[MR imaging of the spinal cord--with special emphasis on the factors influencing spinal cord measurement].

On MR images the spinal cord is seen differently in size depending on imaging parameters and displaying window; consequently the findings may be interpreted erroneously as swelling or atrophy of the spinal cord. The purpose of this paper was to evaluate factors influencing spinal cord size on images and to determine the optimal condition estimating the size of the spinal cord. At first we selected 4 cases suspected of cervical spinal disorders which had been examined by both MRI and myelography with tomography. Sagittal diameter of the spinal cord was measured on a film and it was significantly different of those three. That is, the measurement value was greater on T1 weighted image (T1WI) and smaller on T2 weighted image (T2WI) than myelo-tomography. To evaluate the effect of imaging parameters, image reconstruction and image displaying window quantitatively, studied were the cadaveric cervical spinal cord and gelatin phantom tube with a diameter of 13 mm and 9 mm placed in a saline-filled plastic tube. The measurement value was significantly greater on T1WI and smaller on T2WI than true size of the objects. Numbers of phase encoding (128 and 256) significantly affected the measurement value, both on T1WI and T2WI, as well. Ringing artifact of high or low signal was observed at the boundary area of the objects and saline (so-called truncation artifact). However, when the window-level of displaying image was raised stepwisely the measurement value was proportionally decreased and it reached to real value when the level was adjusted at the mean MR signal intensity of the object and saline.(ABSTRACT TRUNCATED AT 250 WORDS)