Lecture: Visual Field Interpretation

During this lecture, a step-by-step process to interpret visual field testing will be discussed. Neurologic visual field abnormalities will be emphasized and correlated with clinical presentations and near-imaging.

Lecturer: Dr. Karl Golnik


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

DR KARL GOLNIK: This week’s Orbis webinar is going to be on visual field interpretation. I’m Dr. Karl Golnik. I am the neuro-ophthalmologist and professor and chairman at the University of Cincinnati Department of Ophthalmology. So we’ll start talking about visual fields. Now, my objectives for this webinar are to — that when we’re done, you’ll be able to describe pros and cons of different types of perimetry, you’ll be able to list what I call the three anatomic correlates, and you’ll be able to describe and identify common patterns of visual field loss. Now, I would note that this — I’m a neuro-ophthalmologist, so I’m not gonna be talking a lot about glaucoma visual field defects, although if there are questions, we can answer them, and you can type your questions in any time, in the question and answer box. I will not be answering those, however, until the end of the webinar. So we will start. And I usually start with this diagram, which we call the hill of vision. So as you know, at the fovea, we have our most sensitive part of the visual system, and there’s this peak or top of the mountain of the hill of vision there. At our physiologic blind spot, we have a pit of infinite depth, because there are no photoreceptors that overlie the optic disc. Hence the physiologic blind spot. When there are abnormalities in this hill of vision and decreased sensitivity on the greyscale, which is depicted below, we will see shaded areas. The shaded areas, of course, are relatively grey or black, just depending on how significant the decrease in sensitivity is. So as you know, we have different ways of assessing peripheral vision. Of course, assessing a visual field by confrontation is sort of the easy standby. We have Goldman static perimetry, and then computerized automated perimetry. The Goldman static perimetry — really, I’m sorry, that should say kinetic. You can do static Goldman perimetry, but typically people do kinetic perimetry. And we’re gonna talk about each of these in turn. And so I’m gonna ask this question, or try. And I’m gonna have to start the polling here. And the question is: How do you currently use or check peripheral vision? So do you only have the ability to do confrontation, or are you able to use Goldman kinetic — that should say — I’m sorry, Goldman kinetic or static perimetry? Of course, you can do both with a Goldman. Or do you have the use of computerized automated? So I know all of you can probably do confrontational fields, but I’m just curious to see how many can only do confrontation, or only do Goldman, or can do the computerized. Because most of my examples will be computerized automated perimetry. We will look at some Goldman visual fields as well, and of course, hard to really depict the confrontation visual fields. So it looks like I’m getting a fair number of responses here. And I’m gonna end the poll and share the results, so that everyone can see. So it looks like certainly the majority of you have access to computerized automated perimetry, although some use Goldman and some are stuck with confrontation visual fields. And we’ll talk about that as we proceed. Oops. So automated perimetry — and this is one of the older Humphrey machines. I have no financial interest in any of these machines. Typically do use Humphrey field machines. That does not mean they’re better than Octopus visual field machines. Just — at least in the United States — they’re probably by far the most common type of automated perimetry. So the good news about these are they’re very sensitive. They’re technician — relatively technician-independent. So you can do a field on this machine wherever you might be living, and email it to me, and I can interpret it, and I can then repeat that test on the patient in my own clinic, and get pretty good comparison from one to the next. It’s good for detecting scotomas, and we’ll mention that in more detail momentarily. And it’s good for following changes. So you can follow changes more precisely because it is computerized and standardized. But I say here — garbage in, garbage out. What that means is that if the patient can’t do the test, you’re still gonna get a result. No matter what happens. The patient will beep or not beep. And so if they are unreliable, you still get a result. And so some of my time I spend having to try to explain why there is a quote-unquote “abnormality” on the visual field, and it may just be because the patient cannot do the test. And that’s part of the art of interpreting these automated visual fields. And so here is a printout of a visual field, and I’m gonna — I can see my video here of myself. I’m gonna get rid of that. Hopefully you won’t have to look at that. And so here’s a visual field printout from the Humphrey visual field. This is a 10-2. And so when we talk about the printout, you shouldn’t just automatically look at the greyscale, which is this circle over here. But you want to look up at the top to note what type of test this is. This is the central 10-2. So this is just the 10 degrees on each side from center. It tells you about the fixation monitor, the blind spot fixation target, which is central. Further down or over to the right in this printout, you’ll see it tells you the stimulus size. So the stimulus size by default is size III. It ranges from I to V. Size V would be 64 square millimeters. Size IV, 16, size III, 4 square millimeters. And in this case, white. You can change the color to blue or red. I’m not gonna be talking about that during this presentation. Suffice it to say I personally never change it from the color. I do sometimes order size I, size III, and size V, although typically, again, the most common would be size III. It tells you the background illumination. For some reason, in the United States, that sometimes is a question on the board examination, which really makes no sense to me. It’s a full threshold testing. And then there’s a part where — what we call the reliability coefficients. So it tells you the fixation losses. So that simply means at the beginning of the test it plots where your physiologic blind spot is. And it then later in the test will show a light stimulus in that blind spot. If you beep when it shows you a light, that it’s determined to be in your blind spot, that’s a fixation loss. Because the only way you can see that light is if you’re not looking where you should be. False positive errors means that it shines a light in an area — I’m sorry. False positive means it doesn’t shine a light, and you beep the beeper. And false negative is it shows a light where it’s determined you can see the light, and that it makes it brighter, and you don’t beep. So that would be a false negative. So those are all indications of just how good the patient is doing with the test. Now, if there are some false positives and false negatives, that doesn’t mean the test is necessarily worthless. You have to look at the results of the test and how high a rate of false positive and false negative errors there are. Whether or not you see a pattern on the test that can be helpful. Certainly the more false positive or negative errors there are, it may be difficult to follow a change over time. But if you’re trying to make a diagnosis, which is oftentimes the case in neuro-ophthalmology, you have to look at that area, but again, there may be an obvious pattern on the test. And we’ll spend a lot of time today talking about those patterns. And then unfortunately the test — the Humphrey field machine comes from the manufacturer with the default as the fovea off. So if you don’t change that during the testing, next to fovea, where here it says 34 decibels or DB, it’ll simply say off. All of my visual fields, I ask the technician to turn the fovea on, and then at the beginning of the test, it’ll check the foveal sensitivity. Which can be helpful when you’re seeing patients — and especially with things like unexplained vision loss. So I always recommend turning the fovea on at the beginning of the test. The technician is prompted. There’s a screen that they can do. It takes 2 seconds to turn it on. It takes maybe 15 seconds to plot the foveal sensitivity. So I think it’s highly worthwhile to do that. And then the bottom right hand corner here, we see a few things. MD, which stands for mean deviation. Those of you who look at yields a lot will be able to tell that this is not actually the mean deviation of this plot. I blew this up from a different visual field. But the mean deviation is how far is this visual field from normal age-matched visual field. Excuse me. I have a bit of a cold. So in our example, in the field, you probably can see that the mean deviation on the actual visual field you’re looking at, the greyscale, is -2.13. So normal age match would be 0.00. And I don’t use the mean deviation too much. Sometimes I’ll use it to follow it. Sometimes I’ll show patients where we’re following for improvement that — hey, your mean deviation is improving. That means you’re getting better. The PSD, pattern standard deviation, the short-term fluctuation, and the corrected pattern standard deviation are measures of reliability within the test. So you’re looking for the short-term fluctuation to be less than 2 decibels. I don’t use the other two much from a practical standpoint. But I sometimes will look at the short-term fluctuation. What about Goldman perimetry? I think that this is something that personally I have not ordered a Goldman visual field in more than a decade. It is technician-dependent. And the problem with that is that you have to be pretty good. So when was a fellow in training, I did Goldman visual fields on every patient I saw, pretty much. I think most of our Goldman visual field technicians are retired. And so we really don’t have anybody in my practice who’s very good at it. That said, if you have a good technician, Goldman perimetry certainly is valuable. It’s becoming increasingly less available — at least in the United States. It’s not real good for detecting scotomas, which are areas of abnormality surrounded by normality. The reproducibility is definitely gonna be dependent on the technician. So in your own practice, if you have a good technician, it might be fairly reproducible. However, if you do a Goldman field in your office and you send that to me, and I repeat it on the patient, the reproducibility certainly can be in question. It’s a bit more patient-friendly. So you can slow down, depending on the patient. And if the patient is having trouble, reinstruct the patient, and so on, so it might be a little bit more patient-friendly. Nevertheless, I’ve not ordered one in over a decade. And this is a plot from the Goldman visual field. So I won’t explain exactly how the test is done, but basically each of these different colors represent different isopters. So the isopter is simply a line connecting points denoting areas of equal sensitivity to light. So the light stimulus is moved from the outside in, and each of these little marks or these little hashmarks or checks are where the patient saw the light and beep the buzzer. You then proceed by bringing the light in around every side. You make all of these little hashmarks. And then you connect the dots and color them. It’s your choice what colors you want to use, but you can see the different isopters here. So the brown is a bigger, brighter stimulus. The purple is the biggest, brightest stimulus. And you can see that we didn’t continue the purple, because this is normal down here. But up above, the purple is somewhat depressed. It’s not a nice normal visual field. And then when there is a scotoma, and a scotoma is simply defined as an area of abnormality within an area of normality, so here you can see that in the purple, right in the center here, there’s this purple spot. That means the patient could not see the biggest, brightest stimulus, although they could see it out to the side. They could not see it in here. So there is a scotoma. And the smaller the stimulus, the bigger the scotoma. Which makes sense. Into this red, this was a very dim stimulus. And this is the only place that this patient saw this dim stimulus. Right in here. They didn’t see it anywhere else. So there’s no red shading, because all of this area, everything outside this red circle, was not seen. So this is a big central scotoma. Dense central scotoma. Now, confrontational visual fields — and I show some automated perimetry, some greyscales here — to try to demonstrate to you how bad confrontational visual field testing is. So this is a patient who was sent to me, as patients often are, with a known pituitary tumor. And the question from the neurosurgeon was simply: We know this patient has a pituitary tumor. Please evaluate their visual fields by confrontation. Even though I knew there was a pituitary tumor, his visual field — this patient’s visual fields were normal by confrontation. I could not detect an abnormality. And yet, on the automated perimetry, there’s an obvious bitemporal visual field defect. So I say at the bottom of this slide: If the main issue is loss of vision, loss of peripheral vision, you have to get formal perimetry. Now, if you just don’t have formal perimetry, and you’re stuck with confrontation, and that’s the best you can do, well, that’s the best you can do. How can you make your confrontational fields a bit more sensitive? There are a couple of ways to do it. One, you can simultaneously show fingers in different quadrants. In other words, instead of just showing a finger or some number of fingers in one quadrant, do it simultaneously, and in someone with a subtle field defect, they might only see the fingers in the field with the normal vision. You can also use a red color, so the neurologist sometimes will carry those pins with the big red ball on the top. You can use a bottle of drops with a red top. And you show that red top right across the midline, right across the vertical or right across the horizontal midline, and ask the patient: Is it the same redness across the midlines? And I’ll demonstrate why you would do that in a moment, when we talk about patterns. That makes it very subjective, and the patient has to try to tell you — is it a little bit pinker or more red across the midline. I never usually do that, because I have automated computerized perimetry. So I’m gonna move on now. That’s the basics of checking the peripheral vision. I’m gonna move on to talking about what I call the three anatomic correlates, which I get from one of my former mentors, Dr. Steve Newman, at the University of Virginia. So the first anatomic correlate is that we have this papillomacular bundle. And it’s depicted right where I have this big arrow. And that are these fibers that run from the fovea to the optic disc. They’re smaller fibers that may be more sensitive to compression, inflammation, and so if you have a problem with the papillomacular bundle, you’re going to see central scotomas. In each of these visual fields is a central scotoma. The one on the left is a very obvious central scotoma. Now, is there some abnormality in the greyscale up here? Yes. But the main problem with this field on the left is the central scotoma. And the center visual field — same thing. The left is sort of — excuse me. The field on the right is sort of a special case. There is a central depression, but it is contiguous, it is in contact, with the physiologic blind spot. We call this a cecocentral scotoma. And we’ll talk a bit more about that in a moment. So here’s another one of these polling slides. Let me just see if I can’t find my… There it is. Polling. I’m gonna open this up. And the question is — hold on. Let me… Huh. So it’s not letting me ask this question. Oh, wait a minute. Sorry. I’ve got it. I’m gonna launch the poll. Okay. So the poll is launched. And the question is: What sort of things cause central scotomas? Optic neuritis would be A. B, Ethambutol toxicity, C, folate deficiency, D, all of the above. So I’m gonna give you a moment or two to consider this. And then we’ll end the poll and show you the results. And then we’ll give you about 5 more seconds. 5, 4, 3, 2, 1. And here are the results. So 2/3 of you said all of the above. You’re right. The others are all right too. But all of these things can cause central scotomas. So optic neuritis certainly can. Ethambutol toxicity typically would be — since it’s a medication you take orally, it would give you bilateral central scotomas. Folate deficiency is a cause of nutritional optic neuropathy, again, giving you bilateral central scotomas. So the best answer is D, all of the above. Okay. Very good. Get out of there. So optic neuritis in this patient, who has a slightly swollen right optic disc, and typical optic neuritis has this MRI with plaques. In the United States, usually typical optic neuritis, mild disc swelling, or no disc swelling and plaques means this patient is fairly likely, at least 3 out of 4 of these patients will develop multiple sclerosis over the following 15 years. Compressive lesions can do this too. I apologize for the photo on the bottom right. That’s not the best photo. But this disc is, again, mildly swollen. There’s a central scotoma, and a tumor pushing on the globe and on the optic nerve. And here’s this case that I mentioned earlier, of cecocentral. In this case, bilateral cecocentral scotomas. When you see bilateral cecocentral scotomas, you should be in general thinking about things like toxic things, ethambutol, for instance, would be common, methanol poisoning, nutritional problems. In particular, folic acid deficiency, folate deficiency, and vitamin B12 deficiencies. And then hereditary problems. So dominant optic atrophy, Leber’s hereditary optic neuropathy can cause bilateral cecocentral scotomas. So the first anatomic correlate we just talked about — papillomacular bundle, small fibers sensitive to a variety of conditions that we just talked about. So when you look at the field test, when you look at the greyscale, you want to look centrally. The second anatomic correlate is that nerve fibers temporal to the fovea, temporal to the fovea, arc around the fovea. So they don’t run right through, right? Because for you to have your good central visual acuity, you don’t want those nerve fibers to be obscuring the foveal photoreceptors. So they run around it. And what that does is it creates this anatomic separation of the visual information. So if there’s a problem anywhere below the horizontal midline or at the disc below the horizontal midline, you’re going to be affecting the nerve fibers and the vision above the horizontal midline, and vice versa. If there’s a problem superiorly, you’re gonna cause a problem inferiorly. And these will respect the horizontal midline. So all three of these visual fields are examples of field defects where there is some respect to the horizontal midline. Now, the first one is obvious. It looks like an almost complete inferior altitudinal defect. The second one — you’ll argue: Wait a minute. What about up here? There’s some junk up here. But the most important point is that there are a number of points right along — and you can look at the numbers, instead of just the greyscale, which you should always do. But right in here there are what we call a step across the horizontal midline. And centrally, that’s more important, that’s more sensitive, so although this might look sort of constricted, clearly there are steps across the horizontal midline in this case. And 3, which really maybe at first glance looks generally constricted, you can see that right over here, nasally, for a couple of points — and again, you can look back at the number scale — you’ll see steps across the horizontal midline. Sorry. And of course, one of the most common conditions that causes visual field defects that respect the horizontal midline is glaucoma. In this patient with the obvious notching, inferiorly, and loss of nerve fiber layer, you expect to see a superior field defect that respects the horizontal midline. There’s a little extra hint of glaucoma over here, with this partially resolving Drance hemorrhage. So glaucoma, as I like to kid my glaucoma specialist friends, is just a small subset of neuro-ophthalmology, just another form of optic neuropathy. Remember that anterior ischemic optic neuropathy is also a problem that affects vision at the optic disc. So there’s disc swelling, like in this patient, with the superior swelling of the left optic disc. You would expect an inferior visual field defect, which for reasons that are unclear are common with anterior ischemic optic neuropathy. This photo on the left is a trypsin digest of the back of the eye, and so protein has been digested away, and what you’re seeing here — this is all the choriocapillaris. This is where the optic nerve used to be, central retinal artery and nerve. And you can see here these posterior ciliary arteries, which tend to be clumped in superior and inferior groups. Thought to be possibly why you tend to see the altitudinal swelling of the disc, which, again, you don’t have to see it, but you might see superior — most commonly, superior disc swelling. You may see inferior disc swelling. You may see 360-degree. But if you see superior swelling, it’s probably gonna be anterior ischemic optic neuropathy. And optic disc drusen, another condition that affects the optic nerve at the optic disc. And so you can see field defects that respect the horizontal midline. And so in this patient, with very obvious optic disc drusen both above and below, it gives you this field defect, which is an almost complete inferior altitudinal kind of a defect. Some sparing, centrally. And then a less impressive but still superior arcuate field defect, because of the compressive effects of the optic disc drusen. So optic disc drusen, accumulations of calcified mucopolysaccharides at the optic disc head. These can be diagnosed if you can’t see them. Now, I’m not going to get B-scan ultrasonography in the patient we just saw, because those drusen are obvious, but sometimes you’ll see quote-unquote “buried” optic disc drusen. You can identify those with B-scan ultrasonography. Here’s the optic nerve shadow. Here’s something suspicious that looks like it may be drusen, and indeed when you change the gain and you get rid of the soft tissue signal, you see this prominent calcification at the optic nerve head, and so in a patient who’s got an elevated optic disc that might even look somewhat swollen, your suspicion of buried optic disc drusen can be confirmed with B-scan ultrasonography. Here is a patient that was sent to me. I did not order the CAT scan to diagnose the drusen, but you can see at both optic discs calcification. This person had a headache or something, and they said — oh, by the way, there’s something going on with the optic nerves, and the patient was sent to me to try to figure that out. They had obvious optic disc drusen. I don’t recommend CAT scan to diagnose disc drusen. Too expensive, and you may not get a cut right through the optic disc, so do the B-scan ultrasonography. So here’s a field, and usually if I see a field that looks like the field here on our left, the inferior altitudinal defect, my first question is: What does that optic disc look like, for the reasons that we just talked about. So there could be some retinal problem, I suppose, but I’m gonna want to look at the disc and see: Is there swelling? Is there pallor? Is there cupping? Are there drusen? So here’s the disc that corresponds to this visual field test. And I was surprised. I said — that disc looks normal. Perhaps slight temporal pallor, compared to the nasal part of the disc. There’s no cupping. There’s no drusen. There’s no swelling. There’s no pallor. It just looks like a normal fundus. And so I show this, because, as you’ll remember, as those fibers go from the optic disc and continue along the optic nerve, there’s still that segregation of the inferior and superior bundles. So you can have lesions like this big olfactory groove meningioma that’s pushing on the optic nerve, as it exits the optic canal, and you can see field defects that respect the horizontal midline from compressive lesions, because that segregation of the superior and inferior bundles of axons is maintained until that optic nerve gets close to the optic chiasm, and then twists and forms the nasal and temporal bundles. The nasal bundles then decussating in the chiasm. So the second anatomic correlate is the nerve fibers temporal to the fovea arc around to get to the disc, so there’s an anatomic separation of the information, the visual information, at the disc. So look at the horizontal midline, in addition to the central portion of the visual field. And then the third anatomic correlate is that nasal nerve fibers cross in the chiasm. So that simply means that — as we just mentioned — as you get back towards your optic chiasm, you have your nasal and temporal fibers. The nasal fibers decussate. The temporal fibers don’t. The temporal fibers serve the nasal field of vision. The nasal fibers serve the temporal field of vision. So if there is a problem at the chiasm, or behind the chiasm, then you’re going to affect either here or here, and you’re going to get visual field defects that respect the vertical midline. And here you can see this obvious step across the vertical midline. So whenever you see a field like this, where one of the fields looks like there’s respect to the vertical midline, your first question should be: What does the other field look like? Is there respect to the midline? So here’s a patient we saw already, when I was talking about confrontational visual fields and how bad they are. This was a person who had a known pituitary tumor, and had normal fields by confrontation. And there are steps across, and certainly this is pretty obvious. Now, you might look at the field on the right and say — wait a minute. Certainly there’s this stuff over here, but it extends. Well, that’s okay. The main point is there is a step across the vertical midline here, and there’s a relative step here. It’s darkly shaded, lightly shaded. So this is a bitemporal visual field defect, and ooh, I’m gonna skip this question, because I just told you the answer. I have to be careful about that. So the lesion in this case is going to be in the optic chiasm, which would be answer B. And I should have shown — I have to remember so I don’t do that the next time too. So let’s look at the next — so here’s the explanation for this, if it’s not obvious. Again, here’s your optic chiasm. If there’s pressure on the chiasm, then here is the optic chiasm draped over this pituitary macroadenoma. Rather homogenous-looking on the MRI. It’s got this kind of a mushroom shape. Does that remind you of anything? It might remind you of a choroidal melanoma, because the choroidal melanoma breaks through Bruch’s membrane. But that Bruch’s membrane kind of acts as a tether, if you will, so you can see a mushroom shape. Well, the same thing goes for the diaphragm sellae. So here’s where the diaphragma sellae is constricting this, so you get this mushroom shape. You know this tumor originated in the sella, and has grown upward, to compress the optic chiasm. Just for comparison, here’s a more normal-looking chiasm, the way it should appear. This is actually a partially empty sella. You may have heard that term. Here’s the pituitary gland down here. This would be termed partially empty sella, which can be a very normal finding. So here’s a visual field defect. And as I just said, the first thing you do, when you look at the field, if you see one that has respect to the vertical midline, and clearly this one has respect. It crosses a little bit here, but there are many steps across this midline, where, when you look at the numbers, you can see the numbers are all zeroes or very close to zero. So there’s a big step across the vertical midline. You look at the other eye and say — whoa, wait a minute. I don’t really see. Could that be a step, maybe? I don’t know. I don’t really see any really obvious steps. This field just looks sort of generally crummy. And so the question is — and I think I have a polling question for you. Let me just see. I do. So let’s look back at this, because I’m gonna ask you: Where is this lesion? Or lesions? Because this is a little bit weird-looking. So let’s ask the question. Hold on. I’m gonna open up my… Let’s see. Is it this one? Lesion is… Hold on. I’m gonna go with this one. Wait a minute. Hold on a second. Sorry. Okay. I’ll launch the poll. Okay. So here are your choices. Where is the lesion? Is it in both of the optic nerves? Is it the nerve and chiasm? Is it the optic tract? Or could it be the optic radiations? So optic radiations, by definition, would be after the lateral geniculate body. The optic tract is between the chiasm and the lateral geniculate body. And… All right. I’m giving you guys a little more time to vote. And we’re gonna show you the poll. 5 more seconds. 5, 4, 3, 2, 1. I’m gonna end the poll. And show you the results. So very good. So certainly the vast majority of you got this question right. This is a problem with the nerve and the chiasm. So this is a problem where the nerve — let me just go. I’m gonna go back to those visual fields, I think. Or maybe… Do I have… Hold on. Let me go backwards. So this is a problem. You’ve got the temporal field defect in the left eye. You’ve got general depression in the right eye. So this is a problem with the right optic nerve, as it is approaching the chiasm, and including the chiasm, but you’re just getting the temporal fibers that are coming over from the other side. So this is sometimes called an anterior chiasmal syndrome, where you get one nerve and the crossing fibers from the contralateral side. Sometimes it’s called the anterior chiasmal syndrome. Sometimes junctional scotoma. Sometimes a syndrome of the distal optic nerve. They all mean the same things. So very good, those of you who got that right. So let’s look at another visual field. Notice this is a little bit tougher, because it’s a Goldman visual field, and in general, at least my residents are not used to looking at Goldman visual fields. But if we look at this, and I think one of the tricks to looking at Goldman visual fields is: Look at the smallest isopter first. So the smallest isopter is gonna be, in this case, just this little red bit here. So basically if you just look at this, that means all the rest of the field is gone to this smallest stimulus. So there’s a big central scotoma. You can tell that from the other isopters. When you look at the other side, if you look at just the smallest isopter, what do you see? You see this. With exquisite respect to the vertical midline. So this is the right eye. So this person has lost all of their temporal field. And so they have exactly the same thing our last patient had. They have a big central scotoma, because the left optic nerve is being compressed. And they’ve lost most but not all of the temporal field on the other side. So this is an anterior chiasmal syndrome, but this is a left anterior chiasmal syndrome. So you’ve got the left optic nerve, plus most but not all of the fibers crossing from the other eye in the anterior chiasm. So left anterior chiasmal syndrome. So this one is a bit tougher. So, now, this is an Octopus visual field. There’s not a big difference between the Octopus and the Humphrey. The Octopus visual fields plot the physiologic blind spot as white. So here’s the normal physiologic blind spot in the right eye. Here’s the normal physiologic blind spot in the left eye. But otherwise the shaded areas are where the patient’s not seeing the light stimulus as well. So in this case, it’s a little bit funny-looking, but the thing that’s, I think, striking is that the field defect is in the same place in each eye. Whenever you see that, you should be thinking homonymous. I didn’t mention that term earlier. I should have. Homonymous hemianopsia means that you’re back behind the optic chiasm, and now you’re affecting the visual system on the same side, and there’s some respect to this vertical midline. This is a funny pattern. It’s sort of a wedge shape, so there’s a wedge here, and there’s a wedge here. Let me just see if I have a — this may be… So this is a homonymous hemianopsia, and I don’t think I have a question, so this is a lateral geniculate body field defect. And this is a lateral choroidal artery infarct that gives you this wedge look. At the bottom I say an anterior choroidal artery infarct gives the opposite. That means that if you have an infarction from the occlusion of the anterior choroidal artery, this part that’s grey is spared. So you have this funny-looking shade here, and shading here, in both eyes. So when you see this wedge sort of look, you should think lateral geniculate body, because of the anatomic organization of the lateral geniculate body in layers. So here’s another visual field defect. So you should look at this very quickly and say… Ah-ha. There’s clear respect to the vertical midline. This is homonymous, because it’s in the same place, so I would describe this as a superior right homonymous hemianopsia. You could say a superior right homonymous quadrantanopsia, if you wish. And I think I have a question for this one. So hold on. Let me get the polling ready here. So here’s the polling question. What do you think is the most common cause of this kind of homonymous — oh, wait a minute. Sorry, I didn’t launch it. This. Now I launched it, so if you voted, vote again. What do you think is the most common cause of a homonymous hemianopsia in a 28-year-old? And I’ll give you a few seconds to vote here. And I’m getting… And I think that certainly all of these things that I’ve listed could cause homonymous hemianopsia. All right. I’m gonna give you 5 more seconds. 5, 4, 3, 2, 1. Okay. I’m gonna end the polling. Everyone’s voted. And let’s show the results. Good. Okay. So I think that the group that said trauma — I think that’s probably correct. That’s probably the most common reason. It may depend on where you live, I suppose. But I think in my practice, I probably see more trauma. Now, can MS cause this? Sure. You can have a big plaque in the retrochiasmal visual pathway, and in the optic neuritis treatment trial, when they followed patients over 15 years, they found as many as 12% to 15% of patients had some sort of homonymous visual field defect during the course of follow-up. So you can definitely see plaques from MS in the retrochiasmal visual pathways. Tumor can definitely do it. But tumors in 28-year-olds aren’t that common. And strokes can do it too. Also not that common in a 28-year-old. But certainly could do it. In this case, the problem was trauma. A screwdriver in this case. Long story. It was a Phillips head screwdriver, in case you were wondering. So here’s another homonymous hemianopsia. This is an inferior. It’s not really quadrantic. Certainly the inferior right quadrant is out. Except for one little spot right there. But it does extend above the midline. I probably would just say that this is primarily a right inferior homonymous hemianopsia, or inferior greater than superior, I guess. And I think I’ve got another question. So most likely cause — let me get my polling thing here. Hold on a second. Okay. Here we go. So most likely cause in a 63-year-old male. So an inferior right homonymous or inferior greater than superior homonymous right hemianopsia. Gonna give you a few seconds here to vote. So the choices I listed here are stroke, tumor, non-organic — non-organic as a term means that the person is faking or is malingering or has some other reason to not have a true visual field defect. All right. A lot of people have voted. And I’m going to give you 5 more seconds. 5, 4, 3, 2, 1. Let’s see what the audience says. Okay. So certainly the majority of you are correct. But certainly the most common cause in a 63-year-old is gonna be stroke. Tumor certainly can do it, but tumors, again, nowhere near as common as strokes. Probably no matter where you live. No one said non-organic, and one person said other. I’m not sure what the other would be. Certainly there are other things that could do this. Interestingly enough, this visual field — and let me just… This visual field is a field that I did. Because I needed this for a book chapter, frankly. And my point is here that you can — anybody can do what they want in this test. I was pretty good. I did start beeping in the bottom right there, when I remembered I wasn’t supposed to, and that’s why you have this little area right here. But some years ago, there was a study, a report in, I think it was the Journal of Ophthalmology. The title of the study was Field of Dreams or Dreamed-Up Fields. And basically what a neuro-ophthalmologist did, John Keltner — asked a bunch of medical students. He showed them greyscales, like this, or a central scotoma. Or a bitemporal field defect, and he said do this on the field test. And pretty much every time they did it perfectly well. So the results of this test are not gospel. I rarely see someone who would fake a homonymous hemianopsia. But I do see people who fake unilateral hemifield defects. So just because it shows up on the test does not mean it’s necessarily real. All right. Here’s a wacky-looking visual field. This is a Goldman visual field. Again, if you look at the smallest isopter, which would be the blue, you can see there’s a step across this vertical midline. There’s a step across the vertical midline. There’s something more than just — more going on, though, than just homonymous field defect, because if it were just homonymous, you would have a normal field over to the right. So if you ignore the left side of this visual field for a second, and just look at the right side, you would say — oh, it looks like there is an inferior sort of partial quadrantic loss on both sides. You see that? If you ignore the right side for a moment, you would say — oh, there’s nothing here. So in the right eye, there’s complete loss of the visual field to the left. In the left eye, there’s almost complete loss, except for this shape over here. And this is the temporal crescent. The temporal crescent is the visual field that’s represented in the anteriormost occipital lobe. Where is it over here? There is no nasal crescent. Right? Your nasal field only goes out to 60 degrees. So this is extra, if you will. Extravisual field. So it’s the only place in the retrochiasmal pathway where there’s monocular representation of the vision. And that’s the anteriormost occipital lobe. So here’s a schematic. And here’s the anteriormost occipital lobe. This corresponds to the temporal crescent, a monocular representation of the field. So if you see a field defect where there’s sparing of the crescent, it means there’s been a problem in the occipital lobe, in the occipital lobe. And way back here is where the central vision is. And so when we talk about macular sparing, you can have a stroke here, and spare the occipital tip, and you’ll have sparing of the macular representation. Here’s the field that I actually mentioned earlier to you. This is a field — you look at — this is a patient who had an accident at work, a refrigerator repairman, the hose blew off the refrigerator, hit him above the eye, caused a little laceration, but he lost vision to the right in that eye. His laceration healed up, and he had repeated visual fields that looked like this. And you’ll notice — what did I say? When you see something that respects the vertical midline, look at the other eye. Nothing. He has no symptoms in the left eye. He’s got normal vision. He’s got normal vision here. His symptom simply is: I can’t see off to the right. And by confrontation, same thing. Automated fields — this is probably the third one he’s done. He’s already seen a couple of ophthalmologists. He’s already had an MRI. All of it’s normal. His examination is otherwise normal, and there is no relative afferent pupillary defect. So the question is: Is this possible? And the answer is… Well, it is possible to have a unilateral temporal field defect. So if there were a lesion in the distal optic nerve, as it approaches the chiasm, but not in the chiasm, there’s some segregation of the nasal and temporal bundles. So if you were to have a tumor, let’s say, and it was pushing on the distal optic nerve, not on the anterior chiasm, you might see, usually, a superotemporal field defect, but you should see a relative afferent pupillary defect. Because the basis of a relative afferent pupillary defect is asymmetric field loss. Right? This person clearly has asymmetric field loss. So given that there’s no relative afferent pupillary defect, and of course, I already know his MRI is normal… How do you prove that he’s doing this on purpose? I told him I wanted to repeat the test one more time. And we did it with both eyes open. Now, when we did this visual field test in his left eye, what did we do to his right eye? We blinded him, right? We put a patch over his eye. And what happens? You get a field that looks like this. So here’s what happened when we left both eyes open. So with both eyes open, he still is purposely not beeping the buzzer when lights are on the right, because he’s thinking that… Oh, I can’t see to the right in my right eye. And so he’s doing this on purpose. Now, the lawyer — his lawyer didn’t quite understand, but I reassured the lawyer that he was malingering. All right. A couple last visual fields. So this one is interesting, and whenever you see a field that looks kind of like this, where there doesn’t seem to be a pattern, erase the physiologic blind spots with your mind. So erase this spot. And erase this spot. If you erase those two spots, what you’re gonna see here are rather superimposable fields. There’s a problem up and to the left, looking very similar. There’s a problem over here, to the right, looking similar. There’s a problem down here, looking similar. So this person has — excuse me — so-called patchwork quilt kind of field defect. This is very superimposable. You should be thinking bilateral, homonymous hemianopsia. And this person had a shower of emboli during coronary artery bypass surgery to both occipital lobes. And he has multiple little occipital infarcts, causing these field defects. So it’s a homonymous hemianopsia. Sorry. Excuse me. Just got back from a long trip, and my ears never popped and I got a cold, and I’m jetlagged. Here’s another field. Extremely superimposable. I can’t even tell which eye, without looking at the eye. This one happens to say left over here. Very superimposable. So this is a bilateral homonymous hemianopsia. If you look here, you’ll say to the right it’s almost complete. But right down here, it’s not. And right here, it’s not. Same over here. If you look to the left, you’ll see there’s primarily a superior left quadrantanopsia. On both sides. Another patient, after coronary artery bypass surgery, who had bilateral occipital lobe strokes. And I show you this field just because it’s a little bit scary. So this visual field is in a young 38-year-old fireman. Whose symptom really didn’t have much to do with his peripheral vision. He was seeing flashing lights to the left and down. And we did the — his exam was normal. His fields by confrontation were normal. We did automated perimetry, and it showed this. So he had this kind of non-specific-looking field defect here, but also down here. He’s a young, healthy guy. Very reliable. And I said… Jeez. You know, it’s possible this is some kind of homonymous hemianopsia. Let’s get an MRI just to be on the safe side. And he had this large glioblastoma, which killed him. But the point is… That’s not a real significant field defect, and I was fairly surprised when we found this lesion causing that field defect. So a little bit of a wake-up call. So the third anatomic correlate is that nasal nerve fibers cross in the optic chiasm, so you want to look closely at the vertical midline. And then I wanted to show this, because this doesn’t really fit into our three anatomic correlates, and that is the big physiologic blind spot. So the question is: Why do you have a blind spot to begin with? Because no photoreceptors overlie the optic disc. When a disc is swollen, you see enlargement of the blind spot. Why? And it’s not because there’s some obscuration. It’s because the swollen nerve pushes the retina, adjacent retina, aside. So there’s a bigger blind spot. So a big physiologic blind spot doesn’t really fit any of those anatomic correlates that we just talked about. So in summary, when you look at visual fields, certainly you want to look at the reliability coefficients. You want to look to make sure the person is reliable. That said, if the person is not that reliable, and there are obvious field defects, homonymous hemianopsia, it doesn’t mean you disregard the visual field testing. You’re gonna look for patterns. You’re gonna look centrally. You’re gonna look at the horizontal midline. You’re gonna look at the vertical midline. And then hopefully characterize the field defect and make the correct diagnosis and order the appropriate imaging studies. So I think I will leave it at that. And I know there are some questions here. Let me just see if I can open up the questions. All right. So… Let’s start. All right. Okay. Here’s a question. It says: Do you use 10-0 Humphrey visual field in real practice as suggested for severe generalized depressed field? I use the 10-2, and yes. So the time I would use a — there are a couple of times I would use a — if you have someone, say, with end-stage glaucoma, or for whatever reason, they have very constricted visual fields, you’re not getting any information from all the rest of that field, then you want to look centrally at the remaining field for change. So the 10-2 is gonna give you more information with that remaining visual field. So that’s one reason I use the 10-2. The other reason I’ll use a 10-2 is if the person’s symptoms sound like there’s a little central or paracentral scotoma. So if someone says — like, if I’m reading, and I can’t see part of the word, for instance, that’s gonna be a really small paracentral scotoma, and that might get lost in a 24-2. Because remember, other than the amount of visual field checked with a 24-2, the grid density of the stimuli are tighter. So the 2-degree grid density on a 10-2, and a 6-degree grid density on a 24-2. So if you have a small scotoma, and you do a 24 or a 30-2, you might miss it, whereas you might find it on a 10-2. The next question is: What is the best visual field test for neuro-ophthalmology? Well, that I guess gets to what I just said. So it depends. My routine visual field test is a 24-2 fast pack. Why? I want to get the information as quickly as I can. The longer the field test is, the more difficult it is for the patient. Depending on your patient, the more time you make them do the test, the more garbage you’re gonna get out of it. So I like to get as much information as quickly as I can. I don’t think the 30-2 gives me much more information. I’m looking for patterns. So I get a 24-2 fast pack — is my common test. Okay. Then there’s a question: Isn’t this better to be called quadrantanopsia, rather than hemianopsia? Yeah, I don’t really care. I don’t really like the term quadrantanopsia, frankly. But it’s perfectly okay to use it. It is a homonymous hemianopsia, either way about it. So you call it a — I would call it a homonymous quadrantanopsia, versus a unilateral quadrantanopsia. I don’t really care what you call it. When do you use stimulus V and I? So stimulus size V — I will use if I get a field and it just looks terrible. It just looks really bad, and yet the patient has some decent vision. I will use a size V, because you’re going to — it’s easier for the patient to see it. So someone’s got a bad visual field defect, but yet you want to follow that defect for change. If you do a size III and it’s black, you can’t follow that, unless, of course, there’s a miraculous recovery, or a bigger recovery. So in someone who has an optic neuropathy, and their size III field is black, I’ll get a size V, because I may be able to follow that better for change. A size I — I rarely, rarely order a size I. Sometimes if you’re looking for something really subtle, say… Some sort of toxic optic neuropathy, medication-induced, hydroxychloroquine, something like that, you might order a size I, to make it really sensitive. Will visual field perimetry test for visual inattention? Yes. It can check for visual inattention. Although that’s a much longer story, and tough to answer in a short period of time. The question is about — non-edge points to consider in 32 only or also in 24… Um… Not really sure what is meant by non-edge points. Certainly, I mean, if you order a 24-2, you’re missing that temporally superior and inferior — you’re missing 6 degrees. So could you miss a field defect out there? Sure. Most of my glaucoma specialists are probably — their standard tests are probably 30-2. For neuro-ophthalmology, I’m looking for patterns, and so a 24-2, I think, is adequate. I’m not sure that answers your question. What are steps and where exactly — okay. So a step simply means that there is a difference in response across either the vertical — excuse me — or the horizontal midline. So when I look for steps, if I see three contiguous points across a vertical or horizontal midline, three points where there’s a big difference, and let me see if I can’t… I think I can go back. So let me just find it. So, for instance, I hope you can still see this — so here a step would be — and it’s probably better… Let me see if I can’t… Hold on a minute. Let me find… So, for instance, on this visual field test, and I’m gonna blow this up — if you look across the vertical midline, here, you’ll see 4 versus 26. 8 versus 22. 0 versus 30. So this is a huge step across the midline. Meaning a big difference in response. The closer in — so if I see three points here with at least a 7 decibel difference, so, for instance, up here — 3 and 17. Well, that’s a 14 difference. 3 and 8. That’s not as interesting. 6 and 16. 10 points. 2 and 27. 25 points. So the closer you get to center, the more sensitive as well. So if I see a couple points centrally, across the midline — interesting. That’s a step. If I see three from the outside in, that’s a step. With at least 7 DB difference. That’s gonna be interesting to me. All right. Let’s see. Any particular pattern for optic neuropathy, neuritis? No. So when I was in training, I was taught — oh, you have to have a central scotoma if you have optic neuritis. Certainly not true. You can see any field defect in optic neuritis. And in the optic neuritis treatment trial — and I didn’t show you that slide — we’ll probably talk about optic neuropathies, perhaps in the next webinar that I do — you can see any field defect. You can see homonymous, you can see anything — horizontal, vertical, midline, I don’t care what it is — you can see it. There is no particular pattern. When vision is reduced to less than 360, can you still assess the field? Sure. So you can assess the field. You can assess it by confrontation. With, instead of counting fingers — of course, you can see if they can count fingers. But if they can’t count fingers, then you see if they can see movement. And you could try a size V. So that would be a time where you might try a size V test object on the automated perimetry. What should we look for when neurologists send an idiopathic intracranial hypertension patient? So the most common field defect — of course, big blind spots are common. I don’t really consider that much. It’s not an optic nerve problem. It shows that there’s swelling of the nerve. So certainly you’re gonna look for big blind spots. That does not mean there’s any optic nerve dysfunction. Remember, the reason there’s a big blind spot is that the retina is being pushed aside. But the most common field defects, early field defects, in idiopathic intracranial hypertension, are inferonasal steps. So loss of field inferonasally, across the horizontal midline. Again, because papilledema is a problem at the disc. And so that fits our second anatomic correlate. So you’re gonna look — that’s most common. But you can see inferior arcuates, inferior and superior arcuates, generalized constriction — but early on, you’re looking for enlargement of the blind spot, and inferonasal steps. Any difference between junctional and Traquair scotoma? Oh, I’m gonna say I don’t know, because I can’t remember the definition of the Traquair scotoma. Sorry. It’s not a term that I use. Junctional, though, should only be used to refer to a field defect where there’s an optic nerve pattern on one side and a temporal field loss on the other side. And so you’ve prompted me. I’m gonna have to look up Traquair scotoma. Because that’s in my distant memory from residency, and that was a long time ago. What is your minimum visual acuity to perform a Humphrey visual field test? So my minimum visual acuity — I think it depends a lot on the patient and a lot on what you think. So here’s a time when let’s say the person is counting fingers, but they’re counting fingers briskly in all four quadrants. You could consider a size V test object. Hand motion, I’m probably not gonna even try automated perimetry, if they’re hand motion or worse. Is there any difference between the Humphrey visual field frequency doubling? Frequency doubling is — I look at that — it’s faster. It’s more of a screening test. There have been certainly papers out there, and I have certainly seen screening frequency doubling tests where it looked like there’s some obvious neurologic defect, and when you repeat it with Humphrey fields, with full threshold testing, it goes away. So I think if you’re screening with frequency doubling — but personally, if I were to find something with frequency doubling, I would repeat it before I go order an expensive imaging study. Can you please explain to us the comprehensive field with the red cap? So it’s not really different than saturation of the color. It takes that into use. So the idea is that if you have a field defect that respects the vertical or the horizontal midline, that in that field, the person may not see the red top as bright red. And so it’s a way that you can try to increase your sensitivity of confrontational visual fields. I don’t use it a lot myself, because I have easy access to automated perimetry. If you don’t have access to automated perimetry, though, you can use it. Basically you just hold that red top up, just right on one side, and then the other side of the horizontal and of the vertical midline, and see if there’s a difference in color perception. The problem is patients say the darnedest things, and so you make it very subjective. Let’s see. Oh, more questions here. Kindly talk more on AION visual field changes. So anterior ischemic optic neuropathy, and particularly if we’re talking about non-arteritic, the most common would be field defects that respect or have some sort of a step across the horizontal midline. Because the problem is at the disc. Inferior, for whatever reason — I don’t know the answer. Inferior field defects and inferior altitudinal field defects are more common. And does that mean you can’t see something else? No. You could see a complete loss of field, if the person’s count-fingers in the eye. But for whatever reason, inferior field defects are more common in anterior ischemic optic neuropathy. Is it recommended to do a visual field on a patient whose pupils have been pharmacologically dilated? So I don’t typically dilate pupils. You can. As long as you give them the right correction, taking into account that you’ve paralyzed their accommodation. Some of my partners often do them dilated. I don’t do them dilated unless the pupils are really tiny. So if the pupils are like 1 millimeter, 1.5, maybe 2 millimeters, I might dilate the pupils. It’s perfectly okay to do it, if they’re dilated. You just have to account for accommodation paralysis. Do you ever do a Bjerrum’s field? I’m not sure what that is, so I guess the answer is no. Classical optic disc drusen — I’m assuming that means what sort of field defect. So you can see enlargement of the blind spot if the drusen are really prominent, but inferonasal, sort of like glaucoma, sort of like papilledema, inferonasal field defects — most common early field defects. I didn’t mention and probably will in future lectures, on optic neuropathies, that disc drusen tend to spare central vision, kind of like glaucoma. So that you would not see a central scotoma with drusen. If you do, think of something else, because it’s not the drusen. Is there a specific field defect in normotensive glaucoma? I don’t think I would say specific, but I tend to see patients with normotensive glaucoma who’ve got what look like sort of central scotomas, but they split fixation. So there are these small scotomas that split fixation, either inferiorly or superiorly. And they’re sometimes sent to me sort of as — gee, hey. This doesn’t look like glaucoma. But in my experience, that’s by far the most common type of field defect I see with low tension glaucoma. Now, there’s probably a referral bias, because if they have some other more glaucomatous-looking field defect, they might not get to me, as a neuro-ophthalmologist. But I see that frequently, and I can look in advance and say — this person is gonna have normal pressure glaucoma, without even looking at the patient. Possible patterns of functional visual loss on visual field. You know, I think the only thing that I’ve rarely seen is homonymous hemianopsia. I do see that temporal hemifield defect from time to time, because it’s usually in a patient who has some medical background or has a relative who had a stroke, and they complain that they lost vision to one side. And even the patient who’s had the stroke doesn’t realize — it’s really they’ve lost vision in both eyes to the same side. In other words, they have a homonymous hemianopsia. But I’ve seen all sorts of different things. I think like small central scotoma — very unlikely. Typically the most common patterns in non-organic are very constricted visual field. You know, the so-called tunnel visual field, or they just don’t beep, and it’s all black, and then probably third most would be the temporal hemifield loss.Yeah. Someone said something about — yeah, Bjerrum scotoma I’ve heard of. I’m not sure if there was a question regarding that. Is there a situation where you request a full field test in Humphrey? So there are other — of course, there are other tests, other than the one I mentioned, the 24-2 and the 10-2, which are my sort of two most common things I order. There’s the full field 120-point screening test. So yeah, I think there are times where you would order a full-field test, if you were concerned about the temporal crescent, which is pretty rare. You might do that. I personally rarely order a full field test, unless the patient says — you know, I’ve got something way out in the periphery, and I just don’t think it’s gonna show up on a 24-2. Should automated perimetry be done with or without presbyopic correction? So it should be done — assuming we’ve not dilated the patient, yes, it should be done with the appropriate age presbyopic correction. Yes. The answer is you want to correct for their age, if they’re in the presbyopic age. And I think that is the end of the questions. And I think we’re a little bit over time. So thanks, everybody, for tuning in to the second neuro-ophthalmology Orbis webinar. We’re planning on trying to do maybe a sequence of these, of a more complete sequence of neuro-ophthalmology sort of curriculum — probably the next one will be on optic neuropathy, but that may be the topic of more than one webinar. So I think — let me just see if there’s any other questions. I’m gonna go ahead and sign off, so thanks, everybody. Have a good day, evening, night, whatever time it is, wherever you are.

Download Slides


September 12, 2016

Last Updated: October 31, 2022

1 thought on “Lecture: Visual Field Interpretation”

Leave a Comment