Neuroradiology: Surveying Today's Field

Podcast Transcript

Introduction: Neuro Pathways, a Cleveland Clinic podcast exploring the latest research discoveries and clinical advances in the fields of neurology, neurosurgery, neuro rehab, and psychiatry.

Glen Stevens, DO, PhD:

Imaging is at the forefront of diagnosis, monitoring and management of almost every neurological disease. In this episode of Neuro Pathways, we're discussing the latest in the field of neuroradiology, including applications of the latest technologies and approaches across neurological subspecialties.I'm your host, Glen Stevens, neurologist/neuro-oncologist in Cleveland Clinic's Neurological Institute. I'm very pleased to have Dr. Doksu Moon join me for today's conversation. Dr. Moon is a neuro-radiologist in Cleveland Clinic's Department of Diagnostic Radiology. Doksu, welcome to Neuro Pathways.

Doksu Moon, MD:

Thank you, Glen.

Glen Stevens, DO, PhD:

You probably know this already, but I'm going to talk a little bit about CT scans and The Beatles, 'cause I think this is fascinating.We recently had some family visiting. We took them to the Rock & Roll Hall of Fame. They had a Beatles exhibit going on, which was great because they're real Beatles fans. Probably a day or two later, I listened to a talk from a neurosurgeon who was talking about the Beatles and the development of the CT scanner, which I hadn't known, but I'm sure that's 101 for you guys.

As you know, the Beatles recorded under a company called EMI. EMI was Electronics and Musical Industries. They were obviously on the musical side of things. In 1967, The Beatles generated 30% of the revenue for the company. They were generating almost $700 a second in revenue, at that point. The powers that be at EMI, were also investing in medical equipment research. They came across an individual who worked for them, Godfrey Hounsfield, which is a very famous name that you know quite well, who then ended up going forward and developed the first CT scanner, for which he won the Nobel Prize. Now EMI, from what I can tell, didn't reap all the rewards of the CT scanners that are everywhere today. They passed it along to GE and Siemens, who then built it up and then developed the MRI scanners. But I think it's just kind of fascinating, and I'm sure I've embellished it too much, but I think it's really cool that The Beatles were involved with the development of the CT. Is this common lore in the-

Doksu Moon, MD:

Yes. I remember hearing about that and doing a little bit of research. Apparently, back in the seventies, the Cleveland Clinic was the beneficiary of that. From what I understand, we had the second CAT scanner in the United States, at that time. It was produced by EMI. And then apparently, there was a big manufacturer here called Picker International. I think they've since merged with some other companies. But yeah, this was, believe it or not, a hotbed of CT imaging. It was a head-only scanner. I remember hearing that the initial scanners took somewhere like 15 or 30 minutes to scan one brain, without all the computerization that we have now. But yeah, I'm familiar with that story.

Glen Stevens, DO, PhD:

Yeah. I think it's just fascinating stuff. I really like the history of medicine. I always try and tell the residents and medical students, if you can find some interesting tidbits or facts about what you do, it'll just ingrain it and stick it in there and just make your job much more enjoyable. So, I think it's always a great way to go. I'll play more Beatles music now. That's where they probably play at your big meetings.

In a prior episode of a podcast that we did, and we hate to have you rehash somebody else's work, but we had had your friend Steve Jones on. Steve talked about the 7 Tesla MRI scan. So to start today's conversation, we thought we'd just revisit that again, just a little bit, because it's important information that's there. So can you enlighten our audience once again, on what the 7T is and how it's being used and why we might consider using a higher strength magnet?

Doksu Moon, MD:

Well, I think Steve spoke with you sometime at the end of 2020. Since then, we've actually had a 7 Tesla FDA-approved MR Scanner installed, as of November of 2021. We've been scanning patients on there, since that time. In fact, we split the day. In the mornings, we use it to scan clinical patients. In the afternoons, we use it for research. But most of the MRI scanners that are done nowadays, is somewhere between 1.0 Tesla and 3.0 Tesla, with a large number being 1.5 Tesla. The Tesla signifies the magnetic strength of the magnet. Basically, it shows you how much data you can pull out of it. So, the higher the magnetic field, the more data you have, the stronger signal that you have. So, you can usually split it up to different things. You can do it so that you can get higher resolution images. You can potentially do faster imaging, instead of something higher resolution. Or you can do some novel things you can't do with 1.5 or 3 Tesla. So that's where we are right now.

We've had a good half a year to scan patients. So, we've scanned a fair number of epilepsy patients here, since we're a big epilepsy center. We've been able to see abnormalities that are invisible, virtually invisible on 1.5 or 3 Tesla scanners. So, it's been very helpful in that way. When we have a seizure patient who comes here and their MRI may be non-lesional, as we say, or normal. Some of those patients, when we do it on the 7 Tesla, the abnormality there becomes very obvious. Or perhaps they had a small, questionable area, that looked borderline or almost normal. As Steve mentioned in your podcast, this 7 Tesla MRI is a very good problem-solving tool. It's not that useful as a screening tool, just because if you jump to that, you can miss some of the more obvious things, but as a problem-solving tool, it's proven to be quite excellent.

Glen Stevens, DO, PhD:

Now, will it take all my fillings out if I'm in a 7T or I'm okay there?

Doksu Moon, MD:

No, I think you're okay there. Yeah. The magnets only really affect things that are ferromagnetic, so iron or cobalt.

Glen Stevens, DO, PhD:

Can you do the 7T in the spine as well, or is it better for the brain? Is the signal to noise ratio such that it's...

Doksu Moon, MD:

The scanner we have is a whole body scanner, but the scanner isn't the only thing we need to scan. We also need coils, which are basically fancy antennas, to image certain parts of the body. Currently, we only have head coils for that MR Scanner. In the next six to nine months, we should be getting coils for the spine, which will let us do high resolution imaging of the spine, to look for MS plaques or [inaudible 00:07:21] disease. So we're looking forward to that, but as of now, we're just scanning the brain.

Glen Stevens, DO, PhD:

Mm-hmm. do you know the incidents of 7T Tesla use across the country? Is it there's a hundred scanners, there's 10 scanners?

Doksu Moon, MD:

I think there are about 30 scanners in the United States, currently, 30 7 Tesla MRI scanners. So, they're not very common. There's less than one per state. From what I understand from my colleagues in other institutions, most of them are being used majority of your time, for research, almost solely for research. So, we are unique in that we use this for clinical problem solving. Certainly, it's a very important tool for a tertiary medical center such as us, to help triage difficult patients.

Glen Stevens, DO, PhD:

I'm have no doubt that you read thousands and thousands and thousands of MRIs, not just in one day, but-

Doksu Moon, MD:

We're out to here. Yes.

Glen Stevens, DO, PhD:

Sometimes it feels like that in the day, but you become very comfortable with it. How difficult was it when you first started reading the 7 T? Is it really different, looking at it or all the things that you learned, generally, you just applied to it? Did you really see things differently, and you had to sort of retrain your brain of what to look for?

Doksu Moon, MD:

There is more detail. You get to see more detail, normal, nice things and more detail in background noise things. So, it does take a while to retrain, to look at them. You probably need to look at around a dozen or so, with a correlated 1.5 or 3 Tesla magnet, to get used to it. Also, the 7 Tesla magnet does have certain limitations. It's not very good when you evaluate brain next to the skull, for instance, because there's a lot of susceptibility artifact. But for most other uses, it's certainly better.

Glen Stevens, DO, PhD:

Can you do MR spectroscopy with a 7 T or no?

Doksu Moon, MD:

We can, but MR spectroscopy's fallen out of favor just because in the past, it hasn't proven for clinical use, to be that fruitful. So, I don't think we do that many clinical MR spectroscopy scans at the clinic. We do approximately 3,000, I think more like a 1,200 neuro MR scans per week. In a month, I think maybe we do two spectroscopy scans. So yeah, they're a minority.

Glen Stevens, DO, PhD:

I know that we're a big multiple sclerosis center here. The 7 T with multiple sclerosis, any new insights there?

Doksu Moon, MD:

Yes. Multiple sclerosis, imaging-wise is very nonspecific, as you know. So, it looks like a conglomeration of white matter, lesions, which can be caused by dozens of different things, including small vessel ischemic changes. With the 7 Tesla MR scanner, when we have the ultrastructure of the anatomy, we can use it to determine non-specific or borderline patients, using the central vein sign. So, it's been very helpful in that way. About once a week or once every other week, we'll have a problematic patient get scanned on the 7 Tesla to evaluate, does this person have MS or not?

Glen Stevens, DO, PhD:

I'm just curious, and you may not know the answer to this, but we start to hear more and more that MS is not just a white matter, but also a gray matter disease.

Doksu Moon, MD:

Yes.

Glen Stevens, DO, PhD:

Can you see the gray matter disease or is it too gross a level?

Doksu Moon, MD:

If you have the right surface coils and you take the time, you can see the plaques. Even on 3 Tesla, you can see it in the gray matter. One thing people forget is that, yeah, we separate brain into white matter and gray matter, but in the gray matter, there's interconnections, which are made by white matter. Certainly, we see gray matter structures, like the thalamus, be affected in cases like ADEM. So yeah, it's more of a research tool right now, but you can see demyelination of the gray matter cortex.

Glen Stevens, DO, PhD:

Mm-hmm. So let's move away from the 7 T to Functional MRI.

Doksu Moon, MD:

Yes.

Glen Stevens, DO, PhD:

I remember when I was training, there was always a big issue in handedness, especially with epilepsy surgery. We used to do Wadas on everybody. Everybody get a Wada, where you essentially freeze one of the hemispheres and determine where their language is. It seems like the functional MRI, one of the utilities is really to replace the Wada. But you can tell me, are we still doing Wadas on patients?

Doksu Moon, MD:

No, we still do Wadas. They're not nearly as common. And as you referred to, the Wada test is an invasive procedure. Basically, we are doing a cerebral angiogram, and while we're doing it, we anesthetize half the brain and epileptologists do neurologic testing. When I was in training, we used to do four or five of those every week. You probably remember back then.

Glen Stevens, DO, PhD:

Yeah.

Doksu Moon, MD:

Now, from what I understand, we only do one or two a month, in the most problematic patients. So the Functional MRI has pretty much supplanted that. The only relative negative of FMRI is that it doesn't let us figure out which side where memory resides. We don't have a good paradigm for that. But in terms of language localization, it's much better than Wada ever was, because with the Wada test, you put half the cerebral hemisphere to sleep and then you do the test. So, you can tell which side is dominant, or if they're co-dominant. But on the FMRI, we can actually localize it to the area of the brain. We can see what part of the brain is active during the paradigms, which portion is active in controlling language. So even if it's not the same size, we can tell how far away it is from a lesion. For instance, we had a young lady, a 16-year-old who had a subtle cortical malformation or the dorsal right frontal lobe. So, she came here for language localization, to figure out whether it was on the left or right side. On most people, it's on the left side. But when we tested her, we saw that her language was dominant on the right side. But we had enough information to show that the lesion that she had, was about three centimeters away from her speech center, from her frontal curriculum. So that gave enough margin so that the surgeons could think about treating her without an open craniotomy or awake craniotomy.

Glen Stevens, DO, PhD:

I'm trying to remember a name from the past. You may or may not remember the research with it, but Jeff Ross, I remember, was doing something with golf imagery with FMRI.

Doksu Moon, MD:

Yes.

Glen Stevens, DO, PhD:

I think they were looking at what area of the brain was maybe active, or could you improve your golf game by mentally going through it? Do you remember what that looked-

Doksu Moon, MD:

I vaguely remember. Jeff is one of my mentor. I was one of his trainees when he was doing that research. But basically, I think he was evaluating golfers and when they mis-swing or get their yips. I think basically, he was evaluating golf dyspraxia with that. I don't know if he ever carried it further, but it was fascinating research with the FMRI. Yes.

Glen Stevens, DO, PhD:

Well, certainly we use the FMRI in the tumor field, for helping us understand where different areas are in the brain for tumors, for resection. Other areas they're using the FMRI for?

Doksu Moon, MD:

It's basically epilepsy and tumor. In terms of clinical FMRI, we're a very large center. We do approximately a hundred patients a year. So, 100 patients a year. I think it's fairly evenly split between the epilepsy patients and the brain tumor patients. So with the brain tumor patients, obviously it helps the surgeons know if an eloquent area is next to or right on top of the tumor, or if brain plasticity has had enough time to move those areas, to somewhere adjacent or maybe a little bit further away.

Glen Stevens, DO, PhD:

Do you see that?

Doksu Moon, MD:

We see that, not as much as we'd like. But the FMRI is very helpful in localizing where speech is or where hand motor, leg motor is, in relationship to the tumor. So, it is very useful for us and for the surgeons.

Glen Stevens, DO, PhD:

One of the other tools that a lot of the surgeons in our group will use is DTI, or Diffusion Tensor Imaging. Can you just tell us a little bit about that?

Doksu Moon, MD:

Yeah. Diffusion Tensor Imaging is basically a way to look at white matter tracks in the brain. It's using diffusion information. You need a fairly powerful magnet and the right software packages and computers to process it, but it gives a pretty good idea of whether a tumor or tumor edema affects any sort of important tracks, like the corticospinal tracts or maybe the arcuate fasciculus, if it's related to speech. So, it gives a very good guide, again, to the surgeons when they're planning their surgeries.

Glen Stevens, DO, PhD:

Mm-hmm. Now, I know that we don't endear ourselves to the neuro-radiologist when we're always asking you to do a perfusion image, along with the MRI scan. Certainly, it has applications outside of tumor, as well. But what types of things can we use with perfusion imaging, blood flow, those types of things?

Doksu Moon, MD:

Generally, for the most part, we use it in an MRI. That is to evaluate, to see if there's increased vascularity, which suggest that something's a high-grade neoplasm. We get that on basically anyone with an intermediate or high-grade tumor, so the vast majority of our patients, as you know. So we do it very frequently. On a busy tumor day, for instance, we'll do a dozen perfusion study. So we're pretty experienced at reading those. We also use perfusion for stroke imaging. There are different parameters that we evaluate for. Brain tumors, we use CBV. For stroke, we use time to peak or mean transit time, but it'll give us a good idea of how much ischemic penumbra is or isn't. So, it'll help the vascular neurosurgeons or endovascular neurologist figure out whether the patient is appropriate to have thrombolysis or not.

Glen Stevens, DO, PhD:

Occasionally, I'll see a patient from Brain Health, that has meningioma or something. They always have different MRIs when they come from Brain Health. They'll do volumetric MRIs, looking for... Now, tell me a little bit about that, so I can understand that a little bit more. I think they're looking at brain atrophy?

Doksu Moon, MD:

Yes. We're in the phase of getting a lot of data. So especially in patients who are older, who might have memory loss, we do a fair amount of volumetric imaging, that we've started about two years ago.

So the majority of patients with dementia or mild cognitive impairment, can get an MRI scanner with volumetrics. On that, we can evaluate the patient's brain parenchyma, just probably about two dozen different metrics, including the frontal lobes. How much volume there is, in relationship to a patient of their age. Hippocampal formation, whether there's volume loss, a comparative person of that cohort. The hippocampus is very small. It's a centimeter, centimeter and a half. To actually try to measure it out, we tend to be very inaccurate. Even eyeballing it on patients, just because the amount of volume loss increases, even the most seasoned of us are not very accurate. But using the volumetric quantitative MRI, we can have a very good idea of if a patient's or a person's hippocampal formation are the 13th percentile for their age or smaller or larger, which gives us a very good idea of how much atrophy they may have.

Glen Stevens, DO, PhD:

Does that use AI, artificial intelligence, or you guys are putting on the parameters and the software then calculates it?

Doksu Moon, MD:

No, it's a sophisticated computer program that segments out those portions of the brain. So, that is computerized at this point. But in terms of making the final call, we still have a little bit of input into that, as doctors. Maybe sometime in 2035, the computer will do that for us.

Glen Stevens, DO, PhD:

We still need you guys around. I'm not worried about that.

Doksu Moon, MD:

Well, that's good to hear, Glen.

Glen Stevens, DO, PhD:

Anything new in vasculitis?

Doksu Moon, MD:

Yes. In vasculitis, there's been a big push to inflammatory disorders of the brain, so unknown encephalities and also vasculitis. We've started to do a fair amount of intracranial vascular imaging. We've done a few with the 7 Tesla, but the bulk of it, we do with conventional 3.0 Tesla magnets. We can see things such as wall thickening, and we can see enhancement in the vessel wall. These are things that we didn't really know existed even. It happened like 10 years ago, but now we can get a very good idea of whether someone is suffering from atherosclerotic disease or some sort of vasculitis that needs to be worked up.

Glen Stevens, DO, PhD:

I always like to use these as times for personal knowledge development. So, I'm going to ask you some other types of questions here. Talk to me a little bit about evolution of gadolinium over time, because I know that we've changed the type of gadolinium that we've used over time, to allow it to get excreted more in the kidneys. I remember we used to really monitor the kidney function quite a bit, but now not so necessary. Right?

Doksu Moon, MD:

Yeah. Gadolinium, well, for the longest time was thought to be very benign. For the most part, it is, except for patients who have issues with renal insufficiency. In those patients, a very, very small fraction can get something called nephrogenic systemic fibrosis. Okay. There's a database of that. There's only about, I think 450 or 500 patients who've ever gotten that, despite the hundreds of millions of MRIs with contrast, that have been done throughout the years. So, it doesn't have much, in terms of renal function. But with MRI contrast, we've actually changed contrast vendors, just because there's linear and cyclic agents. The linear agents are the ones that are associated with nephrogenic systemic fibrosis. The cyclic agents, because they bind the gadolinium tighter, are not. So, we're currently using a cyclic contrast agent, gadolinium-based agent, gadolinium-based contrast. One other thing about gadolinium that we've realized is that, despite what contrast you use, that there is some tissue deposition over time. It's relatively subtle. So, you can notice in patients who've gotten 50 or maybe a hundred MRIs with contrast, but for the most part, it doesn't really affect anything. These are just very trace amounts in the tissues.

Glen Stevens, DO, PhD:

Yeah. Patients do ask me about that, occasionally. I think that maybe there'll be something written about it and a patient will say, "I've had MRIs for 10 years." Sometimes you can see some subtle changes on the T1 images on those patients. I always tell them that, at this point, we don't know of any... you don't need to be chelated or have anything done. We don't know of any specific disease, although we may be smarter down the road, with that.

Doksu Moon, MD:

Yes. So far, the data shows that there's no harmful effects, despite repeated administration of gadolinium, just like many other medical agents, contrast agents.

Glen Stevens, DO, PhD:

I guess one commercial out there is that, one of the other things that we run into problems with, are pacemakers and other types of devices that are in there. Now, I don't know the percentage of pacemaker that are placed, that are MRI compatible. But I guess my commercial is that, consider looking at MRI-compatible pacemakers for patients.

Doksu Moon, MD:

Yeah. Yeah. We use the term conditional, instead of-

Glen Stevens, DO, PhD:

Oh, okay.

Doksu Moon, MD:

If you take any sort of electronic device, it may be working fine at 1 Tesla, but at 3 Tesla, it may not. And at 7 Tesla, it definitely may not. So, we use the word conditional, but I believe just about all of the pacemaker and AICDs being implanted now, are MR-conditionally safe.

Doksu Moon, MD:

The problem comes in with patients who've had theirs implanted 15, 20 years ago. Where the thoughts of getting MRIs, for a lot of instances, wasn't really on the horizon. We have a very good process in which we can evaluate all the patients, whether they have conditional MRIs or not. We do this in a very reasonable, safe way. And actually, we have dedicated time slots to do these patients at the main campus and at Fairview Hospital and Hillcrest. So depending on the individual's history, we reevaluate it. And if we need to get someone from the company to be there to reset the pacemaker after the MRI scanner, we do arrange all that.

Glen Stevens, DO, PhD:

When I started in medicine, an MRI was an MRI. Now in the brain tumor area, we're using MRIs to do laser interstitial thermal therapy.

Doksu Moon, MD:

Yes.

Glen Stevens, DO, PhD:

Never would've thought of that, many years ago, where we can use the MRI to monitor realtime thermography as we're... It's unbelievable.

Doksu Moon, MD:

It is.

Glen Stevens, DO, PhD:

I've been quite involved with the focused ultrasound. As you know, we had Dr. Nagel and Dr. Lockwood, one of your colleagues on previously, talking about high intensity focused ultrasound, where we're again doing thermal ablations. We're doing low intensity focused ultrasound. Again, all done in an MRI machine.

Doksu Moon, MD:

For blood-brain barrier breakdown.

Glen Stevens, DO, PhD:

With blood barrier breakdown. I mean, it's just untapped.

Doksu Moon, MD:

Yeah.

Glen Stevens, DO, PhD:

Well, Doksu, this has been a great educational event. At least I got to ask the questions that I wanted to get some answers to, although I see you guys every day, all day and you guys are just fantastic. I don't think you're going to get replaced by AI, at least while I'm still working. But remember, I'm old, as things go through. Thanks for joining me today and continued enjoyment of your career, as you move forward in this exciting field.

Doksu Moon, MD:

Thank you very much for having me, Glen.

Conclusion: This concludes this episode of Neuro Pathways. You can find additional podcast episodes on our website, clevelandclinic.org/neuropodcast, or subscribe to the podcast on iTunes, Google Play, Spotify, or wherever you get your podcasts. And don't forget, you can access real-time updates from experts in Cleveland Clinic's Neurological Institute on our Consult QD website. That's consultqd.clevelandclinic.org/neuro, or follow us on Twitter @CleClinicMD, all one word. And thank you for listening.