Richard H. White, MD; Marsia Vermeulen, DO, FACEP
NYU Langone Health/Bellevue Hospital
In our emergency department (ED), we often use ultrasound (US) measurement of the optic nerve sheath diameter (ONSD) as a surrogate marker for increased intracranial pressure (ICP). It is fast, non-invasive, and provides an opportunity to avoid the ever-challenging fundoscopic exam. Anatomically, the optic nerve is surrounded by cerebrospinal fluid and sheathed by all three meningeal layers (dura, arachnoid, and pia mater), forming the optic nerve sheath (ONS). Any changes in intracranial pressure will cause a mechanical obstruction and axoplasmic flow stasis in the optic nerve.1 The resultant swelling of the nerve fibers, leakage of proteins, cellular components and water causes engorgement that is reflected in an increased ONSD.
The standard methodology is to measure the optic sheath diameter 3 mm from the posterior wall of the optic globe with < 5 mm as the most commonly accepted value for a non-elevated ICP.1
There are many studies that support use of US measurement of ONSD to help identify patients with elevated intracranial pressure. A recent meta-analysis published in the Journal of Ultrasound Medicine concluded that ocular sonography has a sensitivity of 95.6% and specificity of 92.3% with a positive likelihood ratio of 12.5 and negative likelihood ratio of 0.05.2
However, a recent study by Wang et al published in JAMA Ophthalmology investigating how ONSD correlates with changes in ICP illustrated results that introduced some controversy.3 The study was attempting to correlate ONSD measurements with lumbar puncture (LP) measurements of ICP in patients with suspected elevated ICP both pre- and post-treatment. In their analysis, they found a Pearson’s correlations coefficient (measure of the linear correlation between two variables X and Y), r of 0.7 (95% CI, 0.425-0.870, P<.001) suggesting a strong correlation between ICP as measured by LP opening pressure and ONSD. However, as highlighted by Dr. Hoffman’s letter to the editor, the accuracy of their data is questionable, thereby challenging clinical use of the ONSD as a correlate to dynamic changes in ICP.4
Figure. Variation in ICP as compared with ONSD (from Wang et al).3
In the Figure we see what appears to be a haphazard association between changes in ONSD and changes in ICP. For example, a change in 200 mm H20 in ICP (a more accurate measurement than a CT scan) had an associated range of ONSD measurements between 0.5-2.4 mm. Certainly, from these results, it is challenging to confidently deduce that changes in ICP are reflected with accuracy by ONSD.
Perhaps an explanation is that our understanding of the anatomy as visualized with ultrasound is incorrect. A recent letter by Drs. Copetti and Cattarossi in Intensive Care Medicine questions whether the optic nerve sheath as seen with ultrasound is anatomically correct.5 Based on images from both CTs and MRIs, the optic nerve is serpiginous as opposed to straight and central as per measurements using ultrasound. This can be seen by placing color Doppler on the ONS and noting that the vascular structures (which run parallel to the sheath) are not straight and central as suspected.
So how can ocular US help us if we are concerned for elevated ICP? One additional method involves visualizing bilateral changes in the optic disc. Sonographic papilledema (visualization of the bulging optic disc) can be used in conjunction with ONSD to help increase our diagnostic accuracy.6 In fact, there have been a number of studies where papilledema actually outperformed ONSD as a surrogate for elevated ICP.7-9 However, as papilledema may take time to develop it may not be evident in the acute phase. To complicate things further, pseudopapilledema (optic disc edema) can occur secondary to causes other than elevated ICP.
As US is becoming more utilized and integrated into our clinical practice as a rapid, non-invasive tool, this one method may not be enough to increase confidence in our clinical decision making. The discrepancy between change in opening pressure and ONSD raises questions of its clinical accuracy. Perhaps we can feel more confident using it in concert with a dilated, engorged optic nerve, assisted with color Doppler, to ensure appropriate measurement and ultrasound identification of bilateral papilledema.