Dynamic time course of typical childhood absence seizures: EEG, behavior, and functional magnetic resonance imaging

Abstract

Absence seizures are 5-10 s episodes of impaired consciousness accompanied by 3-4 Hz generalized spike-and-wave discharge on electroencephalography (EEG). The time course of functional magnetic resonance imaging (fMRI) changes in absence seizures in relation to EEG and behavior is not known. We acquired simultaneous EEG-fMRI in 88 typical childhood absence seizures from nine pediatric patients. We investigated behavior concurrently using a continuous performance task or simpler repetitive tapping task. EEG time-frequency analysis revealed abrupt onset and end of 3-4 Hz spike-wave discharges with a mean duration of 6.6 s. Behavioral analysis also showed rapid onset and end of deficits associated with electrographic seizure start and end. In contrast, we observed small early fMRI increases in the orbital/medial frontal and medial/lateral parietal cortex >5 s before seizure onset, followed by profound fMRI decreases continuing >20 s after seizure end. This time course differed markedly from the hemodynamic response function (HRF) model used in conventional fMRI analysis, consisting of large increases beginning after electrical event onset, followed by small fMRI decreases. Other regions, such as the lateral frontal cortex, showed more balanced fMRI increases followed by approximately equal decreases. The thalamus showed delayed increases after seizure onset followed by small decreases, most closely resembling the HRF model. These findings reveal a complex and long-lasting sequence of fMRI changes in absence seizures, which are not detectable by conventional HRF modeling in many regions. These results may be important mechanistically for seizure initiation and termination and may also contribute to changes in EEG and behavior.

Figure 1.

EEG signal power changes abruptly at the beginning and end of seizures. Average time–frequency dynamics of spike-wave discharges in channel F7 channel from −20 s to +26 s relative to seizure onset. A total of 54 seizures (9 patients) were analyzed: 40 seizures in eight patients during CPT or RTT and 14 seizures in four patients during VFT. Analysis was performed by using short-time Fourier transform (see Materials and Methods). All ictal periods were scaled to the mean seizure duration (6.6 s); and the preictal (−20 to 0 s), ictal (0–6.6 s), and postictal (seizure end to +26 s) periods were then temporally aligned across seizures. The power spectrum was normalized to the range of [0 1] and shown in blue to red scale where the maximum power for each seizure in the range 0–25 Hz from −20 s to +26 s was defined as “1.” Power-frequency maps for all seizures were averaged across patients. The dominant frequency component of the EEG signal was 3–4 Hz, which was only observed during spike-wave discharges. Average time–frequency dynamics in other frontal channels (FP1, FP2, F3, F4, and F8) are shown in supplemental Figure 1 (available at www.jneurosci.org as supplemental material).

Figure 2.

Behavioral impairment during seizures. Percentage correct responses are shown over time (2 s time bins) before, during, and after seizures (shaded region). Data were temporally aligned to match the preictal, ictal, and postictal periods across seizures as described in the text. Performance on the more difficult CPT task declined rapidly for letters presented just before seizure onset and recovered quickly after seizures end. Impaired performance on the RTT task was more transient than on CPT, did not begin until after seizure onset, and was less severely impaired during seizures than the CPT task (F = 15.3, p = 0.017; ANOVA). Results are based on a total of 53 seizures in eight patients: 41 seizures in five patients during CPT and 12 seizures in four patients during RTT. Fluctuations in baseline with CPT appear due to variable and small sample sizes in some CPT bins (e.g., −12 to −10 s bin), since target letters appear on average only once per 4 s with CPT, while they appear every second with RTT.

Figure 3.

Thalamic increases and “default mode” cortical decreases are the most prominent changes seen with conventional HRF modeling in SPM. fMRI increases (warm colors) and decreases (cool colors) are shown resulting from group analysis with second-level random-effects analysis, FDR-corrected height threshold p < 0.05, and extent threshold k = 3 voxels (voxel dimensions = 2 × 2 × 2 mm). Functional data are superimposed on the Montreal Neurological Institute brain template “colin27” (single_subj_T1 in SPM2) displayed in radiological right–left convention. In total, 54 seizures in nine patients (40 in 8 patients during CPT or RTT; 14 in 4 patients during VFT, 3 patients with both CPT/RTT and VFT runs) were analyzed using GLM with canonical HRF in SPM2. The dataset in this analysis was the same as Figure 1. fMRI increases were seen in bilateral thalamus, occipital (calcarine) cortex, and to a lesser extent the midline cerebellum, anterior and lateral temporal lobes, insula, and adjacent to the lateral ventricles. fMRI decreases were seen in the bilateral lateral parietal, medial parietal, and cingulate cortex and basal ganglia.

Figure 4.

Early and late fMRI changes in cortical–subcortical networks associated with absence seizures Percentage fMRI signal changes are shown from group analysis of 51 seizures in eight patients: 37 seizures in seven patients during CPT or RTT and 14 seizures in four patients during VFT (3 patients with seizures during both CPT/RTT and VFT runs). fMRI percentage change increases (warm colors) and decreases (cool colors) are shown, with a display threshold of 0.5%. The ictal time period of seizures was scaled to 6.6 s (mean seizure duration), and the preictal, ictal, and postictal time periods temporally aligned across all seizures. Early fMRI signal increases were seen well before seizure onset (0 s) in medial orbital frontal (OF), frontal polar (FP), cingulate (CG), lateral parietal (LP), precuneus (PC), and lateral occipital (LO) cortex. After seizure onset, fMRI increases progressed to also involve lateral frontal (LF) and temporal (LT) cortex. Following the end of seizures, fMRI increases were seen in the medial occipital (MO) cortex, and lastly in the thalamus (Th). fMRI signal decreases occurred later, and continued well after seizure end. Decreases were seen observed initially in medial/orbital frontal, cingulate, medial, and lateral parietal cortex, followed by decreases in lateral frontal, temporal, and occipital cortex and basal ganglia. See also supplemental Figure 2 and the supplemental video (available at www.jneurosci.org as supplemental material) for more detailed views.

Figure 5.

fMRI time courses during absence seizures differs substantially from model HRF in most regions. Mean fMRI time courses in seven anatomical VOIs (for regions, see supplemental Fig. 5, available at www.jneurosci.org as supplemental material) are shown from −20 s to +26 relative to the seizure onset. Red and blue curves represent the positive and negative HRF models obtained by convolving a boxcar function of 6.6 s (mean seizure duration, shown in gray) with the canonical HRF of SPM2. Green dashed lines are SD. Data are the same as in Figure 4, and were obtained from group analysis of 51 seizures in eight patients.