Optic nerve sheath diameter: window to the soul?
Almost exactly 3 years ago I posted an article looking at the utility of ultrasonographic Optic Nerve Sheath Diameter measurement to assess patients for raised intracranial pressure. That post is here
The idea is sound, the optic nerve sheath is an extension of the subarachnoid space – so if ICP goes up it should expand and this ought to be measurable. So – does it actually work?
Well – yes and no…. there is a pile of studies of various size, measuring various things. So the better question is – who knows all this stuff?
To that – the answer is: Dr Leanne Hartnett, Emergency Physician and Ultrasound Doc from my old alumni – Fremantle Hospital in W. Australia.
Leanne presented a review of the evidence and techniques involved in ONSD assessment last year at the ACEM meeting here in Broome. It was one of those talks that left me scratching my head – the data was there, but when you look at studies beside one another – they just don’t make sense.
For example – here is a summary of the measurements from the largest trials. As you can see the mean in one trial might be completely “abnormal” in another.
However – when you compare ONSD to invasive measures of ICP there seems to be great correlation – as below:
So I managed to catch up with Leanne at the SMACC GOLD conference in March – [ we were both dateless for the gala dinner ]
We got chatting and this podcast is what we decided to do. DOWNLOAD HERE
Here are a few more images to refer to as we go through the numbers and the papers – and try to make sense of it all!
If you are busy and want the punch line:
– more data required… not yet for prime time
– under 5mm = probably OK
– over 6 mm = probably not OK
– And in between 5 and 6 mm = who knows, not us for sure!
About The Author
Casey Parker
I am a GP working in Broome, NW of Western Australia. I work as a hospital DMO (District Med Officer) doing Emergency, Anaesthestics, some Obstetrics and a lot of miscellaneous primary care. Also on the web as @broomedocs | + Casey Parker | Contact
Hi Casey, have you ever thought about looking for venous pulsation in optic nerve? Obviously need to take care not to fry eyeballs with too much power, but might be a more functional indicator of raised intracranial pressure.
No I hadn’t thought of that
Would love to hear your thoughts on how it might work.
Presumably a flat or lost venous flow wave would signify an impedance to flow i.e.. raised ICP.
Any studies / data you know of?
Would love to see some basic images
Will have to ask around the gurus
Casey
Thanks Casey and Leanne for the ongoing conversation. Ultrasound ONSD is a lovely tool and, like most, is good for what it is good for. It could help to rule in and rule out the really obvious cases, such as the alcohol +/- CHI, as you describe. And it gives us something vaguely useful to do while we are waiting for the punter to demonstrate (?post-morbid) lateralising signs or to not wake up. Like other investigations and techniques (eg FAST), it has statistical limitations and sits in the clinical context.
I wondered if our potential trial would use LP opening pressure, rather than sticking in a ventricular drain, as the comparative (if not ‘gold’) standard (laying down, not sitting). So, as an idea, if one was doing a LP, anyway for headache or whatever, maybe flash up the USS and gather a data point. Anyone could do it (?multicentre)(in all that spare time you ED chaps have).
Meanwhile, having never done a ONSD scan myself, the first idea might be to get a bit of training / practice at that technique (again, like keeping a hand in at FAST, even on the low pre-test patients).
Thanks.
Fair point GEoffery
I would imagine a study using IV drains / invasive monitors would be biased toward the ICU-end of disease severity spectrum – not our intended ED population.
So something like opening pressure is more attainable – though I do few LPs in the context of head-injured pts with a CT close by.
Maybe the best outcome would be to track patient-oriented end points, Or else we end up with another trial that compares ONSD to a number (mmHg) but no idea as to whether it actually has clinically important outcome benefits.
Thoughts
Casey
Thanks Casey,
I have been trying this a bit recently – not yet confident enough to make any grand prognostications, but it is surprisingly simple and consistent. I can see it being very useful in sites without CT.
Your series of two US images above nicely demonstrates the problem I have sometimes had, that is, the nerve sheath appearing to have multiple concentric layers to it. I have presumed we should take the diameter of the outermost hypodensity (as occurred in your second image), but have not been sure what the “layers” are.
Also, for training, you can conveniently practice on a normal person sitting or standing, and then position them upside down and get a measurable increase in ONSD.
Thanks again.
Great point – inverted pt training! I am going to do this.
…. paging all interns to my office
C
You may have the answer by now but this article helps: Reproducibility and accuracy of optic nerve sheath diameter assessment using ultrasound compared to magnetic resonance imaging. BMC Neurology201313:187. – http://bmcneurol.biomedcentral.com/articles/10.1186/1471-2377-13-187
Outer to outer echogenic border would seem to be the most consistent margin to aim for. This may also limit variances seen between high res and cheap vascular probe measurement.
Just buy a bloody Volk Lens mate 🙂 ( and hope they can get to the slit lamp!)
New to the site- Will be studying this in the Trauma Bay at the Shock Trauma Center in Baltimore this year, under a AAST Grant! Looking at optic nerve sheath and TCD in the trauma patient to see if we can correlate with CT findings and predict need for interventions- ie- do they need to head to the neurosurgeon or can they safely be triaged to the local hospital. Exciting stuff, wish me luck…
Any idea of the optic disc sizes in kids? I’ve got a 9 year old 1 day post head injury + normal CT still drowsy and vomiting: ONSD 5mm
Regards
Chris.
Not a lot of data out there. A few small series looking at head injury and one I saw looking at blocked VP shunts.
Numbers 4.5mm and 5.0 mm used – but as we say in the podcast – there is a wide range of normal overlapping with pathology.
C