Why 2D Matters: Comparative Evaluation of 2D and 3D T1-Weighted Imaging of the Skull Base and Neck.

Abstract

Background and purpose: 3-dimensional (3D) T1-weighted imaging (T1-WI) MR protocols have been proposed as time-efficient alternatives for skull base and neck imaging, but direct comparisons with conventional 2-dimensional (2D) T1-weighted turbo spin echo (TSE) imaging are limited. This study aims to qualitatively and quantitatively compare 2D and 3D T1-WI of the head and neck.

Materials and methods: A retrospective review was conducted on 21 patients who underwent both 2D and 3D pre-and post-contrast T1-WI during the same scan session (3T Philips 7700). Seven independent radiologists (4.5-35-years experience) evaluated image quality using Likert scales, categorical ratings (2D better, 3D better, equivalent), and yes/no questions. Quantitative measures included ROI intensity values in buccal fat, masseter muscle, and tumors (if present), as well as percent delineation of two small but relevant muscles, tensor veli palatini and superior pharyngeal constrictor muscles on skull base and neck MR, respectively.

Results: Of the 21 MRIs (10 skull base, 11 neck), contrast timing was balanced across sequences. Raters significantly favored 2D T1-weighted TSE for improved fat visualization at skull base foramina, homogeneity of fat signal, and parotid architecture (p<.001). Tumor margins were more clearly defined on 2D neck MR (pre-and post-contrast: p<.001, p=.04), although this was not significantly different at the skull base. 3D sequences showed less pulsation artifact particularly pre-contrast (both p<.001), but more susceptibility artifact (both, p<.001). Quantitatively, 2D images had higher fat-to-muscle (both p<.001) and tumor-to-muscle relative signal intensity ratios (neck: p<.001, skull base: p=.04). Delineation of tensor veli palatini and superior pharyngeal constrictor muscles was significantly better with 2D imaging (both p<.001). Multiplanar reformat capability from 3D imaging did not add diagnostic value in either the neck or skull base (all p<.05).

Conclusions: Despite advances in 3D imaging, 2D T1-weighted TSE sequences continue to offer superior soft tissue contrast and delineation in the complex anatomy of the skull base and neck. While 3D sequences reduce pulsation artifacts and provide thinner slices, their overall diagnostic utility remains less favorable due to poorer tissue contrast and increased susceptibility artifacts particularly at tissue interfaces, especially relevant in the head and neck.

Abbreviations: TSE=turbo spin echo, TFE= turbo field echo, ROI=region-of-interest, 2D=2-dimension, 3D=3-dimension, mDixon=modified Dixon, T1-WI=T1-weighted image, MPR=multiplanar reformat.