Comparison of Diffusion-Weighted MRI Using Single-Shot Echo-Planar Imaging and Split Acquisition of Fast Spin-Echo Signal Imaging, a Non-EPI Technique, in Tumors of the Head and Neck

BACKGROUND AND PURPOSE: DWI using single-shot echo-planar imaging (DWI-EPI) is susceptible to distortions around air-filled cavities and dental fillings, typical for the head and neck area. Non-EPI, split acquisition of fast spin-echo signals for diffusion imaging (DW-SPLICE) could reduce these distortions and enhance image quality, thereby potentially improving recurrence assessment in squamous cell carcinoma (SCC) of the head and neck region. This study evaluated whether DW-SPLICE is a viable alternative to DWI-EPI through quantitative and qualitative analyses.

MATERIALS AND METHODS: The DW-SPLICE sequence was incorporated into the standard 3T head and neck MRI protocol with DWI-EPI. Retrospective analysis was conducted on 2 subgroups: first benign or malignant lesions, and second, posttreatment SCC recurrence. In both subgroups, image quality and distortion were scored by 2 independent radiologists, blinded to the DWI technique and evaluated using mixed-effect linear models. Lesion ADC values were assessed with interclass correlation and Bland-Altman analyses. The delineation geometric similarity of DWI to T1-weighted postcontrast MRI was evaluated using the DSC before and after registration. Recurrence in posttreatment SCC scans was evaluated by the same 2 radiologists blinded to the DWI technique. Recurrence detection rates were then compared between DW-SPLICE and DWI-EPI using mixed logistic regression at 6 months and 1 year postscan follow-up data.

RESULTS: From August 2020 to January 2022, fifty-five benign or malignant lesion scans (55 patients) and 74 posttreatment SCC scans (66 patients) were analyzed. DW-SPLICE scored better on image quality and showed less overall distortion than DWI-EPI (0.04<P < .001). There was high ADC measurement reliability (intraclsss correlation coefficient = 0.93, P < .001), though a proportional bias was also observed (β = 0.11, P = .03), indicating that the bias increases as ADC values increase. DW-SPLICE exhibited greater geometric similarity to T1WI with gadolinium contrast before registration (DSC 0.63 versus 0.47, P < .001) and outperformed DWI-EPI by more accurately identifying recurrences after 1 year (OR  = 0.96, P = .05) but not after 6 months (OR  = 0.72, P = .13).

CONCLUSIONS: DW-SPLICE surpasses DWI-EPI on image distortion and quality and improves diagnostic reliability for detecting recurrent or residual SCC on 3T MRI of the head and neck. Consistent use of 1 method for follow-up is advised, because ADC values are not completely interchangeable. Integrating DW-SPLICE can significantly improve tumor assessments in clinical practice.