Epilepsia, 52(10):e123–e125, 2011 doi: 10.1111/j.1528-1167.2011.03161.x
BRIEF COMMUNICATION
Iatrogenic seizures during intracranial EEG monitoring *Julie A. Khoury, *Katherine H. Noe, *Joseph F. Drazkowski, *Joseph I. Sirven, and yRichard S. Zimmerman Departments of *Neurology and yNeurosurgery, Mayo Clinic, Phoenix, Arizona, U.S.A.
SUMMARY Cerebral edema with declining neurologic status is a known complication of intracranial electroencephalography (EEG) monitoring. The frequency and consequences of iatrogenic edema that is not clinically evident are presently poorly defined. We investigated the potential for intracranial electrodes to cause subclinical cerebral edema, and for such edema to cause iatrogenic seizures. In a retrospective review of 33 adults who had head magnetic resonance imaging (MRI) while undergoing epilepsy
Intracranial electroencephalography (EEG) monitoring can define the epileptogenic zone in subjects undergoing epilepsy surgery evaluation when noninvasive techniques are inadequate. Invasive monitoring is considered the gold standard for seizure localization, although results must be interpreted in conjunction with the clinical history and results of scalp EEG and imaging studies. The procedure is generally safe, but complications can include subdural and epidural hematoma, intracranial hemorrhage, infection, cerebrospinal fluid (CSF) leak, and cerebral edema (Hamer et al., 2002; Bureno et al., 2006; Van Gompel et al., 2008). Complications are usually identified by clinical signs and symptoms (e.g., fever, decline in neurologic examination) or by direct visualization during surgery. Complications also occur without clinical symptoms; however, the frequency and significance of subclinical complications are poorly defined. Prior studies with postoperative computed tomography (CT) or magnetic resonance imaging (MRI) identified subclinical hematoma, hemorrhage, fluid collection, or edema in one of 50 subjects with depth electrodes and 14 of 54 with subdural grids (Ross et al., 1996; Silberbusch et al., 1998). In 20 asymptomatic patients undergoing MRI within 24 h after explantation Accepted May 16, 2011; Early View publication July 8, 2011. Address correspondence to Katherine H. Noe, M.D., Ph.D., Department of Neurology, Mayo Clinic Hospital, 5777 East Mayo Boulevard, Phoenix, AZ 85054, U.S.A. E-mail:
[email protected] Wiley Periodicals, Inc. ª 2011 International League Against Epilepsy
surgery evaluation with intracranial EEG, 28% (6 of 21) depth electrode implantations had subclinical vasogenic edema. Of these, 50% (3 of 6) had nonhabitual electrographic seizures that appear to result from iatrogenic edema. No long-term adverse sequelae were noted, however, if unrecognized, iatrogenic seizures could lead to unnecessary exclusion from definitive surgical intervention for refractory epilepsy. KEY WORDS: Seizures, Epilepsy, Intracranial electroencephalography, MRI.
of subdural electrodes placed via burr hole, 35% had subdural hematoma, 25% cortical contusion, and 25% focal edema identified radiographically (Al-Otaibi et al., 2010). Furthermore, depth and subdural electrodes have been associated with histopathologically defined, subclinical localized inflammatory response (Stephan et al., 2001). Radiologically defined complications have not been linked with long-term adverse clinical outcomes. However, focal edema from electrode implantation could potentially cause de novo seizures, leading to confusion about localization of the true epileptogenic zone. Iatrogenic seizures during intracranial EEG have only rarely been reported in the literature. Engel and Crandall reported five patients undergoing bitemporal depth electrodes between 1977 and 1980 with atypical seizures arising contralateral to typical seizure onset, all with good surgical outcomes (Engel & Crandall, 1983). Malow et al. (1995) reported a case of ‘‘nonhabitual’’ seizures (distinct clinically and electrographically from usual event) with subdural electrodes and a focal subdural hematoma. Fountas and Smith described nonhabitual seizures in 5 (2.7%) of 185 patients undergoing invasive monitoring, which resolved after explantation, and theorized that this might have arisen from mechanical irritation of the cortex by the electrodes or blood products (Fountas & Smith, 2007). Our goal in this study was to better define the frequency of nonhabitual seizures, correlate the EEG findings with MRI-defined complications with electrodes in place, and to determine any impact on outcome.
e123
e124 J. A. Khoury et al.
Methods Following institutional review board approval, admissions to the Mayo Clinic Arizona epilepsy monitoring unit from January 2005 to May 2010 were reviewed to identify subjects undergoing epilepsy surgery evaluation with intracranial EEG and head MRI obtained with intracranial electrodes in place. These 1.5 Tesla MRI studies were part of routine practice to confirm electrode placement prior to planned resective surgery, with images obtained on the day of electrode explantation. The chart was reviewed for clinical history, EEG results, and surgical outcomes. MRI reports and images were reviewed. Statistical analysis with Fisher’s exact test (two-tailed p-value) was performed using SAS software (SAS Institute, Cary, NC, U.S.A.).
Results Thirty-three admissions in 27 patients (16 male, 11 female) had MRI with intracranial electrodes in place. Seven had depth electrodes only, 12 subdural grids and/or strips only, and 14 had a combination of the two. Mean age was 35 years (range 18–71 years). Mean duration of implantation was 11.8 days (range 2–24 days). Mean number of intracranial electrode contacts was 62 (range 8–134). The rationale for intracranial monitoring was nonlesional extratemporal epilepsy in 13 and to refine the extent of resection in 9 cases with large extratemporal lesions. Two had possible bitemporal ictal onset by scalp EEG: one was nonlesional and one had a unilateral temporal low grade tumor. The remaining nine had suspect unilateral temporal onset that was nonlesional. There were no cases of radiographically defined subclinical edema in subjects with subdural electrodes alone. MRI demonstrated perielectrode vasogenic edema in 28% (6 of 21) of depth implantations. Of these, 3 of 6 had nonhabitual electrographic seizures without clinical correlate recorded from the depth electrode with surrounding edema. The clinical background of each subject is summarized in Table 1. Presurgical evaluation for subjects 1 and 2 suggested right
temporal onset; however, as the MRI was nonlesional, invasive EEG monitoring was pursued. In subject 3, noninvasive testing suggested right frontotemporal onset. Intracranial electrodes were placed as this was a posttraumatic case with bilateral frontal and right temporal (but not left temporal) encephalomalacia. In all three subjects, two eight-contact depth electrodes were placed bilaterally under stereotactic guidance into the anterior and posterior mesial temporal lobe via a lateral approach. Subject 1 additionally had a four-contact right frontal subdural strip, and Subject 3 also had two eight-contact depth electrodes placed into each frontal lobe. Subject 1 had 11 typical seizures recorded arising from the right mesial temporal lobe. Four electrographic seizures arising in the left temporal neocortical contacts and lasting several minutes without clinical correlate were captured on days 1, 5, and 7 of implantation. MRI on day 10 showed vasogenic edema surrounding the left temporal depths (Fig. 1A). After a standard right temporal lobectomy the subject is seizure-free for 14 months. Subject 2 had five typical clinical seizures recorded from the right anterior mesial temporal lobe. Twelve subclinical seizures were recorded on days 1 and 2 of implantation arising in the right temporal neocortical contacts, lasting 30 s to several minutes in duration. MRI on day 10 postimplantation demonstrated vasogenic edema around the temporal depths, right greater than left (Fig. 1B,C). Subject 2 had a subsequent selective right amygdalohippocampectomy and had been seizure-free for 15 months. Subject 3 had three typical seizures arising from the right mesial temporal and right mesial inferior frontal lobe. On days 1–3 postimplantation, 27 subclinical seizures arising in the neocortical contacts of the left temporal depth electrodes were recorded. MRI on day 13 demonstrated edema along the temporal depth electrodes bilaterally. Subject 3 is seizure-free 6 weeks after partial right frontal lobectomy. Three of 6 subjects had follow-up MRI between 11 and 27 days (mean 18.67 days) after removal of intracranial electrodes. There was complete resolution of edema in two and minimal edema seen in Subject 2 who had the MRI on day 27 (Fig. 1d). No clinical symptoms or signs were noted
Table 1. Clinical characteristics of subjects with radiologically identified edema associated with intracranial depth electrodes Subject
Age (year)
Seizure etiology
Baseline MRI
1
42
Idiopathic
Nonlesional
2 3
40 48
Idiopathic Posttraumatic
Nonlesional Bifrontal and R inferior temp encephalomalacia
Interictal scalp EEG R temp R temp R frontotemp
Ictal scalp EEG R temp R temp R hemisphere
PET
SISCOM
Neuropsych. testing
Borderline R frontotemp Negative N/A
Indeterminate
Nonlateralized
R mesial temp R frontal, R mesial and neocortical temp
Nonlateralized Nonlateralized
EEG, electroencephalography; PET, positron emission tomography; SISCOM, subtraction ictal SPECT (single-photon emission computed tomography) coregistered to MRI; Neuropsych., neuropsychological; R, right; Temp, temporal; N/A, not available.
Epilepsia, 52(10):e123–e125, 2011 doi: 10.1111/j.1528-1167.2011.03161.x
e125 Iatrogenic Seizures in Intracranial EEG A
B
C
D
appear to be falsely localizing, subclinical seizures resulting from iatrogenic cerebral edema. The findings serve as a reminder to interpret information gleaned from intracranial monitoring carefully in conjunction with the results from noninvasive tests when determining candidacy and strategy for epilepsy surgery. If clinically and electrographically atypical seizures are recorded, further evaluation with MRI may be helpful to determine if mild cerebral edema surrounding the intracranial electrodes is present. It is reassuring that subclinical edema defined by MRI in this case series resolved without any intervention other than removal of the intracranial electrodes, did not result in demonstrable short- or long-term adverse effects, and did not affect seizure-free outcome.
Acknowledgments None.
Disclosure None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
Figure 1. Axial FLAIR MRI images demonstrating vasogenic edema associated with implantation of intracranial depth electrodes. (A) Subject 1. Left temporal edema. (B) Subject 2. Right temporal edema. (C) Subject 2. Bilateral temporal edema. (D) Subject 2. Residual left temporal edema, 27 days after electrode explantation. Epilepsia ILAE
to suggest adverse sequelae from electrode implantation on postoperative visits. Similarly discordant subclinical seizures were not seen in patients with depth electrodes in the absence of cerebral edema on MRI. One patient with a subdural grid and MRI without edema had atypical clinical seizures on postoperative day 1 arising from an area of the grid distinct from that of his typical events. This patient has been seizure-free 2 years after a focal cortical resection. In a combined analysis of all 33 subjects in this case series, the likelihood of atypical seizures was significantly greater in those with cerebral edema on MRI than in those without (p = 0.014, Fisher’s exact test).
Discussion In this small case series, 9% of all intracranial EEG cases and 14% of those with depth electrodes developed what
References Al-Otaibi FA, Alabousi A, Burneo JG, Lee DH, Parrent AG, Steven DA. (2010) Clinically silent magnetic resonance imaging findings after subdural strip electrode implantation. J Neurosurg 112:461–466. Bureno JG, Steven DA, McLachlan RS, Parrent AG. (2006) Morbidity associated with the use of intracranial electrodes for epilepsy surgery. Can J Neurol Sci 33:223–227. Engel J, Crandall P. (1983) Falsely localizing ictal onsets with depth EEG telemetry during anticonvulsant withdrawal. Epilepsia 24:344– 355. Fountas KN, Smith JR. (2007) Subdural electrode-associated complications: a 20 year experience. Stereotact Funct Neurosurg 85:264–272. Hamer HM, Morris HH, Mascha EJ, Karafa MT, Bingaman WE, Bej MD, Burgess RC, Dinner DS, Folvary NR, Hahn JF, Kotagal P, Najm I, Wyllie E, Luders HO. (2002) Complications of invasive video-EEG monitoring with subdural grid electrodes. Neurology 58:97–103. Malow BA, Sato S, Kufta CV, Blaxton TA, Figlozzi CM, Theodore WH. (1995) Hematoma-related seizures detected during subdural electrode monitoring. Epilepsia 36:733–735. Ross DA, Brunberg JA, Drury I, Thomas R. (1996) Intracerebral depth electrode monitoring in partial epilepsy: the morbidity and efficacy of placement using magnetic resonance image-guided stereotactic surgery. Neurosurgery 39(2):327–334. Silberbusch MA, Rothman MI, Bergey GK, Zoarski GH, Zagardo MT. (1998) Subdural grid implantation for intracranial EEG recording: CT and MRI appearance. AJNR Am J Neuroradiol 19:1089–1093. Stephan CL, Kepes JJ, SantaCruz K, Wilkinson SB, Fegley B, Osorio I. (2001) Spectrum of clinical and histopathologic responses to intracranial electrodes: from multifocal meningitis to multifocal hypersensitivity-type meningovasculitis. Epilepsia 42:895–901. Van Gompel JJ, Worrell GA, Bell ML, Patrick TA, Cascino GD, Raffel C, Marsh WR, Meyer FB. (2008) Intracranial electroencephalography with subdural grid electrodes: techniques, complications, and outcomes. Neurosurgery 63:498–506.
Epilepsia, 52(10):e123–e125, 2011 doi: 10.1111/j.1528-1167.2011.03161.x
