Value of fat-suppressed MRI pulse sequences for the diagnosis of malignant tumors of the head and neck region

Purpose: Evaluation of frequency-selective fat saturation (FS) and short-tau inversion recovery (STIR) fat suppression (FU) in MRI of patients with malignant head and neck tumors. Methods: Forty-five patients with biopsy-proven carcinomas of the extracranial head and neck were examined with MRI at 1.0 T. A T2-weighted TSE sequence iwth and without STIR-FU and a T1-weighted SE sequence with and without FS were compared in axial slices. Results: STIR-FU was successful in all and FS in 85-88% of the head examinations (nasopharynx, sinuses, oropharynx, and oral cavity) and 33-46% of the neck examinations (hypopharynx and larynx). When visualization and delineation of tumors were ranked on a four-point scale (0-3), respective mean values for images with/without FU or FS were 2.6/1.9 for T2-TSE in all examinations, 2.2/1.7 (nasopharynx and sinuses) and 1.3/1.4-1.6 (oropharynx, oral cavity, hypopharynx, and larynx) for T1-SE without contrast media administration (CM) and 2.3/2.1 (nasopharynx and sinuses) and 2.4-2.5/1.9-2.0 (oropharynx, oral cavity, hypopharynx, and larynx) with CM. Conclusions: STIR-T2-TSE was a technically reliable pathfinder for localization and extension of both tumors and lymph nodes. FS-T1-SE was technically unreliable in examinations of the hypopharynx and larynx. With CM, FS-T1-SE was most useful for MRI of carcinomas of the oral cavity and oropharynx. In the nasopharynx and sinuses, T1-SE with CM and FS-T1-SE with or without CM were equal for tumor visualization and delineation. (orig.)