Estimating the accuracy of optic nerve sheath diameter measurement using a pocket-sized, handheld ultrasound on a simulation model

• Published in: Critical ultrasound journal Vol. 8; no. 1; p. 18
• Main Authors: Johnson, Garrett G. R. J., Zeiler, Frederick A., Unger, Bertram, Hansen, Gregory, Karakitsos, Dimitrios, Gillman, Lawrence M.
• Format: Journal Article
• Language: English
• Published: Milan Springer Milan 01.12.2016; Springer Nature B.V; SpringerOpen
• Subjects: Accuracy; Correlation coefficients; Critical Care Medicine; Diagnostic Radiology; Diameters; Emergency Medicine; Handheld ultrasound; Imaging; Intensive; Interventional Radiology; Intra-cranial pressure; Medicine; Medicine & Public Health; Nerves; Optic nerve; Optic nerve sheath diameter; Original; Original Article; Point-of-care ultrasound; Portability; Radiology; Sheaths; Simulation models; Ultrasonic imaging; Ultrasonic methods; Ultrasound; Point-of-care ultrasound; Handheld ultrasound; Ultrasound; Optic nerve sheath diameter; Intra-cranial pressure
• ISSN: 2036-3176; 2036-7902; 2524-8987
• DOI: 10.1186/s13089-016-0053-9

Background Ultrasound measurement of optic nerve sheath diameter (ONSD) appears to be a promising, rapid, non-invasive bedside tool for identification of elevated intra-cranial pressure. With improvements in ultrasound technology, machines are becoming smaller; however, it is unclear if these ultra-portable handheld units have the resolution to make these measurements precisely. In this study, we estimate the accuracy of ONSD measurement in a pocket-sized ultrasound unit. Methods Utilizing a locally developed, previously validated model of the eye, ONSD was measured by two expert observers, three times with two machines and on five models with different optic nerve sheath sizes. A pocket ultrasound (Vscan, GE Healthcare) and a standard portable ultrasound (M-Turbo, SonoSite) were used to measure the models. Data was analyzed by Bland–Altman plot and intra-class correlation coefficient (ICC). Results The ICC between raters for the SonoSite was 0.878, and for the Vscan was 0.826. The between-machine agreement ICC was 0.752. Bland–Altman agreement analysis between the two ultrasound methods showed an even spread across the range of sheath sizes, and that the Vscan tended to read on average 0.33 mm higher than the SonoSite for each measurement, with a standard deviation of 0.65 mm. Conclusions Accurate ONSD measurement may be possible utilizing pocket-sized, handheld ultrasound devices despite their small screen size, lower resolution, and lower probe frequencies. Further study in human subjects is warranted for all newer handheld ultrasound models as they become available on the market.