- Email: [email protected]
Continuous EEG–fMRI in patients with partial epilepsy and focal interictal slow-wave discharges on EEG
Available online at www.sciencedirect.com
Magnetic Resonance Imaging 26 (2008) 1089 – 1100
Continuous EEG–fMRI in patients with partial epilepsy and focal interictal slow-wave discharges on EEG Paolo Manganotti a , Emanuela Formaggio a , Anna Gasparini c , Roberto Cerini c , Luigi Giuseppe Bongiovanni b , Silvia Francesca Storti a , Roberto Pozzi Mucelli c , Antonio Fiaschi a,d , Mirko Avesani a,⁎ a
Dipartimento di Scienze Neurologiche e della Visione–Sezione di Neurologia Riabilitativa, University of Verona, 37134 Verona, Italy b Dipartimento di Scienze Neurologiche e della Visione–Sezione di Neurologia Clinica, University of Verona, Verona 37134, Italy c Dipartimento di Scienze Morfologico-biomediche–Sezione di Radiologia, University of Verona, Verona 37134, Italy d I.R.C.S.S. Ospedale S. Camillo, Venezia, Italy Received 30 July 2007; revised 13 February 2008; accepted 21 February 2008
Abstract Purpose: To verify whether in patients with partial epilepsy and routine electroenecephalogram (EEG) showing focal interictal slow-wave discharges without spikes combined EEG–functional magnetic resonance imaging (fMRI) would localize the corresponding epileptogenic focus, thus providing reliable information on the epileptic source. Methods: Eight patients with partial epileptic seizures whose routine scalp EEG recordings on presentation showed focal interictal slowwave activity underwent EEG–fMRI. EEG data were continuously recorded for 24 min (four concatenated sessions) from 18 scalp electrodes, while fMRI scans were simultaneously acquired with a 1.5-Tesla magnetic resonance imaging (MRI) scanner. After recording sessions and MRI artefact removal, EEG data were analyzed offline. We compared blood oxygen level-dependent (BOLD) signal changes on fMRI with EEG recordings obtained at rest and during activation (with and without focal interictal slow-wave discharges). Results: In all patients, when the EEG tracing showed the onset of focal slow-wave discharges on a few lateralized electrodes, BOLD-fMRI activation in the corresponding brain area significantly increased. We detected significant concordance between focal EEG interictal slowwave discharges and focal BOLD activation on fMRI. In patients with lesional epilepsy, the epileptogenic area corresponded to the sites of increased focal BOLD signal. Conclusions: Even in patients with partial epilepsy whose standard EEGs show focal interictal slow-wave discharges without spikes, EEG– fMRI can visualize related focal BOLD activation thus providing useful information for pre-surgical planning. © 2008 Elsevier Inc. All rights reserved. Keywords: Partial seizures; Symptomatic epilepsy; EEG; fMRI; Coregistration; BOLD
1. Introduction Many patients with drug-resistant focal seizures cannot undergo epilepsy surgery because standard magnetic resonance imaging (MRI) scans fail to visualize a clear epileptic source. Localizing the epileptogenic zone is difficult for several reasons. First, the spatial relationship between the epileptic source and irritative and ictal onset zones may be hard to establish, even if they are closely spatially related ⁎ Corresponding author. Tel.: +39 045 8124768; fax: +39 045 8124873. E-mail address: [email protected] (M. Avesani). 0730-725X/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.mri.2008.02.023
[1,2]. Second, seizures are difficult to assess inside the scanner. And third, epileptic discharges may be too weak to elicit MRI changes that will localize the epileptic zone, hence the interest in highly sensitive diagnostic techniques that can record interictal epileptiform discharges (IEDs) and concurrently visualize on neuroimaging the corresponding irritative area thus reliably identifying brain areas where seizures arise and those to which they spread. A noninvasive diagnostic technique that provides valuable information for localizing the brain regions generating interictal epileptiform activity is simultaneous electroenecephalogram (EEG) recording and functional magnetic
1090
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 1. Patient 1. EEG tracing during continuous EEG–fMRI. Note the irritative focus in T3-T5.
resonance imaging (fMRI). The first EEG–fMRI studies using EEG-triggered scanning [3,4] confirmed that EEG– fMRI can noninvasively identify brain activation associated with subclinical IEDs at a high spatial resolution but achieved low diagnostic sensitivity (60%). Later studies [5–8] showed that continuous EEG–fMRI acquisition is better than EEG-triggered recording because it allows offline analysis after artefact subtraction and provides better sensitivity (80%). Others used EEG–fMRI to analyze the various epileptiform syndromes [9–23]. Additional studies using stereotaxic EEG (sEEG) and electrocortical mapping [3,22,24] confirmed the colocalization between IEDs and fMRI activation. All these studies, considering patients whose interictal activity differed in frequency, suggested that the limiting factor for significant blood oxygen-level dependent (BOLD) activation is the frequency of IEDs. Although two studies proposed one IED per minute as the minimum value needed to obtain significant cerebral blood flow activation [3,4] neither of them studied slow-wave IEDs. Slow-wave IEDs have been addressed only in two single case reports of simultaneous EEG–fMRI in patients with focal interictal slow-wave activity [25,26]. In the first paper [25] the investigators used sEEG to study a patient with absence seizures and confirmed the colocalization between IEDs and fMRI activation.
1.1. Purpose Prompted by these previous EEG findings [25,26], we designed this study primarily to verify whether the interictal slow waves originating from the EEG irritative focus are sufficient to increase cerebral blood flow in a spatially related brain area. To do so, using EEG–fMRI, we investigated BOLD responses to IEDs characterized by focal interictal slow-wave activity in patients with partial epilepsy. We also aimed to understand whether in patients with lesional epilepsy, fMRI-BOLD activation corresponded to the epileptogenic area previously identified with standard fMRI. We then investigated whether EEG– fMRI could document the spatial relationship between the irritative zone and ictal onset zone in patients with interictal focal slow-wave discharges arising from a known epileptogenic lesion.
2. Materials and methods 2.1. Patients To obtain a homogeneous study sample, from among patients hospitalized in our epilepsy ward, we selected eight patients with partial epilepsy whose routine EEG recordings
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1091
Fig. 2. Patient 1. EEG tracing during continuous EEG–fMRI. Note the irritative focus in T3-T5.
showed frequent focal (IEDs) manifesting as 5–6 Hz focal EEG activity (slow waves, slow spike waves, highamplitude slow waves) over a few lateralized electrodes. Of the eight patients selected for study, three had a structural lesion on MRI: two had hippocampal atrophies secondary to mesial temporal sclerosis (MTS) and one had a cavernous angiomas. One patient had cryptogenic epilepsy. Finally, four patients had nonlesional (idiopathic) epilepsy. In one patient with MTS (Patient 4), the patient with cryptogenic epilepsy (Patient 2) and the one with idiopathic epilepsy (Patient 7), routine EEG showed slow-wave activity; in the patient with a cavernoma (Patient 3) and the two patients with idiopathic epilepsy (Patients 5 and 8) they showed slow spike waves; and in the other patients with idiopathic epilepsy (Patient 6) and the second patient with MTS
(Patient 1), they showed high-amplitude slow-waves (N130 μV) (Table 1). 2.2. EEG Recordings and analysis During fMRI scanning, EEG signals were continuously recorded through 18 MRI-compatible silver/silver chloride (Ag/AgCl) electrodes placed on the scalp according to the 10/20 system (Fp1, F3, Fz, F7, F4, F8, C3, Cz, C4, P3, Pz, P4, T3, T5, T4, T6, O1, O2). The magnetic resonance (MR)compatible Ag/AgCl-coated electrodes (8 mm in diameter, 0.5-mm thickness) had 2-mm slits to interrupt eddy currents. Each electrode was connected to an input of an MRcompatible differential amplifier, which amplified the voltage between each electrode and a reference electrode (common reference lead; range chosen to avoid saturating
Table 1 Patients enrolled in the study Patient
Epilepsy form
Standard MRI findings
IEDs morphology
Irritative focus
1 2 3 4 5 6 7 8
Lesional Cryptogenic Lesional Lesional Idiopathic Idiopathic Idiopathic Idiopathic
MTS Negative Cavernoma MTS Negative Negative Negative Negative
h.a.s.w. s.w. sp.s.w. s.w. sp.s.w. h.a.s.w. s.w. h.a.sp.s.w.
T3-T5 T3-T5 T5 T4-T6 O1 T4-T6 F7 T4-T6
s.w., slow waves; h.a.s.w., high amplitude slow waves; sp.s.w., spiked slow waves; h.a.sp.s.w., high amplitude spiked slow waves.
1092
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 3. Patient 1. fMRI BOLD activations in the left mesial temporal lobe.
the EEG amplifier: −12.8 and 12.8 mV; bandwidth 0.15– 269.5 Hz), the signal was then input to a computer (software Brain-Quick System Plus, Micromed). Two electrodes were used to record the electrocardiogram (ECG) and electromyogram (EMG) to remove pulse and movement artifacts. The EMG electrode was placed on the right abductor pollicis brevis muscle and the other (ECG) on the precordial area. The MRI-approved amplifier was placed at the end of the magnet bore, while the recording computer was placed outside the magnet room and connected via a fiberoptic cable. To minimize variability in the EEG artefacts due to wire movement caused by mechanical vibration during MRI acquisition, lead wires were immobilized on foam pads. Details of the EEG recording method are given in Bènar et al [27]. An anterior to Fz electrode was used as reference, and a posterior to Fz electrode was used as the ground as detailed in other studies using the same recording system [28]. The EEG artifact induced by the gradient magnetic field was digitally removed offline by an adaptive filter (Micromed). After the raw EEG data were reviewed, a marker was placed at the end of the last volume. The remaining markers were then placed automatically and equidistant to each other.
The data were then shifted in the temporal domain to align each volume to the first one. Finally, average slice and volume artefacts were computed and subsequently subtracted from the data, thus yielding clean EEG signals. EEG artifacts associated with pulsatile blood flow were digitally removed offline using a simple averaging procedure [29,30]. An artifact template was first created from all the occurrences of the cardioballistographic artifact by means of averaging and, afterwards, subtracted from each occurrence. Subsequently, a single electroencephalographer visually reviewed the filtered EEGs and marked the time of onset and duration of each IED. Automatic IED identification was avoided. 2.3. fMRI Acquisition and analysis Images were obtained with a 1.5-Tesla MRI scanner (Vision, Siemens, Erlangen, Germany). At the start of each study, a T1-weighted anatomical MRI was acquired [192 slices; field of view=256×256, scanning matrix 512×512, slice thickness 1 mm, sagittal slice orientation, echo time (TE)=3ms, repetition time (TR)=1990 ms]. All patients then underwent a 24-min fMRI recording session (four
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1093
Fig. 4. Patient 3. EEG tracing during continuous EEG–fMRI. Note the irritative focus in the left temporal region.
concatenated sessions) after giving informed consent. BOLD fMRI data were acquired, using a standard gradient-echo (echo planar imaging) sequence, on the axial orientation, in four runs of 6 min with the patient in the resting state, as described by the Kobayashi team [20] (voxel dimension 3×3×3 mm; 36 slices; matrix 64×64; TE=50 ms, TR=3.7 s; slice thickness=3 mm). For image processing and statistical analysis of the fMRI time series data, we used BrainVoyager QX software (Brain Innovation, Maastricht, Netherlands). Preprocessing of the functional MRI included three-dimensional (3D) motion correction, slice scan time correction (sync interpolation), linear trend removal by temporal highpass filtering (three cycles in time course) and transformation into the Talairach coordinate space. In each subject, activated voxels were identified with a single-subject general linear model approach for time series data [31] To account for the hemodynamic delay, the boxcar waveform representing the rest and task conditions was convolved with an empirical hemodynamic response function [32].
identified the single region of condition-associated BOLD signal changes with a statistical threshold based on the amplitude (Pb.05, corrected for multiple comparisons: Bonferroni test) and extent of the regions of activation. The location of voxels with maximal signal increase was expressed in terms of x, y and z in the Talairach space, and activation volumes were expressed in terms of number of activated voxels. Positive BOLD-fMRI responses were defined as activations. Significant responses were defined as almost five contiguous voxels with Pb.05 over at least two contiguous slices [3,4,11] in two-dimensional (2D) reconstruction. The anatomic localization of BOLD responses was determined by coregistration of the anatomic data and statistical t maps. We analyzed the extent and maximum fMRI response for each study, considering all areas with significative activations. We also determined the locations of maximum activation based on the maximum peak response (maximum t value).
2.4. Statistical analysis of fMRI data A t statistic was used to determine significance on a voxel-by-voxel basis and correlation values were transformed into a normal distribution (Z statistic). The results were displayed on parametric statistical maps in which the pixel Z value is expressed on a colorimetric scale. We
3. Results 3.1. EEG–fMRI studies and bold responses No clinical or EEG manifestations of ictal events developed during scanning session. None of the patients reported
1094
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 5. Patient 3. fMRI-BOLD activations in the left temporal cortex around the poro-encephalic cavity (site of previous surgical treatment of the cavernoma).
discomfort or other adverse reactions related to the EEG recording procedure. In all patients, EEG showed unilateral focal activity during the EEG–fMRI session. IEDs recorded inside the scanner had a localization, amplitude and morphology similar to those in previous routine EEG recordings. In two patients (Patient 1— see Figs. 1–3—and Patient 6), EEG recordings showed focal high-amplitude slow waves and in the other two patients (Patients 2 and 4) slow-wave discharges over the temporal electrodes. In another patient (Patient 7), we detected slowwave activity in the extratemporal (F7) region. In the patient with a cavernoma (Patient 3—see Figs. 4 and 5), focal EEG activity was characterized by focal slow-spiked wave activity, reaching maximal amplitude over the left temporal electrodes (T5). In the other two patients (Patients 5 and 8) we detected slow-spiked waves; in Patient 5 (see Figs. 6–9), in the extratemporal (F3) region and in Patient 8 (see Figs. 10–12), in the temporal region (T4–T6). In Patient 8, EEG tracings (Fig. 10) also showed high-amplitude spiked slow wave activity. The mean frequency was 2.4 IEDs per minute (S.D. 0.17). In all patients, fMRI analysis showed a significant focal BOLD activation in a single activated area related to the EEG irritative focus (Table 2).
3.2. Activation sites In all patients studied, fMRI BOLD activation (Table 3) (fMRI-3D reconstruction) corresponded to the irritative focus on standard EEG recordings. 3.2.1. Temporal lobe activations In six of the eight patients, focal BOLD signal changes reached statistical significance in the temporal regions. In five patients (the patients with MTS, Patients 1 and 4; the patient with cryptogenic epilepsy, Patient 2; in two of the four patients with idiopathic epilepsies, Patients 6 and 8), the significant BOLD changes were located in the mesial temporal lobe, and in one, the patient with a cavernoma (Patient 3) in the neocortical regions (laterally and posteriorly to the resection margins). 3.2.2. Extratemporal activations In two patients fMRI showed extratemporal activation, located in one (Patient 5) in the left occipital and in the other (Patient 7) in the left frontal lobe. The occipital activation corresponded closely to the patient's clinical manifestations and to standard EEG. This patient was admitted to our ward for investigation of hallucinations triggered by eye opening. A standard EEG
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1095
Fig. 6. Patient 5. EEG tracing, during continuous EEG–fMRI, showing an irritative focus in the left temporo-occipital region.
disclosed a left temporal–occipital irritative focus that spread anteriorly when she opened the eyes. When the EEG showed the spread, a polygraphic study showed a right-beating nystagmus and head deviation to the right side. Continuous EEG–fMRI (Fig. 7) demonstrated significant BOLD signal activation in the associative area around the calcarine cortex (Brodmann area:18).
In the patient in whom fMRI showed BOLD activation in the left frontal region (Brodmann area 6), standard EEG showed a left frontotemporal irritative focus, characterized by frequent slow-spiked waves. A recent polygraphic study (recorded during Stage I and II non-rapid-eye-movement sleep) showed polymorphic slow waves (theta-delta) in the left frontal regions, with phase inversion at F7 and without
Fig. 7. Patient 5. fMRI-BOLD activation in Brodmann area 18 in the patient who had an irritative focus in the left temporo-occipital lobe.
1096
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 8. Patient 5. fMRI-BOLD activation in Brodmann area 18 in the patient who had an irritative focus in the left temporooccipital lobe.
spread. These findings suggested that the patient was a good candidate for an EEG–fMRI study, and during the EEG– fMRI sessions described here we confirmed the fMRI activation area within the precentral gyrus. 4. Discussion The main finding in this EEG–fMRI study of patients with partial epilepsy is that focal interictal slow-wave activity was invariably associated with increased focal fMRI-BOLD activation responses in a spatially related brain area. Our study therefore extends current knowledge on epileptic foci localization and confirms previous reports suggesting that EEG–fMRI BOLD activation associated with modelled slow activity might have a role in localizing the epileptogenic region even in the absence of clear interictal spikes [25,26]. All the eight patients with partial epilepsy we enrolled in this study had frequent interictal focal slow-wave activity on standard EEG. In all continuous EEG–fMRI recording sessions, after fMRI artefact removal, we obtained a good-
Fig. 9. Patient 5. fMRI-BOLD activation in Brodmann area 18 in the patient who had an irritative focus in the left temporooccipital lobe.
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1097
Fig. 10. Patient 8. EEG during recording session. Irritative focus on right temporal regions.
quality EEG that allowed us to detect spontaneous IEDs and analyze the related fMRI BOLD activation. The EEG recording left the quality of fMRI almost undistorted, and in all patients, the focal activity seen in the concurrent EEG was associated with a focal increase in the MRI signal. In their focal distribution, these BOLD activations resembled the focal IEDs seen on routine scalp EEG and EEG recorded during EEG–fMRI sessions. An interesting finding came from the patients with lesional epilepsy. These patients, whose standard MRI documented a lesion and whose standard EEG identified an irritative focus, are the ideal candidates for verifying a possible spatial relationship between the epileptogenic and irritative focus [1,2]. In the patient who had undergone surgery to remove a cavernoma (Patient 3), the EEG–fMRI study, by localizing the irritative focus and linking it to fMRI as “active state,” showed a significant BOLD activation signal closely related to the poroencephalic cavity (a residual of previous treatment). Even if this focal BOLD activation presumably arose from a blood vessel (residual cavernoma), undoubtedly, it was obtained by a protocol study linking an active fMRI state to IEDs on EEG. To clarify the relationship, we therefore decided to reach, progressively, a very high specificity level (Pb.0001), as demonstrated in iconography. Even with these high specificity values, the BOLD activation in that site persisted.
Another new finding is the BOLD activation we detected on fMRI in the two patients with MTS. In contrast to others, who studied a series of patients with MTS (five studied with continuous coregistration) [8] and found no significant activation, in two of the eight patients studied (Patient 1 and 4), we detected significant BOLD activation. These interesting results suggest a possible role of simultaneous EEG– fMRI in disclosing focal activation even in the mesial temporal cortex. This cortical area is notoriously difficult to study with standard methods because the deep localization of the irritative area often makes spikes smoother and therefore harder to recognize in recordings from standard EEG scalp electrodes than in sEEG. Hence, IEDs presenting as slowwave discharges could be useful in determining significant BOLD activation in a corresponding area, as previously suggested by Laufs et al. [25]. Useful information that may help to understand BOLD responses in the various forms of epilepsy came also from studying activation in extratemporal IEDs. In the two patients with extratemporal discharges (Patients 5 and 7), we noted a good correlation between clinical–polygraphic data and BOLD activation during fMRI. No significant difference was found between activation in the frontal or occipital lobe. In both zones, after focal IEDs, BOLD activation increased. Why we obtained such good results (8/8 activation — significant concordance between EEG and fMRI data)
1098
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 11. Patient 8. fMRI-BOLD activations in right temporal lobe during recording session.
remains open to question. The most plausible reason is that during enrolment, to obtain the largest possible percentage of activations, we explicitly selected patients whose standard EEGs showed a high IED firing rate, confirmed on EEG during the scanning session. The mean frequency of IEDs in the eight patients was 2.4/min, S.D. 0.17 (Table 2), considerably higher than the one IED per minute Duncan et al. [3] considered as the minimum to obtain a focal BOLD activations. Collectively, these findings confirm the importance of IED firing rates in EEG–fMRI. Possibly the most interesting finding in this study is that conversely, morphology seems less important than IED firing rates in triggering EEG–fMRI-BOLD activation: of the 8 patients we studied, 5 had pure slow-waves on standard EEG (3/8 with high amplitude and 2/8 with a normal voltage) and 3 a slow spike-waves discharges (one with high amplitude). Although these slow spike-wave discharges might have originated from a spike focus smoothed by filtering (unlikely because EEG detected the same IEDs before patients entered the magnet room), we found no difference in the statistical significance of BOLD activation between the two groups. In particular, slow spike-wave IEDs
Fig. 12. Patient 8. fMRI-BOLD activations in right temporal lobe during recording session.
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1099
Table 2 Number of IEDs revealed on EEG during recording session and BOLD activation in 2D-fMRI, with number of contiguous activated slices 1
16
45
61
2.54
2
13
42
55
2.29
3
14
40
54
2.25
4
15
43
58
2.41
5
17
47
64
2.66
6
13
38
51
2.13
7 8
14 16
44 46
58 62
2.41 2.58
Mean S.D.
14.75 1.49 A (IEDs/6I)
43.12 3.04 B (IEDs/18I)
57.88 4.54 A+B (IEDs/24I)
2.41 0.18 C (IEDs/min)
Left mesial temporal lobe Left mesial temporal lobe Left superior temporal lobe — neocortical region Right mesial temporal lobe Left occipital lobe — calcarin cortex Right superior temporal lobe Left frontal lobe Right superior temporal lobe
D (BOLD activation in 2D reconstruction)
2 2 3
2 2
2 2 2
E (n. slices)
A, IEDs recorded during the first concatenated session (lasting 6 min); B, IEDs recorded during the following three concatenated sessions (18 min); A+B, Total IEDs recorded during the four concatenated sessions (24 min); C, IEDs per minute with mean value and S.D.; D, BOLD activation in 2D reconstruction; E, number of contiguous slices.
were no more efficient than slow-wave IEDs in eliciting significant BOLD activation. Hence, we agree as previously suggested [25,26] that slow-wave IEDs, like spikes, can Table 3 Number of activated voxels, coordinates (mean and S.D.) and corresponding localization, with Talairach system, of activated regions in 3D reconstruction Mean “x” Patient –18±1.7 1
Mean “y”
Mean “z”
–7±1.8 –19±1.3
Patient –14±2.3 –33±3.9 2
No. of Corresponding area voxels 177
–6±2.4
522
Patient –59±2.3 –41±2.4 –15±3.6 3
540
–14±1.5 –15±1.2
175
Patient 4
17±3
Patient 5
–3±2.1 –78±1.8
2±2.7 1227
Patient 6
62±3.8 –15±3.1
7±1.7
364
13±3.4
442
–3±2.8
422
Patient –51±3.7 7 Patient 8
–3±4
54±1.6 –36±2.7
Left Cerebrum, limbic lobe, parahippocampal gyrus, Brodmann area 34 Left cerebrum, limbic lobe, parahippocampal gyrus, Brodmann area 30 Left cerebrum, temporal lobe, inferior temporal gyrus, Brodmann area 20 Right cerebrum, limbic lobe, parahippocampal gyrus, Brodmann area 28 Left cerebrum occipital lobe, lingual gyrus, Brodmann area 18 Right cerebrum, temporal lobe, transverse temporal gyrus, Brodmann area 42 Left cerebrum, frontal lobe, precentral gyrus, Brodmann area 6 Right cerebrum, temporal lobe, middle temporal gyrus, Brodmann area 21
elicit a significant cerebral blood flow increase in a spatially related brain area. The slow-wave discharge in five of the eight patients we studied also suggests that voltage has a minor role in determining BOLD activation: even if our study sample is limited, we found no differences in BOLD activations between IEDs with high-amplitude and normal-amplitude slow waves. The significant concordance between EEG and fMRI data (and particularly the absence of multiple activation areas) depends on the high specificity threshold our study design envisaged. In designing this study, we maintained only the single activation with the strongest specificity so that we could verify whether this activation, and this activation alone coincided with the epileptic focus previously documented by standard EEG and used as paradigm of the activation-state during fMRI analysis. The complete EEG–fMRI concordance we achieved in this study suggests that slow-wave IEDs even without spikes may be useful in activating fMRI BOLD responses during the presurgical, noninvasive evaluation of patients with partial, drug-resistant, seizures. References [1] Ebersole JS, Wade PB. Spike voltage topography identifies two types of frontotemporal epileptic foci. Neurology 1991;41:1425–33. [2] Benbadis SR, Gerson WA, Harvey JH, Luders HO. Photosensitive lobe epilepsy. Neurology 1996;46:1540–2. [3] Krakow K, Woermann FG, Symms, Allen PJ, Lemieux L, Barker GJ, et al. EEG-fMRI of interictal epileptiform activity in patients with partial seizures. Brain 1999;122:1679–88.
1100
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
[4] Krakow K, Lemieux L, Messina D, Scott CA, Symms MR, Duncan JS, et al. Spatio-temporal imaging of focal interictal epileptiform activity using EEG-triggered functional MRI. Epileptic Disord 2001;3:67–74. [5] Lemieux L, Salek-Haddadi A, Josephs O, Allen P, Toms N, Scott C, et al. Event-related fMRI with simultaneous and continuous EEG: description of the method and initial case report. Neuroimage 2001;1: 780–7. [6] Lazeyras F, Zimine I, Blanke O, Perrig SH, Seek M. Functional MRI with simultaneous EEG recording: feasibility and application to motor and visual activation. J Magn Reson Imaging 2001;13:943–8. [7] Benar CG, Gross DW, Wang Y, Petre V, Pike B, Dubeau F, et al. The BOLD responses to interictal epileptiform discharges. Neuroimage 2002;17:1182–92. [8] Al-Asmi A, Benar CG, Gross DW, Khani YA, Andermann F, Pike B, et al. fMRI activation in continuous and spike-triggered EEG-fMRI studies of epileptic spikes. Epilepsia 2003;44:1328–39. [9] Boor S, Vucurevic G, Pfleiderer C, Stoeter P, Kutschke G, Boor R. EEG-related functional MRI in benign childhood epilepsy with centrotemporal spikes. Epilepsia 2003;44:688–92. [10] Salek-Haddadi A, Lemieux L, Merschhemke M, Friston K, Duncan JS, Fish DR. Functional magnetic resonance imaging of human absence seizures. Ann Neurol 2003;53:663–7. [11] Salek-Haddadi A, Diehl B, Hamandi K, Merschhemke M, Liston A, Friston K, et al. Hemodynamic correlates of epileptiform discharges: an EEG-fMRI study of 63 patients with focal epilepsy. Brain Res 2006;1088:148–66. [12] Bagshaw AP, Aghakhani Y, Benar CG, Kobayashi E, Hawco C, Dubeau F, et al. EEG-fMRI of focal epileptic spikes: analysis with multiple haemodynamic functions and comparison with gadoliniumenhanced MR angiograms. Hum Brain Mapp 2004;22:179–92. [13] Bagshaw AP, Kobayashi E, Dubeau F, Pike GB, Gotman J. Correspondence between EEG-fMRI and EEG dipole localisation of interictal discharges in focal epilepsy. Neuroimage 2006;30:417–25. [14] Aghakhani Y, Bagshaw AP, Benar CG, Hawco C, Andermann F, Dubeau F, et al. fMRI activation during spike wave discharges in idiopathic generalized epilepsy. Brain 2004;127:1127–44. [15] Aghakhani Y, Kinay D, Gotman J, Soualmi L, Andermann F, Olivier A, et al. The role of periventricular nodular heterotopia in epileptogenesis. Brain 2005;128:641–51. [16] Gotman J, Benar CG, Dubeau F. Combining EEG and FMRI in epilepsy: methodological challenges and clinical results. J Clin Neurophysiol 2004;21:229–40. [17] Gotman J, Grova C, Bagshaw A, Kobayashi E, Aghakhani Y, Dubeau F. Generalized epileptic discharges show thalamocortical activation and suspension of the default state of the brain. PNAS 2005;142:15236–40. [18] Kobayashi E, Bagshaw AP, Jansen A, Andermann F, Andermann E, Gotman J, et al. Intrinsic epileptogenicity in polymicrogyric cortex suggested by EEG-fMRI BOLD responses. Neurology 2005;64:1263–6.
[19] Kobayashi E, Bagshaw AP, Grova C, Gotman J, Dubeau F. Grey matter heterotopia: what EEG-fMRI can tell us about epileptogenicity of neuronal migration disorders. Brain 2006;129:366–74. [20] Kobayashi E, Hawco CS, Grova C, Dubeau F, Gotman J. Widespread and intense BOLD changes during brief focal electrographic seizures. Neurology 2006;66:1049–55. [21] Kobayashi E, Bagshaw AP, Grova C, Dubeau F, Gotman J. Negative BOLD responses to epileptic spikes. Hum Brain Mapp 2006;27: 488–97. [22] Kobayashi E, Bagshaw AP, Benar CG, Aghakhani Y, Andermann F, Dubeau F, et al. Temporal and extratemporal BOLD responses to temporal lobe interictal spikes. Epilepsia 2006;47:343–54. [23] Laufs H, Lengler U, Hamandi K, Kleinschmidt A, Krakow K. Linking generalized spike-and-wave discharges and resting state brain activity by using EEG-fMRI in a patient with absence seizures. Epilepsia 2006; 47:444–8. [24] Benar CG, Grova C, Kobayashi E, Bagshaw AP, Aghakhani Y, Dubeau F, et al. EEG-fMRI of epileptic spikes: concordance with EEG source localization and intracranial EEG. Neuroimage 2006;30:1161–70. [25] Laufs H, Hamandi K, Walter MC, Scott C, Smith S, Duncan JS, et al. EEG-fMRI mapping of asymmetrical delta activity in a patient with refractory epilepsy is concordant with the epileptogenic region determined by intracranial EEG. Magn Reson Imaging 2006;24: 367–71. [26] Avesani M, Milanese F, Formaggio E, Gasparini A, Baraldo A, Cerini R, et al. Continuous EEG-fMRI in pre-surgical evaluation of a patient affected by symptomatic seizures: BOLD activation linked to interictal epileptiform discharges caused by cavernoma. Neuroradiol J 2008;21:183–91. [27] Benar RG, Aghakhani Y, Wang Y, Izenberg A, Al-Asmi A, Dubeau F, et al. Quality of EEG in simultaneous EEG-fMRI for epilepsy. Clin Neurophysiol 2003;114:569–80. [28] Gonçalves SI, de Munck JC, Pouwels PJ, Schoonhover R, Kuiker JP, Mauritis NM, et al. Correlating the alpha rhythm to BOLD using simultaneous EEG-fMRI: inter-subject variability. Neuroimage 2006; 30:203–13. [29] Allen PJ, Polizzi G, Krakow K, Fish DR, Lemieux L. Identification of EEG events in the MR scanner: the problem of pulse artifact and a method for its subtraction. Neuroimage 1998;8:229–39. [30] Allen PJ, Josephs O, Turner R. A method for removing imaging artefact from continuous EEG recorded during functional MRI. Neuroimage 2000;12:230–9. [31] Friston KJ, Holmes AP, Worsley KJ, Poline JP, Frith CD, Frackowiak RSJ. Statistical parametric maps in functional imaging: a general linear approach. Hum Brain Mapp 1995;2:173–81. [32] Boynton GM, Engel SA, Glover GH, Heeger DJ. Linear systems analysis of functional magnetic resonance imaging in human V1. J Neurosci 1996;16:4207–41.
Magnetic Resonance Imaging 26 (2008) 1089 – 1100
Continuous EEG–fMRI in patients with partial epilepsy and focal interictal slow-wave discharges on EEG Paolo Manganotti a , Emanuela Formaggio a , Anna Gasparini c , Roberto Cerini c , Luigi Giuseppe Bongiovanni b , Silvia Francesca Storti a , Roberto Pozzi Mucelli c , Antonio Fiaschi a,d , Mirko Avesani a,⁎ a
Dipartimento di Scienze Neurologiche e della Visione–Sezione di Neurologia Riabilitativa, University of Verona, 37134 Verona, Italy b Dipartimento di Scienze Neurologiche e della Visione–Sezione di Neurologia Clinica, University of Verona, Verona 37134, Italy c Dipartimento di Scienze Morfologico-biomediche–Sezione di Radiologia, University of Verona, Verona 37134, Italy d I.R.C.S.S. Ospedale S. Camillo, Venezia, Italy Received 30 July 2007; revised 13 February 2008; accepted 21 February 2008
Abstract Purpose: To verify whether in patients with partial epilepsy and routine electroenecephalogram (EEG) showing focal interictal slow-wave discharges without spikes combined EEG–functional magnetic resonance imaging (fMRI) would localize the corresponding epileptogenic focus, thus providing reliable information on the epileptic source. Methods: Eight patients with partial epileptic seizures whose routine scalp EEG recordings on presentation showed focal interictal slowwave activity underwent EEG–fMRI. EEG data were continuously recorded for 24 min (four concatenated sessions) from 18 scalp electrodes, while fMRI scans were simultaneously acquired with a 1.5-Tesla magnetic resonance imaging (MRI) scanner. After recording sessions and MRI artefact removal, EEG data were analyzed offline. We compared blood oxygen level-dependent (BOLD) signal changes on fMRI with EEG recordings obtained at rest and during activation (with and without focal interictal slow-wave discharges). Results: In all patients, when the EEG tracing showed the onset of focal slow-wave discharges on a few lateralized electrodes, BOLD-fMRI activation in the corresponding brain area significantly increased. We detected significant concordance between focal EEG interictal slowwave discharges and focal BOLD activation on fMRI. In patients with lesional epilepsy, the epileptogenic area corresponded to the sites of increased focal BOLD signal. Conclusions: Even in patients with partial epilepsy whose standard EEGs show focal interictal slow-wave discharges without spikes, EEG– fMRI can visualize related focal BOLD activation thus providing useful information for pre-surgical planning. © 2008 Elsevier Inc. All rights reserved. Keywords: Partial seizures; Symptomatic epilepsy; EEG; fMRI; Coregistration; BOLD
1. Introduction Many patients with drug-resistant focal seizures cannot undergo epilepsy surgery because standard magnetic resonance imaging (MRI) scans fail to visualize a clear epileptic source. Localizing the epileptogenic zone is difficult for several reasons. First, the spatial relationship between the epileptic source and irritative and ictal onset zones may be hard to establish, even if they are closely spatially related ⁎ Corresponding author. Tel.: +39 045 8124768; fax: +39 045 8124873. E-mail address: [email protected] (M. Avesani). 0730-725X/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.mri.2008.02.023
[1,2]. Second, seizures are difficult to assess inside the scanner. And third, epileptic discharges may be too weak to elicit MRI changes that will localize the epileptic zone, hence the interest in highly sensitive diagnostic techniques that can record interictal epileptiform discharges (IEDs) and concurrently visualize on neuroimaging the corresponding irritative area thus reliably identifying brain areas where seizures arise and those to which they spread. A noninvasive diagnostic technique that provides valuable information for localizing the brain regions generating interictal epileptiform activity is simultaneous electroenecephalogram (EEG) recording and functional magnetic
1090
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 1. Patient 1. EEG tracing during continuous EEG–fMRI. Note the irritative focus in T3-T5.
resonance imaging (fMRI). The first EEG–fMRI studies using EEG-triggered scanning [3,4] confirmed that EEG– fMRI can noninvasively identify brain activation associated with subclinical IEDs at a high spatial resolution but achieved low diagnostic sensitivity (60%). Later studies [5–8] showed that continuous EEG–fMRI acquisition is better than EEG-triggered recording because it allows offline analysis after artefact subtraction and provides better sensitivity (80%). Others used EEG–fMRI to analyze the various epileptiform syndromes [9–23]. Additional studies using stereotaxic EEG (sEEG) and electrocortical mapping [3,22,24] confirmed the colocalization between IEDs and fMRI activation. All these studies, considering patients whose interictal activity differed in frequency, suggested that the limiting factor for significant blood oxygen-level dependent (BOLD) activation is the frequency of IEDs. Although two studies proposed one IED per minute as the minimum value needed to obtain significant cerebral blood flow activation [3,4] neither of them studied slow-wave IEDs. Slow-wave IEDs have been addressed only in two single case reports of simultaneous EEG–fMRI in patients with focal interictal slow-wave activity [25,26]. In the first paper [25] the investigators used sEEG to study a patient with absence seizures and confirmed the colocalization between IEDs and fMRI activation.
1.1. Purpose Prompted by these previous EEG findings [25,26], we designed this study primarily to verify whether the interictal slow waves originating from the EEG irritative focus are sufficient to increase cerebral blood flow in a spatially related brain area. To do so, using EEG–fMRI, we investigated BOLD responses to IEDs characterized by focal interictal slow-wave activity in patients with partial epilepsy. We also aimed to understand whether in patients with lesional epilepsy, fMRI-BOLD activation corresponded to the epileptogenic area previously identified with standard fMRI. We then investigated whether EEG– fMRI could document the spatial relationship between the irritative zone and ictal onset zone in patients with interictal focal slow-wave discharges arising from a known epileptogenic lesion.
2. Materials and methods 2.1. Patients To obtain a homogeneous study sample, from among patients hospitalized in our epilepsy ward, we selected eight patients with partial epilepsy whose routine EEG recordings
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1091
Fig. 2. Patient 1. EEG tracing during continuous EEG–fMRI. Note the irritative focus in T3-T5.
showed frequent focal (IEDs) manifesting as 5–6 Hz focal EEG activity (slow waves, slow spike waves, highamplitude slow waves) over a few lateralized electrodes. Of the eight patients selected for study, three had a structural lesion on MRI: two had hippocampal atrophies secondary to mesial temporal sclerosis (MTS) and one had a cavernous angiomas. One patient had cryptogenic epilepsy. Finally, four patients had nonlesional (idiopathic) epilepsy. In one patient with MTS (Patient 4), the patient with cryptogenic epilepsy (Patient 2) and the one with idiopathic epilepsy (Patient 7), routine EEG showed slow-wave activity; in the patient with a cavernoma (Patient 3) and the two patients with idiopathic epilepsy (Patients 5 and 8) they showed slow spike waves; and in the other patients with idiopathic epilepsy (Patient 6) and the second patient with MTS
(Patient 1), they showed high-amplitude slow-waves (N130 μV) (Table 1). 2.2. EEG Recordings and analysis During fMRI scanning, EEG signals were continuously recorded through 18 MRI-compatible silver/silver chloride (Ag/AgCl) electrodes placed on the scalp according to the 10/20 system (Fp1, F3, Fz, F7, F4, F8, C3, Cz, C4, P3, Pz, P4, T3, T5, T4, T6, O1, O2). The magnetic resonance (MR)compatible Ag/AgCl-coated electrodes (8 mm in diameter, 0.5-mm thickness) had 2-mm slits to interrupt eddy currents. Each electrode was connected to an input of an MRcompatible differential amplifier, which amplified the voltage between each electrode and a reference electrode (common reference lead; range chosen to avoid saturating
Table 1 Patients enrolled in the study Patient
Epilepsy form
Standard MRI findings
IEDs morphology
Irritative focus
1 2 3 4 5 6 7 8
Lesional Cryptogenic Lesional Lesional Idiopathic Idiopathic Idiopathic Idiopathic
MTS Negative Cavernoma MTS Negative Negative Negative Negative
h.a.s.w. s.w. sp.s.w. s.w. sp.s.w. h.a.s.w. s.w. h.a.sp.s.w.
T3-T5 T3-T5 T5 T4-T6 O1 T4-T6 F7 T4-T6
s.w., slow waves; h.a.s.w., high amplitude slow waves; sp.s.w., spiked slow waves; h.a.sp.s.w., high amplitude spiked slow waves.
1092
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 3. Patient 1. fMRI BOLD activations in the left mesial temporal lobe.
the EEG amplifier: −12.8 and 12.8 mV; bandwidth 0.15– 269.5 Hz), the signal was then input to a computer (software Brain-Quick System Plus, Micromed). Two electrodes were used to record the electrocardiogram (ECG) and electromyogram (EMG) to remove pulse and movement artifacts. The EMG electrode was placed on the right abductor pollicis brevis muscle and the other (ECG) on the precordial area. The MRI-approved amplifier was placed at the end of the magnet bore, while the recording computer was placed outside the magnet room and connected via a fiberoptic cable. To minimize variability in the EEG artefacts due to wire movement caused by mechanical vibration during MRI acquisition, lead wires were immobilized on foam pads. Details of the EEG recording method are given in Bènar et al [27]. An anterior to Fz electrode was used as reference, and a posterior to Fz electrode was used as the ground as detailed in other studies using the same recording system [28]. The EEG artifact induced by the gradient magnetic field was digitally removed offline by an adaptive filter (Micromed). After the raw EEG data were reviewed, a marker was placed at the end of the last volume. The remaining markers were then placed automatically and equidistant to each other.
The data were then shifted in the temporal domain to align each volume to the first one. Finally, average slice and volume artefacts were computed and subsequently subtracted from the data, thus yielding clean EEG signals. EEG artifacts associated with pulsatile blood flow were digitally removed offline using a simple averaging procedure [29,30]. An artifact template was first created from all the occurrences of the cardioballistographic artifact by means of averaging and, afterwards, subtracted from each occurrence. Subsequently, a single electroencephalographer visually reviewed the filtered EEGs and marked the time of onset and duration of each IED. Automatic IED identification was avoided. 2.3. fMRI Acquisition and analysis Images were obtained with a 1.5-Tesla MRI scanner (Vision, Siemens, Erlangen, Germany). At the start of each study, a T1-weighted anatomical MRI was acquired [192 slices; field of view=256×256, scanning matrix 512×512, slice thickness 1 mm, sagittal slice orientation, echo time (TE)=3ms, repetition time (TR)=1990 ms]. All patients then underwent a 24-min fMRI recording session (four
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1093
Fig. 4. Patient 3. EEG tracing during continuous EEG–fMRI. Note the irritative focus in the left temporal region.
concatenated sessions) after giving informed consent. BOLD fMRI data were acquired, using a standard gradient-echo (echo planar imaging) sequence, on the axial orientation, in four runs of 6 min with the patient in the resting state, as described by the Kobayashi team [20] (voxel dimension 3×3×3 mm; 36 slices; matrix 64×64; TE=50 ms, TR=3.7 s; slice thickness=3 mm). For image processing and statistical analysis of the fMRI time series data, we used BrainVoyager QX software (Brain Innovation, Maastricht, Netherlands). Preprocessing of the functional MRI included three-dimensional (3D) motion correction, slice scan time correction (sync interpolation), linear trend removal by temporal highpass filtering (three cycles in time course) and transformation into the Talairach coordinate space. In each subject, activated voxels were identified with a single-subject general linear model approach for time series data [31] To account for the hemodynamic delay, the boxcar waveform representing the rest and task conditions was convolved with an empirical hemodynamic response function [32].
identified the single region of condition-associated BOLD signal changes with a statistical threshold based on the amplitude (Pb.05, corrected for multiple comparisons: Bonferroni test) and extent of the regions of activation. The location of voxels with maximal signal increase was expressed in terms of x, y and z in the Talairach space, and activation volumes were expressed in terms of number of activated voxels. Positive BOLD-fMRI responses were defined as activations. Significant responses were defined as almost five contiguous voxels with Pb.05 over at least two contiguous slices [3,4,11] in two-dimensional (2D) reconstruction. The anatomic localization of BOLD responses was determined by coregistration of the anatomic data and statistical t maps. We analyzed the extent and maximum fMRI response for each study, considering all areas with significative activations. We also determined the locations of maximum activation based on the maximum peak response (maximum t value).
2.4. Statistical analysis of fMRI data A t statistic was used to determine significance on a voxel-by-voxel basis and correlation values were transformed into a normal distribution (Z statistic). The results were displayed on parametric statistical maps in which the pixel Z value is expressed on a colorimetric scale. We
3. Results 3.1. EEG–fMRI studies and bold responses No clinical or EEG manifestations of ictal events developed during scanning session. None of the patients reported
1094
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 5. Patient 3. fMRI-BOLD activations in the left temporal cortex around the poro-encephalic cavity (site of previous surgical treatment of the cavernoma).
discomfort or other adverse reactions related to the EEG recording procedure. In all patients, EEG showed unilateral focal activity during the EEG–fMRI session. IEDs recorded inside the scanner had a localization, amplitude and morphology similar to those in previous routine EEG recordings. In two patients (Patient 1— see Figs. 1–3—and Patient 6), EEG recordings showed focal high-amplitude slow waves and in the other two patients (Patients 2 and 4) slow-wave discharges over the temporal electrodes. In another patient (Patient 7), we detected slowwave activity in the extratemporal (F7) region. In the patient with a cavernoma (Patient 3—see Figs. 4 and 5), focal EEG activity was characterized by focal slow-spiked wave activity, reaching maximal amplitude over the left temporal electrodes (T5). In the other two patients (Patients 5 and 8) we detected slow-spiked waves; in Patient 5 (see Figs. 6–9), in the extratemporal (F3) region and in Patient 8 (see Figs. 10–12), in the temporal region (T4–T6). In Patient 8, EEG tracings (Fig. 10) also showed high-amplitude spiked slow wave activity. The mean frequency was 2.4 IEDs per minute (S.D. 0.17). In all patients, fMRI analysis showed a significant focal BOLD activation in a single activated area related to the EEG irritative focus (Table 2).
3.2. Activation sites In all patients studied, fMRI BOLD activation (Table 3) (fMRI-3D reconstruction) corresponded to the irritative focus on standard EEG recordings. 3.2.1. Temporal lobe activations In six of the eight patients, focal BOLD signal changes reached statistical significance in the temporal regions. In five patients (the patients with MTS, Patients 1 and 4; the patient with cryptogenic epilepsy, Patient 2; in two of the four patients with idiopathic epilepsies, Patients 6 and 8), the significant BOLD changes were located in the mesial temporal lobe, and in one, the patient with a cavernoma (Patient 3) in the neocortical regions (laterally and posteriorly to the resection margins). 3.2.2. Extratemporal activations In two patients fMRI showed extratemporal activation, located in one (Patient 5) in the left occipital and in the other (Patient 7) in the left frontal lobe. The occipital activation corresponded closely to the patient's clinical manifestations and to standard EEG. This patient was admitted to our ward for investigation of hallucinations triggered by eye opening. A standard EEG
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1095
Fig. 6. Patient 5. EEG tracing, during continuous EEG–fMRI, showing an irritative focus in the left temporo-occipital region.
disclosed a left temporal–occipital irritative focus that spread anteriorly when she opened the eyes. When the EEG showed the spread, a polygraphic study showed a right-beating nystagmus and head deviation to the right side. Continuous EEG–fMRI (Fig. 7) demonstrated significant BOLD signal activation in the associative area around the calcarine cortex (Brodmann area:18).
In the patient in whom fMRI showed BOLD activation in the left frontal region (Brodmann area 6), standard EEG showed a left frontotemporal irritative focus, characterized by frequent slow-spiked waves. A recent polygraphic study (recorded during Stage I and II non-rapid-eye-movement sleep) showed polymorphic slow waves (theta-delta) in the left frontal regions, with phase inversion at F7 and without
Fig. 7. Patient 5. fMRI-BOLD activation in Brodmann area 18 in the patient who had an irritative focus in the left temporo-occipital lobe.
1096
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 8. Patient 5. fMRI-BOLD activation in Brodmann area 18 in the patient who had an irritative focus in the left temporooccipital lobe.
spread. These findings suggested that the patient was a good candidate for an EEG–fMRI study, and during the EEG– fMRI sessions described here we confirmed the fMRI activation area within the precentral gyrus. 4. Discussion The main finding in this EEG–fMRI study of patients with partial epilepsy is that focal interictal slow-wave activity was invariably associated with increased focal fMRI-BOLD activation responses in a spatially related brain area. Our study therefore extends current knowledge on epileptic foci localization and confirms previous reports suggesting that EEG–fMRI BOLD activation associated with modelled slow activity might have a role in localizing the epileptogenic region even in the absence of clear interictal spikes [25,26]. All the eight patients with partial epilepsy we enrolled in this study had frequent interictal focal slow-wave activity on standard EEG. In all continuous EEG–fMRI recording sessions, after fMRI artefact removal, we obtained a good-
Fig. 9. Patient 5. fMRI-BOLD activation in Brodmann area 18 in the patient who had an irritative focus in the left temporooccipital lobe.
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1097
Fig. 10. Patient 8. EEG during recording session. Irritative focus on right temporal regions.
quality EEG that allowed us to detect spontaneous IEDs and analyze the related fMRI BOLD activation. The EEG recording left the quality of fMRI almost undistorted, and in all patients, the focal activity seen in the concurrent EEG was associated with a focal increase in the MRI signal. In their focal distribution, these BOLD activations resembled the focal IEDs seen on routine scalp EEG and EEG recorded during EEG–fMRI sessions. An interesting finding came from the patients with lesional epilepsy. These patients, whose standard MRI documented a lesion and whose standard EEG identified an irritative focus, are the ideal candidates for verifying a possible spatial relationship between the epileptogenic and irritative focus [1,2]. In the patient who had undergone surgery to remove a cavernoma (Patient 3), the EEG–fMRI study, by localizing the irritative focus and linking it to fMRI as “active state,” showed a significant BOLD activation signal closely related to the poroencephalic cavity (a residual of previous treatment). Even if this focal BOLD activation presumably arose from a blood vessel (residual cavernoma), undoubtedly, it was obtained by a protocol study linking an active fMRI state to IEDs on EEG. To clarify the relationship, we therefore decided to reach, progressively, a very high specificity level (Pb.0001), as demonstrated in iconography. Even with these high specificity values, the BOLD activation in that site persisted.
Another new finding is the BOLD activation we detected on fMRI in the two patients with MTS. In contrast to others, who studied a series of patients with MTS (five studied with continuous coregistration) [8] and found no significant activation, in two of the eight patients studied (Patient 1 and 4), we detected significant BOLD activation. These interesting results suggest a possible role of simultaneous EEG– fMRI in disclosing focal activation even in the mesial temporal cortex. This cortical area is notoriously difficult to study with standard methods because the deep localization of the irritative area often makes spikes smoother and therefore harder to recognize in recordings from standard EEG scalp electrodes than in sEEG. Hence, IEDs presenting as slowwave discharges could be useful in determining significant BOLD activation in a corresponding area, as previously suggested by Laufs et al. [25]. Useful information that may help to understand BOLD responses in the various forms of epilepsy came also from studying activation in extratemporal IEDs. In the two patients with extratemporal discharges (Patients 5 and 7), we noted a good correlation between clinical–polygraphic data and BOLD activation during fMRI. No significant difference was found between activation in the frontal or occipital lobe. In both zones, after focal IEDs, BOLD activation increased. Why we obtained such good results (8/8 activation — significant concordance between EEG and fMRI data)
1098
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
Fig. 11. Patient 8. fMRI-BOLD activations in right temporal lobe during recording session.
remains open to question. The most plausible reason is that during enrolment, to obtain the largest possible percentage of activations, we explicitly selected patients whose standard EEGs showed a high IED firing rate, confirmed on EEG during the scanning session. The mean frequency of IEDs in the eight patients was 2.4/min, S.D. 0.17 (Table 2), considerably higher than the one IED per minute Duncan et al. [3] considered as the minimum to obtain a focal BOLD activations. Collectively, these findings confirm the importance of IED firing rates in EEG–fMRI. Possibly the most interesting finding in this study is that conversely, morphology seems less important than IED firing rates in triggering EEG–fMRI-BOLD activation: of the 8 patients we studied, 5 had pure slow-waves on standard EEG (3/8 with high amplitude and 2/8 with a normal voltage) and 3 a slow spike-waves discharges (one with high amplitude). Although these slow spike-wave discharges might have originated from a spike focus smoothed by filtering (unlikely because EEG detected the same IEDs before patients entered the magnet room), we found no difference in the statistical significance of BOLD activation between the two groups. In particular, slow spike-wave IEDs
Fig. 12. Patient 8. fMRI-BOLD activations in right temporal lobe during recording session.
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
1099
Table 2 Number of IEDs revealed on EEG during recording session and BOLD activation in 2D-fMRI, with number of contiguous activated slices 1
16
45
61
2.54
2
13
42
55
2.29
3
14
40
54
2.25
4
15
43
58
2.41
5
17
47
64
2.66
6
13
38
51
2.13
7 8
14 16
44 46
58 62
2.41 2.58
Mean S.D.
14.75 1.49 A (IEDs/6I)
43.12 3.04 B (IEDs/18I)
57.88 4.54 A+B (IEDs/24I)
2.41 0.18 C (IEDs/min)
Left mesial temporal lobe Left mesial temporal lobe Left superior temporal lobe — neocortical region Right mesial temporal lobe Left occipital lobe — calcarin cortex Right superior temporal lobe Left frontal lobe Right superior temporal lobe
D (BOLD activation in 2D reconstruction)
2 2 3
2 2
2 2 2
E (n. slices)
A, IEDs recorded during the first concatenated session (lasting 6 min); B, IEDs recorded during the following three concatenated sessions (18 min); A+B, Total IEDs recorded during the four concatenated sessions (24 min); C, IEDs per minute with mean value and S.D.; D, BOLD activation in 2D reconstruction; E, number of contiguous slices.
were no more efficient than slow-wave IEDs in eliciting significant BOLD activation. Hence, we agree as previously suggested [25,26] that slow-wave IEDs, like spikes, can Table 3 Number of activated voxels, coordinates (mean and S.D.) and corresponding localization, with Talairach system, of activated regions in 3D reconstruction Mean “x” Patient –18±1.7 1
Mean “y”
Mean “z”
–7±1.8 –19±1.3
Patient –14±2.3 –33±3.9 2
No. of Corresponding area voxels 177
–6±2.4
522
Patient –59±2.3 –41±2.4 –15±3.6 3
540
–14±1.5 –15±1.2
175
Patient 4
17±3
Patient 5
–3±2.1 –78±1.8
2±2.7 1227
Patient 6
62±3.8 –15±3.1
7±1.7
364
13±3.4
442
–3±2.8
422
Patient –51±3.7 7 Patient 8
–3±4
54±1.6 –36±2.7
Left Cerebrum, limbic lobe, parahippocampal gyrus, Brodmann area 34 Left cerebrum, limbic lobe, parahippocampal gyrus, Brodmann area 30 Left cerebrum, temporal lobe, inferior temporal gyrus, Brodmann area 20 Right cerebrum, limbic lobe, parahippocampal gyrus, Brodmann area 28 Left cerebrum occipital lobe, lingual gyrus, Brodmann area 18 Right cerebrum, temporal lobe, transverse temporal gyrus, Brodmann area 42 Left cerebrum, frontal lobe, precentral gyrus, Brodmann area 6 Right cerebrum, temporal lobe, middle temporal gyrus, Brodmann area 21
elicit a significant cerebral blood flow increase in a spatially related brain area. The slow-wave discharge in five of the eight patients we studied also suggests that voltage has a minor role in determining BOLD activation: even if our study sample is limited, we found no differences in BOLD activations between IEDs with high-amplitude and normal-amplitude slow waves. The significant concordance between EEG and fMRI data (and particularly the absence of multiple activation areas) depends on the high specificity threshold our study design envisaged. In designing this study, we maintained only the single activation with the strongest specificity so that we could verify whether this activation, and this activation alone coincided with the epileptic focus previously documented by standard EEG and used as paradigm of the activation-state during fMRI analysis. The complete EEG–fMRI concordance we achieved in this study suggests that slow-wave IEDs even without spikes may be useful in activating fMRI BOLD responses during the presurgical, noninvasive evaluation of patients with partial, drug-resistant, seizures. References [1] Ebersole JS, Wade PB. Spike voltage topography identifies two types of frontotemporal epileptic foci. Neurology 1991;41:1425–33. [2] Benbadis SR, Gerson WA, Harvey JH, Luders HO. Photosensitive lobe epilepsy. Neurology 1996;46:1540–2. [3] Krakow K, Woermann FG, Symms, Allen PJ, Lemieux L, Barker GJ, et al. EEG-fMRI of interictal epileptiform activity in patients with partial seizures. Brain 1999;122:1679–88.
1100
P. Manganotti et al. / Magnetic Resonance Imaging 26 (2008) 1089–1100
[4] Krakow K, Lemieux L, Messina D, Scott CA, Symms MR, Duncan JS, et al. Spatio-temporal imaging of focal interictal epileptiform activity using EEG-triggered functional MRI. Epileptic Disord 2001;3:67–74. [5] Lemieux L, Salek-Haddadi A, Josephs O, Allen P, Toms N, Scott C, et al. Event-related fMRI with simultaneous and continuous EEG: description of the method and initial case report. Neuroimage 2001;1: 780–7. [6] Lazeyras F, Zimine I, Blanke O, Perrig SH, Seek M. Functional MRI with simultaneous EEG recording: feasibility and application to motor and visual activation. J Magn Reson Imaging 2001;13:943–8. [7] Benar CG, Gross DW, Wang Y, Petre V, Pike B, Dubeau F, et al. The BOLD responses to interictal epileptiform discharges. Neuroimage 2002;17:1182–92. [8] Al-Asmi A, Benar CG, Gross DW, Khani YA, Andermann F, Pike B, et al. fMRI activation in continuous and spike-triggered EEG-fMRI studies of epileptic spikes. Epilepsia 2003;44:1328–39. [9] Boor S, Vucurevic G, Pfleiderer C, Stoeter P, Kutschke G, Boor R. EEG-related functional MRI in benign childhood epilepsy with centrotemporal spikes. Epilepsia 2003;44:688–92. [10] Salek-Haddadi A, Lemieux L, Merschhemke M, Friston K, Duncan JS, Fish DR. Functional magnetic resonance imaging of human absence seizures. Ann Neurol 2003;53:663–7. [11] Salek-Haddadi A, Diehl B, Hamandi K, Merschhemke M, Liston A, Friston K, et al. Hemodynamic correlates of epileptiform discharges: an EEG-fMRI study of 63 patients with focal epilepsy. Brain Res 2006;1088:148–66. [12] Bagshaw AP, Aghakhani Y, Benar CG, Kobayashi E, Hawco C, Dubeau F, et al. EEG-fMRI of focal epileptic spikes: analysis with multiple haemodynamic functions and comparison with gadoliniumenhanced MR angiograms. Hum Brain Mapp 2004;22:179–92. [13] Bagshaw AP, Kobayashi E, Dubeau F, Pike GB, Gotman J. Correspondence between EEG-fMRI and EEG dipole localisation of interictal discharges in focal epilepsy. Neuroimage 2006;30:417–25. [14] Aghakhani Y, Bagshaw AP, Benar CG, Hawco C, Andermann F, Dubeau F, et al. fMRI activation during spike wave discharges in idiopathic generalized epilepsy. Brain 2004;127:1127–44. [15] Aghakhani Y, Kinay D, Gotman J, Soualmi L, Andermann F, Olivier A, et al. The role of periventricular nodular heterotopia in epileptogenesis. Brain 2005;128:641–51. [16] Gotman J, Benar CG, Dubeau F. Combining EEG and FMRI in epilepsy: methodological challenges and clinical results. J Clin Neurophysiol 2004;21:229–40. [17] Gotman J, Grova C, Bagshaw A, Kobayashi E, Aghakhani Y, Dubeau F. Generalized epileptic discharges show thalamocortical activation and suspension of the default state of the brain. PNAS 2005;142:15236–40. [18] Kobayashi E, Bagshaw AP, Jansen A, Andermann F, Andermann E, Gotman J, et al. Intrinsic epileptogenicity in polymicrogyric cortex suggested by EEG-fMRI BOLD responses. Neurology 2005;64:1263–6.
[19] Kobayashi E, Bagshaw AP, Grova C, Gotman J, Dubeau F. Grey matter heterotopia: what EEG-fMRI can tell us about epileptogenicity of neuronal migration disorders. Brain 2006;129:366–74. [20] Kobayashi E, Hawco CS, Grova C, Dubeau F, Gotman J. Widespread and intense BOLD changes during brief focal electrographic seizures. Neurology 2006;66:1049–55. [21] Kobayashi E, Bagshaw AP, Grova C, Dubeau F, Gotman J. Negative BOLD responses to epileptic spikes. Hum Brain Mapp 2006;27: 488–97. [22] Kobayashi E, Bagshaw AP, Benar CG, Aghakhani Y, Andermann F, Dubeau F, et al. Temporal and extratemporal BOLD responses to temporal lobe interictal spikes. Epilepsia 2006;47:343–54. [23] Laufs H, Lengler U, Hamandi K, Kleinschmidt A, Krakow K. Linking generalized spike-and-wave discharges and resting state brain activity by using EEG-fMRI in a patient with absence seizures. Epilepsia 2006; 47:444–8. [24] Benar CG, Grova C, Kobayashi E, Bagshaw AP, Aghakhani Y, Dubeau F, et al. EEG-fMRI of epileptic spikes: concordance with EEG source localization and intracranial EEG. Neuroimage 2006;30:1161–70. [25] Laufs H, Hamandi K, Walter MC, Scott C, Smith S, Duncan JS, et al. EEG-fMRI mapping of asymmetrical delta activity in a patient with refractory epilepsy is concordant with the epileptogenic region determined by intracranial EEG. Magn Reson Imaging 2006;24: 367–71. [26] Avesani M, Milanese F, Formaggio E, Gasparini A, Baraldo A, Cerini R, et al. Continuous EEG-fMRI in pre-surgical evaluation of a patient affected by symptomatic seizures: BOLD activation linked to interictal epileptiform discharges caused by cavernoma. Neuroradiol J 2008;21:183–91. [27] Benar RG, Aghakhani Y, Wang Y, Izenberg A, Al-Asmi A, Dubeau F, et al. Quality of EEG in simultaneous EEG-fMRI for epilepsy. Clin Neurophysiol 2003;114:569–80. [28] Gonçalves SI, de Munck JC, Pouwels PJ, Schoonhover R, Kuiker JP, Mauritis NM, et al. Correlating the alpha rhythm to BOLD using simultaneous EEG-fMRI: inter-subject variability. Neuroimage 2006; 30:203–13. [29] Allen PJ, Polizzi G, Krakow K, Fish DR, Lemieux L. Identification of EEG events in the MR scanner: the problem of pulse artifact and a method for its subtraction. Neuroimage 1998;8:229–39. [30] Allen PJ, Josephs O, Turner R. A method for removing imaging artefact from continuous EEG recorded during functional MRI. Neuroimage 2000;12:230–9. [31] Friston KJ, Holmes AP, Worsley KJ, Poline JP, Frith CD, Frackowiak RSJ. Statistical parametric maps in functional imaging: a general linear approach. Hum Brain Mapp 1995;2:173–81. [32] Boynton GM, Engel SA, Glover GH, Heeger DJ. Linear systems analysis of functional magnetic resonance imaging in human V1. J Neurosci 1996;16:4207–41.