Advances in MRI for 'cryptogenic' epilepsies

Nearly one-third of patients with focal epilepsy experience disabling seizures that are refractory to pharmacotherapy. Drug-resistant focal epilepsy is, however, potentially curable by

surgery. Although lesions associated with the epileptic focus can often be accurately detected by MRI, in many patients conventional imaging based on visual evaluation is unable to pinpoint

the surgical target. Patients with so-called cryptogenic epilepsy represent one of the greatest clinical challenges in many tertiary epilepsy centers. In recent years, it has become

increasingly clear that epilepsies that are considered cryptogenic are not necessarily nonlesional, the primary histopathological substrate being subtle cortical dysplasia. This Review

considers the application of new advances in brain imaging, such as MRI morphometry, computational modeling and diffusion tensor imaging. By revealing dysplastic lesions that previously

eluded visual assessments, quantitative structural MRI methods such as these have clearly demonstrated an increased diagnostic yield of epileptic lesions, and have provided successful

surgical options to an increasing number of patients with 'cryptogenic' epilepsy.

In drug-resistant epilepsy, the most important predictor of favorable surgical outcome is complete resection of the lesion detected by MRI

Epilepsies that are considered cryptogenic are not necessarily nonlesional

The most common histopathological finding in cryptogenic epilepsy is focal cortical dysplasia

Quantitative structural image analysis could reveal dysplastic lesions in patients considered to have cryptogenic epilepsy on the basis of visual evaluation by MRI

The continued development of new imaging modalities and computational methods aimed at revealing the so-called nonlesional epilepsies will enable surgery in an increasing number of patients

A. Bernasconi and N. Bernasconi wrote the article, and contributed to the editing and reviewing of the article. All the authors provided contributions to discussions of the content, and A.

Bernasconi, B. C. Bernhardt and D. Schrader researched data for the article.

Anyone you share the following link with will be able to read this content: