Mapping interictal discharges using intracranial EEG-fMRI to predict postsurgical outcomes

Abstract

Various subjective and objective methods have been proposed to identify which interictal epileptiform discharge (IED)-related EEG-functional MRI (fMRI) results are more likely to delineate seizure-generating tissue in patients with drug-resistant focal epilepsy for the purposes of surgical planning. In this intracranial EEG-fMRI study, we evaluated the utility of these methods to localize clinically relevant regions preoperatively and compared the extent of resection of these areas to postoperative outcome. Seventy patients admitted for intracranial video-EEG monitoring were recruited for a simultaneous intracranial EEG-fMRI study. For all analyses of blood oxygen level-dependent responses associated with IEDs, an experienced epileptologist identified the most clinically relevant brain activation cluster using available clinical information. The maximum cluster (the cluster with the highest z-score) was also identified for all analyses and assigned to one of three confidence levels (low, medium or high) based on the difference of the peak z-scores between the maximum and second maximum cluster (the cluster with the second highest peak z-value). The distance was measured and compared between the peak voxel of the aforementioned clusters and the electrode contacts where the interictal discharge and seizure onset were recorded. In patients who subsequently underwent epilepsy surgery, the spatial concordance between the aforementioned clusters and the area of resection was determined and compared to postoperative outcome. We evaluated 106 different IEDs in 70 patients. Both subjective (identification of the clinically relevant cluster) and objective (maximum cluster much more significant than the second maximum cluster) methods of culling non-localizing EEG-fMRI activation maps increased the spatial concordance between these clusters and the corresponding IED or seizure onset zone contacts. However, only the objective methods of identifying medium and high confidence maps resulted in a significant association between resection of the peak voxel of the maximum cluster and postoperative outcome. Resection of this area was associated with good postoperative outcomes but was not sufficient for seizure freedom. On the other hand, we found that failure to resect the medium and high confidence maximum clusters was associated with a poor post-surgical outcome (negative predictive value = 1.0, sensitivity = 1.0). Methods to identify higher confidence EEG-fMRI results are needed to localize areas necessary for good postoperative outcomes. However, resection of the peak voxel within higher confidence maximum clusters is not sufficient for good outcomes. Conversely, failure to resect the peak voxel in these clusters is associated with a poor post-surgical outcome.

Keywords: electroencephalography; epilepsy; imaging; markers; seizures; surgery.

Figure 1

Selection of the clinically relevant BOLD response from all available clusters. (A) Example of a blood oxygen level-dependent (BOLD) activation map associated with left posterior hippocampal interictal epileptiform discharges. Yellow and blue dots indicate the contacts where interictal discharges or seizure onset was observed, respectively. In this example, the Max BOLD cluster (red circle) is also the Closest BOLD cluster (yellow circle) to the interictal epileptiform discharge (IED) contacts and was selected as the clinically relevant (CR) BOLD cluster (dark green circle). The Max’ BOLD cluster (orange circle) was observed contralateral to the discharge and extends superiorly. (B) Composition of the clinically relevant BOLD cluster for all 106 IED types studied. Note, a single cluster may be identified as both the Max or Max’ cluster and Closest cluster (denoted Max and Closest cluster or Max’ or Closest cluster, respectively) or belong to only one group (denoted Max Only, Max’ Only or Closest Only). SOZ = seizure onset zone.

Figure 2

Distances between the BOLD clusters of interest and electrode contacts. (A) Example of a blood oxygen level-dependent (BOLD) activation map associated with interictal epileptiform discharges (IEDs) recorded from the right anterior and posterior hippocampus. Yellow and blue dots indicate the contacts where interictal discharges or seizure onset was seen, respectively. In this example, the Closest BOLD cluster (yellow circle) to the contacts where IEDs were seen, and not the Max BOLD cluster (red circle), was selected as the clinically relevant (CR) BOLD cluster (dark green circle). The Max’ BOLD cluster (orange circle) was located in more superior slices. Distances (indicated by white hatched arrows) were measured between the peak voxel of each BOLD cluster of interest and both the IED contacts as well as seizure onset zone (SOZ) contacts (71.5 mm to the Closest BOLD cluster, 102.0 mm to the Max BOLD cluster and 102.8 mm to the Max’ BOLD cluster). (B) Mean distance between the BOLD cluster of interest and the associated electrode contacts where IEDs were recorded. (C) Mean distance between the BOLD cluster of interest and paired (ipsilateral) SOZ contacts. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Error bars represent standard mean error.

Figure 3

Classification of maximum cluster confidence. (A) Top, middle and bottom rows show examples of low, medium and high confidence interictal epileptiform discharge (IED)-associated blood oxygen level-dependent (BOLD) activation maps, respectively. Yellow and blue dots indicate the contacts where interictal discharges or seizure onset was seen, respectively. The Maximum and Closest BOLD clusters are indicated by a surrounding dark red circle or yellow circle, respectively. (B) Mean distances from the peak voxel of the IED-associated Maximum BOLD clusters (low, medium or high confidence) to the IED (top) and seizure onset zone (SOZ) contacts (bottom). Note that higher confidence clusters are necessarily also included in the group mean calculation for the lower confidence groups. That is, the medium confidence group is a subset of the low confidence group, and the high confidence group is a subset of both the medium and low confidence groups. Significant differences are indicated by asterisks, ns = non-significant differences. Error bars represent standard error.

Figure 4

Classification of spatial concordance between the Max BOLD cluster and the area of resection. Top left: An example of a subject where the Max blood oxygen level-dependent (BOLD) cluster was fully concordant with the area of resection (i.e. peak voxel was resected) and the subject is seizure-free (Engel IA). Top right: An example of a subject where the Max BOLD cluster was fully discordant with the area of resection (i.e. no cluster was resected) and the subject continues to experience seizures (Engel IVB). Bottom left: An example of a subject where the Max BOLD cluster was partially concordant with the area of resection (i.e. peak voxel not resected, but overlap with Max BOLD cluster) and the subject is seizure-free (Engel IA). Bottom right: An example of a subject where the Max BOLD cluster is partially discordant with the area of resection (i.e. no part of the Max BOLD cluster was resected, but another less significant cluster was resected) and the subject continues to experience seizures (Engel IVC).