l zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Original Contribution zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA MAGNETIC RESONANCE IMAGING OF BONE AFTER RADIATION PETER A. REMEDIOS, PATRICK M. COLLETTI, JANAK K. RAVAL, ROBERT C. BENSON, LINDA Y. CHAK, WILLIAM D. BOSWELL, JR., JAMES M. HALLS Department of Radiology, Los Angeles County-University of Southern California Medical Center, 1200 N. State Street, Los Angeles, CA 90033 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQP Magnetic resonance imaging (MRI) was performed in 22 patients at various times (O-3) years) following radia- tion therapy to the spine. T, and T2 weighted images were obtained at 0.5 Tesla. Increased signal was seen after 800-6000 rads (8-60 Gy). Marrow effects corresponded to radiation ports. Recurrent tumor was clearly separated from fatty replacement. This was much better seen on T, weighted images. Five patients that had MRI during their course of radiotherapy (XRT) did not have increased signal on T, images of the bone marrow. The earli- est fatty marrow change was seen nine days following completion of 3000 rads (30 Cy) XRT over one month’s duration. One patient who received 800 rads (8 Gy) to the upper thoracic spine for eosinophilic granuloma had no radiation effects on MRI when imaged 16 days following completion of XRT given over five days. Fatty mar- row change was seen in this patient on zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA MRI six months later. MRI was particularly useful in defining the extent of prior radiation effects when repeat therapy was needed. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFE Keywords: Spinal radiation; Bone marrow radiation. INTRODUCTION Magnetic resonance imaging of the spine is an excellent method for evaluating bony metastases.‘-’ Radiation therapy has been shown to appear as in- creased signal intensity on T,weighted imaging corre- sponding with radiation ports.4 It is unclear, however, what the influence of time or dose is on the appearance of fatty marrow change. This study is a retrospective evaluation of twenty-two patients who underwent MRI during or after their spinal radiation therapy. Attempts are made to categorize the various factors that play a role in determining those patients whose MRI reveals increased signal intensity on T, imaging from those who did not. METHODS Twenty-two consecutive patients were studied ret- rospectively who had MRI during or following radi- ation therapy to the spine. MRI was performed on a 0.5 T magnet (Picker International, Highland Heights, Ohio). Typical imaging parameters were T, : 550 msec TR, 26 msec TE with four repetitions in a 256’ matrix; Tz: 2000 msec TR, 100 msec TE or 1500 msec TR, 60 msec TE with four repetitions in a 1282 matrix with the latter T, technique for better spinal cord visualization. Sections were 5 mm thick in the sagittal plane. Suspicious areas were imaged axially with T, weighting. The medical and radiation therapy records of these patients were reviewed. Data were collected concerning age, sex, radiation port area(s), and the dates and doses of radiation therapy received. The MRI images were studied for radiation effects as well as for focal lesions suggestive of metastatic deposits. RESULTS Twenty-two patients, 9 male and 13 female, age ranges 5-75, were studied. Spinal imaging was per- formed in 19, pelvic imaging in 5; two patients had MRI of both areas. Twenty-one of twenty-two pa- tients had malignant involvement of the vertebrae. zyxwvutsrqpo RECEIVED 10/7/87; ACCEPTED 1 l/.5/87. Presented at the 5th Annual meeting of the Society for Magnetic Resonance Imaging, San Antonio, TX, March 1987. Address correspondence to Peter A. Remedios Department, of Radiology, Saint John’s Hospital and Health Center, 1328 22nd Street, Santa Monica, CA 90404. zyxwvutsrqponmlkjihgfedcba 301