No abstract

Over a 2-year period 20 patients who presented with masses in the parotid gland were evaluated with magnetic resonance (MR) imaging. T1-weighted images were obtained on a high-resolution, thin-section MR imaging system. When "cystic-appearing" lesions were found, T2-weighted images were obtained in order to better characterize the tumor. As in other areas of the body, MR images of parotid tumors are not usually histologically specific. MR findings may be distinctive in rare cases and define the internal architecture of complex parotid masses. Although poor tumor margination was a clue to malignancy, this was not a consistent finding. The real advantage of MR imaging in evaluating parotid masses was its ability to accurately reveal the extraparotid or intraparotid location of a tumor and demonstrate the relationship of the tumor to the facial nerve. Small and medium-sized mass lesions could be seen as superficial or deep to the facial nerve. Larger masses producing some distortion of the normal course of the nerve made identification of the nerve more difficult, if not impossible. In malignant tumors with gross invasion of the facial canal, MR images can show the extent of nerve involvement.

Ten healthy subjects and 44 patients with diseases of the tongue or oropharynx were studied with magnetic resonance (MR) imaging. Axial, coronal, and sagittal images with a thickness of 4 mm were obtained with a pixel size of 0.75 X 0.75 mm on a 256 matrix. Nineteen of the patients underwent computed tomography (CT). Nine of those patients later had surgery, and the specimens were obtained for organ sectioning. These three studies as well as clinical history and physical examination findings were correlated. MR imaging was equal to or better than CT in those patients having both examinations. However, neither CT nor MR allowed recognition of histologic features or detection of microscopic spread of disease. Direct coronal and sagittal imaging planes on MR imaging allowed visualization of intrinsic tongue musculature, not possible with CT; this was important in recognizing subtle tumor extension. For these reasons, MR is the imaging method of choice for studying diseases of the tongue and oropharynx.

Forty patients with disorders of the larynx or hypopharynx were studied with magnetic resonance (MR) imaging. Axial, coronal, and sagittal sections, 4 mm thick, were obtained. Twenty-eight of the patients underwent computed tomography (CT) scanning; 17 underwent surgery, and specimens were obtained for organ sectioning. Correlation was made between these three studies as well as with clinical history, physical examination, and endoscopic photography. In 13 patients who underwent all three studies, the depiction of cartilage invasion, adenopathy, and intraorgan and extraorgan spread of disease was compared. MR consistently showed superior soft-tissue definition and extent of disease compared with CT. Neither CT nor MR was able to depict histologic detail or microscopic spread of disease. Both studies were also less effective in the postoperative or postirradiated neck. The use of direct coronal and sagittal imaging planes on MR allowed the visualization of intrinsic laryngeal musculature, which was important in the recognition of subtle tumor extension. For these reasons, surface coil MR imaging is currently the imaging study of choice at our institution for disorders of the larynx and hypopharynx.

Magnetic resonance imaging (MRI) of the facial nerve was evaluated by studying normal volunteers and patients with diseases of the facial nerve with a 0.3 Tesla permanent-magnet MRI system with special surface coils. The normal MR images were correlated with the anatomy of thin cryosection specimens of fresh cadavers. The seventh nerve was followed from its nucleus in the brainstem through the temporal bone to the parotid gland bed. The entire labyrinth and tympanic portions, as well as the geniculate ganglion, could be shown with appropriate scan planes. Examples of brainstem diseases affecting the facial nerve and nucleus, facial neuromas, parotid tumors involving the facial nerve, and other diseases were studied. MRI is a technique that allows unique evaluation of the entire course of the facial nerve. It produces superior images of the facial nerve with high-contrast resolution. Unlike computed tomography, there is no beam-hardening artifact from the temporal bone or exposure to ionizing radiation and contrast agents. MRI also allows visualization of the main trunks of the facial nerve in the parotid bed not possible with any other imaging technique.

To date, high-resolution computed tomographic (CT) scanning has proved to be the most valuable preoperative study for the evaluation of the anatomic extent of head and neck cancers. Now, with the development of magnetic resonance (MR) imaging, the prospect of having an equally reliable test without associated radiation exposure has emerged. This study is designed to compare MR images of the larynx with corresponding histologic sections obtained by step-serial sectioning of whole organ specimens. Images were generated by a 0.3 Tesla permanent magnet MR imager. Additional correlation with CT scans was obtained when possible. CT resolution proved to be slightly superior in the axial plane but markedly inferior to MR imaging in the coronal and sagittal planes. Overall, MR imaging appeared to have good resolution of the larynx.

The use of surface coils on subjects, and internal engineering improvements have combined to greatly improve the signal-to-noise ratio on the magnetic resonance imager at our institution over the past two and one-half years. This has resulted in a marked improvement in image resolution, as well as reduction in the time required and related cost of these studies. Magnetic resonance (MR) examinations at our institution now cost less than comparable computed tomography (CT) studies. This article summarizes our experiences in imaging the larynx and tongue base. Eleven cases are presented to illustrate how the direct multiplane capacity of MRs and their increased soft tissue contrast provide some definite advantages over CT in determination of extent of disease in these regions.

No abstract