Katerina Grebenyuk, MS4
The Article: Oberfoell, S., Murphy, D., French, A., Trent, S. and Richards, D. (2017), Inter‐rater Reliability of Sonographic Optic Nerve Sheath Diameter Measurements by Emergency Medicine Physicians. J Ultrasound Med, 36: 1579-1584.
The Idea: It is thought that optic nerve sheath diameter (ONSD) correlates with the intracranial pressure (ICP), with an ONSD greater than 5mm predicting an ICP greater than 20 mmHg. Because invasive approaches to measuring ICP require a lot of time, expertise, and have a risk of complications, the point-of-care ultrasound in the ED is a unique opportunity to screen for elevated ICP in the early acute setting. For point-of-care ultrasound in the ED to be a useful screening tool of elevated ICP, however, there must be an established reliability of measuring ONSD by ED providers, especially because the ONSD is a small structures and even subtle measurement differences may alter interpretation. This study set out to estimate and compare the degree of agreement of sonographic ONSD measurements by residents and ultrasound fellowship-trained EM physicians.
The Study: This study was cross-sectional in design, and recruited both resident and ultrasound fellowship-trained EM physicians to measure the ONSD using a computerized ruler on five still-frame sonograms of adult eyes, that included the retrobulbar optic nerve.
Primary Endpoint: to estimate and compare the inter-rater reliability (IRR) of ONSD measurements by ultrasound fellowship-trained and resident EM physicians on the same still-frame sonographic image. The IRR was estimated using the intraclass correlation coefficient (ICC).
Secondary Endpoint: to estimate and compare the ONSD measurements between the two groups on each image.
The Results: 61 EM physicians were enrolled between November 2014 and January 2015. There were 16 were PGY-1, 12 were PGY-2, 14 were PGY-3, and 9 were PGY-4 residents and 10 ultrasound trained-physicians. The ICC among resident EM physicians was 0.50 (95% confidence interval [CI] 0.25-0.89), and the ICC among ultrasound fellowship-trained EM physicians was 0.73 (95% CI 0.44-0.96). The ICC for PGY-1s was 0.44 (95% CI 0.19-0.87), for PGY-2s was 0.41 (95% CI .16-0.86), for PGY-3s was 0.69 (95% CI 0.41-0.95), for PGY-4s was 0.40 (0.13-0.86). Given the overlapping confidence intervals surrounding the point estimates of ICC’s calculated for each group, no significant difference in IRR was observed between the two groups.
The Takeaway: This study revealed a few important points. First, the IRR for ultrasound fellowship-trained and resident EM physicians showed strong and moderate agreement (ICC 5 0.73 and ICC 5 0.50), respectively. Even though ultrasound fellowship-trained physicians had a higher ICC, the wide confidence intervals and significant overlap between the two groups suggest that the difference between the two is not statistically significant. The large degree of variability observed among EM physicians in measuring ONSD is concerning for widespread implementation of this technique to screen for elevated ICP, especially because they removed the step of image acquisition, which is thought to be the more difficult component of establishing ONSD measurements. Second, the study also showed that more reliable ONSD measurements are not acquired by residency training alone, as reflected by similar ICCs among PGY-1 and PGY-4 residents, and there not being a significant difference between ultrasound fellowship-trained and resident EM physicians. Third, this study did not even address the accuracy of the acquired measurements, due to there being a lack of gold standard neuroimaging to correspond with sonographic ONSD measurements. Therefore, the limited ability of EM physicians to make reliable measurements of ONSD using point-of-care ultrasound in an ideal setting may limit its applicability as a screening tool for elevated ICP in the ED.