NeuroradiologyDiffusion Tensor Imaging of the Lateral Rectus Muscle in Duane Retraction SyndromeAbdel Razek, Ahmed Abdel Khalek MD*; Helmy, Eman Mohamed MD*; Maher, Hala MD†; Kasem, Manal Ali MD‡ Author Information From the *Department of Diagnostic Radiology, Mansoura Faculty of Medicine, Mansoura; †Department of Diagnostic Radiology, Aswan Faculty of Medicine, Aswan; and ‡Mansoura Ophthalmic Center, Mansoura Faculty of Medicine, Mansoura, Egypt. Received for publication December 21, 2018; accepted January 29, 2019. Correspondence to: Ahmed Abdel Khalek Abdel Razek, MD, Department of Diagnostic Radiology, Mansoura Faculty of Medicine, Mansoura 13551, Egypt (e-mail: [email protected]). The authors declare no conflict of interest. Journal of Computer Assisted Tomography: 5/6 2019 - Volume 43 - Issue 3 - p 467-471 doi: 10.1097/RCT.0000000000000859 Buy Metrics Abstract Aim This study aimed to assess metrics of diffusion tensor imaging in evaluating microstructural abnormalities of the lateral rectus muscle in Duane retraction syndrome (DRS). Patient and methods A prospective study was conducted on 27 patients with DRS and 16 age- and sex-matched controls who underwent diffusion tensor imaging of orbit and forced duction test (FDT). Fractional anisotropy (FA) and mean diffusivity (MD) of the lateral rectus were calculated by 2 observers. Results Fractional anisotropy of the lateral rectus in patients (0.62 ± 0.07 and 0.59 ± 0.06) was significantly higher (P = 0.001) than that in controls (0.49 ± 0.06 and 0.51 ± 0.06). Selection values of 0.53 and 0.52 as cutoff points of FA of the lateral rectus to differentiate patients from controls revealed areas under the curve of 0.92 and 0.86 and accuracy values of 84.8% and 80.4% by both observers, respectively. Mean diffusivity of the lateral rectus by both observers in patients (1.19 ± 0.13 and 1.23 ± 0.19 × 10–3 mm2/s) was significantly lower (P = 0.001) than that in controls (1.54 ± 0.18 and 1.49 ± 0.16 × 10–3 mm2/s). Selection values of 1.35 and 1.40 × 10–3 mm2/s as cutoff points of MD of the lateral rectus to differentiate patients from the control groups revealed areas under the curve of 0.93 and 0.85 and accuracy values of 91.3% and 80.4% by both observers, respectively. Interobserver agreement for MD and FA of the lateral rectus by both observers were excellent (r = 0.870 and, 0.959). Diffusion tensor imaging metrics of the lateral rectus muscle did not differ significantly between patients with unilateral and bilateral disease (P = 0.05) and patients with DRS type I and type III (P = 0.05). Diffusion tensor imaging metrics of the lateral rectus muscle differed significantly between FDT grades I and II versus grades III and IV, and these metrics were well correlated with the degree of FDT. Conclusion Diffusion tensor imaging metrics are valuable noninvasive tools in evaluating the microstructural abnormalities of the lateral rectus in DRS and are well correlated with degree of FDT. Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.