Optic nerve sheath diameter as a predictor of acute cerebellar infarction in acute vertigo patients without brain computed tomography abnormalities

Abstract

Background and objectives: To date, no study has investigated the ability of optic nerve sheath diameter calculated from non-contrast brain computed tomography to predict acute cerebellar infarction in patients with acute vertigo. The aim of our study was to evaluate the predictive utility of optic nerve sheath diameter for diagnosing acute cerebellar infarction in patients with acute vertigo without computed tomography abnormalities. Methods: We retrospectively enrolled patients with acute vertigo without computed tomography abnormalities who underwent magnetic resonance imaging including diffusion-weighted imaging at our emergency department between January 2016 and December 2017. Two emergency physicians independently measured optic nerve sheath diameter at 3 mm (ONSD3) and 10 mm (ONSD10) behind the globe in each patient. Final magnetic resonance imaging reports with clinical progress notes were used as the reference standard. A multivariate logistic regression analysis, receiver operating characteristic curves, and intra-class correlation coefficients were calculated to estimate predictive value. Results: A total of 34 patients (16.1%) were diagnosed with acute infarction and 177 patients (83.9%) were diagnosed with peripheral vertigo. Mean ONSD3 (p < 0.001) and ONSD10 (p < 0.001) were independent predictive factors for distinguishing acute infarction and peripheral vertigo. ONSD3 (cut-off = 4.22 mm) had 100% (95% confidence interval = 89.7–100.0) sensitivity and 97.7% (95% confidence interval = 95.1–99.6) specificity with area under the receiver operating characteristic curve of 0.988 (95% confidence interval = 0.978–1.0), while ONSD10 (cut-off = 3.63 mm) had 100% (95% confidence interval = 89.7–100.0) sensitivity and 87.6% (95% confidence interval = 81.8–92.0) specificity with area under the receiver operating characteristic curve of 0.976 (95% confidence interval = 0.959–0.997). There were good inter- and intra-observer agreements for both sides of ONSD3 and ONSD10 (intra-class correlation coefficient range = 0.652–0.773). Conclusion: Optic nerve sheath diameter, in particular OSND3, is a feasible predictive marker for acute infarction in patients with acute vertigo without computed tomography abnormalities. This information can assist decision-making in ordering brain magnetic resonance imaging for the assessment of acute vertigo.