David L. Daniels 1 James S. Hyde J. Bruce Kneeland A. Jesmanowicz W. Froncisz Thomas M. Grist Peter Pech Alan L. Williams Victor M. Haughton Received February 22, 1985; accepted after re- vision August 5, 1985. This work was supported in part by a grant from General Electric Medical Systems and by grant RR01 008 from the National Institutes of Health. 1 All authors: Department of Radiology, Medical College of Wisconsin , Froedtert Memorial Lutheran Hospital, 9200 W. Wi sconsin Ave., Milwaukee, WI 53226. Address reprint requests to D. L. Daniels. AJNR 7:129-133, January/February 1986 0195-6108/86/0701-0129 © American Society of Neuroradiology The Cervical Nerves and Foramina: Local-Coil MR Imaging 129 A detailed study of the magnetic resonance (MR) appearance of the cervical nerves and neural foramina is presented. Using a "butterfly" local coil and a 1.5-T GE Signa MR system, the authors obtained 3-mm-thick axial, parasagittal, and 45° oblique MR images of the cervical neural foramina in four normal volunteers. For the oblique images, subjects were rotated 45° toward the right posterior oblique. Neural and vascular structures in the foramina were identified by correlation of the MR images with corre- sponding cryomicrotomic sections from four cadavers. The 45° oblique images were more useful than axial or parasagittal images for demonstrating anatomic relations in the neural foramina. MR imaging with local coils demonstrates the complex anatomy of cervical neural foramina in various planes. The myelographic and computed tomographic (CT) evaluation of the cervical neural foramina is not optimal. Positive-contrast myelography carries the risk of undesirable side effects and imperfect sensitivity or specificity for root sheath lesions [1, 2]. The CT evaluation of cervical nerves may be difficult, especially at the C7 - T1 level or in the presence of minor degenerative changes that narrow the foramen and obscure the fat [3]. Intrathecal or intravenous contrast enhancement has not improved the sensitivity of CT sufficiently to earn it wide acceptance [3,4]. CT images reformatted in 45° oblique planes, which show anatomic relations optimally in the neural foramen [5], have less resolution than does direct axial CT. Magnetic resonance (MR) imaging has the potential to demonstrate cervical nerves without osseous artifacts and in several planes. Because of the deficiencies of CT, the use of MR imaging in the clinical evaluation of cervical disk disease is anticipated. We describe the normal appearance of the neural foramina in local-coil MR images. Materials and Methods Cervical spine specimens were removed from four fresh-frozen cadavers, embedded in styrofoam boxes with carboxymethyl cellulose gel, and sectioned with a heavy-duty sledge cryomicrotome (LKB 2250, Gaithersburg, MD) in axial , parasagittal , or 45° oblique planes [6 , 7]. The surfaces of the specimens were photographed as they were sectioned. MR images were obtained with a GE Signa 1.5-T system and a modified loop-gap resonator [8-10] as the receiver coil. The loop-gap resonator had a two-loop, single-gap "butterfly" configuration (Hyde J, unpublished data). Four normal volunteers (age range, 25-35 years) without clinical evidence of cervical radiculopathy were recruited for imaging . Sagittal and axial images were used for localization. For axial and parasagittal images, the volunteers were positioned supine on the Signa table with the loop-gap resonator under the neck. For the oblique images they were rotated 45° toward the right posterior oblique so that their sagittal plane formed an angle of 45° with the surface of the coil and tabletop. Technical factors included a repetition time (TR) of 800 msec, an echo time (TE) of 25 msec, a 128 x 256 or 256 x 256 matrix (pixel sizes of 1.2 x 0.6 mm and 0.6 x 0.6 mm , respectively), a 16- cm field of view, two excitations, and a 3-mm slice thickness. Two volunteers also had parasagittal images with a TR of both 400 and 2500 msec.