CASE REPORT Misleading EEG Lateralization Associated With Midline Shift Gena R. Ghearing, Sergiu Abramovici, Alexandra Popescu, and Maria E. Baldwin Department of Neurology, University of Pittsburgh, Pittsburgh, Pennsylvania, U.S.A. Summary: Midline discharges, lateralized periodic discharges, and seizures have been described with ipsilateral lesions that result in midline shift (MLS). Periodic discharges and seizures arising contralateral to a known lesion have not previously been described as a sign of MLS. We present four patients with focal brain lesions, resulting in MLS and epileptiform discharges arising from the contralateral hemisphere. Patient 1 underwent a right anterior temporal lobectomy. On postoperative day 2, computed tomography demonstrated a right to left MLS of 12 mm, and EEG was remarkable for left temporal nonconvulsive status epilepticus. Patient 2 experienced a subarachnoid hemorrhage, which was more prominent on the left. Computed tomography after craniotomy demonstrated left to right MLS of 6 mm, and EEG was remarkable for right lateralized periodic E lectroencephalography remains an important clinical tool in the evaluation and dynamic monitoring of patients with focal cerebral lesions, especially those with fluctuating or depressed neurologic status. The spectrum of epileptiform discharges varies between sporadic sharp waves, lateralized periodic discharges (LPDs), focal seizures, or status epilepticus, in the setting of an acute lesion. Monitoring with EEG not only allows for the identification of subtle or subclinical seizures and provides prognostic information but also may help identify comorbidities such as progressive mass effect and edema. Understanding EEG features suggestive of midline shift (MLS) may permit earlier therapeutic intervention. CASE PRESENTATIONS These four cases were retrospectively analyzed, and the patients were followed using serial clinical evaluation, standard 5-mm slice computed tomographic (CT) brain imaging, and 22channel digital continuous video-EEG recordings. Midline shift was measured at the level of the foramen of Monroe on the axial CT scan by drawing a line through the midline in the sagittal plane and then drawing a second perpendicular line to the septum pellucidum. This displacement of the septum pellucidum was measured in millimeters and recorded as MLS. The EEG manifestation contralateral to the lesion and the degree of MLS as measured in millimeters on CT at different time points along the patient’s hospitalization course are summarized in Table 1. The authors have no funding or conflicts of interest to disclose. Presented as a poster 3.227 at the 68th Conference of the American Epilepsy Society, Seattle, WA, December 8, 2014. Address correspondence and reprint requests to Gena R. Ghearing, MD, University of Pittsburgh Epilepsy Center, 810 Kaufmann Building, 3471 Fifth Avenue, Pittsburgh, PA 15213, U.S.A.; e-mail: ghearinggr@upmc.edu. Copyright Ó 2017 by the American Clinical Neurophysiology Society ISSN: 0736-0258/17/0000-0001 DOI 10.1097/WNP.0000000000000385 clinicalneurophys.com discharges. Patient 3 had a right subdural hematoma and underwent craniotomy for evacuation. On postoperative day 3, computed tomography demonstrated a right MLS of 7 mm, and EEG was remarkable for left temporal nonconvulsive status epilepticus. Patient 4 had traumatic brain hemorrhages with maximal left frontotemporal involvement. Six days after the trauma, computed tomography was significant for left to right MLS of 9 mm, and EEG showed right lateralized periodic discharges. Epileptiform discharges and seizures occurring contralateral to a known lesion may be an indicator of MLS. Key Words: Midline shift, Focal lesion, Contralateral epileptiform activity. (J Clin Neurophysiol 2017;0: 1–4) Patient 1 was a 47-year-old woman who was admitted for a right amygdalohippocampectomy and anterior temporal lobectomy for the treatment of her refractory epilepsy. On postoperative day (POD) 2, the patient had altered mental status, right leg shaking, and perioral automatisms. Computed tomography demonstrated a right to left MLS of 12 mm as a result of a postoperative subdural fluid collection and edema. The EEG was remarkable for left temporal intermittent rhythmic delta activity and left temporal periodic spikes, which intermittently evolved into rhythmic 3-Hz spikes and sharp waves (Fig. 1A), consistent with left temporal nonconvulsive status epilepticus. The nonconvulsive status epilepticus resolved after loading with lorazepam, but rhythmic delta activity with superimposed left temporal sharp waves continued to be present. Rare left temporal seizures occurred over the next 2 days, and the patient was loaded with phenytoin in addition to her home seizure medication regimen. On POD 6, the measured MLS decreased to 9 mm with no surgical intervention, and the contralateral rhythmic delta activity and LPDS resolved with only sporadic sharp waves recorded in that region (Table 1). The patient recovered without deficits. She has continued to have her habitual seizures. Patient 2 was a 78-year-old man, whom had a subarachnoid hemorrhage secondary to a ruptured left posterior communicating artery aneurysm. After left craniotomy for aneurysm clipping on the day of admission, he experienced increased altered mental status. On POD 0, CT demonstrated a left to right MLS of 6 mm and the EEG was remarkable for right LPDs, which were stimulus induced and maximal in the right temporal region (Fig. 1B). The patient was maintained on levetiracetam, and no intervention was performed. On POD 1, there was a mild decrease in the MLS of approximately half a millimeter, and the previously recorded contralateral LPDs resolved (Table 1). The patient’s decreased mental status persisted despite resolution of periodic discharges and treatment for hydrocephalus. Care was withdrawn on POD 5. Journal of Clinical Neurophysiology Volume 0, Number 0, Month 2017 1 Copyright Ó by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited. G. R. Ghearing, et al. TABLE 1. Misleading EEG Lateralization Evolution of Midline Shift and EEG Changes During Hospitalization Patient 1 Day Patient 2 MLS, mm EEG MLS, mm EEG 0 d d 6 1 2 3 4 5 6 d 12 13 10 d 9 d Left temporal RDA evolving into left NCSE Six left temporal seizures and left LPDs, RDA Left LPDs and 2 left temporal seizures Left LPDs Left temporal spikes 6 d 8 d d d Right LPDs; generalized slowing, lower voltage on left Generalized slowing, maximal left Generalized slowing, maximal left d d d Patient 3 Day 0 1 2 3 4 5 6 Patient 4 MLS EEG MLS, mm EEG 11 9 d 7 6 d 0 d d Right-sided slowing Left temporal NCSE Left temporal NCSE Right-sided slowing Right-sided slowing 9 8 d d d d d Right LPDs Right frontocentral sharp waves d d d d d LPDs, lateralized periodic discharges; MLS, midline shift; NCSE, nonconvulsive status epilepticus; RDA, rhythmic delta activity. Patient 3 was a 47-year-old woman who was admitted for a right frontotemporal subdural hematoma after a fall, initially associated with a right to left MLS of 11 mm. She underwent right craniotomy and evacuation of the hematoma 6 days after admission. On POD 1, there was a right to left MLS of 9 mm on the postoperative CT, but because the patient did not have a significant worrisome clinical manifestation, an EEG was not ordered. On POD 2, the patient complained of left face and arm twitching, and the patient was started on phenytoin. A 25-minute EEG showed expected right side slowing, but no changes were seen in the contralateral hemisphere. On POD 3, the patient complained of left face and arm twitching again. A CT demonstrated a right to left MLS of 7 mm. An EEG performed after the CT was remarkable for occasional right temporal sharp waves and nearly continuous, evolving left temporal 5- to 7-Hz rhythmic sharp theta activity, consistent with nonconvulsive status epilepticus (Fig. 1C). The seizures resolved after loading with lorazepam, levetiracetam, and lacosamide, and continuous EEG monitoring concluded that the patient was seizure free for 48 hours on POD 6. On POD 6, there was no MLS (0 mm) and contralateral epileptiform activity resolved completely (Table 1). She recovered well and remained seizure free on two seizure medications. Patient 4 was a 50-year-old woman with a left frontotemporal intracerebral hemorrhage after head trauma. A small right frontal contusion was also present. She had an acute deterioration in mental status 6 days after the trauma. Computed tomography was significant for an acutely expanding left frontotemporal hematoma and a left to right MLS of 9 mm. Her EEG demonstrated right frontocentral LPDs (Fig. 1D). The patient underwent emergent evacuation of the expanding hematoma with subsequent clinical, imaging, and electrophysiological improvements. After the surgery, the MLS decreased to 8 mm and the 2 contralateral epileptiform activity changed from LPDs to sporadic sharp waves. DISCUSSION This four case series demonstrates MLS occurring with the appearance of epileptiform activity contralateral to a known lesion. In all four patients in this series, as the MLS gradually decreased, the contralateral epileptiform decreased or even resolved. A previous study found that seizures after intracerebral hemorrhage were a marker of progressive MLS.1 Another study of EEGs in intracerebral hemorrhages could not confirm seizures being associated with increased MLS, but it did demonstrate an association between seizures and expanding hemorrhages.2 Others have reported ipsilateral or bilateral periodic discharges and midline epileptiform discharges as a marker of MLS after acute subdural hematoma.3 These four cases presented in this series suggest that LPDs or focal seizure activity contralateral to the lesion may also serve as an EEG marker for emerging MLS. This series is limited because it is a retrospective review of only four patients, and it remains unproven that MLS is actually causing the periodic discharges or seizures. The MLS in patient 3 was actually already improved when the seizures were found on EEG, but the patient was not on continuous EEG until after the CT of the head on POD 3, and subtle or subclinical seizures before this may have been missed. The four patients described are a heterogenous group with different etiologies for the MLS. In addition, once seizures were recorded, these patients were treated with seizure medications, making it difficult to comment on the relationship of MLS and expected time course for the EEG changes. Patients 2 and 4 also had bilateral imaging Journal of Clinical Neurophysiology Volume 0, Number 0, Month 2017 clinicalneurophys.com Copyright Ó by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited. G. R. Ghearing, et al. Misleading EEG Lateralization FIG. 1. All EEGs are displayed using a longitudinal bipolar montage with a low-frequency filter of 1 Hz, highfrequency filter of 70 Hz, notch filter at 60 Hz, and a sensitivity of 7 mV/mm. A, EEG of patient 1 demonstrating left temporal periodic spikes with evolution during a clinical seizure with oral automatisms. Computed tomography of head of patient 1 demonstrating a 12-mm right to left midline shift. Both CT and EEG were acquired on postoperative day 2. B, EEG of patient 2 demonstrating stimulation-induced right lateralized periodic discharges. Computed tomography of the head of patient 2 demonstrating a 6-mm left to right midline shift. Both CT and EEG were acquired on postoperative day 0. C, EEG of patient 3 demonstrating left temporal seizure activity characterized by rhythmic sharply contoured 7 to 8 Hz theta and alpha activity. Computed tomography of the head for patient 2 demonstrating a 7-mm right to left midline shift. Both CT and EEG were acquired on postoperative day 3. D, EEG of patient 4 demonstrating right lateralized periodic discharges. Computed tomography of the head for patient 4 demonstrating a 9-mm left to right midline shift. Both CT and EEG were acquired 6 days after the head trauma, immediately after the emergent surgery. abnormalities. However, the most prominent abnormalities were contralateral to the onset of the LPDs and seizures in these patients, and patients 1 and 3 did not have any additional abnormality in the contralateral hemisphere to explain the periodic discharges or seizures. Further research is required for adequate quantification and characterization of the relationship between the severity of contralateral epileptiform activity and the degree or progression of MLS. clinicalneurophys.com The periodic epileptiform discharges and seizures contralateral to the patient’s known lesion could be caused by disturbed cortical function related to mass effect from MLS. A pathologic evaluation of subdural hematoma showed that the cortex underlying the fluid collection is often protected and free of abnormality, whereas there is flattening of the contralateral gyri, which were compressed against the skull, and this could explain the contralateral EEG abnormalities.4 Also, three of Journal of Clinical Neurophysiology Volume 0, Number 0, Month 2017 3 Copyright Ó by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited. G. R. Ghearing, et al. four of these patients had seizures or periodic discharges, which were maximal in the contralateral temporal region. This suggests the possibility that these discharges may represent secondary epileptogenesis or a mirror focus, although no seizures were recorded ipsilateral to the lesion in these four patients. It is also possible that these contralateral EEG changes could be explained as a marker of diffuse cerebral dysfunction or disruption of a bilateral network. Although the etiology of contralateral epileptiform changes in MLS remains uncertain, periodic discharges or seizures appearing contralateral to a known lesion should prompt evaluation for MLS or new unexpected lesions. 4 Misleading EEG Lateralization REFERENCES 1. Vespa PM, O’Phelan K, Shah M, et al. Acute seizures after intracerebral hemorrhage: a factor in progressive midline shift and outcome. Neurology 2003;60:1441–1446. 2. Classen J, Jetté N, Chum F, et al. Electrographic seizures and periodic discharges after intracerebral hemorrhage. Neurology 2007;69:1356–1365. 3. Rudzinski LA, Rabinstein AA, Chung SY, et al. Electroencephalographic findings in acute subdural hematoma. J Clin Neurophysiol 2011;28:633–641. 4. Kaplan HA, Huber W, Browder J. Electroencephalogram in subdural hematoma; a consideration of its pathophysiology. J Neuropathol Exp Neurol 1956;15:65–78. Journal of Clinical Neurophysiology Volume 0, Number 0, Month 2017 clinicalneurophys.com Copyright Ó by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.