[Study of spondylodiscitis with magnetic resonance. Use of gadolinium-DTPA]

Eighteen patients suffering from spondylodiscitis were examined with MRI and 22 pathologic levels were identified: 1 cervical, 6 thoracic and 15 lumbosacral. As for etiology, 4 cases were tubercular, 3 strepto-staphylococcal, 3 brucellar 1 strepto-bacteroides and 11 of unknown origin. Five patients underwent MR follow-up 4-9 months after the clinical onset. SE T1- and T2-weighed sequences were always performed and 11 patients, with 13 pathologic levels, also underwent SE T1-weighted sequences with i.v. Gd-DTPA administration. No correlation could be made between the signal intensity of the involved vertebral bodies and disks and etiology. On SE T1-weighted sequences 91% of the vertebral bodies were hypointense, whereas the intervertebral disks were hypointense in 55% of cases and isointense in 45%. On SE T2-weighted sequences, the vertebrae appeared hyperintense in 65% of the cases while the disks were hyperintense in 55% and isointense in 41%. After Gd-DTPA injection signal intensity increased in 100% of cases, both at the body and at the disk. In 6 cases an inflammatory infiltration of the intracanalar structures was found, whose identifications was possible, in 4 cases, only after contrast medium injection. In the follow-up of 5 cases at 4-9 months the vertebral bodies remained hypointense in 3 cases on T1-weighted sequences, while in 2 cases signal intensity tended to normalize. Signal normalized in 4 cases on T2-weighted sequences. In 2 follow-up examinations disk structures were never depicted, neither on T1- nor on T2-weighted sequences. To conclude, T1-weighted sequences appear to be more sensitive than T2-weighted sequences to demonstrate inflammatory processes in the vertebral bodies, while the two sequences are equivalent to evaluate the disks. Gadolinium-enhanced T1-weighted sequences were more sensitive than the T1- and T2-weighted ones, especially to assess the extent of inflammatory processes into the spinal canal and to evaluate their persistence. The authors suggest the use of unenhanced and enhanced SE T1-weighted sequences only at least in the follow-up of spondylodiscitis.