Ping Lu1,2, Yan Sha3, Hailin Wan2, Feng Wang2, Guohong Tian4. 1. Shanghai Institution of Medical Imaging, Zhongshan Hospital, Fudan University, 180 Fenglin Road, Shanghai, 200032, China. 2. Department of Radiology, Eye & ENT Hospital of Fudan University, 83 Fenyang Road, Shanghai, 200031, China. 3. Department of Radiology, Eye & ENT Hospital of Fudan University, 83 Fenyang Road, Shanghai, 200031, China. cjr.shayan@vip.163.com. 4. Department of Ophthalmology, Eye & ENT Hospital of Fudan University, 83 Fenyang Road, Shanghai, 200031, China.
Abstract
PURPOSE: Through a comparison with the axial orientation, we aimed to evaluate the role of coronal high-resolution diffusion-weighted imaging (DWI) in acute optic neuritis based on diagnostic accuracy and the reproducibility of apparent diffusion coefficient (ADC) measurements. METHODS: Orbital DWI, using readout-segmented, parallel imaging, and 2D navigator-based reacquisition (RESOLVE-DWI), was performed on 49 patients with acute vision loss. The coronal (thickness = 3 mm) and axial (thickness = 2 mm) diffusion images were evaluated by two neuroradiologists retrospectively. The sensitivity, specificity, and accuracy were calculated through diagnostic test; the inter- and intra-observer reliabilities were assessed with a weighted Cohen's kappa test. In addition, the agreement of ADC measurement among observers was evaluated by the intra-class correlation coefficient (ICC), coefficient of variation (CV), and Bland-Altman plots. Comparison of ADC values was also performed by unpaired t test. RESULTS: Among the 49 patients, 47 clinically positive optic nerves and 51 clinically negative optic nerves were found. The sensitivity, specificity, and accuracy were 85.1/87.2%, 90.2/94.12%, and 87.8/90.8%, respectively, for coronal RESOLVE-DWI and 83.0/85.1%, 66.7/76.5%, and 75.5/79.6%, respectively, for axial RESOLVE-DWI. The inter-observer kappa values were 0.710 and 0.806 for axial and coronal RESOLVE-DWI, respectively, and the intra-observer kappa values were 0.822 and 0.909, respectively (each P < 0.0001). Regarding the reproducibility of ADC measurements on axial and coronal RESOLVE-DWI, the ICCs among observers were 0.846 and 0.941, respectively, and the CV values were 7.046 and 4.810%, respectively. Bland-Altman plots revealed smaller inter-observer variability on coronal RESOLVE-DWI. ADC values were significantly lower in positive group (each P < 0.0001). CONCLUSION: Higher specificity and better reproducibility of ADC measurements were found for coronal RESOLVE-DWI, which demonstrated the feasibility of the use of coronal RESOLVE-DWI to examine acute optic neuritis patients.
PURPOSE: Through a comparison with the axial orientation, we aimed to evaluate the role of coronal high-resolution diffusion-weighted imaging (DWI) in acute optic neuritis based on diagnostic accuracy and the reproducibility of apparent diffusion coefficient (ADC) measurements. METHODS: Orbital DWI, using readout-segmented, parallel imaging, and 2D navigator-based reacquisition (RESOLVE-DWI), was performed on 49 patients with acute vision loss. The coronal (thickness = 3 mm) and axial (thickness = 2 mm) diffusion images were evaluated by two neuroradiologists retrospectively. The sensitivity, specificity, and accuracy were calculated through diagnostic test; the inter- and intra-observer reliabilities were assessed with a weighted Cohen's kappa test. In addition, the agreement of ADC measurement among observers was evaluated by the intra-class correlation coefficient (ICC), coefficient of variation (CV), and Bland-Altman plots. Comparison of ADC values was also performed by unpaired t test. RESULTS: Among the 49 patients, 47 clinically positive optic nerves and 51 clinically negative optic nerves were found. The sensitivity, specificity, and accuracy were 85.1/87.2%, 90.2/94.12%, and 87.8/90.8%, respectively, for coronal RESOLVE-DWI and 83.0/85.1%, 66.7/76.5%, and 75.5/79.6%, respectively, for axial RESOLVE-DWI. The inter-observer kappa values were 0.710 and 0.806 for axial and coronal RESOLVE-DWI, respectively, and the intra-observer kappa values were 0.822 and 0.909, respectively (each P < 0.0001). Regarding the reproducibility of ADC measurements on axial and coronal RESOLVE-DWI, the ICCs among observers were 0.846 and 0.941, respectively, and the CV values were 7.046 and 4.810%, respectively. Bland-Altman plots revealed smaller inter-observer variability on coronal RESOLVE-DWI. ADC values were significantly lower in positive group (each P < 0.0001). CONCLUSION: Higher specificity and better reproducibility of ADC measurements were found for coronal RESOLVE-DWI, which demonstrated the feasibility of the use of coronal RESOLVE-DWI to examine acute optic neuritispatients.
Authors: Scott Kolbe; Caron Chapman; Thanh Nguyen; Clare Bajraszewski; Leigh Johnston; Michael Kean; Peter Mitchell; Mark Paine; Helmut Butzkueven; Trevor Kilpatrick; Gary Egan Journal: Neuroimage Date: 2009-01-07 Impact factor: 6.556
Authors: Simon J Hickman; Claudia A M Wheeler-Kingshott; Stephen J Jones; Katherine A Miszkiel; Gareth J Barker; Gordon T Plant; David H Miller Journal: AJNR Am J Neuroradiol Date: 2005-04 Impact factor: 3.825