Intravoxel incoherent motion diffusion-weighted imaging in nasopharyngeal carcinoma: comparison of the turbo spin-echo and echo-planar imaging techniques.

Background: Traditional echo-planar imaging with intravoxel incoherent motion (EPI-IVIM) exhibits significant magnetic susceptibility artifacts and geometric distortions, which limits its application in nasopharyngeal carcinoma (NPC). This study aimed to compare the qualitative and quantitative indicators between turbo spin echo with intravoxel incoherent motion (TSE-IVIM) and EPI-IVIM in patients with NPC and to provide optimal scanning strategies based on the relationships among these indicators. Methods: A cross-sectional study was conducted between March 2022 and August 2022. Patients with pathologically confirmed NPC underwent pretreatment staging magnetic resonance (MR) examinations with both TSE-IVIM and EPI-IVIM. IVIM images were subjectively and objectively assessed for anatomical structures including nasopharyngeal lesions, nasal concha, the spinal cord, and the temporal pole. Subjective evaluation indicators (including susceptibility artifacts, geometric distortion, lesion conspicuity, and overall image quality on a five-point scale) were compared using the Wilcoxon signed rank test. Quantitative indicators, including signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), apparent diffusion coefficient (ADC), and IVIM-derived parameters (true diffusion coefficients, perfusion fraction, pseudodiffusion coefficient), were compared using the Wilcoxon signed rank test. Bland-Altman plots and coefficient of variation (CV) were used to compare the reproducibility and robustness of the ADC and IVIM-derived parameters between TSE-IVIM and EPI-IVIM. Results: A total of 30 patients (24 males and 6 females) aged between 29 and 71 years (mean age 47.6±12.3 years) were included in the study. TSE-IVIM demonstrated significantly better subjective evaluation indicators for nasopharyngeal lesions, nasal concha, and temporal poles as compared to EPI-IVIM (all P values <0.001), with no significant difference observed in the spinal cord. The SNR of TSE-IVIM was significantly lower than that of EPI-IVIM in the spinal cord and temporal pole (SNR: P=0.027 and P=0.026, respectively). For nasopharyngeal lesions and nasal concha, TSE-IVIM showed no significant difference in SNR (P=0.926 and P=0.428, respectively) but had a significantly higher CNR compared to EPI-IVIM (P=0.003 and P=0.01, respectively). The perfusion fraction (f) value for TSE-IVIM in nasopharyngeal lesions was significantly lower than that for EPI-IVIM (P=0.004), while no statistically significant differences were found in the diffusion coefficient (D) or pseudodiffusion coefficient (D*) values (P=0.914 and P=0.644, respectively). The 95% limits of agreement (LoAs) in nasopharyngeal lesions and nasal concha were wider and had a larger CV compared to those in the spinal cord and temporal pole. Conclusions: TSE-IVIM provided better subjective image quality scores and a significantly higher CNR, with no significant reduction in SNR. It demonstrated superior image quality and more stable quantitative indicators for nasopharyngeal lesions and nasal concha although magnetic sensitivity artifacts were more noticeable. Due to the different f values and wide LoAs observed between the two sequences in nasopharyngeal lesions, TSE-IVIM is recommended for follow-up in patients with NPC.