The challenge & management of slow flow dural shunts

The challenge & management of slow-flow dural shunts transcript
Presenter: Gordon Plant

Well, you’ve heard from all people actually do things. I’m a diagnostician and so you’re going to hear a little bit more about the difficulties in diagnosis of this particular problem, which we haven’t discussed so far, namely slow-flow dural fistulas: how we think about these pathophysiologically, how we think they arise, how they present clinically, and what are the significant complications that can arise.

Dural fistulas were quite a late arrival on the neurological scene. I know Michael Aminoff very well and when he was senior registrar at Maida Vale, he wrote up 16 cases of intracranial dural fistula, a disorder that was very little understood at the time. Certainly with intraspinal dural fistulas even more recently we have become aware of how they present and how to manage them. In general terms, they arise from these smaller dural arteriolar vessels and they shunt into dural venous sinuses. They’re less aggressive than high-flow fistulas and that’s basically how this distinction is made, what the clinical consequences are, and there’s much less likely to be a history of trauma in the case of carotico-cavernous fistulas.

In the cases we’re talking about which present ophthalmologically the dural sinus involved is, of course, the cavernous sinus. And the distinction between low and high flow is really based on the severity of the physical signs rather than the precise underlying anatomy.

‘How do they arise?’ Well I can just say ‘we don’t know’ and leave it at that, but there are theories and most of the theories are based around the pathogenesis involving venous sinus thrombosis. It has been suggested that venous sinus thrombosis gives rise to venous hypertension, which opens up pre-existing arteriovenous communications or that angiogenesis is involved, so the hypoxia produced by the venous hypertension results in the generation of new vessels. It’s also been suggested that new vessels can arise in an organising thrombus and that’s another way that these might arise.

The problem is we don’t see these fistulas when they first arise, it is very unusual to actually see this process happening. In a large series of dural arteriovenous fistulas most of them were associated with dural sinus thrombosis but not all, and it wasn’t possible to explain all of these on any such theory. The other problem is of course that arteriovenous malformations -of any kind- are associated with venous thrombosis and the reason is, as Virchow would tell us, it’s the injury to the vessel wall that is the problem. If we have high flow in a vein, it seems counter-intuitive we have a high flow situation, we’re less likely to get thrombus, but it’s the endothelial injury that gives rise to it. Just because you see thrombosis in relation to a fistula: “yes, we have a chicken-and-egg problem”, we don’t know which came first.

The chicken and egg problem has of course been solved we now know for sure that eggs existed long before chickens. But in the case of the dural sinus thrombosis we’re not sure, but there are some individual cases. I mean, this for example was a case which was very well worked up with angiography at the time of a dural sinus thrombosis and there were no fistulas; and then later the patient had a recurrence of raised intracranial pressure and fistulas were found. But that doesn’t tell us whether they arose de novo or whether they are opening up of pre-existing communications.

‘So how do they present?’ Well, in terms of the the orbital presentation, the ophthalmic features of these dural fistulas: they can all be summed up as being manifestations of venous hypertension. And this goes for the red eye. Here’s a before and after picture and a more severe case with chemosis. Because of this, many of the patients are misdiagnosed as having conjunctivitis. I know in Moorfields eye casualty no one should be sent home without ocular lubricants because you never know, they may help but we do see a lot of these patients who their only manifestation is a red eye that they get misdiagnosed.

The dilated superior ophthalmic vein (‘SOV’) is almost universal. I have only seen one orbital slow-flow dural fistula that did not have a dilated superior ophthalmic vein. Maybe others have seen them, it is an extremely useful screen and of course, if you have orbital ultrasound available, this will tell you for sure. But there are catches, because if there is stasis, it’s very straightforward if the SOV has flow going out through the orbit, the SOV remember is one of the emissary veins, which have no valves in them, so as we make a valsalva and as we stand up, and lie down, blood can flow either direction in the emissary veins in the skull, and the SOV is one of those, quite happy to take blood in either direction, and if you have pulsatile flow anteriorly from the cavernous sinus, then that’s straightforward. But if there is stasis, then it may actually be invisible on ultrasound.

We have dealt with varices, Geoff has removed some thrombosed varices for us in treated patients. We also have tinnitus, it’s not common with these slow-flow fistulas, much more likely to occur with high flow. If it is there, you can modify it with jugular or common carotid compression. But remember that tinnitus is also a symptom of raised intracranial pressure, which is one of the complications as we’ll discuss in a moment, of this type of fistula.

Diplopia: here you can see a patient with a dural fistula. All of the muscles are congested. This doesn’t seem to affect the function of the extraocular muscles What it does do, is lead to a misdiagnosis as dysthyroid eye disease. If a patient has diplopia with a dural fistula, it’s more likely to be due to a cranial neuropathy (a VI nerve palsy or IV) and these happen in the cavernous sinus. One not uncommon situation is a patient who presents like a microvascular VI, which recovers and they develop the signs of the fistula afterwards. That could be related to a partial thrombus occurring in the cavernous sinus causing the VI nerve palsy, not sufficient to cause venous hypertension and any other signs resolves, but the fistula follows. Here are some exceptions: the so called white-eyed. This means the fistula is draining through the back door into the petrous venous system, so the patient doesn’t develop the eye signs but they may still present to ophthalmology because they may develop cranial nerve palsies. The white eye fistula is of course, only of significance to ophthalmologists, because they’re expecting the eye to be red. This type of fistula, because it’s slow-flow, it’s very unusual for it to be bilateral (to have bilateral signs) but it can switch. As you know, the two cavernous sinuses are not separate structures, they are connected and the flow pattern can change from one eye to the other quite spontaneously.

‘What are the complications?’ Well, as I’ve said thrombosis can occur as a result of the endothelial damage in the venous structures, and sometimes the patient will get suddenly worse because of thrombosis of the superior ophthalmic vein. This is a very unusual event in any other situation other than infections in the orbit and so on, it’s an unusual thing to happen and it can be quite dramatic when it does.

Glaucoma: here’s the Goldmann equation and it is of course the venous pressure that is relevant. When I’m trying to explain pseudotumor cerebri to ophthalmologists, when I point out that the skull is a rigid more or less spherical structure that is constantly generating fluid, which drains out into the veins, they suddenly start to understand why patients with venous hypertension get raised intracranial pressure, because of course the same thing happens in the eye. If there is visual loss from glaucoma due to raised intraocular pressure, we take this as a reason for needing to deal with those patients urgently. Raised intracranial pressure is however, very unusual with these cases. I think this is because they are draining predominantly anteriorly through the orbit that we we don’t see this.

For these slow-flow fistulas, generally, one of the most serious, potentially dangerous complications is from the cortical venous drainage of the brain itself, because this can lead to venous infarction, or a progressive deficit, or even a fluctuating deficit (which is one of the things we particularly see with spinal cord dural fistulas), it can lead to intra-cerebral haemorrhage and seizures. Fortunately, this is quite unusual again with this type of fistula. So here, this is not an orbital cavernous dural fistula, but this can be happening in the brain; oedema and haemorrhage due to venous hypertension, affecting the drainage (the cortical drainage) or even the deep cerebral drainage. Once again, it’s something that is quite unusual with this type of fistula.

As for management, I’m not going to say very much, We confirm the diagnosis of course, but management can only be planned following catheter angiography. We can do many studies to confirm the diagnosis so that we know what we’re dealing with, etc. But you have, if you’re going to do, if you want the patient to have management and I think there are cases that can be managed conservatively, if they are going to have active management, then they need to have a catheter study and that’s where we involve our other colleagues.

So, I think what’s important about this type of fistula, is this crosses over between ophthalmology and neurology that our trainees need to know how these fistulas can present and how they can be misdiagnosed and be aware of the more serious complications.

Q & As section

Question 1
Do you recommend carotid massage to these patients if you’re contemplating conservative management?

Answer
This is something that was talked about by my senior colleagues when I was training and we’ve never put it into practice. The idea was that you ask the patient to occlude the ipsilateral common carotid using the contralateral arm, so if they develop a hemiparesis the arm will fall away. And part of the rationale for this was, when the management involved carotid angiography, sometimes the fistula would seem to close following the procedure. I think if that’s going to happen and I think this fistula is quite likely to close spontaneously, which many of them do. We don’t advocate that, we monitor the patients and if they’re getting worse then we’ll refer them on and we’ll see if they close spontaneously all well and good. I think it’s possibly a slightly hazardous thing to ask people to do.

Question 2
I’ve had a couple of patients who’ve taken quite a long time to settle down from an eye point of view after treatment of their CCS but they do, whereas the VI nerve palsy doesn’t seem to recover. I just wondered what your experience was of those patients that present with VI nerve palsy; Do they usually recover or do they require intervention?

Answer
Well quite often when they recover as I say, it happens very early in the story, sometimes even before the fistula has fully developed and that will behave more like a micro vascular VI and recover. When it’s the only presentation, of course it’s going to have gone on for longer than that, as with the white-eyed fistulas. We had a patient recently with a very painful fistula associated with the IV (which never ever recovered). I think we’re kind of selecting out here the ones they can recover spontaneously, we’re selecting out here the ones that don’t.

Question 3
I was wondering what tips you might have for avoiding misdiagnosing a slow-flow fistula presenting as a red eye, because I’m terrified of that happening.

Answer
They often don’t have much chemosis (so that’s really getting at a higher flow) but they’re really snaky vessels that are big all the way up and you may not see the normal flow in them, which you see and there’s sometimes pulse and they’re always there 360 degrees, unlike exposure vessels which tend to be inferior.