Podcast Transcript: Diagnosis & Treatment of Tarlov Cysts

[Dr. Fleming]
You covered so much there, and so much to unpack with the significance of the Tarlov cysts. A difficult population and often their imaging is read as essentially negative, and perhaps the only thing that's mentioned out of the ordinary is Tarlov cyst, so a lot of these patients somewhat naturally latch on to, "This has to be the cause for my symptoms." As you said, it is some of the time. Tell us about how you tease that out a patient who's come to you and said, "I have a Tarlov cyst?" How do you tease out, what are the symptoms? Is the cyst causing it? How do we go from there?

[Dr. Murphy]
First of all, I get all my patients - their GP has to send the referral, their family doctor. I have to get a formal referral from that person, not the person referring themselves. I practice now in Canada, in Toronto, where we have a socialized system. My salary doesn't change whether I treat 1 or 200. We get the referral, then I review their imaging, then I have a consultation with them. Reviewing outside imaging in itself is a pain because you've got to figure out how to read that CD or how to download that thing from Public Health or whatever. Then, we talk.

I divide the patients into those who have pain from something else and those who may have pain from their cystic dilatation of distal or caudal sacral nerve roots. I look for other causes of low back pain, radicular pain. There will be a handful of people who have a congenital anomaly. They will have some ligamentous laxity, and they will have perineural cysts all the way down. On sagittal T2, I'll look at the C-spine, T-spine, C of each little neuroforaminal has a light bulb in it, some dilated perineural cyst.

[Dr. Fleming]
Neurolactasia, yes.

[Dr. Murphy]
Then you know already, there's something more systemic going on here. Then I look in the sacrum for degenerative disease, for set joint disease in particular. Many of these people cannot sit. I look anterior to the sacrum for fibroids, ovarian masses. It's amazing how many have been missed, or piriformis syndrome, or ischial vine tuberosities, or irregularities, or spicules of bone or other things, and prior trauma to the sacrum, where somebody might have fallen out of a tree when they were a kid, but now at 52, they've got an abrupt angulation at S34 and they've got pain from that.

The funny things happen with that. Somebody goes on a Zempic, and previously they had a little cushion, and now the cushion is gone, and that fracture is painful. You have to go through step by step to figure these things out. In terms of the anatomy of the sacral abnormality, because it's usually S234, Tarlov cysts are a particular narrow-necked entity with minimal communication with the subarachnoid space around the distal caudal nerve roots. There are meningeal diverticula, which are wide-necked communications which are really form fruits of sacral agenesis or digenesis.
You'll have these big dilated distal, sacral, dural, CSF-filled sacs that directly communicate with the nerve roots around the distal spinal canal. Sometimes you'll see a lipoma of the phylum terminale going into them, and you do not treat those. They were built like this. It's the way that person is built. Their pain may be from something else. There will be people with erodent loss or Marfan's or Loeys-Dietz who have enormous duralectasias, and sadly, there is nothing you can do for them. That's a case of waiting for reincarnation. In those patients, sometimes I see, I'm thinking of two or three in particular.

Particularly one young lady who had tattoos everywhere, really quite terrifying to see, and she was the nicest, kindest person. She was more tattoos than surface area, almost, but when she lifted a heavy object, she would Valsalva. This would push fluid into her presacral dilatation, giving her transient intracranial hypotension. Then, it would recover over the next half hour or so. She eventually died. Some of these people will commit suicide because they have chronic, untreatable, unmanageable pain and no hope. The risk of suicide in this group overall is quite high.

A lot of these folks they have this underestimated, misunderstood, chronic pain at S234, in their pelvis, in their vaginas, in their perineums, down their legs. They've been told they're not symptomatic by orthopedic surgeons, spine surgeons, that this isn't a thing. They've been put on horrible drugs by spine guys who just want to put them on Lyrica and Gabapentin, which are cynical drugs because they do nothing for the underlying condition. They're anticonvulsants that were designed for epilepsy, that Park Davis figured out could be sold for other things even more. They're shit drugs. They're so over-prescribed.

They get depressed. If they're in some place where they don't have a good family doctor and they've no one to communicate and they're isolated, they will kill themselves. It's not uncommon. I hear this once or twice a year. You have to take these things seriously. They're existential for some patients. Now, the ones that I treat, I can define as the cause of the symptoms. They will be narrow-necked. Usually, they've eroded bone around them. The signal on T2 is going to be higher in the cyst than the adjacent subarachnoid space because it's not recessing. It's not moving. You've got static CSF. It'll be slightly brighter on T2.

The bone erosion is fascinating because you're talking about CSF pressure in a cyst and pulsating with each cardiac output, because the brain's a pump. This erodes cortical bone over time. You're talking about millimeters of water causing bone erosion. That's really unusual. Normally, you're talking about mercury and sphygmomanometers and that level of pressure, not millimeters of water. The bone erodes, I think, the distended dura around the cyst, is the cause of the pain because it's referred from the dural innervation of the cyst. Often, when I pass my needle through that dura, they reproduce their pain. That's really reassuring.

If I'm not sure, I will do nerve root blocks around the nerve that is compressed. Usually, this is S2. The S2 adjacent nerve that's compressed can be flattened and ribbon-like between the cyst and the adjacent bone. When you deflate the cyst and treat the cyst, you see that formerly ribbon-like nerve becomes plump again and slightly oval. The symptoms go away. As the cyst retracts and shrinks, that nerve will re-expand. I firmly believe that these patients, I treat them a bit like gardening. You don't finish your garden in one day. If you try and do everything at the one time, you can hurt people.
Our job is not to make people worse. I will stage things if I have to. Aspirate the cyst, put in half or one cc of 1% lidocaine to see if there's symptoms. If that works, then I use that two-needle technique that I described, which is really simple.

[Dr. Fleming]:
Yes. I wanted to hear you talk about it. I think this is one of the compelling things.

[Dr. Murphy]:
Yes. The first one or two patients I treated, I caused them more pain than anything else I've done. When you try and aspirate liquid out of a closed space, you can't. Then at two, three, in the morning, I just thought of how we get detergent out of a big detergent bottle. There's a vent. How we used to drink beer by putting two holes in the can. That's how I thought I'll put in two needles and take the stylet of both out. One goes deep, one is relatively superficial. Aspirate from the deep one, get an air fluid level. If the air fluid level stays over about three, four, five minutes, you know it's a narrow neck cyst that's not going to refill, and you can safely put Baxter to seal fibrin into it.

I add probably half a cc of 1% lidocaine up front just to decrease post-operative pain because the fibrin can cause some inflammation. Now, is this on-label or off-label? To discuss that. Baxter's fibrin has been used for years as a dural sealant for spine surgery. I think it's arguable. It's a bit gray. We do need a better dural sealant. All this work on CSF leaks tells us we need a better dural sealant. I've been fortunate enough to be asked to work on a better one that looks interesting. It works and it's safe. That's the technique. Two needles, one deep, one superficial, air fluid level, weight, then put in fibrin.

I put in about 80% of the volume of CSF that I aspirated because there will be some inflammatory reaction. How do I get the needles in? The lamina usually will have a defect in it from the chronic expansion of the cyst and the erosion of the bone. I find that defect and put the needles in through there. I find the place of most thinned lamina, and I just use conventional 18-gauge spinal needles, and I get them to go through that thinned lamina by spinning them at my fingertips. Very little pressure. Just spin them, and it will go through the lamina.

You put in one, they experience their typical pain. You put in the other, then you aspirate. That's all there is to it. What's the retreatment rate? About 20 to 30% after a few years. Because I don't move my job. I've only had two jobs. 10 years at Hopkins, 15 years in Toronto, I have a longitudinal follow-up on people. Over time, you see the cyst shrink and shrink and shrink and go away. These folks will let you know if they have recurrences or other things happen. That's all there is to it. It's pretty simple.

[Dr. Fleming]
Yes.

[Dr. Murphy]
It's just about patient selection.

CSF Leak vs. Tarlov Cyst

[Dr. Fleming]
Fantastic. You mentioned the somewhat related entity of CSF flow abnormalities. Obviously, this is a hot topic, especially within the neuroradiology community.

[Dr. Murphy]
Yes.

[Dr. Fleming]
Can you talk a little bit about how this relates to the broader emphasis towards improved diagnosis and treatment of CSF abnormalities?

[Dr. Murphy]
Follow the money, right? I actually think we have an ethical issue with over-treatment of people's headaches and it being turned into a cash cow by centers who are commercializing that. We need to be careful with that. There are more trial and assist patients out there looking for care than there are people with CSF leaks. Yet all the focus is on CSF leaks. Why is that? It's money. I've seen this once or twice before. Y90 is a good example. I think those institutions that are turning these patients into their product line and charging outrageous sums should be subject to some ethical oversight.
I think it's a problem. We have a big CSF leak program at Toronto Western Hospital. Richard Farb, my gifted colleague, is probably the father of all this. He doesn't go to many meetings, doesn't speak at many meetings, has trained most of the people doing this. He's a great guy. He's like, "I'm just going to do my work." He's really great at it. Mehran, the person we just recruited, trained with him and me, and he does it too. I see this energy to pursue every headache as a potential CSF leak. There are a lot of headaches in the world.

[Dr. Fleming]
Absolutely. Yes.

[Dr. Murphy]
There needs to be better patient selection.

[Dr. Fleming]
I agree with that. I trained in an institution that at the time had a very vigorous blood patching program. The neurologist who was there at the time was one of these experts in CSF hypotension. A lot of the patients that you've alluded to multiple times now, these patients have been put through it and bounced around.

[Dr. Murphy]
Yes, yes.

[Dr. Fleming]
It's a difficult situation where you have a patient who has been to tertiary, quaternary care facilities. They're saying they're hanging their hope on this as a diagnosis. I have to agree with you, though. I think that sometimes what can be a nebulous diagnosis is made de facto, that has to be the diagnosis. That's a difficult thing, I think, about CSF leaks and Tarlov cysts in particular. Sometimes patients are--

[Dr. Murphy]
Yes, to separate out the Tarlov cysts, I think that we can define with a dermatome distribution what is causing the problem. When you start to get into this whole weird world of cranial cervical instability and chronic headaches, and imaging negative CSF leaks and headaches being so common, and so many reasons for headaches. I don't mean to dismiss the suffering of people with these problems or any of the other functional abnormalities that people suffer from. These are terrible complaints to have. They are terrible diseases to have. The commercialization of them is an ethical issue. We've had this before in medicine. It's not uncommon.

If you look at the great bad procedures in medicine over the years. I would include typhoplasty in that. Here's a procedure with very little benefit over conventional vertebroplasty. The destruction of functional tubercular and the overselling of an expensive technology, and the minimal height restoration it achieves by comparison to a simpler technology, where the same restoration of height can be achieved by high-viscosity cements. Ultimately, a lot of these things are about us, not the patient. The patient cares about the implant. I care about my aortic valve replacement more than I do the surgery that happened six years ago. I live with that valve.

People will live with their PMMA in their vertebrae. Why are we still using PMMA? Dentists stopped using that 30 years ago. I think we have to question everything we do systematically, rigorously, look for our unconscious or implicit biases, and challenge ourselves to do better. We exist at the apex of Maslow's hierarchy of needs in medicine. If we're doing things like this here, what's happening in the rest of the world? How transferable are the things that we are doing to other medical cultures? South of the U.S. border, south of the Mediterranean, east of the Alps.

I constantly look at what we do with a certain level of functional circumspection to make sure that I'm trying to do better. I worry about this whole headache thing. I've sat in the back of those meetings. One person was asked about how many allergic reactions they'd seen to fibrin glue. This person said, "Well, I've-- Oh, I've injected, probably, Fibrin 2000 times." I'm thinking that means you don't have a target. You don't know where you're putting it. You don't have a specific location to put that fibrin.
If you're putting fibrin at every level, whether it's a perineural cyst in somebody's T-spine, you don't know what you're doing. As neurodivergentalists, if I'm off by a millimeter, something goes haywire. I think we have to look at the technique, we have to look at the implant, we have to look at the logic, we have to look at the culture of what's going on, and get more specific and get more focused. It's always about the implant. It's always about the mechanism of action.

[Dr. Fleming]
That's so thought-provoking, everything that you just said. It makes me think about, we talked about, the commercial aspect. How does that inform what has a future for improvement, for treatment for Tarlov cysts?

[Dr. Murphy]
We need a better-- First of all, I'm fascinated by CSF. We don't think about liquids. We don't think about liquids much. I'm fascinated by synovium. I know that sounds bizarre. Do you think the synovial fluid at occipital C1 is the same as the synovial fluid in your ankle?

[Dr. Fleming]
No, probably not.

[Dr. Murphy]: No, it wouldn't be. If you've got a car, you're going to have gearbox oil in the gearbox, you're going to have axle stuff on the axle, you're going to have different fluids lubricating the engine. When we look at Synovest, why is it the same everywhere? Should we not actually study the synovium, the synovial fluid in different locations, and see what is specific to that? When you think about a Tarlov cyst, why is the CSF like that, and what is it doing, and what can it tolerate being in there with it? Now, sadly, nerves only like CSF. If there's a nerve in that cyst or in the wall of that cyst and I put fibrin in there, it's going to get pissed off.

I need to come up with a fibrin sealant, a dural sealant, that does not cause a neuritis of some kind by being a foreign body. When people are treating CSF leaks, they need to have a material that does that as well. Adhesives to dura shuts the hole because fibrin's not very adhesive, and actually solves the problem, and be really specific on what they're doing.

Hormone-Associated Immunologic Risks of Fibrin Use