ORIGINAL RESEARCH
Stimulus-Induced Generalized Epileptiform Discharges: An Unrecognized EEG Pattern in Refractory Nonconvulsive Status Epilepticus Carles Gaig,* Alex Iranzo,* Ana Tercero,* Susan T. Herman,† and Joan Santamaria*
Purpose: To report an unrecognized EEG pattern occurring in patients with coma with nonconvulsive status epilepticus (NCSE) treated with thiopental and characterized by a high amplitude generalized sharp wave that was induced by isolated or repetitive low frequency photic or tactile stimulation (stimulus-induced generalized epileptiform discharges). Methods: Review of clinical and EEG records of 4 patients with stimulusinduced generalized epileptiform discharges identified among patients admitted to an intensive care unit (ICU) of our institution who underwent EEG monitoring between July 2011 and January 2013. Results: Four patients had refractory NCSE secondary to hepatic encephalopathy, brain tumor, immunomediated encephalopathy, or anti-NMDA encephalitis. All of them were treated with several antiepileptic drugs and anesthetics, but stimulus-induced generalized epileptiform discharges were only recorded after control of NCSE with thiopental. The discharges occurred without associated clinical manifestations. Two patients died in the ICU, and two had a favorable outcome. Stimulus-induced generalized epileptiform discharges were not recorded in 240 critically ill patients admitted to the ICU who had an EEG recorded during this period. Twenty-two of them were in NCSE, and only one was treated with thiopental. Conclusions: Stimulus-induced generalized epileptiform discharges is an EEG pattern occurring in patients with NCSE treated with thiopental. The clinical relevance and pathophysiology of stimulus-induced generalized epileptiform discharges remain to be clarified. Key Words: Stimulus-induced generalized epileptiform discharges, Nonconvulsive status epilepticus, Thiopental, EEG, Coma. (J Clin Neurophysiol 2014;31: 580–585)
I
n the critically ill patient in coma, auditory, tactile, and painful stimulation are carried out to assess EEG background reactivity. Absence of reactivity is considered in some clinical settings to be a marker for unfavorable neurological outcome (Fugate et al., 2010; Rossetti et al., 2012). Sensory stimulation in the patient with coma can also induce abnormal clinical and EEG responses including myoclonus (Niedermeyer et al., 1977), seizures and burst-spiking discharges (van Cott et al., 1996; Dan and Boyd, 2006; Hirsch et al., 2008), and rhythmic, periodic, or ictal EEG patterns (SIRPIDs,
From the *Neurology Service, Hospital Clínic de Barcelona, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain; and †Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, U.S.A. Presented previously at the LXV meeting of the Spanish Neurological Society as an Oral Communication, November 19–23, Barcelona, Spain. Address correspondence and reprint requests to Carles Gaig, MD, Neurology Service, Hospital Clinic de Barcelona, C/Villarroel 170, Barcelona 08036, Spain; e-mail: [email protected]. Copyright Ó 2014 by the American Clinical Neurophysiology Society
ISSN: 0736-0258/14/3106-0580
580
or stimulus-induced rhythmic, periodic, or ictal discharges) (Hirsch et al., 2004). Herein, we describe an unrecognized EEG pattern induced by sensory stimulation and occurring in four patients with coma with refractory nonconvulsive status epilepticus (NCSE) treated with thiopental. This EEG phenomenon consisted of a single generalized sharp wave that was evoked by isolated or repetitive low frequency photic or tactile stimulation. To our knowledge, these stimulusinduced generalized epileptiform discharges (SIGEDs) have not been described previously.
METHODS The 4 patients were identified between July 2011 and January 2013. All four patients were admitted to an intensive care unit (ICU) of the Hospital Clinic of Barcelona because of different critical illnesses that were complicated by refractory NCSE. Stimulusinduced generalized epileptiform discharges were detected in all of them during EEG routine monitoring. EEGs were recorded digitally using 21 electrodes placed according to the International 10-20 system. In our institution, auditory (hand claps, patient’s name), tactile (light touch, pain), and photic (repetitive flashes at 0.5, 1, 5, 10, 15, and 20 Hz) stimulations are performed in a routine way in critically ill patients admitted to the ICUs and undergoing EEG monitoring. During routine photic stimulation, SIGEDs were noticed for first time in all four patients. Clinical files and EEG records of these four patients with SIGEDs were reviewed. In addition, presence of NCSE, anesthetics drugs administered, and occurrence of SIGEDs were reexamined in 240 critically ill patients admitted to the ICUs and undergoing EEG recording during the same period of time when the 4 patients with SIGEDs were identified.
RESULTS Clinical features of the four patients with SIGEDS are summarized in Table 1. Case 1 was a 62-year-old man with cirrhosis and hepatocellular carcinoma admitted to the ICU because of massive duodenal bleeding and hepatic encephalopathy. Intensive care unit admission was complicated by NCSE. Patient 2 was an 80-yearold man with recurrent nonconvulsive seizures secondary to a right fronto-temporal brain tumor. Case 3 was a woman aged 29 years who was admitted to the ICU because of NCSE of undetermined etiology that responded to immunotherapy (methylprednisolone 1 g daily for 5 days and cyclophosphamide 1 g), and consequently, an immunomediated encephalopathy was suspected. Patient 4 was a 26 year-old-woman with anti-NMDA receptor encephalitis transferred
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
TABLE 1.
Stimulus-Induced Epileptiform Discharges
Clinical Characteristics of the Patients With Stimulus-Induced Generalized Epileptiform Discharges Case 1
Case 2
Case 3
Case 4
Gender, age (years) Underlying condition
Male, 62 Hepatic encephalopathy
Male, 80 Right fronto-temporal lobe tumor
Other relevant medical history
Hepatitis C: cirrhosis; hepatocellular carcinoma
None
Preexisting epilepsy Cause of ICU admission
No Massive duodenal bleeding and coma
No Recurrent nonconvulsive seizures and stupor
Female, 29 Encephalopathy responsive to immunotherapy Episode of cerebellar ataxia of undetermined etiology; idiopathic thrombocytopenic purpura No Recurrent nonconvulsive seizures and stupor
Refractory NCSE
Yes; NCSE was detected on the second day in the ICU LEV, PHE Midazolam Propofol Thiopental 12 days
Yes; NCSE on ICU admission
Yes; NCSE on ICU admission
No Motor hyperactivity and agitation with stereotyped facial and limb movements and stupor* Yes; NCSE was detected on ICU admission*
VPA, LEV Midazolam Propofol Thiopental 14 days
CNZ, LEV, VPA, LAC Propofol Midazolam Thiopental 8 days
LEV, VPA, PHE, LAC Midazolam Propofol Thiopental 12 days*
10 days
14 days
8 days
Unknown*
In the eighth day of admission in the ICU; 2 days† None (Thiopental stopped 2 days before, and propofol and midazolam 4 days before) At least 5 days (SIGEDs still present in the last EEG performed before death) Propofol
In the fourth day of admission in the ICU; 7 days None (Thiopental stopped 2 days before, and propofol and midazolam 9 days before) One day (SIGEDs was recorded in the first and last EEG before death)
In the first day of admission in the ICU; 7 days Thiopental infusion active
In the seventh day of admission in the ICU*; 2 days† None (Thiopental stopped 2 days before, and propofol and midazolam 4 days before)‡
Three days
Two days. SIGEDs were recorded despite sedation with midazolam
None
None
Midazolam
In coma, with brainstem reflexes preserved
In coma, with brainstem reflexes preserved
In coma, with brainstem reflexes preserved
In coma, with brainstem reflexes preserved
Death (26 days after ICU admission) because of repeated duodenal bleeding and pneumonia; therapeutic effort limitation)
Death (1 day later SIGEDs detection and 15 days after ICU admission)
Good recovery. Regaining the ability to conduct a normal life but secondary epilepsy was left
Recovery. Independent for activities of daily living but mental sequel precludes resume previous level of function
AED administered Anesthetics before SIGEDs (listed in order of administration) Days in the ICU when SIGEDs were detected Duration of NCSE when SIGEDs were detected Onset and duration of thiopental anesthesia Active anesthetics while SIGEDs were first recorded
Duration of SIGEDs
Administration of other anesthetics while SIGEDs were recorded Neurological examination when SIGEDs were recorded Outcome
Female, 26 Anti-NMDA receptor encephalitis None
*Case 4 was admitted in an ICU of another hospital for 2 months, from where she was transferred to the medical ICU of our center, and NCSE was detected on the EEG performed on admission. †In case 1 and 4, thiopental was stopped because of severe hypernatremia (158 and 183 mEq/L, respectively; normal range, 135–145 mEq/L). ‡In case 4, plasma levels of thiopental was 19.7 mg/mL (normal range, 20–40 mg/mL) 72 hours after its suppression while SIGEDs were still recorded. AED, antiepileptic drugs; CNZ, Clonacepam; LAC, Lacosamide; LEV, Levetiracetam; NCSE, nonconvulsive status epilepticus; PHE, Phenytoin; SIGEDs, stimulus-induced generalized epileptiform discharges; VPA, Valproate.
to our medical ICU from another center. On admission, she was found to have recurrent nonconvulsive seizures on the EEG. All four patients suffered from NCSE that was refractory to multiple antiepileptic drugs and anesthetics, such as propofol and Copyright Ó 2014 by the American Clinical Neurophysiology Society
midazolam, and therefore, all of them were eventually treated with thiopental, which effectively controlled the NCSE. Stimulus-induced generalized epileptiform discharges were initially recorded after thiopental anesthesia in all four patients. Duration of thiopental 581
C. Gaig et al.
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
therapy lasted from 2 to 7 days. When SIGEDs were firstly recorded, thiopental had been stopped 2 days before in 3 patients; in 2 of them as a consequence of severe hypernatremia and in the other case because of resolution of NCSE. In the remaining patient (case 3), thiopental was still being administered when SIGEDs were firstly recorded to maintain the control of NCSE while immunotherapy was administered. Active infusion of other anesthetics was absent when SIGEDs were initially recorded. At the time of SIGEDs detection, admission in the ICU of these patients lasted from 8 to 62 days, and refractory NCSE was present for 8 to 14 days (Table 1). In all 4 patients, SIGEDs were identified during repetitive photic stimulation at rates of 0.5 and 1 Hz (Figs. 1A, 1B and 2; Table 2). At these low frequencies, each flash of light elicited a high amplitude, generalized sharp wave. This sharp wave was synchronized with the
flash. The latency between the flash and the sharp wave was constant in each patient and ranged from 180 to 340 milliseconds (Table 2). Because light stimulus was administered in repetitive trains, strings of these generalized sharp waves occurred at the same frequency of the flash (at 0.5 or 1 Hz), as occurs with photic driving, and therefore, they resembled periodic or repetitive discharges. The SIGEDs were usually not evoked by flashes at frequencies higher than 1 Hz (e.g., 2, 3, 4, 5, 10, 15, or 20 Hz), although occasionally the first 1 to 2 stimulus or isolated stimuli within the repetitive train of flashes were able to elicit a sporadic sharp wave (Figs. 1B and 1C). Stimulus-induced generalized epileptiform discharges had a regular morphology and consisted of a sharp-and-wave complex with two, three, or sometimes more phases (Table 2 and Fig. 2). The duration of the sharp-and-wave complex was stereotyped in each
FIG. 1. Stimulus-induced generalized epileptiform discharges (SIGEDs). A, Stimulus-induced generalized epileptiform discharges evoked by photic stimulation at rate of 0.5 Hz in patient 1. During periods without photic stimulation (second half of the epoch), some isolated diffuse sharp waves appear spontaneously (not induced by stimulation). B, In patient 2, sharp waves are induced by photostimulation at rates of 0.5 and 1 Hz. At 2 Hz, only the initial flashes are able to induce generalized sharp waves. C, In the same patient, photic stimulation at 5 and 10 Hz does not induce generalized sharp waves except for the first or some sporadic flashes of the series (asterisks). D, Stimulus-induced generalized epileptiform discharges elicited by tactile stimulation consisting in repetitive gentle tapping on the left cheek (arrows) in patient 3. EKG, electrocardiogram; EOG, electrooculogram; PHO, photostimulation. 582
Copyright Ó 2014 by the American Clinical Neurophysiology Society
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
Stimulus-Induced Epileptiform Discharges
FIG. 2. Stimulus-induced generalized epileptiform discharges (SIGEDs) in all four cases. Illustrative example of SIGEDs induced by photic stimulation in a bipolar (A) and referential montage to ipsilateral ear (B) of the four patients to show their variable distribution and morphology. Stimulus-induced generalized epileptiform discharges induced by tactile stimulation (C; arrows indicate gentle touch in left forearm in case 1 [just after administration of a bolus of propofol] light tapping on the top of the nose in case 2, and repetitive tapping on the chest in cases 3 and 4). EKG, electrocardiogram; EOG, electrooculogram; PHO, photostimulation. patient and ranged from 150 to 540 milliseconds. Although generalized, they were of maximal amplitude in the anterior area in two patients and in the central or posterior region in the other two. Stimulus-induced generalized epileptiform discharges were not associated with myoclonic jerks, clonic movements, or other clinical manifestations in any case. Tactile stimulation also induced SIGEDs (Figs. 1D and 2). A single gentle touch or slow and soft repetitive tapping (e.g., 1 tap each 1 or 2 seconds) in different body areas (e.g., shoulders, sternum, and upper or lower limbs), and particularly in the face (e.g., forehead, cheek, and nose), evoked a single generalized sharp wave, as occurred with photic stimulation (Table 2). In contrast, rapid repetitive touching (e.g., several light taps per second) did not elicit the diffuse sharp wave. Auditory stimulation (e.g., hand clapping) did not induce SIGEDs except in 1 patient (case 3). In this case, some isolated hand clapping elicited occasionally a diffuse sharp wave. Stimulus-induced generalized epileptiform discharges induced by either photic or tactile stimulation had the same form, amplitude, and distribution in each patient. Copyright Ó 2014 by the American Clinical Neurophysiology Society
At the time SIGEDs were recorded, background EEG was suppressed in three patients (Table 2). In the remaining patient, the background was nearly continuous with occasional (,10% of the record) periods of suppression accordingly to the updated ACNS nomenclature (Hirsch et al., 2013). Background activity was nonreactive in all four patients. Spontaneous sharp waves were recorded in all patients (Table 2). In three patients, they were generalized and in one lateralized. These spontaneous epileptiform discharges were more heterogeneous in morphology and amplitude than SIGEDs, and they had a clinical correlate in only 1 patient (case 1), in whom spontaneous generalized periodic sharp waves were associated with clonic contractions of the left hand. Two patients died in the ICUs (cases 1 and 2, Table 1; postmortem studies were not performed). In the other 2, NCSE resolved after immunotherapy, and the patients were discharged from the hospital (cases 3 and 4). In one patient, SIGEDs were recorded for the first and last time in the EEG performed the day before he died. In the other 3 patients, SIGEDs were present in the EEGs performed 583
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
C. Gaig et al.
TABLE 2.
Electroencephalographic Features of Stimulus-Induced Generalized Epileptiform Discharges Case 1
Photic stimulation (at rates of 0.5 and 1 Hz) eliciting SIGEDs Tactile stimulation (isolated gentle touch or repeated tapping at rates of 1 Hz or less) eliciting SIGEDs Auditory stimulation (hand clamps) eliciting SIGEDs Latency from stimulus to SIGEDs, milliseconds* Morphology of SIGEDs Maximal amplitude of SIGEDs†
Duration of SIGEDs Background EEG‡ Symmetry Predominant frequency Anteroposterior gradient Variability Reactivity Voltage Continuity Sporadic epileptiform discharges
Case 2
Case 3
Case 4
Yes
Yes
Yes
Yes
Yes. In the forehead, both shoulders, and left limbs
Yes. In mouth, nose, and forehead
Yes. In the forehead, cheek, nose, chest, and all limbs
Yes. In the face, trunk, and all limbs except right leg
No
No
Yes, occasionally
No
210
340
240
180
Biphasic or triphasic 120–150 mV in central and posterior areas (phase reversal C3-P3 and C4-P4) 400 ms
Biphasic or triphasic 90–160 mV in anterior areas (Fp1, Fz, Fp2, F3, and F4)
Poliphasic (3–5 phases) 110–140 mV in central areas (Fp1, Fz, and Fp2)
Biphasic or triphasic 90–100 mV in central areas (phase reversal C3-F3 and C4-F4)
200–300 ms
500–540 ms
150–170 ms
Symmetric Theta Absent Absent Absent Suppressed (,10 mV) Suppression GPDs
Symmetric Theta Present Absent Absent Normal Discontinuous Abundant lateralized nonrhythmic and nonperiodic
Symmetric Theta Absent Absent Absent Normal Bursts suppression Abundant generalized nonrhythmic and nonperiodic
Symmetric Theta Present Absent Absent Low (,20 mV) Nearly continuous Abundant generalized nonrhythmic and nonperiodic
*Latency was measured from the stimulus to the peak of the sharp wave in standard longitudinal bipolar 10-20 recording. †Maximal amplitude was measured in standard longitudinal bipolar 10-20 recording in the channel in which the sharp wave was of highest amplitude, measured from peak to trough. ‡EEG background activity is described according to the ACNS nomenclature. GPDs, generalized periodic discharges; SIGEDs, stimulus-induced generalized epileptiform discharges.
from 2 to 5 days after their initial recording. During this period of time, in two patients SIGEDs appeared even under anesthesia with propofol and midazolam administered in doses sufficient to produce a generalized EEG suppression. Stimulus-induced generalized epileptiform discharges were not recorded in 240 critically ill patients admitted to the ICUs during the same period of time when the 4 patients with SIGEDs were identified. Twenty-two of these 240 patients had NCSE, and only 1 was treated with thiopental. Stimulus-induced generalized epileptiform discharges were not recorded in this patient who had a postanoxic NCSE.
DISCUSSION We have reported a distinct EEG pattern occurring in patients with refractory NCSE treated with thiopental that consisted of a single generalized sharp wave of high amplitude that was regularly induced by isolated or repetitive low frequency photic or tactile stimulation. These SIGEDs occurred when the NCSE was effectively controlled by thiopental anesthesia, with an EEG background ranging from continuous generalized low voltage EEG activity to burst-suppression pattern or generalized slowing with short hypoactive periods. Stimulus-induced generalized epileptiform discharges 584
were recognized for the first time during photic stimulation at rates of 0.5 and 1 Hz in all 4 patients, whereas flashes at rates higher than 1 Hz usually failed to elicit this EEG pattern. Repetitive flashes of light at low rates (0.5 Hz and 1 Hz) were the stimulus that elicited more clearly and easily SIGEDs, allowing us to identify this EEG pattern. Photic stimulation is usually not routinely performed during the EEG recording in patients admitted in the ICUs and may explain why SIGEDs has been previously unrecognized. To detect this EEG pattern, it is essential to include low frequency photic stimulation within the activating maneuvers in the EEG performed in the critically ill patient. Stimulus-induced generalized epileptiform discharges is likely an infrequent EEG pattern in the critically ill patient because it was recorded in only 4 of 244 (1.6%) patients admitted to the ICUs and undergoing EEG recording during a period of 18 months. This EEG pattern, however, may not be uncommon in patients with NCSE treated with thiopental. Stimulus-induced generalized epileptiform discharges were seen in 4 of 26 (15.5%) patients in NCSE assessed during this period and in 4 of 5 (80%) patients in NCSE treated with thiopental. The clinical relevance of SIGEDs remains to be elucidated. This EEG pattern has no associated clinical manifestations and prognosis in our patients varied from favorable outcome to death. The four patients reported here presented prolonged NCSE requiring treatment with several antiepileptic drugs and anesthetics including Copyright Ó 2014 by the American Clinical Neurophysiology Society
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
propofol and midazolam; but in all of them, SIGEDs were recorded only after thiopental sedation. This EEG phenomenon lasted up to 7 days after discontinuation of thiopental, which could be related to the presence of long-lasting plasma levels of this barbiturate as a consequence of its prolonged half-life (Russo and Bressolle, 1998). In contrast, the etiology of NCSE did not influence the appearance of SIGEDs because the four patients reported here suffered from diverse disorders, including diffuses metabolic and immunological diseases, as well as focal brain lesions. The mechanism generating SIGEDs is uncertain. Flashes at frequencies higher than 1 Hz usually did not elicit SIGEDs. This observation could suggest the presence of a refractory period in which another sharp wave cannot be generated. Because SIGEDs were not recorded in patients with NCSE treated with other GABAergic anesthetics such as propofol or midazolam, it can be speculated that a thiopental-related specific effect in combination with a neuronal hyperexcitability inherent to NCSE (Chen and Wasterlain, 2006) are the cause of SIGED. Stimulus-induced generalized epileptiform discharges differ from other stimulus-induced EEG patterns in the critically ill patients, such as stimulus-induced rhythmic, periodic, or ictal discharges or postanoxic stimulus sensitive myoclonus (Hirsch et al., 2004; Niedermeyer et al., 1977). In contrast to these EEG patterns, SIGEDs are made up of a single sharp wave that was time-locked to the stimulus, had no clinical manifestations, and were not related to any specific etiology. Stimulus-induced generalized epileptiform discharges, however, may provide some clues to understand the underlying mechanism generating these other stimulus-induced EEG patterns.
Copyright Ó 2014 by the American Clinical Neurophysiology Society
Stimulus-Induced Epileptiform Discharges
In summary, SIGED is an EEG pattern occurring in the patient with coma in NCSE treated with thiopental. Stimulus-induced generalized epileptiform discharges consist of a single generalized sharp wave of high amplitude that is regularly elicited by isolated or repetitive low frequency photic or tactile stimuli. Routine photic stimulation with repetitive flashes at rates of 1 Hz or lower during the EEG in critically ill patients is necessary to identify this EEG pattern, still of undetermined clinical relevance, and pathophysiology. REFERENCES Chen JW, Wasterlain CG. Status epilepticus: pathophysiology and management in adults. Lancet Neurol 2006;5:246–256. van Cott AC, Blatt I, Brenner RP. Stimulus-sensitive seizures in postanoxic coma. Epilepsia 1996;37:868–874. Dan B, Boyd SG. Stimulus-sensitive burst-spiking in burst- suppression in children: implications for management of refractory status epilepticus. Epileptic Disord 2006;8:143–150. Fugate JE, Wijdicks EF, Mandrekar J, et al. Predictors of neurologic outcome in hypothermia after cardiac arrest. Ann Neurol 2010;68:907–914. Hirsch LJ, Claassen J, Mayer SA, Emerson RG. Stimulus-induced rhythmic, periodic, or ictal discharges (SIRPIDs): a common EEG phenomenon in the critically ill. Epilepsia 2004;45:109–123. Hirsch LJ, Pang T, Claassen J, et al. Focal motor seizures induced by alerting stimuli in critically ill patients. Epilepsia 2008;49:968–973. Hirsch LJ, LaRoche SM, Gaspard N, et al. American clinical Neurophysiology Society’s Standardized critical care EEG Terminology: 2012 version. J Clin Neurophysiol 2013;30:1–27. Niedermeyer E, Bauer G, Burnite R, Reichenbach D. Selective stimulus-sensitive myoclonus in acute cerebral anoxia. Arch Neurol 1977;34:365–368. Rossetti AO, Carrera E, Oddo M. Early EEG correlates of neuronal injury after brain anoxia. Neurology 2012;78:796–802. Russo H, Bressolle F. Pharmacodynamics and pharmacokinetics of thiopental. Clin Pharmacokinet 1998;35:95–134.
585
Stimulus-Induced Generalized Epileptiform Discharges: An Unrecognized EEG Pattern in Refractory Nonconvulsive Status Epilepticus Carles Gaig,* Alex Iranzo,* Ana Tercero,* Susan T. Herman,† and Joan Santamaria*
Purpose: To report an unrecognized EEG pattern occurring in patients with coma with nonconvulsive status epilepticus (NCSE) treated with thiopental and characterized by a high amplitude generalized sharp wave that was induced by isolated or repetitive low frequency photic or tactile stimulation (stimulus-induced generalized epileptiform discharges). Methods: Review of clinical and EEG records of 4 patients with stimulusinduced generalized epileptiform discharges identified among patients admitted to an intensive care unit (ICU) of our institution who underwent EEG monitoring between July 2011 and January 2013. Results: Four patients had refractory NCSE secondary to hepatic encephalopathy, brain tumor, immunomediated encephalopathy, or anti-NMDA encephalitis. All of them were treated with several antiepileptic drugs and anesthetics, but stimulus-induced generalized epileptiform discharges were only recorded after control of NCSE with thiopental. The discharges occurred without associated clinical manifestations. Two patients died in the ICU, and two had a favorable outcome. Stimulus-induced generalized epileptiform discharges were not recorded in 240 critically ill patients admitted to the ICU who had an EEG recorded during this period. Twenty-two of them were in NCSE, and only one was treated with thiopental. Conclusions: Stimulus-induced generalized epileptiform discharges is an EEG pattern occurring in patients with NCSE treated with thiopental. The clinical relevance and pathophysiology of stimulus-induced generalized epileptiform discharges remain to be clarified. Key Words: Stimulus-induced generalized epileptiform discharges, Nonconvulsive status epilepticus, Thiopental, EEG, Coma. (J Clin Neurophysiol 2014;31: 580–585)
I
n the critically ill patient in coma, auditory, tactile, and painful stimulation are carried out to assess EEG background reactivity. Absence of reactivity is considered in some clinical settings to be a marker for unfavorable neurological outcome (Fugate et al., 2010; Rossetti et al., 2012). Sensory stimulation in the patient with coma can also induce abnormal clinical and EEG responses including myoclonus (Niedermeyer et al., 1977), seizures and burst-spiking discharges (van Cott et al., 1996; Dan and Boyd, 2006; Hirsch et al., 2008), and rhythmic, periodic, or ictal EEG patterns (SIRPIDs,
From the *Neurology Service, Hospital Clínic de Barcelona, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain; and †Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, U.S.A. Presented previously at the LXV meeting of the Spanish Neurological Society as an Oral Communication, November 19–23, Barcelona, Spain. Address correspondence and reprint requests to Carles Gaig, MD, Neurology Service, Hospital Clinic de Barcelona, C/Villarroel 170, Barcelona 08036, Spain; e-mail: [email protected]. Copyright Ó 2014 by the American Clinical Neurophysiology Society
ISSN: 0736-0258/14/3106-0580
580
or stimulus-induced rhythmic, periodic, or ictal discharges) (Hirsch et al., 2004). Herein, we describe an unrecognized EEG pattern induced by sensory stimulation and occurring in four patients with coma with refractory nonconvulsive status epilepticus (NCSE) treated with thiopental. This EEG phenomenon consisted of a single generalized sharp wave that was evoked by isolated or repetitive low frequency photic or tactile stimulation. To our knowledge, these stimulusinduced generalized epileptiform discharges (SIGEDs) have not been described previously.
METHODS The 4 patients were identified between July 2011 and January 2013. All four patients were admitted to an intensive care unit (ICU) of the Hospital Clinic of Barcelona because of different critical illnesses that were complicated by refractory NCSE. Stimulusinduced generalized epileptiform discharges were detected in all of them during EEG routine monitoring. EEGs were recorded digitally using 21 electrodes placed according to the International 10-20 system. In our institution, auditory (hand claps, patient’s name), tactile (light touch, pain), and photic (repetitive flashes at 0.5, 1, 5, 10, 15, and 20 Hz) stimulations are performed in a routine way in critically ill patients admitted to the ICUs and undergoing EEG monitoring. During routine photic stimulation, SIGEDs were noticed for first time in all four patients. Clinical files and EEG records of these four patients with SIGEDs were reviewed. In addition, presence of NCSE, anesthetics drugs administered, and occurrence of SIGEDs were reexamined in 240 critically ill patients admitted to the ICUs and undergoing EEG recording during the same period of time when the 4 patients with SIGEDs were identified.
RESULTS Clinical features of the four patients with SIGEDS are summarized in Table 1. Case 1 was a 62-year-old man with cirrhosis and hepatocellular carcinoma admitted to the ICU because of massive duodenal bleeding and hepatic encephalopathy. Intensive care unit admission was complicated by NCSE. Patient 2 was an 80-yearold man with recurrent nonconvulsive seizures secondary to a right fronto-temporal brain tumor. Case 3 was a woman aged 29 years who was admitted to the ICU because of NCSE of undetermined etiology that responded to immunotherapy (methylprednisolone 1 g daily for 5 days and cyclophosphamide 1 g), and consequently, an immunomediated encephalopathy was suspected. Patient 4 was a 26 year-old-woman with anti-NMDA receptor encephalitis transferred
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
TABLE 1.
Stimulus-Induced Epileptiform Discharges
Clinical Characteristics of the Patients With Stimulus-Induced Generalized Epileptiform Discharges Case 1
Case 2
Case 3
Case 4
Gender, age (years) Underlying condition
Male, 62 Hepatic encephalopathy
Male, 80 Right fronto-temporal lobe tumor
Other relevant medical history
Hepatitis C: cirrhosis; hepatocellular carcinoma
None
Preexisting epilepsy Cause of ICU admission
No Massive duodenal bleeding and coma
No Recurrent nonconvulsive seizures and stupor
Female, 29 Encephalopathy responsive to immunotherapy Episode of cerebellar ataxia of undetermined etiology; idiopathic thrombocytopenic purpura No Recurrent nonconvulsive seizures and stupor
Refractory NCSE
Yes; NCSE was detected on the second day in the ICU LEV, PHE Midazolam Propofol Thiopental 12 days
Yes; NCSE on ICU admission
Yes; NCSE on ICU admission
No Motor hyperactivity and agitation with stereotyped facial and limb movements and stupor* Yes; NCSE was detected on ICU admission*
VPA, LEV Midazolam Propofol Thiopental 14 days
CNZ, LEV, VPA, LAC Propofol Midazolam Thiopental 8 days
LEV, VPA, PHE, LAC Midazolam Propofol Thiopental 12 days*
10 days
14 days
8 days
Unknown*
In the eighth day of admission in the ICU; 2 days† None (Thiopental stopped 2 days before, and propofol and midazolam 4 days before) At least 5 days (SIGEDs still present in the last EEG performed before death) Propofol
In the fourth day of admission in the ICU; 7 days None (Thiopental stopped 2 days before, and propofol and midazolam 9 days before) One day (SIGEDs was recorded in the first and last EEG before death)
In the first day of admission in the ICU; 7 days Thiopental infusion active
In the seventh day of admission in the ICU*; 2 days† None (Thiopental stopped 2 days before, and propofol and midazolam 4 days before)‡
Three days
Two days. SIGEDs were recorded despite sedation with midazolam
None
None
Midazolam
In coma, with brainstem reflexes preserved
In coma, with brainstem reflexes preserved
In coma, with brainstem reflexes preserved
In coma, with brainstem reflexes preserved
Death (26 days after ICU admission) because of repeated duodenal bleeding and pneumonia; therapeutic effort limitation)
Death (1 day later SIGEDs detection and 15 days after ICU admission)
Good recovery. Regaining the ability to conduct a normal life but secondary epilepsy was left
Recovery. Independent for activities of daily living but mental sequel precludes resume previous level of function
AED administered Anesthetics before SIGEDs (listed in order of administration) Days in the ICU when SIGEDs were detected Duration of NCSE when SIGEDs were detected Onset and duration of thiopental anesthesia Active anesthetics while SIGEDs were first recorded
Duration of SIGEDs
Administration of other anesthetics while SIGEDs were recorded Neurological examination when SIGEDs were recorded Outcome
Female, 26 Anti-NMDA receptor encephalitis None
*Case 4 was admitted in an ICU of another hospital for 2 months, from where she was transferred to the medical ICU of our center, and NCSE was detected on the EEG performed on admission. †In case 1 and 4, thiopental was stopped because of severe hypernatremia (158 and 183 mEq/L, respectively; normal range, 135–145 mEq/L). ‡In case 4, plasma levels of thiopental was 19.7 mg/mL (normal range, 20–40 mg/mL) 72 hours after its suppression while SIGEDs were still recorded. AED, antiepileptic drugs; CNZ, Clonacepam; LAC, Lacosamide; LEV, Levetiracetam; NCSE, nonconvulsive status epilepticus; PHE, Phenytoin; SIGEDs, stimulus-induced generalized epileptiform discharges; VPA, Valproate.
to our medical ICU from another center. On admission, she was found to have recurrent nonconvulsive seizures on the EEG. All four patients suffered from NCSE that was refractory to multiple antiepileptic drugs and anesthetics, such as propofol and Copyright Ó 2014 by the American Clinical Neurophysiology Society
midazolam, and therefore, all of them were eventually treated with thiopental, which effectively controlled the NCSE. Stimulus-induced generalized epileptiform discharges were initially recorded after thiopental anesthesia in all four patients. Duration of thiopental 581
C. Gaig et al.
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
therapy lasted from 2 to 7 days. When SIGEDs were firstly recorded, thiopental had been stopped 2 days before in 3 patients; in 2 of them as a consequence of severe hypernatremia and in the other case because of resolution of NCSE. In the remaining patient (case 3), thiopental was still being administered when SIGEDs were firstly recorded to maintain the control of NCSE while immunotherapy was administered. Active infusion of other anesthetics was absent when SIGEDs were initially recorded. At the time of SIGEDs detection, admission in the ICU of these patients lasted from 8 to 62 days, and refractory NCSE was present for 8 to 14 days (Table 1). In all 4 patients, SIGEDs were identified during repetitive photic stimulation at rates of 0.5 and 1 Hz (Figs. 1A, 1B and 2; Table 2). At these low frequencies, each flash of light elicited a high amplitude, generalized sharp wave. This sharp wave was synchronized with the
flash. The latency between the flash and the sharp wave was constant in each patient and ranged from 180 to 340 milliseconds (Table 2). Because light stimulus was administered in repetitive trains, strings of these generalized sharp waves occurred at the same frequency of the flash (at 0.5 or 1 Hz), as occurs with photic driving, and therefore, they resembled periodic or repetitive discharges. The SIGEDs were usually not evoked by flashes at frequencies higher than 1 Hz (e.g., 2, 3, 4, 5, 10, 15, or 20 Hz), although occasionally the first 1 to 2 stimulus or isolated stimuli within the repetitive train of flashes were able to elicit a sporadic sharp wave (Figs. 1B and 1C). Stimulus-induced generalized epileptiform discharges had a regular morphology and consisted of a sharp-and-wave complex with two, three, or sometimes more phases (Table 2 and Fig. 2). The duration of the sharp-and-wave complex was stereotyped in each
FIG. 1. Stimulus-induced generalized epileptiform discharges (SIGEDs). A, Stimulus-induced generalized epileptiform discharges evoked by photic stimulation at rate of 0.5 Hz in patient 1. During periods without photic stimulation (second half of the epoch), some isolated diffuse sharp waves appear spontaneously (not induced by stimulation). B, In patient 2, sharp waves are induced by photostimulation at rates of 0.5 and 1 Hz. At 2 Hz, only the initial flashes are able to induce generalized sharp waves. C, In the same patient, photic stimulation at 5 and 10 Hz does not induce generalized sharp waves except for the first or some sporadic flashes of the series (asterisks). D, Stimulus-induced generalized epileptiform discharges elicited by tactile stimulation consisting in repetitive gentle tapping on the left cheek (arrows) in patient 3. EKG, electrocardiogram; EOG, electrooculogram; PHO, photostimulation. 582
Copyright Ó 2014 by the American Clinical Neurophysiology Society
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
Stimulus-Induced Epileptiform Discharges
FIG. 2. Stimulus-induced generalized epileptiform discharges (SIGEDs) in all four cases. Illustrative example of SIGEDs induced by photic stimulation in a bipolar (A) and referential montage to ipsilateral ear (B) of the four patients to show their variable distribution and morphology. Stimulus-induced generalized epileptiform discharges induced by tactile stimulation (C; arrows indicate gentle touch in left forearm in case 1 [just after administration of a bolus of propofol] light tapping on the top of the nose in case 2, and repetitive tapping on the chest in cases 3 and 4). EKG, electrocardiogram; EOG, electrooculogram; PHO, photostimulation. patient and ranged from 150 to 540 milliseconds. Although generalized, they were of maximal amplitude in the anterior area in two patients and in the central or posterior region in the other two. Stimulus-induced generalized epileptiform discharges were not associated with myoclonic jerks, clonic movements, or other clinical manifestations in any case. Tactile stimulation also induced SIGEDs (Figs. 1D and 2). A single gentle touch or slow and soft repetitive tapping (e.g., 1 tap each 1 or 2 seconds) in different body areas (e.g., shoulders, sternum, and upper or lower limbs), and particularly in the face (e.g., forehead, cheek, and nose), evoked a single generalized sharp wave, as occurred with photic stimulation (Table 2). In contrast, rapid repetitive touching (e.g., several light taps per second) did not elicit the diffuse sharp wave. Auditory stimulation (e.g., hand clapping) did not induce SIGEDs except in 1 patient (case 3). In this case, some isolated hand clapping elicited occasionally a diffuse sharp wave. Stimulus-induced generalized epileptiform discharges induced by either photic or tactile stimulation had the same form, amplitude, and distribution in each patient. Copyright Ó 2014 by the American Clinical Neurophysiology Society
At the time SIGEDs were recorded, background EEG was suppressed in three patients (Table 2). In the remaining patient, the background was nearly continuous with occasional (,10% of the record) periods of suppression accordingly to the updated ACNS nomenclature (Hirsch et al., 2013). Background activity was nonreactive in all four patients. Spontaneous sharp waves were recorded in all patients (Table 2). In three patients, they were generalized and in one lateralized. These spontaneous epileptiform discharges were more heterogeneous in morphology and amplitude than SIGEDs, and they had a clinical correlate in only 1 patient (case 1), in whom spontaneous generalized periodic sharp waves were associated with clonic contractions of the left hand. Two patients died in the ICUs (cases 1 and 2, Table 1; postmortem studies were not performed). In the other 2, NCSE resolved after immunotherapy, and the patients were discharged from the hospital (cases 3 and 4). In one patient, SIGEDs were recorded for the first and last time in the EEG performed the day before he died. In the other 3 patients, SIGEDs were present in the EEGs performed 583
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
C. Gaig et al.
TABLE 2.
Electroencephalographic Features of Stimulus-Induced Generalized Epileptiform Discharges Case 1
Photic stimulation (at rates of 0.5 and 1 Hz) eliciting SIGEDs Tactile stimulation (isolated gentle touch or repeated tapping at rates of 1 Hz or less) eliciting SIGEDs Auditory stimulation (hand clamps) eliciting SIGEDs Latency from stimulus to SIGEDs, milliseconds* Morphology of SIGEDs Maximal amplitude of SIGEDs†
Duration of SIGEDs Background EEG‡ Symmetry Predominant frequency Anteroposterior gradient Variability Reactivity Voltage Continuity Sporadic epileptiform discharges
Case 2
Case 3
Case 4
Yes
Yes
Yes
Yes
Yes. In the forehead, both shoulders, and left limbs
Yes. In mouth, nose, and forehead
Yes. In the forehead, cheek, nose, chest, and all limbs
Yes. In the face, trunk, and all limbs except right leg
No
No
Yes, occasionally
No
210
340
240
180
Biphasic or triphasic 120–150 mV in central and posterior areas (phase reversal C3-P3 and C4-P4) 400 ms
Biphasic or triphasic 90–160 mV in anterior areas (Fp1, Fz, Fp2, F3, and F4)
Poliphasic (3–5 phases) 110–140 mV in central areas (Fp1, Fz, and Fp2)
Biphasic or triphasic 90–100 mV in central areas (phase reversal C3-F3 and C4-F4)
200–300 ms
500–540 ms
150–170 ms
Symmetric Theta Absent Absent Absent Suppressed (,10 mV) Suppression GPDs
Symmetric Theta Present Absent Absent Normal Discontinuous Abundant lateralized nonrhythmic and nonperiodic
Symmetric Theta Absent Absent Absent Normal Bursts suppression Abundant generalized nonrhythmic and nonperiodic
Symmetric Theta Present Absent Absent Low (,20 mV) Nearly continuous Abundant generalized nonrhythmic and nonperiodic
*Latency was measured from the stimulus to the peak of the sharp wave in standard longitudinal bipolar 10-20 recording. †Maximal amplitude was measured in standard longitudinal bipolar 10-20 recording in the channel in which the sharp wave was of highest amplitude, measured from peak to trough. ‡EEG background activity is described according to the ACNS nomenclature. GPDs, generalized periodic discharges; SIGEDs, stimulus-induced generalized epileptiform discharges.
from 2 to 5 days after their initial recording. During this period of time, in two patients SIGEDs appeared even under anesthesia with propofol and midazolam administered in doses sufficient to produce a generalized EEG suppression. Stimulus-induced generalized epileptiform discharges were not recorded in 240 critically ill patients admitted to the ICUs during the same period of time when the 4 patients with SIGEDs were identified. Twenty-two of these 240 patients had NCSE, and only 1 was treated with thiopental. Stimulus-induced generalized epileptiform discharges were not recorded in this patient who had a postanoxic NCSE.
DISCUSSION We have reported a distinct EEG pattern occurring in patients with refractory NCSE treated with thiopental that consisted of a single generalized sharp wave of high amplitude that was regularly induced by isolated or repetitive low frequency photic or tactile stimulation. These SIGEDs occurred when the NCSE was effectively controlled by thiopental anesthesia, with an EEG background ranging from continuous generalized low voltage EEG activity to burst-suppression pattern or generalized slowing with short hypoactive periods. Stimulus-induced generalized epileptiform discharges 584
were recognized for the first time during photic stimulation at rates of 0.5 and 1 Hz in all 4 patients, whereas flashes at rates higher than 1 Hz usually failed to elicit this EEG pattern. Repetitive flashes of light at low rates (0.5 Hz and 1 Hz) were the stimulus that elicited more clearly and easily SIGEDs, allowing us to identify this EEG pattern. Photic stimulation is usually not routinely performed during the EEG recording in patients admitted in the ICUs and may explain why SIGEDs has been previously unrecognized. To detect this EEG pattern, it is essential to include low frequency photic stimulation within the activating maneuvers in the EEG performed in the critically ill patient. Stimulus-induced generalized epileptiform discharges is likely an infrequent EEG pattern in the critically ill patient because it was recorded in only 4 of 244 (1.6%) patients admitted to the ICUs and undergoing EEG recording during a period of 18 months. This EEG pattern, however, may not be uncommon in patients with NCSE treated with thiopental. Stimulus-induced generalized epileptiform discharges were seen in 4 of 26 (15.5%) patients in NCSE assessed during this period and in 4 of 5 (80%) patients in NCSE treated with thiopental. The clinical relevance of SIGEDs remains to be elucidated. This EEG pattern has no associated clinical manifestations and prognosis in our patients varied from favorable outcome to death. The four patients reported here presented prolonged NCSE requiring treatment with several antiepileptic drugs and anesthetics including Copyright Ó 2014 by the American Clinical Neurophysiology Society
Journal of Clinical Neurophysiology Volume 31, Number 6, December 2014
propofol and midazolam; but in all of them, SIGEDs were recorded only after thiopental sedation. This EEG phenomenon lasted up to 7 days after discontinuation of thiopental, which could be related to the presence of long-lasting plasma levels of this barbiturate as a consequence of its prolonged half-life (Russo and Bressolle, 1998). In contrast, the etiology of NCSE did not influence the appearance of SIGEDs because the four patients reported here suffered from diverse disorders, including diffuses metabolic and immunological diseases, as well as focal brain lesions. The mechanism generating SIGEDs is uncertain. Flashes at frequencies higher than 1 Hz usually did not elicit SIGEDs. This observation could suggest the presence of a refractory period in which another sharp wave cannot be generated. Because SIGEDs were not recorded in patients with NCSE treated with other GABAergic anesthetics such as propofol or midazolam, it can be speculated that a thiopental-related specific effect in combination with a neuronal hyperexcitability inherent to NCSE (Chen and Wasterlain, 2006) are the cause of SIGED. Stimulus-induced generalized epileptiform discharges differ from other stimulus-induced EEG patterns in the critically ill patients, such as stimulus-induced rhythmic, periodic, or ictal discharges or postanoxic stimulus sensitive myoclonus (Hirsch et al., 2004; Niedermeyer et al., 1977). In contrast to these EEG patterns, SIGEDs are made up of a single sharp wave that was time-locked to the stimulus, had no clinical manifestations, and were not related to any specific etiology. Stimulus-induced generalized epileptiform discharges, however, may provide some clues to understand the underlying mechanism generating these other stimulus-induced EEG patterns.
Copyright Ó 2014 by the American Clinical Neurophysiology Society
Stimulus-Induced Epileptiform Discharges
In summary, SIGED is an EEG pattern occurring in the patient with coma in NCSE treated with thiopental. Stimulus-induced generalized epileptiform discharges consist of a single generalized sharp wave of high amplitude that is regularly elicited by isolated or repetitive low frequency photic or tactile stimuli. Routine photic stimulation with repetitive flashes at rates of 1 Hz or lower during the EEG in critically ill patients is necessary to identify this EEG pattern, still of undetermined clinical relevance, and pathophysiology. REFERENCES Chen JW, Wasterlain CG. Status epilepticus: pathophysiology and management in adults. Lancet Neurol 2006;5:246–256. van Cott AC, Blatt I, Brenner RP. Stimulus-sensitive seizures in postanoxic coma. Epilepsia 1996;37:868–874. Dan B, Boyd SG. Stimulus-sensitive burst-spiking in burst- suppression in children: implications for management of refractory status epilepticus. Epileptic Disord 2006;8:143–150. Fugate JE, Wijdicks EF, Mandrekar J, et al. Predictors of neurologic outcome in hypothermia after cardiac arrest. Ann Neurol 2010;68:907–914. Hirsch LJ, Claassen J, Mayer SA, Emerson RG. Stimulus-induced rhythmic, periodic, or ictal discharges (SIRPIDs): a common EEG phenomenon in the critically ill. Epilepsia 2004;45:109–123. Hirsch LJ, Pang T, Claassen J, et al. Focal motor seizures induced by alerting stimuli in critically ill patients. Epilepsia 2008;49:968–973. Hirsch LJ, LaRoche SM, Gaspard N, et al. American clinical Neurophysiology Society’s Standardized critical care EEG Terminology: 2012 version. J Clin Neurophysiol 2013;30:1–27. Niedermeyer E, Bauer G, Burnite R, Reichenbach D. Selective stimulus-sensitive myoclonus in acute cerebral anoxia. Arch Neurol 1977;34:365–368. Rossetti AO, Carrera E, Oddo M. Early EEG correlates of neuronal injury after brain anoxia. Neurology 2012;78:796–802. Russo H, Bressolle F. Pharmacodynamics and pharmacokinetics of thiopental. Clin Pharmacokinet 1998;35:95–134.
585
