Navigating MRI Procedures in Shunt Patients with NPH

Normal pressure hydrocephalus (NPH) is a pathologic condition whereby excess CSF is retained in and around the brain despite normal intracranial pressure. MRI-safe programmable shunt valves allow for fluid drainage adjustment based on patients’ symptoms and radiographic images. Approximately 75% of patients with NPH improve after shunt surgery regardless of shunt type or location.

In this episode, Aaron Berkowitz, MD, PhD, FAAN, speaks with Kaisorn L. Chaichana, MD, author of the article “Management of Normal Pressure Hydrocephalus” in the Continuum® June 2025 Disorders of CSF Dynamics issue.

Dr. Berkowitz is a Continuum® Audio interviewer and a professor of neurology at the University of California San Francisco in the Department of Neurology in San Francisco, California.

Dr. Chaichana is a professor of neurology in the department of neurological surgery at the Mayo Clinic in Jacksonville, Florida.

Additional Resources

Read the article: Management of Normal Pressure Hydrocephalus

Subscribe to Continuum®: shop.lww.com/Continuum

Earn CME (available only to AAN members): continpub.com/AudioCME

Continuum® Aloud (verbatim audio-book style recordings of articles available only to Continuum® subscribers): continpub.com/Aloud

More about the American Academy of Neurology: aan.com

Social Media

facebook.com/continuumcme

@ContinuumAAN

Host: @LyellJ

Guest: @kchaichanamd

Full episode transcript available here

Dr Jones: This is Dr Lyell Jones, Editor-in-Chief of Continuum. Thank you for listening to Continuum Audio. Be sure to visit the links in the episode notes for information about earning CME, subscribing to the journal, and exclusive access to interviews not featured on the podcast.

Dr Berkowitz: This is Dr Aaron Berkowitz, and today I'm interviewing Dr Kaisorn Chaichana about his article on management of normal pressure hydrocephalus, which he wrote with Dr Jeremy Cutsforth-Gregory. The article appears in the June 2025 Continuum issue on disorders of CSF dynamics. Welcome to the podcast, and please introduce yourself to our audience.

Dr Chaichana: Yeah, thank you for having me. I'm Kaisorn Chaichana. I'm a neurosurgeon at Mayo Clinic in Jacksonville, Florida. Part of my practice is doing hydrocephalus care, which includes shunts for patients with normal pressure hydrocephalus.

Dr Berkowitz: Fantastic. Well, before we get into shunt considerations and NPH specifically, which I know is the focus of your article, I thought it would be a great opportunity for a neurologist to pick a neurosurgeon's brain a bit about shunts. So, to start, can you lay out for us the different types of shunts and shunt procedures, the advantages, disadvantages of each type of shunt, how you think about which shunt procedure should be used for which patient, that type of thing?

Dr Chaichana: Yeah. So, there are different types of shunts, and the most common one that is used is called a ventricular peritoneal shunt. So, it has a ventricular catheter, it has a catheter that tunnels underneath the skin and it goes into the peritoneum where the fluid goes from the ventricular system into the peritoneum. Typically, the shunts are in the ventricle because that is the largest fluid-filled space in the brain. Other terminal areas include the atrium, which is really the jugular vein, and those are called ventricular atrial shunts. You can also have ventricular pleural shunts, which end in the pleural space and drain flui into the pleural space. Those are pretty much the most common ventricular shunts. There's also a lumboperitoneal shunt that drains from the lumbar spine, similar to a lumbar drain into the peritoneum. For the lumbar shunts, we don't typically have a lumbar pleural or lumbar atrial shunt just because of the pressure dynamics, because the lumbar spine is below the lung and as well as the atrium. And so, the drainage pattern is very different than ventricular peritoneal which is top to bottom. The most common shunt, why we use the ventricular peritoneal shunt the most, is because it has the most control. So, the peritoneum is set at a standard pressure in the intraabdominal pressure, whereas the ventricular atrial shunt depends on your venous return or venous pressure and your ventricular pleural shunt varies with inspiration and expiration. So, the easiest way for us to control the fluid, the ventricular system is through the ventricular peritoneal shunt. And that's why that's our most common shunt that we use.

Dr Berkowitz: Fantastic. So, as you mention in the article, neurologists may be reluctant to offer a shunt to patients with NPH because many patients may not improve, or they improve only transiently; and out of fear of shunt complications. So, of course, as neurologists, we often only hear about a patient’s shunt when there is a problem. So, we have this sort of biased view of seeing a lot of shunt malfunction and shunt infection. Of course, we might not see the patient if their shunt is working just fine. How common are these complications in practice, and how do you as a neurosurgeon weigh the risks against the often uncertain or transient benefits of a shunt in a patient with NPH who may be older and multiple medical comorbidities? How do you think about that and talk about it with patients?

Dr Chaichana: When you hear about shunt complications, most of the shunt complications you hear about are typically in patients with congenital hydrocephalus. Those patients often require several shunt revisions just from either growing or the shunt stays in for a long time or the ventricular caliber is a lot less than some with normal pressure hydrocephalus. So, we don't really see a lot of complications with normal pressure hydrocephalus. So that shunt placement in these patients is typically pretty safe. The procedure’s a relatively short procedure, around 30 minutes to 45 minutes to place a shunt, and we can control the pressure within the shunt setting so that we don't overdrain---which means too much fluid drains from the ventricular system---which can cause things like a subdural, which is probably the most common complication associated with normal pressure hydrocephalus. So, to obviate those risks, what we do is typically insert the shunt and then keep the shunt setting at a high setting. The higher the setting, the less it drains, and then we bring it slowly down based on the patient's symptoms to try to minimize the risk of this over drainage in the subdural hematoma while at the same time benefiting the patient. So, there's a concern for shunt in patients with normal pressure hydrocephalus. The concern or the complication risks are very low. The problem with normal pressure hydrocephalus, though, is that over time these patients benefit less and less from drainage or their disease process takes over. So, I do recommend placing this shunt as soon as possible just so that we can maximize their quality of life for that period of time.

Dr Berkowitz: So, if I'm understanding you, then the risk of complication is more sort of due to the mechanical factors in patients with congenital hydrocephalus or sort of outgrowing the shunt, their pressure dynamics may be changing over time. And in your experience, an older patient with NPH, although they may have more medical comorbidities, the procedure itself is relatively quick and low-risk. And the actual complications due to mechanical factors, my understanding, are just much less common because the patient is obviously fully grown and they're getting one sort of procedure at one point in time and tend to need less revision, have less complication. Is that right?

Dr Chaichana: Yeah, that's correct. The complication risk for normal hydrocephalus is a lot less than other types of hydrocephalus.

Dr Berkowitz: That's helpful to know. While we're talking about some of these complications, let's say we're following a patient in neurology with NPH who has a shunt. What are some of the symptoms and signs of shunt malfunction or shunt infection? And what are the best studies to order to evaluate for these if we're concerned about them?

Dr Chaichana: Yeah. So basically, for shunt malfunction, it’s basically broken down into two categories. It's either overdrainage or underdrainage. So, underdrainage is where the shunt doesn't function enough. And so basically, they return to their state before the shunt was placed. So that could be worsening gait function, memory function, urinary incontinence are the typical symptoms we look for in patients with normal pressure hydrocephalus and underdrainage, or the shunt is not working. For patients that are having overdrainage, which is draining too much, the classic sign is typically headaches when they stand up. And the reason behind that is when there's overdrainage, there's less cerebrospinal fluid in their ventricular system, which means less intracranial pressure. So that when they stand up, the pressure differential between their head and the ground is more than when they're lying down. And because of that pressure differential, they usually have worsening headaches when standing up or sitting up. The other thing are severe headaches, which would be a sign of a subdural hematoma or focality in their neurological symptoms that could point to a subdural hematoma, such as weakness, numbness, speaking problems, depending on the hemisphere.

How we work this up is, regardless if you're concerned about overdrainage or underdrainage, we usually start with a CAT scan or an MRI scan. Typically, we prefer a CAT scan because it's quicker, but the CAT scan will show us if the ventricular caliber is the same and/or the placement of the proximal catheter. So, what we look for when we see that CAT scan or that MRI to see the location of the proximal catheter to make sure it hasn't changed from any previous settings. And then we see the caliber of the ventricles. If the caliber of the ventricles is smaller, that could be a sign of overdrainage. If the caliber of the ventricles are larger, it could be a sign of underdrainage. The other thing we look for are subdural fluid collections or hydromas or subdural hematomas, which would be another sign of lower endocranial pressure, which would be a sign of overdrainage. So those are the biggest signs we look for, for underdrainage and overdrainage. Other things we can look for if we're concerned of the shunt is fractured, we do a shunt X-ray and what a shunt x-ray is is x-rays of the skull, the neck and the abdomen to see the catheter to make sure it's not kinked or fractured. If you're really concerned, you can't tell from the x-ray, another scan to order is a CT of the chest and abdomen and pelvis to look at the location of the catheter to make sure there's no brakes in the catheter, there's no fluid collections on the distal portion of the catheter, which would be a sign of shunt malfunction as well.

Other tests that you can do to really exclude shunt malfunction is a shunt patency test, and what that is a nuclear medicine test where radionucleotide is injected into the valve and then the radionucleotide is traced over time or imaged through time to make sure that it's draining appropriately from the valve into the distal catheter into the peritoneum or the distal site. If there's a shunt malfunction that's not drainage, that radioisotope would remain stagnant either in the valve or in the catheter. There's overdrainage, we can't really tell, but there will be a quick drainage of the radioisotope. For shunt infection, we start with an imaging just to make sure there's not a shunt malfunction, and that usually requires cerebrospinal fluid to test. The cerebrospinal fluid can come from the valve itself, or it can come from other areas like the lumbar spine. If the lumbar spine is showing signs of shunt infection, then that usually means the shunt is infected. If the valve is aspirated with- at the bedside with a butterfly needle into the valve and that shows signs of shunt infection, that also could be a sign of infection.

Dr Berkowitz: That's very helpful. You mentioned CT and shunt series. One question that often comes up when obtaining neuroimaging in patients with a shunt, who have NPH or otherwise, is whether we need to call you when we're doing an MRI to reprogram the shunt before or after. Is there a way we can know as a neurologists at the bedside or as patients carry a card, like with some devices where we know whether we have to call and bother our neurosurgery colleagues to get this MRI? Or if the radiology techs ask us, is this safe? And is the patient's shunt going to get turned off? How do we go about determining this?

Dr Chaichana: Yeah, so unfortunately, a lot of patients don't carry a card. We typically offer a card when we do the shunt, but that card, there's two problems with it. One is it tells the model, but the second thing is it has to be updated any time the shunt is changed to a different setting. Oftentimes patients don't know that shunt setting, and often times they don't know that company brand that they use. There are different types of shunts with different types of settings. If there's ever concern as to what type of shunt they have, an x-ray is usually the best bet to see with a shunt series, or a skull x-ray. A lateral skull x-ray usually looks at the valve, and the valve has certain radio-dense markers that indicate what type of shunt it is. And that way you can call neurosurgery and we can always tell you what the shunt setting is before the MRI is done. Problem with an MRI scan if you do it without a shunt x-ray before is that you don't know the setting before unless the patient really knows or it's in the patient chart, and the MRI can need to change the setting. It doesn't usually turn it off, but it would change the setting, which would change the fluid dynamics within their ventricular system, which could lead to overdrainage or underdrainage. So, any time a patient needs MRI imaging, whether it's even the brain MRI, a spine MRI, or even abdominal MRI, really a shunt x-ray should be done just to see the shunt setting so that it could be returned to that setting after the MRI is done.

Dr Berkowitz: So, the only way to know sort of what type of shunt it would be short of the patient knowing or the patient getting care at the same hospital where the shunt was placed and looking it up in the operative reports would be a skull film. That would then tell us what type of shunt is there and then the marking of the setting. And then we would be able to call our colleagues in neurosurgery and say, this patient is getting an MRI this is the setting, this is the type of shunt. And do we need to call you afterwards to come by and reprogram it? Is that right?

Dr Chaichana: That's correct, yeah.

Dr Berkowitz: Is there anything we would be able to see on there, or it's best we just- best we just call you and clarify?

Dr Chaichana: The easiest thing to do is, when you get the skull x-ray, you can Google different types of shunts or search for different shunts, and they'll have markers that show the type of shunt it is as well as the setting that it's at. And just match it up with the picture.

Dr Berkowitz: And as long as it's not a programmable shunt, there's no concern about doing the MRI. Is that right?

Dr Chaichana: Correct. So, if it's a programmable shunt, even if it's MRI-compatible, we still like to get the setting before and make sure the setting after the MRI is the same. Nonprogrammable shunts can't be changed with MRI scans, and those don't need neurosurgery after the MRI scan, but it should be confirmed before the scan is done.

Dr Berkowitz: Very helpful. Okay, so let's turn to NPH specifically. As you know, there's a lot of debate in the literature, some arguing, even, NPH might not even exist, some saying it's underdiagnosed. I think. I don't know if it was last year at our American Academy of Neurology conference or certainly in recent years, there was a pro and con debate of “we are underdiagnosing NPH” versus “we are overdiagnosing NPH.” What's your perspective as a neurosurgeon? What's the perspective in neurosurgery? Is this something we're underdiagnosing, and the times you shunt these patients you see miraculous results? Is this something that we're overdiagnosing, you get a lot of patients sent to that you think maybe won't benefit from a shunt? Or is it just really hard to say and some patients have shunt-responsive noncommunicating hydrocephalus of unclear etiology and either concurrent Parkinson's disease, Alzheimer's, cervical lumbar stenosis, neuropathy, vestibular problems, and all these other issues that play into multifactorial gait to sort of display a certain amount of the percentage of problem in a given patient or take overtime? What's your perspective if you're open to sharing it, or what's the perspective of neurosurgery? Is this debated as it is in neurology or this is just a standard thing you see and patients respond to shunt to some degree in some proportion of the time? And what are the sort of predictors you see in your experience?

Dr Chaichana: Yeah, so, for me, I'd say it's too complicated for a neurosurgeon to evaluate. We rely on neurology to tell us whether or not they need a shunt. But I think the problem is, obviously, a part of the workout for at least the ones that I like to do, is that I want them to have a high-volume lumbar puncture with pre- and postgait analysis to see if there's really an objective measure of them improving. If they have an objective measure of improvement---and what's even better is that they have a subjective measure of improvement on top of the objective measure of improvement---then they benefit from a shunt. The problem is, some patients do benefit even though they don't have objective performance increases after a high-volume shunt. And those are the ones that make me the most worrisome to do the shunt, just because I don't like to do a procedure where there's no benefit for the patient. I do see, according to the literature as well, that there's around a 30 to 40%, even 50%, increase in gait function, even in patients that don't have large improvements following the high-volume lumbar puncture. And those are the most challenging patients for us as neurosurgeons because we'll put the shunt in, they say we're no better in terms of their gait, no better in terms of their urinary incontinence. We try to lower their shunt down to a certain setting and we're kind of stuck after that point. The good thing about NPH, though, is that, from the neurosurgery side, the shunt, like I said, is a pretty benign, low-risk procedure. So, we're not putting the patient through a very severe procedure to see if there's any benefit. So, in cases where we try to improve their quality of life in patients that don't have a benefit from high-volume lumbar puncture, we give them the odds of whether or not it's improving and say it might not improve. But because the procedure’s minimally invasive, I think it's a good way to see if we can benefit their quality of life.

Dr Berkowitz: Yeah, it's a very helpful perspective. Yeah, those are the most challenging cases on our side as well, right. If the patient- we think they may have NPH, or their gait and/or urinary and/or cognitive problems are- at least have a component of NPH that could be reversible, we certainly want to do the large volume lumbar puncture and/or consider a lumbar drain trial, all discussed in other articles and interviews for this issue of Continuum, But the really tough ones, as you said, there is this literature on patients who don't respond to the large-volume lumbar puncture for some reason but still may be shunt responsive. And despite all the imaging predictors and all the other ways we try to think about this, it's hard to know who's going to benefit. I think that's really a helpful perspective from your end that, as you say in the very beginning of your article, right, maybe there's a little bit too much fear of shunting on the neurology side because when we hear about shunts, it's often in the setting of complication. And so, we're not sort of getting the full spectrum of all the patients you shunt and you see who are doing just fine. They might not improve---the question is related to NPH---but at least they're not harmed by the shunt, and we're maybe overbiased and/or seeing a overly representative sample of negative shunt outcomes when they're actually not that common in practice. Is that a fair summary of your perspective?

Dr Chaichana: Yeah, that's correct. So, I mean, complications can occur---and anytime you do a surgery, there are risks of complications---but I think they're relatively low for the benefit that we can help their quality of life. And the procedure's pretty short. So, the risk, it mostly outweighs the benefits in cases with normal pressure hydrocephalus.

Dr Berkowitz: Very helpful perspective. So, well, thanks so much again. Today I've been interviewing Dr Kaisorn Chaichana about his article on management of normal pressure hydrocephalus, which he wrote with Dr Jeremy Cutsforth-Gregory. This article appears in the most recent issue of Continuum on disorders of CSF dynamics. Be sure to check out Continuum Audio episodes from this and other issues, and thank you to our listeners for joining us today.

Dr Monteith: This is Dr Teshamae Monteith, Associate Editor of Continuum Audio. If you've enjoyed this episode, you'll love the journal, which is full of in-depth and clinically relevant information important for neurology practitioners. Use the link in the episode notes to learn more and subscribe. AAN members, you can get CME for listening to this interview by completing the evaluation at continpub.com/audioCME. Thank you for listening to Continuum Audio.