MRI-negative refractory partial epilepsy: Role for diffusion tensor imaging in high field MRI

Summary

Objective

Our aim is to use the high field MR scanner (3 T) to verify whether diffusion tensor imaging (DTI) could help in locating the epileptogenic zone in patients with MRI-negative refractory partial epilepsy.

Method

Fifteen patients with refractory partial epilepsy who had normal conventional MRI, and 40 healthy volunteers were recruited for the study. DTI was performed on a 3 T MR scanner, individual maps of mean diffusivity (MD) and fractional anisotropy (FA) were calculated, and Voxel-Based Analysis (VBA) was performed for individual comparison between patients and controls.

Result

Voxel-based analysis revealed significant MD increase in variant regions in 13 patients. The electroclinical seizure localization was concurred to seven patients. No patient exhibited regions of significant decreased MD. Regions of significant reduced FA were observed in five patients, with two of these concurring with electroclinical seizure localization. Two patients had regions of significant increase in FA, which were distinct from electroclinical seizure localization.

Conclusion

Our study's results revealed that DTI is a responsive neuroradiologic technique that provides information about the epileptogenic areas in patients with MRI-negative refractory partial epilepsy. This technique may also helpful in pre-surgical evaluation.

Introduction

About a third of patients with focal epilepsy are refractory to antiepileptic drugs (Kwan and Brodie, 2000). For these patients, the identification of the epileptogenic zone is of the most pressing interest because this can directly impact the possibility of surgical resection of the focus. Surgical treatment of patients without any obvious abnormalities on cerebral imaging is generally associated with a less favorable outcome (Cascino et al., 1992). In the last few decades, MR imaging techniques have been becoming increasingly important in localizing the seizure focus and in managing patients in a noninvasive manner (Hajek et al., 2003). However, in about 20–30% patients with refractory partial epilepsy, apparent lesions that are responsible for seizure onset cannot be found in conventional MRI (Duncan, 1997). These “MRI-negative” patients consist of an important subgroup for clinical treatment. Therefore, much effort has been made to accurately identify focal abnormalities prior to possible epilepsy surgery in these patients. Consequently, delineation of the epileptogenic zone through intracranial recording is usually required, but it is perceived to be more difficult in MRI-negative cases.

Diffusion tensor imaging (DTI) is a new MRI technique that has been recently applied in locating the epileptogenic zone in patients with epilepsy (Arfanakis et al., 2002, Assaf et al., 2003, Dumas de la Roque et al., 2005, Thivard et al., 2005, Thivard et al., 2006, Yu et al., 2006). This technique provides three-dimensional information about tissue water diffusion in each image voxel, offering a new method to quantify not only the magnitude of water diffusivity (mean diffusivity, MD) but also the extent to which diffusion has directionality (fractional anisotropy, FA) in vivo (Pierpaoli et al., 1996). Several reports referred to the role of DTI in locating epileptogenic zone in different types of epilepsy (Eriksson et al., 2001, Rugg-Gunn et al., 2001). In a large number of patients with acquired non-progressive cerebral lesions and partial seizures, DTI abnormalities were identified. The findings showed that most of these abnormalities concurred with those identified on the visual inspection of conventional MRI (Rugg-Gunn et al., 2001). Another study by Eriksson et al. (2001) found areas of reduced anisotropy in 17 out of 22 patients, and areas of increased diffusivity in 10 out of 22 patients with malformations of cortical development (MCD). They also noticed that diffusion alterations were found not only within the areas of visible MCD but also in some instances beyond them. Diffusion abnormalities were also reported in hippocampal malformation (Assaf et al., 2003, Salmenpera et al., 2006) and bilateral limbic diffusion abnormalities (Concha et al., 2005) of temporal lobe epilepsy (TLE), in the subtle cortical malformation of frontal lobe epilepsy (Okumura et al., 2004), as well as in lesions of late post traumatic epilepsy (Gupta et al., 2005). These studies suggested that DTI can assist in the lateralization of the epileptogenic zone in patients with apparent lesions.

Unfortunately, only a few studies addressed the use of DTI in MRI-negative patients with epilepsy, and the results are said to be inconsistent. Rugg-Gunn et al. found diffusion abnormalities only in 27% patients (8 out of 30 patients) with “cryptogenic” partial seizure. In addition, six out of eight patients who have diffusion abnormalities and increased diffusivity corresponded to epileptiform abnormalities (Rugg-Gunn et al., 2001). The same group also reported an abnormal diffusion in the right frontal lobe of a patient with refractory epilepsy. After surgically resecting this area, the patient had a good clinical outcome, suggesting the significant role of DTI in identifying occult epileptogenic cerebral lesions (Rugg-Gunn et al., 2002). In another study, 16 patients who had partial seizure were studied. The results showed that 9 out of the 14 MRI-negative patients (64.3%) had areas of diffusion alterations, which were consistent with intracerebral EEG localization (Thivard et al., 2006). Furthermore, the inconsistency of previous studies is partly due to the heterogeneity of patient groups and limited techniques, namely, low resolution and signal-to-noise ratio on a relative low field MR system (1.5 T). With the development of the high field MRI technique, a MR scanner can now provide a better estimate of the diffusion tensor by improving the resolution and signal-to-noise ratio (Alexander et al., 2006). To our knowledge, no study has been conducted to evaluate the diffusion abnormalities of MRI-negative patients who suffer from partial refractory epilepsy, through a high field MR scanner (3 T).

Thus, the objectives of the present study are to use the high field MR scanner (3 T) and to verify whether DTI could help in locating the epileptogenic areas in patients with MRI-negative refractory partial epilepsy.