. What is the best… One more. What is the best type of MRI to diagnose upbeat nystagmus? So upbeat nystagmus is a pretty non-specific finding. It usually means brain stem, cerebellum. So an MRI with and without contrast of the brain. Attention, brain stem/cerebellum. Nystagmus, attention brain stem/cerebellum, with and without gadolinium. So that I think is the last question. If so, thank you all for attending the Orbis webinar on neuroimaging. Please — I think they’ll ask you for feedback. Please give me feedback. In the feedback, if you wish to recommend other neuro-ophthalmology, I guess, or other webinars entirely, I’m sure we’ll be doing another neuro-ophthalmology webinar in the future. So please let us know. And have a good day or night, depending on where you live in the world. Thank you.

Download Slides


January 29, 2018

Last Updated: October 31, 2022

Leave a Comment