Transcript - The Role of Magnetic Resonance Imaging
in the Diagnosis and Management of
Relapsing-Remitting Multiple Sclerosis

HOLLY ATKINSON, MD: Hello, and welcome to MS Conversations. I'm your host, Dr. Holly Atkinson. Today's conversation is going to focus on the role of MRI in the diagnosis and management of multiple sclerosis. Joining me today are Dr. George Hutton and Dr. James Miller. Dr. George Hutton is an assistant professor of neurology at the Baylor College of Medicine in Houston, Texas, and Dr. James Miller was, until recently, director of the Multiple Sclerosis Center at Columbia-Presbyterian Medical Center in New York. Welcome, both of you.

Jim, I want to start with you, because you obviously remember practice long before MRI was invented, and of course, MRI imaging has come a long, long way. What do you see as its role in the diagnosis of MS today? How important is it?

JAMES R. MILLER, MD: It's a critical in diagnosis of multiple sclerosis. It's the most important piece in our armamentarium that we have, because it gives us information both on the multiplicity of lesions and the intensity or the number of lesions, which is also important, even for prognostic purposes, as well as diagnosis.

HOLLY ATKINSON, MD: George, you've obviously used this in your training all along. How important is it to you?

GEORGE J. HUTTON, MD: I would agree. It's a very important aspect to the diagnosis. We always go back to the basics of diagnosing based on clinical symptoms and clinical presentation, but the MRI clearly is very supportive of the diagnosis and, in certain cases, helps to make the diagnosis when the clinical setting is not fully apparent.

It's a double-edged sword to some degree. Sometimes we see cases where MRI was done for different reasons -- perhaps for headache, and it shows finding suggestive of MS, and of course, then you have to go back and decide if clinically things fit with that diagnosis. In other cases, you have someone who clearly has MS, and perhaps the -- clinically, I should say, and the MRI does not necessarily look fully like MS normally presents.

So there are some difficulties with the diagnosis using MRI.

HOLLY ATKINSON, MD: Well, George, let's talk about how you determine disease burden. There are three measures. Tell us about those.

GEORGE J. HUTTON, MD: Well, historically, the most established measure has really been the T2 lesion load -- the number of hyperintensities, the size of hyperintensities seen on T2 imaging. And these are the classic bright white spots that we see that most people are familiar with. Those, however, are somewhat nonspecific and they can represent areas of inflammation and edema, areas of demyelination, as well as acute areas of loss. So they don't necessarily help us as much as some of the other modalities. T1 imaging is now being used more and more to look at hypointense areas, and these are felt to represent some more chronic areas. And then there's gadolinium enhancement, which we look at on T1 imagining, that gives us much more information about the acuity of lesions. So that those lesions which show contrast enhancement show an area of breakdown of the blood-brain barrier and are representative of acute lesions.

HOLLY ATKINSON, MD: Maybe you can talk to us a little more about that, Jim, to explain to us the histopathology of what's going on that's coupled with the findings.

JAMES R. MILLER, MD: Well, as George indicated, the bright spots on T2 or now we use mostly the flare image for that determination, particularly in the hemispheres, is really something that any type of lesion in the brain will show. It's not specific. There are certain characteristics of these lesions which are helpful in the diagnosis if they tend to center around the ventricles, that makes it very likely. If there are small subcortical lesions, that's also an important finding in inflammatory disease, particularly multiple sclerosis, but these really don't correlate directly with any particular type of pathology. You can see these and they represent edema or a large part of it will represent edema. Some of it will represent older lesions that are essentially burned out. And some of the them will mirror the acute lesions that you'll see on the T1 scan, which because of the leakage of the paramagnetic substance into the brain, is a sign of acute inflammation.

HOLLY ATKINSON, MD: Jim, what is the predictive value of these lesions?

JAMES R. MILLER, MD: Well, the predictive value is something that we're working on right now. We've, in the early days of MRI scanning and multiple sclerosis, we would find patients who had a lot of lesions but who were doing clinically very well, and it took us quite a while to understand this, and now we've come to learn, over the course of time and repeated studies of people like this and also people just starting out with the diagnosis of MS, that the number of lesions in the hemispheres, particularly, can be highly predictive of what's going to happen to the person in the future, if not necessarily what's happening to them right then. So there is a great predictive value, and certainly, with having a few lesions at the first attack of MS, you can predict with high confidence that they're going to go on and develop a second attack and have clinically definite MS.

The converse is also true. If someone presents with an inflammatory attack and has no lesions on the brain MRI scan -- on the T2 or flare-weighted scans -- then it's highly unlikely, although still not impossible, that they are going to go on to develop multiple sclerosis. And that's been a comfort to us, because in the past, even those people we couldn't say definitely or likely whether they were going to have multiple sclerosis or not.

HOLLY ATKINSON, MD: Now we've talked about the three different subtype of lesions, George. Did they offer a different kind of predictive value in terms of clinical course, or do you take all three of them together as a picture?

GEORGE J. HUTTON, MD: There is some development in that understanding as well. Gadolinium lesions, for example, it's very likely from studies that we've seen that they do lead to further gadolinium-enhancing lesions, further T2 lesions down the road. And so that someone who does presented with a large number of gad-enhancing lesions is predictive of someone who will develop more lesions, at least. And hence, from what Dr. Miller said, yes, very likely will also develop worse disease.

T1 hyperintensities (that we've talked about briefly as well) are more likely to represent areas that will show persistent deficits on the imaging, and therefore more likely also to have some clinical accompaniment to that lesion. These are areas of more chronic loss. And one thing we haven't really touched upon that I think's important to remember, too, is we're talking mostly about brain MRI so far, but spinal cord MRI -- everything's sort of exaggerated even from what we've talked about. Small lesions in a certain critical area in the spinal cord can have very obvious symptoms that could persist, and so I think it's very important to talk and to look at spinal cord MRI, too.

JAMES R. MILLER, MD: That's a very important point. We've concentrated a lot on head MRI scans in our thinking about MS, and that leads to one of those puzzles I mentioned where people would have a lot of lesions in the brain and still clinically look very good, and that's because their spinal cord wasn't that much affected, and conversely, you have to look in the spinal cord because people can have signs of attack and have nothing in the brain, and yet there'd be a lesion in the spinal cord which could be quite small that's causing a lot of difficultly.

HOLLY ATKINSON, MD: George, earlier on, you had mentioned, obviously, using MRI in difficult cases where the diagnosis is not clear-cut clinically. Let's talk about that a little bit. How do you use the findings on MRI to argue for evidence for -- let's talk about for MS initially, and then when you find evidence against MS.

GEORGE J. HUTTON, MD: Okay. This all assumes, again, that we've got a clinical situation, that we're suspecting the disease. And I'd just like to make that point: we're talking about MRI, but clinically we want to make sure we're dealing with the right scenarios. I think a trap people fall into and a reason for consultation to an MS center very frequently is the abnormal MRI is someone who doesn't have a clinical syndrome of MS, and that's sort of a different question.

But for someone you're suspecting MS and you're doing an MRI for that reason, there are some specific things we'd like to see: the size of the lesions generally larger than 5 mm (so that we may see smaller lesions in MS, but if they're all less than 5 mm, I think you must consider the possibility that this may not be MS). Location's already been referred to. Periventricular lesions, brainstem lesions, subcortical lesions. But again, you'd like to see more periventricular lesions. Orientation -- perpendicular to the ventricles is a very classic orientation. Shape -- ovoid shape as opposed to linear shape as opposed to all shaped like comma-shaped, for example. Things like this. So there are some characteristics we look for, and this is mostly on the T2 imaging when I mention these types.

HOLLY ATKINSON, MD: Evidence against, Jim?

JAMES R. MILLER, MD: Well, certainly the converse of whatever George said, for the positive evidence, but also the things that you worry about, as I mentioned, are no lesions, which is very important, despite the clinical thing, although sometimes I've made the diagnosis of multiple sclerosis because of repeated attacks without and a positive spinal fluid for oligoclonal bands, without the presence of MRI scan changes, we can only assume that they're so small and subclinical in terms of viewing them that these people have negative scans. It's very rare. It's very rare.

The other thing that worries you is if there's meningeal enhancement, as long as the MRI scan is done before the spinal tap. You have to remember that after a lumbar puncture, people do get enhancement of the meninges on flare imaging, and so that could be misleading.

HOLLY ATKINSON, MD: Let's turn for the remainder of our time to the use or MRI in the management of MS. Diagnosis is something we think about all the time when we talk about imaging, but obviously there's a role there for management as well. How have you been using it, Jim?

JAMES R. MILLER, MD: Well, this has become much more important in recent years. We've turned to the concept of adding on therapies to our baseline therapies and when people aren't doing as well as we would like them to do, even though they make be taking one of the basic medications: the interferons or glatiramer acetate that's available. And we're trying to look at ways to judge prognosis.

In the past, I often did not do follow-up MRI scans because it didn't change my treatment, but now that it does, we've started to look at it, and this leads us to one of the reasons that we talk about in regularizing the scans so there is a comparison from one time to the next. Nobody has exact definitions of when you worry about scans changing, when -- this is when people are doing well clinically, but the idea is that we need to turn to this, because we know that MRI changes are much more frequent than the clinical changes, and we also know that prognostically, those changes will be important in the future. So we have to use the MRI scan to come to some conclusions about when to change the medication.

HOLLY ATKINSON, MD: George, Jim said "to regularize it." Give us a summary of best practices for the practicing neurologist out there. What does regular mean?

GEORGE J. HUTTON, MD: Groups have gotten together and tried to make some suggestions here, and unfortunately, each group has differed a little bit, and there's some uncertainty and some controversy as to how often they should be done in terms of what frequency of scanning brain, what frequency of scanning the spine, and what we've suggested, really, is to consider after the diagnosis, repeating brain MRI at about one year, and then perhaps at year intervals, at least during the period of time where someone is active and having frequent relapses, for example.

I would tend to decrease that frequency in someone who's had a chronic course and perhaps is not changing clinically too much.

Now spine MRI certainly is more controversial and unfortunately even in our small group, we weren't able to agree on a frequency there. I think if someone has spinal cord symptoms, certainly you should look at a spinal cord MRI, but for routine monitoring, it's really not something that we could highly recommend.

JAMES R. MILLER, MD: I think this just has to do with the different volumes of central nervous system tissue. You get (so to speak) more "bang for buck" in looking for changes if you look at the brain because it's a much bigger volume than the spinal cord. This is all in people who are clinically asymptomatic or not very active. If you have a lot of activity, you're going to take a stand based on the clinical activity of people going downhill. You're not going to wait for the MRI scan to tell you, yes, it's time to go ahead.

But on the converse of that, when people are doing well and you see a lot of changes in the MRI scan, that's basically the worth of follow-up scans.

HOLLY ATKINSON, MD: We have just a few minutes left, so I'd like to concentrate on some summary statements for the neurologist. What do you think is the most important thing for them to keep front and center when using MRI, either in the diagnosis or management?

JAMES R. MILLER, MD: Well, I think one of the things to keep in mind is that really is it only an adjunct to diagnosis. At the onset of our talk, I sort of made it sound like that's the only thing I looked at, but as George said, it's very important to keep the clinical situation in mind. I think it's important to try to regularize it. Use the same views. Some of the views that are considered pretty standard are the axial and sagittal flare views, also the axial and sagittal T1 views and a gadolinium axial view in the brain, and then in the spinal cord, T2 is probably a better choice than flare, because flare is subject to CSF pulsation artifacts. So in the brainstem and in the spinal cord, T2 imaging is important along with a gadolinium-enhancing T1 scan.

HOLLY ATKINSON, MD: George, take-home points?

GEORGE J. HUTTON, MD: Along with the measures we've talked about, and this is sort of an additional point we didn't exactly cover is atrophy should be considered, as well. We talk about looking a lesion volumes and lesion size, but it's really important to remember that atrophy is a common phenomenon in this disease. I like to explain to my patients we all lose brain volume with age, but unfortunately it happens at a more frequent rate with multiple sclerosis. So it's important to be aware of, it's important to know how to look for it, and it's not something you need fancy tests for, you know, that they do in some of the studies. You can learn from visual analysis just to be able to tell, by looking at the size of the ventricles and the size of the sulci if there is significant atrophy in your patients.

JAMES R. MILLER, MD: I would also add that the scans of the sagittal scans of the brain and looking at the corpus callosum is also a very important indication of atrophy, because that thins out quite typically in multiple sclerosis cases and in very few other situations.

GEORGE J. HUTTON, MD: And atrophy and corpus callosal atrophy, to some degree, particularly are best correlated with a clinical measure, and that happens to be with cognitive dysfunction. So it's really the best correlate we have, and once you remember that and start looking for it, I think it does make an impact on your practice and understanding the aspect of your patients' care.

Secondly, you have to look at your own films, and that's of utmost importance. It's really important to get that report back from a radiologist. However, you do need to see your own films and interpret them.

JAMES R. MILLER, MD: Even in a radiology group that services our MS center regularly, we still find things that they don't pay attention to because they're screening so many other diagnoses as well, that are important to us. So it is important to check the scans yourself to look for what you want to see in multiple sclerosis.

HOLLY ATKINSON, MD: We're almost out of time, but one last question. Future of MRI?

JAMES R. MILLER, MD: More sensitivity. More ability to look at things that we can't see now. Magnetic transfer is becoming a very important tool as well as diffusion imaging, and what we hope to do is to be able to define more what really is demyelination as opposed to just the big white blobs that we see, and we're beginning to get there, at least on an experimental basis. It's not yet ready for prime time, but these are things that we're looking for: more sensitivity, being able to pick up abnormalities in what we now think is normal white matter. We know they're there from the experiments. We just want to be able to do that clinically, as well. Spectroscopy is one of the tools that's being used quite extensively now for that.

HOLLY ATKINSON, MD: Well, thank you both for joining me today. George and Jim, nice to have you here.

JAMES R. MILLER, MD: You're very welcome.

GEORGE J. HUTTON, MD: Thank you.

HOLLY ATKINSON, MD: And thank you for joining me in this conversation.