Ahmed Abdel Khalek Abdel Razek1. 1. Department of Diagnostic Radiology, Mansoura Faculty of Medicine, Mansoura 13551, Egypt. Electronic address: arazek@mans.edu.eg.
Abstract
BACKGROUND: Differentiation between residual head and neck squamous cell carcinoma and post-radiation changes is difficult with routine computed tomography and magnetic resonance imaging. PURPOSE: To assess the reliability and reproducibility of mean diffusivity (MD) and fractional anisotropy (FA) parameters of diffusion tensor imaging in differentiation residual head and neck squamous cell carcinoma from post-radiation changes. MATERIAL AND METHODS: A retrospective analysis of diffusion tensor imaging of 43 patients with head and neck squamous cell carcinoma cancer after radiotherapy. The MD and FA of the lesion were calculated by the same reader at two-time points. RESULTS: There was a significantly lower difference (P = 0.001) in MD of both readings between residual head and neck squamous cell carcinoma (1.48 ± 0.06 and 1.47 ± 0.07 × 10-3 mm2/s) and post-radiation changes (1.72 ± 0.08 and 1.71 ± 0.11 × 10-3 mm2/s). The FA of residual head and neck squamous cell carcinoma of both readings (0.41 ± 0.09 and 0.42 ± 0.09) shows a significantly higher difference (P = 0.001) than post-radiation changes (0.17 ± 0.04 and 0.16 ± 0.03). There was excellent intra-reader agreement between both readings using MD (K = 0.958) and FA (K = 0.987). The threshold MD and FA used for differentiating the residual from post-radiation changes of both readings was 1.61, 1.65 × 10-3 mm2/s and 0.27, 0.25 with an area under the curve of 0.991, 0.934, 0.993 and 0.990 respectively. CONCLUSION: MD and FA of diffusion tensor imaging are non-invasive reliable and reproducible parameters that can help in differentiation residual head and neck squamous cell carcinoma from post-radiation changes.
BACKGROUND: Differentiation between residual head and neck squamous cell carcinoma and post-radiation changes is difficult with routine computed tomography and magnetic resonance imaging. PURPOSE: To assess the reliability and reproducibility of mean diffusivity (MD) and fractional anisotropy (FA) parameters of diffusion tensor imaging in differentiation residual head and neck squamous cell carcinoma from post-radiation changes. MATERIAL AND METHODS: A retrospective analysis of diffusion tensor imaging of 43 patients with head and neck squamous cell carcinoma cancer after radiotherapy. The MD and FA of the lesion were calculated by the same reader at two-time points. RESULTS: There was a significantly lower difference (P = 0.001) in MD of both readings between residual head and neck squamous cell carcinoma (1.48 ± 0.06 and 1.47 ± 0.07 × 10-3 mm2/s) and post-radiation changes (1.72 ± 0.08 and 1.71 ± 0.11 × 10-3 mm2/s). The FA of residual head and neck squamous cell carcinoma of both readings (0.41 ± 0.09 and 0.42 ± 0.09) shows a significantly higher difference (P = 0.001) than post-radiation changes (0.17 ± 0.04 and 0.16 ± 0.03). There was excellent intra-reader agreement between both readings using MD (K = 0.958) and FA (K = 0.987). The threshold MD and FA used for differentiating the residual from post-radiation changes of both readings was 1.61, 1.65 × 10-3 mm2/s and 0.27, 0.25 with an area under the curve of 0.991, 0.934, 0.993 and 0.990 respectively. CONCLUSION: MD and FA of diffusion tensor imaging are non-invasive reliable and reproducible parameters that can help in differentiation residual head and neck squamous cell carcinoma from post-radiation changes.