Disappearing Brain Lesions Following Seizures: A Critical Reappraisal
Abstract number :
2.125
Submission category :
Year :
2000
Submission ID :
1277
Source :
www.aesnet.org
Presentation date :
12/2/2000 12:00:00 AM
Published date :
Dec 1, 2000, 06:00 AM
Authors :
Maurice Dematteis, Philippe Kahane, Jean-Francois Le Bas, Alim-Louis Benabid, Grenoble Hosp, Grenoble, France.
Rationale. Reversible CT and MRI images following seizures have been repeatedly reported, sometimes with surprising clinical and radiological descriptions, suggesting that some of these images were not related to seizures. Because these transient images may represent a non invasive tool for localizing the epileptogenic zone, we looked for predicting criteria. Methods. 9 patients with MRI reversible images following seizures were selected and an extensive work-up was performed. Results. Among the 9 subjects, 5 were not epileptic (NE) and had, in addition to seizures, an other cause which could induce transient images. The remaining 4 subjects were epileptic (E) and had no other cause than seizures. Clinical and radiological features of the 2 groups were compared. Seizure frequency (> or < 5/day) preceding imaging was higher in E than in NE (p=O.O48, Fisher's exact test). Although not statistically significant, radiological features were different: in E, MRI showed extended cortical images which were single in 3/4, without white matter involvement. In NE, images were often multiple (3/5), nodular (2/5) or annular (2/5), involving also white matter with surrounding edema in 3. Interestingly, mixed cryoglobulinemia was found in 3 NE. A careful MRI follow-up showed that images totally disappeared in all E, while a marked but incomplete resolution was noticed in 3/5 NE. Follow-up duration was not different in the 2 groups (E: 62?10 vs NE: 66?4 months). Conclusion. Epileptic patients have an homogeneous MRI pattern which may correspond to seizure-related images, while images seen in NE may rather correspond to epileptogenic lesions. In view of these findings, a critical reappraisal of literature data seems in accordance with the 2 suggested mechanisms.